Log of the 380 SL's life with us.

Sunday–Friday, July 22–27, 2007

Saturday, July 28, 2007

Monday, July 30, 2007

Tuesday, July 31, 2007

...And the noontime pickup was fairly uneventful. I got the electric vacuum pump hooked up and it worked well, according to my test pilot. I used a spare main vacuum line from some diesel (115?) that had a check valve and an accessory tap on it, I hooked it to the booster and fed the pump to the tap, and just corked the end that normally goes to the vacuum pump. (Pilot also reports that the CD player works nicely, and that the PO forgot to remove a disc from the player.) We stopped a few blocks away and I tried bypassing the fuel pump relay, but it was not the problem. It is apparently the (expensive!) pump itself, further diagnosis will probably confirm this.

The tow car (Chicken Wagon) was unable to drag the dead weight up the graveled hill near the house (inadequate traction). I had to detach and quickly go get the 4wd Chevy pickup. But otherwise the trip, though slow and on back roads, was uneventful: just the way I like them!

When I got home after work I put the car on a charger to replenish the battery from its episode with the hazard lights and the vacuum pump. Let the games begin!

Wednesday, August 1, 2007

I restored the vacuum system to normality from yesterday's jury-rigging. I jacked up the rear of the car and removed the plastic shield for the fuel pump assembly. The rubber boots for the electrical connections are bad; the connections, though packed with dirt, were still good. I checked, and power was not getting back to there, even with the fuel pump relay bypassed. The ground line had good continuity to the car's frame, it's a power feed problem. I disconnected the cable and clipped the battery charger directly onto the fuel pump. It ran, drawing about 5 A. With it running I tried the key. The car started instantly. When I turned off the fuel pump the car died instantly. Looks like it could be a wiring problem. That's much better than a bad $200–250 fuel pump, though it could be a bit labor-intensive. The engine 'ticks' when running. Lifter noise?

I could tell, just by smell, that the catastrophic converter is working properly. That's good.

I also note that the car's got one of those wretched 85 MPH speedometers. I remember those! I wonder if I have a spare that I could swap out? Maybe someday.

This car's a day or two from being drivable. I need to get it licensed and insured!

Thursday, August 2, 2007

Anyway, I put the relay on the bench and it didn't feed power to the pump output unless the start pin was asserted. It might be supposed to give it a burp of a second or so at key-on as well, though I only infer this from the 560 SLs' behaviors. (And obviously it also didn't feed power when the engine was running.) I opened it up to find the usual sort of vintage circuit: two CMOS 4013 dual-D flip-flops, two 555 timers, four transistors, nine diodes, two electrolytic capacitors, two trim pots, one thermistor, two relays, and the usual handful of resistors and capacitors. Highly repairable, in other words. The solder was starting to look a little oxidized, I suspect that this thing will succumb to a resoldering. The relay box is labeled with "5300/min 8Zyl.12V", a Bosch logo and "896377", "Kickdown", and has what looks like a date code on it: 27-02-1997. If so this certainly isn't the original! And, if so, isn't much of an advertisement for Bosch electrical part longevity. Prominently missing is its supposed Mercedes part number: 000 545 36 05, listed in the EPC for '82 only. (Other part numbers might also work, 06 05 and 15 05 substitute for each other through '81, and 53 05 is listed for '83+. The relay was hard to find in the EPC, for future reference it's callout 107 in section 54, subgroup 045.)

So I resoldered it, first giving it a good wire-brushing with the Dremel to remove oxidation. When reinstalled in the car, it started and ran. That's $155 I don't need to spend! There was still no key-on burp, so I guess that's a feature of later cars.

I then cleaned off the differential fill plug and removed it. (It has a non-factory gasket on it, and the plug has been chiseled, though the hex socket is still in good shape.) The oil was low, but not dangerously so. I put in 1/2 quart and put the plug back in, lowered the car, and took it for a little test drive.

It drove well, the engine and transmission seemed as it should, the brakes worked well. (Perhaps not perfectly, it did pull a bit.) It seemed to have the requisite amount of power for the 3.8l motor, though I've no prior experience with them. There was some rattling from the RF corner of the car on the gravel washboard, there may be some loose suspension. I'll have a look at that sometime.

I tried the cruise and it didn't work, though I recall the PO saying something about that it should be disconnected because he thought it could self-destruct trying to use it when it was flakey. It may be unplugged somewhere. [It is.]

...At lunch I went to get it licensed. I expected there to be some sort of impediment, but there was not. $252 later I had plates in hand, title to be mailed in a few weeks. No inspections were required, which was surprising.

...After work we used the car to drive up the hill to the neighbors' place for dinner. We took the top off, by hand. The car worked well. It was a bit reluctant to start, like it doesn't want to fire at first. We'll keep an eye on that.

Friday, August 3, 2007

Since the soft top was 'up', I spent some time playing with the release handle linkages to get them adjusted better, and I oiled all the joints. I did find that there is a broken pivot bolt in the soft top bow mechanism, that explains why it doesn't feel right to put up and down. After that I washed the car and bolted on the license plates. The paint sure is a weird mix of very good and very bad.

I drove the car to work, it drove well. Having had to start it a few times today, it seems that the cold-start idle speed is too low. That's its reluctant starting. Will need to look into what it's got for a cold-start system.

Saturday, August 4, 2007

I pulled the under-dash panel and the steering wheel. The latter was a big PITA as the Allen head in the big retaining bolt started to strip, as usual. I was able to remove the horn pad and get a big cold chisel in there, then used the BFH to break it loose from the edge. Stupid design, the earlier big nut was better IMHO. One surprise was that I found a keyless entry system under the dash, a Crimestopper CS-845RKE complete with two relays. That probably explains the problems with the door! If the mechanism in the door turns out to be salvageable I may replace the CS-845 with a unit to which I have a remote, I've wanted to try adding keyless entry to one of these vacuum-only lock systems. (Replacement remotes for these things are usually quite expensive.) Anyway, with the steering wheel off the instrument cluster could be removed. I note that this car has an electronic speedometer, the only mechanical connection to the instrument cluster is the vacuum line to the Economy gauge. That being so, there's no way my spare 107 instrument cluster's non-85 MPH speedometer, actually from a 116 sedan, will work, as it is mechanical and not electronic. I also note that there is no lamp in the O₂-sensor indicator socket. That's ominous.

...Later in the day I got a chance to replace the electrolytic capacitors in the clock. As always, the parts are cheap and easy to replace. Easy, that is, after you dig the instrument cluster out of the car and the clock mechanism out of the instrument cluster! (This time, at least, I didn't break off the tachometer needle.) Reassembled, the clock is ticking again on the bench. During reassembly I cleaned the dust off the inside of the window and off the instrument faces, it looks nice. (I'll probably reinstall it tomorrow.)

Sunday, August 5, 2007

I pulled the trunk lamp assembly and found its guts all deformed from heat. I cut up a pop can to make a narrow nozzle for the heat gun and heated the center spring-holding post to soften it up, then I bent it back into position. After some trimming of plastic spurs and miscellaneous adjustment and lubrication I reassembled the thing and found that it was working again. Due to the application of heat I had to glue the center contact back in to the assembly, its little plastic snap-in slot got melted. Back in the car the lamp works again, though its switch is a little flakey (as so many are). A little jiggling gets it turned on, though. I think that's acceptable. I may work on it more later.

Since I was under the dash I pulled out the cruise control amplifier. It was unplugged, no surprise.

When reinstalling the instrument cluster I found that when the missing O₂-sensor lamp was replaced it was on all the time. More fun! When I started the car I also tried unplugging the idle actuator motor, and there was no effect on the car's idle. So that system is defunct too. I'm thinking I'll try a resoldering of the idle speed controller first, I don't see how it can hurt even if it doesn't help.

As church-time was looming I got the car buttoned up enough to be drivable again.

...When I got home again the car was fully warmed up and I again tried unplugging the idle actuator. RPM soared, so the idle system is intermittent.

I popped off the fuel tank connector (under a plug beneath the stowed soft top) and jumpered the two connections to ground. The needle pegged and the light came on, so the bulb is good. OK, I've seen this before, this cluster doesn't have a lamp-test for the fuel-out lamp! (The only lamp, of those you can't turn on at will, that's missing a lamp test feature.) I added one, it only takes adding a diode and a resistor to the instrument cluster, to the back side of the big connector inside the cluster. A generic diode such as a 1N4001 and an 18-ohm resistor are in series and connected between pins 8 and 1, anode (bar) towards pin 8. This replicates the circuit of the (for example) brake warning lamp. I pioneered this on the 240D I sold a while ago, I believe it's a nice enhancement.

Next I addressed the cruise control amplifier. It is a 14-pin VDO servo unit with a black connector, marked:

11.81
12V
412 205/1/2
001 545 76 32
Mercedes-Benz

Bosch
Geblaserregler 12V
1 147 328 026
Made in Germany
000 822 11 03

Once the stripper had done its job I washed it off of both units and dried them with compressed air. The cruise amplifier exhibited the expected signs of degraded solder joints, but the idle [blower!] controller looked pretty good. There may have been some signs of cracking at the big connector or perhaps it was just scratches, it's hard to tell. Regardless, I resoldered it.

With the idle [blower!] controller reinstalled the car started just as before. [That's what I would expect given that I was actually messing with the heater fan!] It did respond normally to unplugging the idle actuator, but the engine was also still hot. I will try this experiment again in the morning. It could be that the idle is low even with the actuator fully in the open position, so that while unplugging it has no apparent effect it's not a sign of a problem with the idle control system. Its cold-start reluctance may be the warm-up regulator, or the cold start injector. It's also possible that it is fuel pressure leaking down. More diagnosis to follow.

In the evening I resoldered the cruise control amplifier and put it in the bench test harness. It tried to work, but it seemed sensitive to flexure so obviously I'm not done resoldering yet. It also only seemed capable of driving the motor one direction, so I pulled the two cooked transistors and threw them on my old Heathkit IT-121 transistor tester. The NPN was still OK but the PNP was toast. (That's unfortunate, because I'm sure I could come up with a power NPN from the junk box, PNP's are a lot rarer. I may need to order some parts.) I think, however, that given the opportunity I will replace both transistors. It depends on what, if anything, surfaces in the junk box. I theorize that the cooking was caused by bad solder joints in the two inverter transistors that run the H-bridge such that the top and bottom legs ended up on simultaneously (at least part of the time).

Monday, August 6, 2007

Swapping these in was a bit of a pain, and upon closer examination the NPN transistor didn't look cooked, it was just coated in cooked varnish from the other side of the (shared) heat sink. So I only installed a new PNP, using new heat sink grease. With it installed the amplifier behaved properly on the bench, there was no sign of sensitivity to vibration or flexure. So I installed it in the car.

Under the hood I found that not only was the cruise actuator unplugged, but the car-side connector shell was also missing. I dug a replacement connector shell out of the junk box and installed it back into the car. The car's schematic was indispensable to get the color-coded wires back into the correct holes.

...On the way to work today the cruise control operated properly. Good.

Tuesday, August 7, 2007

As I remove panels and trim the car's starting to not look so good, and it's only going to get worse before it gets better!

...I opened up the ACC pushbutton array, and the solder joints are indeed bad inside. In the evening I did some poring over the car's schematic and augmented the diagnostic checklist to have some test information for the ACC system used in these cars. (We currently have five cars using this particular system, it's about time I made up a checklist for them.)

Wednesday, August 8, 2007

I then reinstalled the repaired ACC panel. Finally some positive cosmetic treatment!

On the way to work it became apparent that the heater blower was working properly, and that the checklist was faulty, as was my quick functional test. (I'd forgotten the effect of the cold engine lockout sensor.) However the vent control wasn't right, all vents operated all the time. Could be a vacuum supply problem. (Let's hope so!) Bad actuators are no fun to replace in this car.

...At lunch I stopped at McParts and bought some replacement 2721 lamps, and two cans of "Toreador Red" touch-up paint, $3 on clearance. The cap was a surprisingly close match, we'll see how it looks in real life.

Thursday, August 9, 2007

The center vent actuator (a nightmare to replace) actually seemed to be working properly once the defrost leak was stemmed, but when I looked down in the center vent I could see that something was wrong. I pulled off the rubber vent manifold and could then see that the ring on the flap that the actuator end snaps to had broken and the actuator was no longer connected. I cleaned the remains, fabricated a folded-over piece of thin sheet metal with appropriately-sized holes drilled in it, and glued this over the stub using Shoe Goo. I'll let that set up and then try reattaching the actuator.

Even though the center vent control line is referenced to power and not ground, if you're doing the testing key-off (as I was) you can feed positive voltage to it to make it go: the unpowered car circuitry acts as a virtual ground. While I was there I figured out what was wrong with my blower test checklist item. (You have to jumper two pins, not just one.) This time it passed!

With all that to the stopping point I got out the rattle-can of touch-up paint and sprayed some onto a couple of the big missing chunks. This was a color test, and it didn't look too bad in the morning light (shade). It's a tiny bit redder than it ought to be, but I think it's going to work out well enough. I sprayed a bunch into the cap and used a small strip of sheet metal to dab it onto the rock chips and etc. I also dabbed some clear spray onto the scratch in the clear coat that some kind parking lot-mate gave me recently. It's lumped up, but less visible than it was. I'll block-sand it down flush later once it's dry.

...When I got home from work I popped the cluster out again and attached the now-dry center vent tab to its actuator using a short screw and a nut that I dabbed paint into the threads to serve as a lock. To improve visibility I removed the center vent eyeballs, that really eased the job. When I tested it the tab bound a bit, but a few licks with a file took off the high corner that was catching on the air horn. Works great! With the eyeballs out I was able to thoroughly clean the grime and dust off of them, I did the inside of the center vent air horn too.

I then managed to get the defrost actuator speed nuts off (two of the three broke) and found that the actuator seems to have come unclipped from the flaps inside the ductwork! The original plan was to see if I could swap the actuator guts with a good used one that I have that has a twist-lock mount, and failing that I was going to temporarily lock the flap in the closed position, but there's no point in doing the former and no way to do the latter. There's nothing for it except to pull the dash and open up the heater box. That is not going to be fun, I'm not sure when I'll do that. Not soon.

The rotted foam hose was very exposed at this point, I cut a length of foam pipe insulation and put it on. Then I reinstalled the center vent parts and buttoned up the car. We're going to be gone for a week so I enlisted some aid to put the hard top back on. It doesn't go on very easily, it seems that either it or the car itself is deformed. Probably the top got dropped at some point, its 'chrome' trim is a bit dinged. It's manageable, just not as easy as the other three SL's I have experience with. Once it's locked down it's fine.

Friday, August 10, 2007

Sunday, August 19, 2007

However that's not something you'd ordinarily do to a nice car, but a cheap SL isn't so much that you'd worry about it. And an SL is as small as you'd want to go for a practical toad for a family of three, plus some crap. It would also be highly unusual, which is good for me.

There are difficulties, largely that the car has an automatic transmission. Though vintage MB's can be towed, it's not fast or far, the rear pump is apparently not up to the task of keeping things lubricated. I am very much against having to take a wrench to the flex disks in order to rig for towing. My wife would not be amused. It would really have to be simple to attach and detach, wrenching and/or trailers/dollies are right out. I do not want to leave the car idling while towing it.

Two ideas:

1. Replace the transmission with a manual. Not easy, the parts probably don't even all exist.
2. Hook up an auxiliary electric hydraulic pump. It oughtn't need much flow. This would keep the guts of the transmission lubricated during the long haul. I'm assuming that something could be plumbed in at the cooler lines, and/or the test gauge ports.

Interesting food for thought.

Monday, August 20, 2007

It was just cool enough today that I'd have expected some small amount of heat from the ACC, though while it seemed to operate normally enough I wasn't getting any heat. I'll need to check out the monovalve, it wouldn't surprise me if it was torn. It's also possible that it wasn't quite cold enough to try to heat, I didn't try messing with the thermostat dial.

Tuesday, August 21, 2007

002 545 33 32
Leerlaufdrehzahrlregelung
Idle Speed Control       8Zyl.
412.202/010 /001       12V

Opened, it revealed a LM2901 quad comparator, two LM2902 quad op-amps, three signal transistors, one power transistor, sixteen diodes, six electrolytic capacitors, and the usual handful of resistors and capacitors. Two of the resistors were mounted on stakes and looked like tunable items. (Idle speed being one, I presume.) This circuit was also varnish-coated, so I hit it with the paint stripper too. The soldering didn't really look that bad, but it doesn't take all that long to redo. There were a few spots that looked questionable.

I also pulled the monovalve and had a look at it. It looked good, the diaphragm was not torn. I cleaned the screen off with a bristle brush and put it back together. The coil continuity was good, the plunger was not hydrolocked.

The stupid hood pad fell down again, so I had to glue it back up. What a pain!

The car started easily, and idled normally. We'll see how it behaves.

...Some have suggested I check into Remco's towing products. They have a driveshaft disconnector, and a transmission lube pump system! (As well as tow bars, etc.) That lube pump system looks very interesting, I bet one of them can be adapted to this car even if they don't officially offer it. Unfortunately it's $600! The tow bar is also $575, plus some $200+ for the part that permanently attaches to the car. A braking unit for the toad is more than $800. The driveline disconnect, if applicable, is more than $650. This is not a cheap hobby!

Driving the car today the idle seemed as before: nothing really wrong with it. Still seems a touch high, if it bothers me enough I could try tuning the resistors inside the controller. Maybe someday. The ACC was definitely heating today, it might even have been overdoing it. On MIN it switched to cooling (vent only), so mostly it seems to be behaving itself. The foam tube kept coming off, I think it's about 1/2" too short. I'll glue on the offending end when I get home, that may have been its only problem.

Once home I had some free time, so I glued on the loose end of the foam tube. I then pulled out the idle speed actuator and cleaned it out with carb cleaner. A lot of black gunk came out. Tested on the bench it seemed to behave properly, so I reinstalled it. The rubber mounting hoses are a bit loose, but I don't think they're leaking yet. They're still flexible, so I don't think I have to replace them. With the engine running a spray of carb cleaner around the valve didn't result in an RPM change, proof of sorts. I note that the vacuum advance line to the distributor is purposely disconnected, I'm going to have to go through the engine vacuum plumbing some time to put it all right. There was a missing screw on one of the exhaust manifold heat shields so I dug up a suitable 10 mm screw and installed it. That should help prevent nasty rattling noises, which I hate.

Next up was the AC system. I pulled off the R134a conversion fittings, there was only a residual puff of pressure. I moved the Schrader valves from the conversion fittings back to the R12 fittings, then connected up the vacuum pump and put a quick vacuum on it. It seemed to hold vacuum, so I then put about 60 PSI of propane into the system, that didn't leak down either (not that I gave it much time, but I was only looking for profound leakage) and the ACC system was able to engage the compressor with that little bit of pressure. With the preliminaries out of the way I vacuumed the system for more than 1/2 hour. It took some minutes before the vacuum pump started making the rattling noise that indicates that a deep vacuum has been pulled. I don't know if there was moisture or not, but vacuuming can't hurt, and a half hour is about the minimum that'll do any real good removing moisture. Once vacuumed I then charged the system with the test refrigerant: 8 oz. of Isobutane/Propane (R600a/R290) mix, followed by a propane charge until the system seemed to be operating correctly. (This takes some time to sneak up on, it's not like you just put in 3 cans of R12 and call it good.) Unfortunately evening was approaching and it wasn't really all that hot, not good conditions for charging a system this way. I did see 39 °F vent temperatures, but that's no great feat under these conditions. Still, it did seem to be working, though it was short cycling due to low evaporator temperatures. Time will tell if I have it charged right, and if the system works correctly and doesn't leak.

The lights were next. I've been afraid that I might be caught out in the dark with them messed up as they are. I pulled the offending bucket assembly out of the car and found that the 6-pin connector was shifted by one position, completely explaining why it wasn't working correctly. The exposed pins were tarnished, suggesting that this has been wrong for more than a little time. Plugged in correctly, the lights all worked right. One fog light was still out, but I found that it was missing a spade connector inside its shell. I replaced that and then all the lights worked as they should. One of the mounting screws for the bucket assembly has a broken clip in the car's sheet metal, it doesn't grab too well but seems to work well enough.

It was time to quit for the night, but I was really having trouble putting the air cleaner back on. I rerouted the offending fuel line, which cured the problem. While in the area I note that this car does not have the same E.H.A. valve as the later SL's, I think it still uses the same warmup regulator as the 450 SL did. It has an O₂ sensor, so there must be some electronic control of mixture possible.

Wednesday, August 22, 2007

Since I'm driving it I also outfitted the car with a calculator, notebook, and pen. I filled it for the first time yesterday, and if the trip odometer was cleared at the last fillup the car got 13.8 MPG. Not great, but long spells of sitting are not good for mileage. Also, if the O₂ sensor system isn't working the car should be running rich, which won't help.

I called the PO, and he's going to look for the receipts for the engine work. He left a couple of CD's in the car, we'll swap at some convenient time.

We've had rain lately and the trunk is damp. That'll have to be addressed soon before real trouble starts. The car was garaged at its PO's; sadly, that can't happen here.

Going to work, the car started with great reluctance. I need to look at the cold-start system soon.

...At lunch I picked up some more carb and brake cleaner, and a piece of 1" foam pipe insulation to perhaps replace the missing insulation on the AC/fuel heat exchanger. (We'll see if it fits.)

Thursday, August 23, 2007

When cold, unplugging the idle actuator still results in no change. (It's probably open all the way already.) Timing? Cold injection boost?

I cut a length of 1" foam pipe insulation and put it on the AC/fuel heat exchanger. It fits close enough. There is a clamp missing that holds the heat exchanger to the retaining bracket on the fender.

...At lunch I found that the U-Pull had a 1982 380 SEL, so I grabbed a few tools from the lab and headed on out. I found that it was still fairly complete but, like this SL, does not have the E.H.A. valve I need for the 560 SL though it does have a lot of engine parts for this car. (It also still used the old-style thermostatic idle valve system, not the electronic system.) Disappointed, I did manage to scrounge around and get some AC valve caps and brake bleeder covers to replenish my stock. I did grab the four unusual relays from the SE's fuse box, including what looks like the identical fuel pump relay to this car, and a wiper blade for the 300 SDL. $10. I also got a piece of grille plastic that I may be able to use for patching the missing chunk, and some bracket bits and screws for miscellaneous fabrication purposes.

It was moderately warm today, and the AC was functional on my errand. It did cycle on and off a bit more than it probably should, I may have it overcharged. With the defrost vent stuck open it's hard to tell just how well it's working as most of the cold air ends up going to a moderately useless place.

I also found a link to a servicing document on the 722 automatic transmission. Not exactly for this car, but I bet it's close. I might be able to glean some toad information out of it.

Friday, August 24, 2007

The weather's getting nicer again, so I took the top back off the car. With the top rail exposed I was able to install the extra turn lock gasket that I had laying around. Still need another, though.

...At lunch I bought a cheap towel to shove over one of the defroster ducts. That kicked up the AC output a bit. I'm thinking of cutting the towel into strips and sewing it into narrow tubes that I can put pea gravel in. Those would lay down in the defroster duct grilles to block them off. They'd be heavy enough to not blow around, and small enough to not be too obtrusive. It's that or pull the dash, and I'm not ready to do that yet.

Saturday, August 25, 2007

The front license plate always looked funny to me, too high, and I decided that its mounting bracket was upside down. When I pulled it off I found that it was pretty mangled, so I welded up the cracks and holes in it. I did a ratty job (dirty thin sheet metal is no picnic to weld on) but it's now solid again, if not beautiful. The crummy welding will not show. I refreshed the black paint and hung it back on the car, right-side up this time. I then put back the license plate. Looks a lot better!

I had another session of gluing the vinyl back onto the fuse box kick panel. The first section has a couple of small wrinkles, but nothing too bad. I'm sure it'll be just fine.

...At lunch time I was out on errands and ran across a yard sale that supplied me a black sweatshirt to wedge into the other defroster duct. Free! The AC blows nice and strong now. I don't know who I'm kidding about sewing up something nice for temporary use, I bet this crap stays wedged over the dash until I fix it properly.

...In the evening we drove up to the neighbors', and the headlight alignment is all wonky. One side is lighting up trees, the other is for finding lost contact lenses...

Sunday, August 26, 2007

I have noticed that the kickdown switch doesn't seem to work. (One test is when I leave the house I usually hold the transmission in L (first gear) going down the hill. If I momentarily shift to S and then back to L it'll shift to second and stay there. If I put my toe under the pedal and poke the switch it should drop back to first but it doesn't, unlike our other cars with the four speed automatic transmission.) I pulled the fuel pump/kickdown relay again and checked continuity from the kickdown pin to ground. The switch and solenoid, when operated, measures 15Ω, which seems right to me. Either the relay's not supplying power or the solenoid just isn't working mechanically.

Who pop-rivets in place of screws? I pulled the driver's door panel off to see what kind of mess was in there. That, of course, required a drill. (I'm missing five of the seven little chromed trim screws that hold on the two chromed cover plates. The pop-riveted hole is, of course, ruined for the proper-sized screw.) With the door apart I found that the keyless entry installation in fact wasn't all that bad, I think it can be used as-is. But there were other problems. The roll pin for the interior door release handle was working out of place, and was too short anyway. The handle was only being held by one of its two ears, that's why it felt so sloppy. I drove out the roll pin and replaced it with a longer chunk of galvanized framing nail that was close in diameter. Though a tiny bit looser than the roll pin the net effect is much better since it makes contact full-length along all bearing surfaces. The actuation ear on the handle was starting to wear where the rod attaches so I put a small brass washer over the worn spot. There were some gouges in the handle's paint so I touched them up. When I reinstalled the handle assembly I found that it was adjustable in position and when shifted all the way forward the door no longer stuck shut, and would open easily. With that adjusted I found that the vacuum valve for the locks also needed some adjustment to operate in both directions without leaking. (It wouldn't always lock before.) I then put some weatherstrip glue in the cracked A-pillar seal and ran the window up to clamp it in place. I tightened the front window channel mounting screws. (They've been stripped, unfortunately, though I could still get a bit of a grip on them.) With that tight the mirror doesn't wobble anymore. With the window up I found a loose chunk of plastic and a metal spring clip in the bottom of the door. Not sure what they're for, will investigate further. The plastic looks like a drip shield or something. If I can't figure it out on my own or with the help of the manuals there's always the other door to compare it to. (It's due for surgery too.) It is obvious that the window glass needs a bit of adjustment in order to operate correctly, I believe its misalignment is responsible for most of the damage to the window seals on the A-pillar and the hard and soft tops. (The major chewage on the top seal of the passenger side is due to the broken bow bolt that lets the center of that seal drop down.)

...In the evening I had a final (?) session of gluing the vinyl back on the fuse box surround panel. It isn't perfectly smooth, but it has come out better so far than any of my earlier attempts. The full-face clamping in the wood vise helps.

Monday, August 27, 2007

I then tackled the driver's-side window adjustment, etc. I printed out the instructions and then put up the soft top. I had to do some more gluing of torn window seals, and the adjustments are fiddly. I don't think they're quite right but they're better than they were, insofar as ripping up the seals is concerned. I imagine that there will be plenty of wind leaks, but to fix that would be a whole 'nother level of commitment! (Mostly monetary: new seals all around.) The stripped bolts in the front window guide/mirror mount are bothering me. Helicoil? Will need to think about it.

The broken bolt in the passenger side soft top mechanism is troublesome. I couldn't get it loose, so I soaked it in Kroil. There was torn rubber on that side too that I started to glue up, and I replaced a missing screw. I also glued down a couple of flaps of the soft top itself.

No clue yet as to the identity of the plastic piece or the metal clip found in the door.

Setback! I tried to go to work and it wouldn't start. I smelled gas, it wasn't a recurrence of the fuel pump problem. I tried to test for spark and got nothing. Though the coil was getting power, the ignition module didn't seem to be grounding the other side. That's a disappointment. I ended up taking the Chicken Wagon, I threw my tools into the trunk.

...At lunch I swung by the U-Pull to have another look at that 1982 380 SEL. The engine was gone already, but the ignition module was still there! I also nabbed the coil, a coil cover (which my car is missing), and the sunvisors. (Those aren't at all a good fit, but who knows, I may be able to do something with them. The lighted mirrors are in perfect condition, and we have four cars that use them. At the very least they'll be spares.) I also got some tiny chromed (?) trim screws from the long strips that cover the door openings and from the foam pieces at the rears of the side windows, perhaps they will stand in for the missing screws on the chromed door covers. And I got a few other things like an idle control relay (that the car couldn't have been using) and a door lock pump. $12 total, $14 with admission.

...Home again after work, I couldn't wait to see if the ignition module worked. I plugged it in place of the original and grounded the case with a clip lead (just in case it needed that). It started! I'm going to investigate more tomorrow, but right now it looks like maybe I dodged a bullet. (New ignition modules aren't particularly cheap.) Perhaps the module died due to the extended key-on session with the windows this morning. It shouldn't have, but who knows. If it is bad and I can get it open I may try fixing it as an experiment.

Tuesday, August 28, 2007

So I opened up the bad one, it has a cover held on with four screws. Inside were some beefy old-school electronics: a possibly TO-3 power transistor embedded in the aluminum case, looks like it's riveted in behind a heavy plate, a 16-pin TO-3 IC on the board with a Bosch logo and labeled 0086 and 62S, what appears to be a potted thing that may or may not be an 8-pin SIP IC, two power diodes, a zener (?) diode, two signal diodes, a 32Ω 10W power resistor mounted to the case, a 0Ω 2W (?) wire-wound (actually ribbon) emitter degeneration resistor thermally clipped to the case near the power transistor, and a handful of resistors and capacitors. Two of the resistors are up on stakes, probably tunable components. It all smells a bit funny, and it's a bit cooked on the transistor/emitter resistor end of things. I got out the Fluke on its diode test function and while the B-C and B-E forward junction voltages measured 0.5 V, OK for a NPN silicon transistor, the B-E reverse voltage was 1 V and the B-C reverse was about 2.3 V. Not good, I'd say.

Throwing caution to the winds, I opened the good one. Its date code is 49 91, much newer. (Am I surprised?) Inside it was much the same, minus the cooked look and smell. It had the same 20086 IC in it, with additional marks of 1492 (date?) and B879828.1, but the mystery potted item looks like it was a resistor network that was blown out onto the board in the newer design. (Using smaller components.) I pulled two of the wires off the power transistor legs and measured it with the Fluke. The only difference was that the B-C reverse connection measured open. The B-E reverse voltage was the same 1V, which is odd but I guess OK. (I read something somewhere that indicated there were B-E anomalies when the transistor has a built-in E-C clamping diode.) Obviously the power transistor in the bad ignition module has suffered some damage.

I put the good one back together and put it in the car. It works. As a side-benefit I found the missing piece of the front grille, so I should be able to glue it back.

The lesson? Don't leave the key on for hours at a time! (Or unplug the ignition module first.)

...After work I used a cold chisel, BFH, and anvil to chisel the rivet bumps off of the transistor chamber cover plate. Released from its prison, revealed was a TO-3 package (but with three pins, not the usual two) riveted to the plate. It was marked 1022 and 1/128. It has a Bosch logo on it, so no doubt there's zero chance of looking up an official replacement part. Its collector is connected to the case, which is normal for TO-3. I think that I could try substituting a suitable NPN power transistor intended for ignition use, provided I could find one with the E-C clamp diode included. (NTE2302? NTE2331? These are $10 items.) It couldn't hurt, except for the cost, and might be instructive. It could also smoke instantly or only after awhile, in which case one had better not travel without carrying the other unit in the car!

Reading shows that the module's case is intended to have a good thermal mount to the car's fender, but mine does not. I should look into that. Some have suggested that this was the root cause of my problem, not leaving the ignition on.

Wednesday, August 29, 2007

I then pulled the ignition coil out and repaired its mount. One of the captive nuts was pulling through the fender wall so I hammered it back down. While I was there I rerouted the wiring to the coil so that I could put on the plastic coil cover from the donor car. (All wires exit the cover from the same side, and none of the ring terminals can hang over the side.)

The AC/fuel heat exchanger is supposed to clamp to this same mount, but the clamping strap was gone. I fabricated a new one from the AC/fuel heat exchanger mount from the donor car, using the grinder, BFH, pliers, and the anvil. It clamps nicely around the new foam insulation, and doesn't even look too bad. With all this done everything is much more solidly mounted than before.

Unfortunately during all this I managed to break a nipple off of the vacuum valve on the vapor tank, so I had to pull that off and repair it. I used cyanoacrylate glue to tack it back into place, then potted it in Shoe Goo for strength. We'll see if it holds up once it has set, it may not be strong enough.

...Surfing, I'm having nearly zero luck coming up with a transistor selection guide that will help me find a suitable replacement transistor. Stupid internet. The few electronic ignition transistors I've managed to run across are all Darlington pairs, and usually in TO-220 cases.

Thursday, August 30, 2007

I began gluing up the driver's door skin and pocket. I peeled back the door skin at the bottom, first removing some staples, and then slid a sheet of mylar (floppy disk) coated with weatherstrip cement behind it. That should close up the small tears that are in the lower rear corner. I then washed the shattered door pocket plastic with acetone and glued a large sheet of thin sheet metal (from a tin can that was in the trash fire) to the inside with Shoe Goo. This metal was thin and malleable enough to conform to the curve, yet not quite so thin and weak as beer can material would have been. It spans from just below the ribbed top edge to partway past the curve at the bottom, over about 1/3 of the length. It has a compound curve because of its size and location (the break was extensive). Clamps were used to force it to conform more-or-less to the correct shape imposed by the pocket's plastic. Several sessions will be required in order to restore this pocket to good condition, there is also a chunk out of the corner where the (missing) attachment screw should go.

I then removed the door panel on the passenger side. As expected, the missing lock knob piece is not to be found. However, inside the bottom of the door was what looks like a window channel slider, and the roller from the broken door check. The spring-loaded lock coupling is bent, and the springs (typically) are broken. I should be able to come up with something from the junk box to fix this.

Neither door had any of the vapor barrier left. Fools.

I then removed the broken door check. That was, as is often the case, very difficult as the pin in the door frame was frozen. Much work with the BFH and prybars was required. Some more work with touch-up paint will now also be required. The check assembly itself was held in the door with only two 5 mm Allen screws, unlike everything else I've seen which uses three 10 mm hex bolts.

Oooh, I like the 107 door check! It's steel, with real coil springs and two rollers that roll over the thick shaped metal arm that's attached to the door frame. Eminently repairable, in other words, no wonder it's so expensive! The axle pin, a long rivet, for one of the rollers has disappeared. I should be able to replace it with a bolt or something and it'll be good as new.

...I was out late, and used the headlights. Still off, I think the left side lows are low and outside, and the right are a bit to the outside. The right high is a bit low, but I'm not sure if it's the dual lamp or the high-only. More dialing in required, naturally.

I also noticed that the key doesn't operate the central locking, it doesn't seem to push the lever quite far enough. The driver's door locks and unlocks fine. Needs some kind of adjustment!

More surfing has led me to the conclusion that you can't get an ignition transistor these days that is not a Darlington pair. Even if you could it would probably not be cheap. I wonder if the existing driver IC would be happy pushing into a Darlington pair? Might be worth a buck or two to try. BU941 is one such transistor, something like $2.50. I may even have an igniter (driver transistor) from some Japanese car laying around that I could try out.

Just as I had gotten into bed it started to rain, so I ran out and did my putting-up-top-naked-in-the-rain dance.

Friday, August 31, 2007

I then fixed the door check. I found a 1.5" bolt with M8×1.25 threads that fit the roller perfectly. When in position half the roller would be on the non-threaded part, so that was good too. I threaded one of the two rivet holes in the spring-loaded arm in the check and drilled the other hole out to fit the non-threaded shank. I then greased the roller and assembled it, using a lock nut to secure it in place. (You can't tighten it into the check because the roller wouldn't spin if you do.) Then I greased up the other roller and the hinge joints and assembled it with the arm extended, as you do for the other cars' door checks. There, that's $170 I don't have to spend!

OK, what I don't like about the 107 door check is getting it into and out of the door. What a PITA. It came out with only moderate difficulty yesterday because it was in two pieces, but today... I had to take the check back apart and assemble it with the arm retracted, then dismount and shift the window lift mechanism in order to get enough room to twist the thing through the tiny maze of access holes to get it into place. Then I had to he-man the arm back out in order to get it into the frame hinge. However, once it was all installed it worked well. It took longer to install the check than it did to fix it.

Inside the door bottom I found another one of those plastic shields, to go along with the window slider I'd already found. So much for using this door as a pattern for the other!

I put another dab of weatherstrip glue on the torn passenger-side soft top window seal, since it was up, and took another stab at headlight adjustment. Some fool had managed to get some kind of plastic material gooped into the screw heads at the top of one of the headlights, I had to take them apart and use a propane torch to soften it up enough to dig it out. Sheesh! I left the headlight bezels off until I can finish the dialing-in. (And I really need to do some paint touch-up under there too, both sides.)

...At lunch I swung by my favorite RV place to pick up a trailer light connector for the boat, and found they've sold out and were blowing things out at half price to empty the shelves. I couldn't resist, and bought one of Blue Ox's Aladdin aluminum tow bars for half price. If I don't end up using it to toad-ify something (this car?) I could probably turn it over without losing too much. There are no stock baseplates (the car-side mating piece) for cars this old. I'm going to have to get one for something else and cut it up to make one if I want to make a middle-aged MB toad.

There were several cloudbursts today. The soft top, though not in great condition, nonetheless seems to work pretty well at shedding water. (It is tricky to put up and down with that broken bolt.) The rear window is intact, but is very cloudy. Really quite hard to see out of. Coming home in the dark, the headlight aim is getting better. I think the right low beam is still aimed out a bit, and the left high beam is also aimed out. We'll get there!

Saturday, September 1, 2007

Sunday, September 2, 2007

I also dug out the extra broken and rust-covered 107 hinge cover that I'd found under the seat of the 450 SL, which I cleaned off and glued (using gap-filling cyanoacrylate glue). It isn't good enough ever to be used on one of the nice 560 SL's, not that either needs it, but it's plenty good enough for this car, for now. It installed nicely on the passenger side of the car, and in spite of its ratty condition really helps improve the door-open view. (Most of the hinge covers are still missing, but those top covers are the most prominent.)

I pulled the passenger-side door pocket off its panel to use as a model. The door panel looks good, I shouldn't need to do anything. The pocket is cracked, but not shattered like the other one. A single Shoe Goo session should do for it. (Both pockets need some weatherstrip cementing to reattach the skins along the tops.) I cleaned the break area with acetone, cut two more stiffening strips for the ribs, and glued it back together. (I didn't use a backing metal plate for the pocket itself this time.) While it was setting up I scrounged up a chunk of black ABS plastic to stand in for the missing chunk and roughly trimmed one end of it to fit the broken edge. Then I Shoe-Gooed it into place. When it dries I'll trim the other end to size, and probably fill in one more small chunk before calling the structural repairs good.

...Later in the day I did some more reconstruction work on the pocket's corner. A trimming and one more Shoe-Gooing should mean it's ready for vinyl gluing.

Monday, September 3, 2007

I pulled the driver's door handle off in order to adjust it as it wouldn't lock and unlock the central locking properly. The paint under the handle was weird, it had slumped off the body of the car into a ripple that broke off when the handle came off. It's like the original paint didn't stick to the primer, and was free to shift around a bit while it was still soft. (And perhaps that's exactly what has happened.)

Anyway, I got sidetracked a bit and did a bunch of touch-up painting since I had to treat under the door handle anyway. I took off the rust over the front headlight bucket with the Dremel and painted over that, and dotted over any other paint chips I could find in the under-light area. There were missing screws in the over-lamp trim on one side, that took a bit of time to deal with. I also did some miscellaneous touch-up in the door frame area, and along one edge of the trunk lid.

Back at the door, the operating 'ear' on the handle didn't seem overly worn, so I pulled the latch and squeezed the steel of the lock's mating channel together a little closer and put it back. (That results in a little more lock travel for any given key travel.) With the handle reinstalled the door would then lock and unlock fully with the key.

I had some thought of us all taking this car out today so I thought I'd dig out one of the Kangol lap belts to put in this car for tethering Daniel's seat in the back. Not! I don't have another Kangol lap belt. All this time I thought I had a full set on the shelf, but all I had left was the fronts. I'm going to have to go back to the yard to see what's left there, I guess.

...Out today (family in two cars) the door still was hesitant, once, to open from the exterior handle. Will try lubing it first, it doesn't seem like there's an adjustment problem. Had a pleasant little drive over to Cd'A today on the back roads.

Tuesday, September 4, 2007

For starters I glued down another section of door pocket skin. One or two more sessions for this pocket, then it's on to the other one.

I put the passenger-side door central locking slider back together. I'd thought the overcentering springs were broken as this is often the case, but they seem only to have come out of their holes. I did some bending of the slider's metal to tighten it up where it had gotten deformed, and then lubricated it so that it wouldn't bind and bend again and reattached the springs. Some experimentation found how it was supposed to be configured. Looking at a potential donor mechanism from my parts pile was also helpful. After all this, and a lubrication of the latch mechanism, it seems to be working correctly.

I also tightened the loose driver's-side window mechanism's mounting bolts, and lubricated the door latch some more to see if that'll help its intermittent stickiness.

...I noticed on the U-Pull web site that they'd just gotten in a 108 body car, a 280 SE as it turns out. I found that it had a complete set of Kangol (KL-2, 1968) rear seat belts, which are my favorites for tying down child seats. Two of them had BW-100 Roll-a-belt retractors on them. I got the whole set of three, but one of the bolts had frozen up and I couldn't get it out. My usual trick of using the BFH and cold chisel to cut out the captive nut didn't work, the metal brace they used was much too thick. I cut the stitches instead and pulled the belt out of the bracket. It may be I'll never be able to use it for anything, but the buckle was in nice shape and I did want that for a spare. $6 (including admission) for the belting this car needs.

On a side note, the 108's engine does indeed use an injection pump that looks almost exactly like a diesel pump. Supposedly only the governor (on the end) is actually different.

I found a spec. for a Motorola chip that might be usable for replacing the entire guts of the ignition module:

MC79076

Reluctance or Hall Sensor input, high energy ignition control circuit used in 12 V battery systems; drives external Darlington used to control coil current, internal zener for flyback protection of external Darlington, spark advance, RPM detect, coil current dwell, and coil current limiting.

From the looks of it, that chip and a relatively common Darlington ignition transistor, plus a few discrete components will replace all of what's there normally. I'll consider that in extremis.

...In the evening I glued down another section of door pocket skin. Last one on this side?

Wednesday, September 5, 2007

I then glued the loose window glide back onto the glass in the passenger door, I used vinyl caulking. I hope that it will stick well enough to work, but not too well!

I selected a pair of seat belt segments from yesterday's booty and washed them in a bucket of soapy water. The water came out filthy! Several rinses later the water was coming out clean, the belts looked clean, and they were nice and supple. I hung them outside to dry. It's amazing how much dirt can get caked into a seat belt.

...And why am I doing all this part rebuilding anyway? A new door pocket is $200. A new door panel is $250. A new door handle is $70. That's why!

And no, the vinyl caulking didn't hold when I tried it after work. Plan B? Will probably need to pull the glass to do that. But first...

Sick of the persistent rattle in the driver's door, and worried about the continually-loosening mirror, I pulled the window lift, glass, and guides out of the door. (Pulling the window lift completely exposes the door check, that's what I should have done on the other side.) The secret to removing the guides is to turn them sideways first. One of the big rivets in the window lift mechanism had a little play in it, so I took it to the anvil and peened it tighter. That should eliminate the rattle. Once the glass and guides were out the problem with the mirror became evident: Some fool had put in too-short bolts, which promptly stripped the pot metal when tightened. I re-tapped the holes M6×1.0 (same as before) and found some longer bolts. If they don't hold I'll get the next-larger size bolt, possibly in US threads, and re-tap for that. (I should be able to do that without taking it back apart, now that I know what everything looks like.) The hidden screws that held the window channel to the mirror body were also loose, that didn't help either. The mysterious black plastic shield found inside the door is now obviously something that goes over the trailing bottom edge of the glass. It looks like it just clips in place, so that's what I did. I'm still not sure what it's for, it's obviously a drip guide, but why?

The top lip of the door where the outside belt clips on had a lot of paint chipping off and minor surface rust starting, and not just due to the clips' teeth. Rather than leave this rust magnet open I scraped it off, masked the door with tape, and used the rattle can of touch-up paint on it. Two coats, we'll let it dry until morning. That's when I'll start putting the door back together. The paint shouldn't show at all once it's back together.

Thursday, September 6, 2007

Because of the missing spring the original plan of reassembling the door was off. I temporarily put the mirror back on the car and put everything away so I could drive it again. (Sans window.) The window's mounting flange was starting to rust at one end, so I wire-brushed it off and gave it some primer and paint. That'll only slow it down, I doubt it'll stop it. (To do that I'd have to remove it from the glass and do some welding as well as more involved painting. Not gonna happen!)

I also did the last (?) gluing of the skin back onto the door pocket.

...At lunch I bought a near-perfect match for the missing spring. $1.50.

...After work, wanting to get the window back in quickly in case of bad weather, I reassembled the door check with its new spring. The check had always been missing its two plastic washers at the hinge, I was able to find one in the junk box but not a second. I made another one out of a piece of thick plastic.

I noticed that the lips of the window opening on the door still had a bit of surface rust, so I sanded and rattle-canned them with gray paint. There was another chip or two in the color coat, so I cleaned them off and rattle-canned that too. I then greased the window guides and tried to put the window back, but that is not easy. It is very fiddly, and the order in which you put things back matters a lot. I had to remove the mirror from the forward window guide in order to get things assembled in the right order, that aluminum trim strip doesn't go back otherwise. (Memo to self: next time remove the mirror first. It's easy to do and really eases the job.) I didn't really reach a good stopping point by the time I ran out of light, nothing is bolted down and the door doesn't even close fully because the trim strip is on too far forward. I believe that I'll be able to get the window back in before going to work tomorrow, though perhaps not the window lift mechanism. This stuff is very time-consuming!

Friday, September 7, 2007

While the glue cured I went back out and put the reconstructed door pocket back on its panel.

After that the glue had set enough that I could resume reassembling the door. It went together straightforwardly enough, but it is very time-consuming to get everything lined up right. There's a lot of tucking while assembling that is required on the door seals front and back. I was even able to get the window lift greased and back in, the secret there is to remove the motor first and reinstall it once the lift is inside the door. The window moves correctly now, but it is very dirty and greasy, and it needs to be adjusted again. I ran out of time to do this.

...The door and window work well, the rattles are gone. (Though the refurbished door catch strap 'grunches' a bit due to the worn roller. That's too bad, but a better repair would have been either much more difficult, or much more expensive.) The window is smoother than before, the new grease helped I'm sure. It just needs a window adjustment again and it could be buttoned up. That would be nice. Oh, and I need to test the keyless actuator first to make sure it works. Sigh. It never ends!

I must say that on the whole the door hardware (such as the window regulator and the check strap) in the 107 is of significantly higher quality than the stuff found in, for example, the 123.

Saturday, September 8, 2007

When I lifted the soft top while making the window adjustments I noticed a fair amount of rust getting started under the soft top cover and on the body channel for its seal. I wire-brushed it all off and painted over it with black paint. That should stop it, or at least slow it down considerably. The seal was torn in one place, I used weatherstrip cement to glue it back together.

The combined trunk/horse collar hinge was stiff, so I sprayed a bunch of PB Blaster and oil into there. I made quite a mess, but you just can't quite get at the hinge. It's now looser than it was, but not yet what it should be.

Rust! I pulled up the trunk seal above the left taillight and found that the channel was holed. I pulled the taillight and the interior trim and hit everything with a wire brush. The corner above the light is rotted away, the rest was only surface rust. I painted the surface rusted areas, including the cover for the vacuum trunk lock. They'll all be fine, but what to do about the missing metal... I really should weld it up, but that's not a small job! And it is a bit risky, too. Let us not mention that the welder is trapped inside the garage right now, I'm not sure the cables could reach far enough to do the job.

Contemplating the welding-ness of it all, I pulled the other taillight and found a bit of rust starting there too, but nothing too bad yet. I wire-brushed it off and painted over it. I had to remove the chrome trim strip that the trunk knob pokes through in order to finish brushing off the rust on the bad side. (And to do that requires that the trunk latch be removed. Everything cascades, it seems.)

Since I was in the area I removed the broken battery cover strap and welded it back together. At least that was easy! I think I can do some welding to restore the holed channel to watertightness. I'll probably not attempt to do more on the inside where the material is eaten away. I don't think it is significant, except cosmetically.

I freed up my afternoon, backed the car up to the garage, and had at it! I welded some narrow strips of heavy-gauge sheet metal over the holes, and then ground them to more-or-less conform to the profile it should have. (On the outside, at least.) Weld and grind, weld and grind. Repeat until done. It took a couple of hours to put back what was at most a couple of square inches of metal. Water-soaked towels draped over the entire area contained the welding spatter, and kept the heat from getting into the paint at the trunk opening. I did a lot of scraping and poking around to ensure there weren't additional holes coming through. I found a couple, in fact, which I welded over and ground down. I then painted the outside with black spray paint. I scraped out the rust flakes that were 'inside' using my finger, compressed air, and a magnet. I then saturated this inside area with the spray paint, shooting into the wall interior through the holes that shouldn't be there. I hope this will cut down on the rust's progress.

After a late lunch I put a color coat on the outside where it needed it and reassembled everything. That took a long time. I glued the trunk seal back in place with a bead of weatherstrip cement. I replaced a couple of dead lamps while I was there and patched the reverse lamp lens that had a piece out of it. What I did was to take a ruined taillight assembly and knock out a few chunks of the clear, which I then ground roughly to shape and glued in place with cyanoacrylate glue. I then ground the bottom edge flat and potted it in Shoe Goo where it meets the gasket. While it's very obvious that this has been patched, if you're not looking for it you might not see it. The damage was in the lower inside corner, and was not large, about 2×0.5 inch.

Sunday, September 9, 2007

The rear of the metal channel clamped over the window was rusting on this side too, so I did the same brushing and painting trick to it. With luck it'll slow things down.

I then cleaned off the glass where the glide had fallen off and cut a piece of inner tube to serve in place of the decomposed fabric that had helped wedge the glass into the slider. The slider couldn't be forced over the rubber onto the glass without an excessive amount of force, I was afraid of breaking the window. I lubed the rubber with Xylene and put it in place on the glass and forced the slider over it. It slipped right on with only moderate force, and once the Xylene had evaporated the slider was attached solidly to the glass. I expect it to hold for a long time. I trimmed off the excess rubber, then cleaned the sliders and rails with solvent and lubed them with grease. I also smeared some more grease on the window lift's moving parts. The window went back into the door almost as easily as it came out. Getting it adjusted took a bit of work, but eventually I got it set more-or-less right. I couldn't adjust the up/down travel too accurately because of the broken pivot bolt in the soft top, but I think I got it close. (You can't run the window all the way up when the soft top is up, because the center of the top rail sags down.)

The rest was just buttoning it back up. On the whole, it went a lot faster than the other side. Experience counts! I did a bit of touch-up painting while I was there. Naturally. That funny bit of plastic drip channel didn't really want to go on like it did on the other side, I suppose it'll fall off by itself pretty quickly. Oh well.

The peeling-apart molding over the passenger sunvisor was dealt with by putting Shoe Goo into the split and clamping it back in place. We had to drive off before the glue had fully set, so I took off the clamp and replaced it with duct tape.

Monday, September 10, 2007

The plastic plugs for the bolt holes unscrewed easily, and the two belt segments bolted down with no trouble. I put a dab of oil on the bolt threads first. The deck snapped back down easily, the belts sneak up nicely along the back wall. I wound up the Roll-a-belt and put it on the hook end.

I even found $0.25 during the procedure!

...I (re-)filled up for the first time today, it got exactly 16 MPG. Not great, but not awful either. About normal for these, based on my recollection of a sample of one! At lunch I stopped at Harbor Freight and bought a cheap Pitman arm puller, I'm hoping I can adapt it to press out the broken-off soft top bow pivot bolt head. (I have no idea if this will even come close to working.)

It's been three weeks since I charged the AC and it's still working. Though warm today it's cooled off enough, however, that you no longer want AC. Mostly.

Tuesday, September 11, 2007

I went to the car and liberated the good locking handle to serve as a pattern. I then got out a scrap of thick-wall 3" muffler pipe and cut out a strip, longer and wider than necessary, to become the finger pull. The material is thinner than the plastic, but it's a lot stronger! The curve of the pipe gives an approximation of the shaped plastic finger pull. I got a piece of thick strapping, extra long, and welded it to the inside of the pull to make the lever part. It made a nice big T. The strap wasn't quite thick enough so I ran a bead across one side for the length of the lever-to-be then ground it flat(-ish). The extra length of strap makes a nice handle for use while working on the piece. I drilled the two holes in the lever, then started grinding away the excess strap to shape the lever. Once that was approximately right I started shaping the finger pull, grinding off the excess, etc. I went slow, and held it up often to compare with the pattern by eye. Once it was the right size I started shaping the edges and taking off the sharp and rough patches. When 'done' I cut off the excess strap and smoothed the back end of the lever. The final result wasn't an exact match to the original, the finger pull ended up a little bit short, but I believe it is good enough. I'm actually quite pleased.

I then went to the car and liberated the bad locking handle to see if the new one fit. I needed to do a bit more grinding on the lever part to get it to fit into the handle assembly and have the correct range of motion, I also made sure the finger pull fit through the trim piece. I then painted it with primer and black paint; I used the shop oven at 200 °F to speed drying. Putting the roll pin back I managed to drop the whole thing into the dirt, so I had to wipe it off and paint it again.

I then put the handles back into the car. Finally! It looks good, and works great.

All that's left is to put the doors back together. They need new plastic moisture barriers. Oh, and I also need to test the keyless actuator to make sure it's OK. However, there's no time left to do any of that this morning.

Wednesday, September 12, 2007

The door looks good, works well, shuts solidly with no rattles, and has a lock knob that both looks good and works. Success!

I went to finish the other side, but first I wanted to test the keyless door lock solenoid. I used the new Harbor Freight multimeter I bought yesterday to keep in this car, $2.88, and found that the solenoid has no continuity. When power is applied, either polarity, nothing happens. Oh well! I can fix this later if I wish to install keyless entry.

I put the driver's door back together. I had to replace five of the plastic panel clips on this side. I used a chrome armrest trim piece from a 123 to substitute for the missing one, it isn't exactly right but it looks pretty good. The riveted hole for the rear chrome door jamb trim piece had been drilled out so the correct small screw didn't fit anymore. I suppose I should weld it over and re-drill, but I think I'll wait on that! The screw is just sitting in that hole right now. I had to use four more of the 126 trim screws on this side.

This door also looks good, works well, and shuts solidly with no rattles. More success! Both doors now shut like they ought to, with that nice solid Mercedes 'thunk'.

...After work, given that I need to do some other things in the mornings for awhile, I tackled the front grille. I pulled it off and found a bit of rust starting against the hood, so I brushed that off and rattle-canned it with the color paint. The car looks pretty sad without the grille on it, I'm hoping to get it back on before I have to drive it to work tomorrow.

The plastic finned grille had a chunk off one end of a fin, which I had found inside the fender, and a half-out chunk from the inside of the big front star. I removed these from the chrome surround, cleaned them off, and used cyanoacrylate glue to tack them back together. I used Shoe Goo to back the repairs up, this is applied where it won't show and is responsible for the strength of the repair. I cut some chunks of scrap gray Mercedes grille material to serve as backing pieces. Acetone was used to clean everything before gluing.

The chromed grille surround was getting some visible rust starting on its inner surfaces, so I put it in a bucket to try the electric rust-removal trick. We'll see how it goes, it takes time to operate. I'm not sure how it will interact with the chrome. As of course I can't paint the result I may try some clear spray to protect the reclaimed spots from moisture and air. I checked the bucket after an hour or so and preliminary results are good: the water's getting nasty looking, the orange rust is gone from the metal and has been replaced by black, and the chrome appears unaffected. I rearranged the sacrificial anode to 'see' more of the rust and resumed the process. We'll let it run overnight.

Thursday, September 13, 2007

It's done a fine job on the part that fit into the bucket. But the grille is large and looks like it needs about four treatments. The other end can go into the bucket and won't need long to treat, but there is some rust inside along the top and bottom. I think I can just lay the grille on its face and fill the channels with electrolyte and suspend sacrificial anode wire in them. We'll see. I put the other end of the grille into the bucket for now, we'll deal with the middle parts later.

I put some metallic gray paint on the glued cracks in the plastic grille pieces. That disguises them a bit.

Still dark outside. I went to the car and tested the kickdown solenoid (using the handy new Harbor Freight meter), which can be diagnosed from the (still-exposed) fuel pump relay socket. When actuated it measures about 15 Ω from pin 87k to 31 and infinity when not actuated, which sounds about right, and when the meter is connected on the 10 A scale between pins 87k and 30 it records 870 mA when actuated and you can hear a quiet 'click' from the transmission. If the fuel pump relay is enabling the 87k output and the kickdown switch isn't working there's nothing to blame except the transmission itself. (The handy new cheap meter is very slow to react, unlike the decidedly not-cheap Fluke that I usually use.)

So I got some phone wire and lassoed pins 31 and 87k of the fuel pump relay and plugged it back in. I hooked the voltmeter to the wires and started the car. There was power, and when I poked the kickdown switch with my toe I could see the voltage drop a few hundred millivolts indicating that the solenoid was drawing current. Well, if it's not doing what I expect (forcing a downshift from 2–1 when I poke the switch with my toe) either the tranny is ill or it's not supposed to do that. (This trick does work on everything else I've tried it on, but MB has incorporated a lot of minor variations into their various transmissions.) I'll test it again on the road this morning, but most likely I'm done with this.

Oh look, it's light. Time to do something else.

...After work on the way home I tried out the kickdown switch. It does work, in that when driving along up a gentle hill at about 38 MPH in fourth gear I can poke the switch with my toe and get it to switch to third. Let up and it switches back to fourth, etc. But it doesn't seem to want to force a second-to-first shift like everything else I've tried, including our 560 SL's.

At home I checked the grille, and the other side cleaned up nicely. I then laid it on its face on the floor and propped it up so one side was level then filled the trough with electrolyte. I clamped a wire to two pieces of wood down in the trough and hooked it up. Done this way it'll take four sessions to clean it out, I can't get the trough full enough to span the rust end-to-end.

...After three hours in the soup I decided that it'd been long enough to do some good on the one side, so I rocked it to the other and moved the wire. The other side will get the treatment overnight.

I found an interesting link involving slipping shifts for these transmissions: Here In case the link disappears, here is the text:

Mercedes Benz 380SL Transmission Slip

Q. I have a 1983 Mercedes Benz 380SL.  V-8 and automatic transmission
   and 107,000 miles.  The original transmission was rebuilt due to a
   3–4 slip.  This occurred when it was cold and as the day went on,
   the slip went away.  But when cold, it was very pronounced.  After
   the rebuild, things were the same.  The modulator was changed and
   adjusted.  Now the shifts were too hard and slippage was still
   there when cold.

   I went to another shop and they did not want to use the old
   transmission so we purchased a used transmission with 80,000 miles.
   This one had slippage in 2–3 range.  The supplier sent a third one
   and still have 2–3 slip, but only when it heats up.  When cold, all
   shifts are fine.  When heated up, 2–3 slip is very pronounce, about
   500–700 RPM slip.

   When I really get on the car and wind out the shifts, all are fine.
   Under normal driving, all shifts are perfect except 2–3.  How can
   three transmissions continue with problems?  What interfaces with
   the transmission and engine that may be causing this?  HELP!!!

   Bob

A. There is a Technical Service Bulletin issued for this problem.
   Your transmission shop can easily check and repair it.

BULLETIN No.: 268
DATE: 1995
TRANSMISSION: 722.3, .4, .5
SUBJECT: Slipping or Soft Shifts
APPLICATION: Mercedes Benz, Porsche

Slipping or Soft Shifts

If a test of modulator pressure shows zero pressure at an idle in
drive, try adjusting the modulator. If there is still no pressure, it
may be due to excessive clearance between the end of the modulator and
the modulator valve assembly.

To inspect for this condition, remove the modulator. Start the engine
and press in on the modulator valve assembly (Figure One). If there is
now pressure showing on the pressure gauge, make a shim and install it
into the modulator valve (Figure Two).

Inspect the assembly to make certain that all clearance has been
removed.  With the shim installed in the modulator valve, reattach the
modulator to the case.  Remove the valve body and carefully push the
modulator valve assembly against the modulator.  Make certain that you
use the exact location as shown in Figure Three.  There should be no
clearance and zero preload.

Friday, September 14, 2007

...I came home at lunch, bringing home the new generator and a swamp cooler in the back of the truck, and put the re-rusting end back into the bucket. I took the convertible back to work, and I installed the junkyard fuel-pump relay in to test it and see if it helps the starting. (It worked, no change.)

I have noticed that the smog pump doesn't run at idle, perhaps it is not supposed to?

...After seeing Seussical with the family this evening I was coming home in the dark. The right low headlight is still aimed down and out, as is the left high headlight. I really need to take a screwdriver out at night sometime and finish this.

Saturday, September 15, 2007

Once the grille had dried, I masked the chrome parts and then rattle-canned gray primer on the inside, then gray metallic paint. After that had dried I sprayed clear enamel over the corroded (but now rust-free) shiny exterior parts and put the grille back together and onto the car. It looks good! The grille rib repair was weak and it cracked during reassembly, so I put a little Shoe Goo on top of the joint too, towards the back where it won't show much.

While I was there at the front of the car it is obvious the AC system is leaking. There's oil all over the receiver/drier, and when I looked closely I could see little bubbles coming from the center of the safety plug that's screwed into it. More to do there!

In the evening we took this car to our event, and I tweaked the low-beam headlights again. Still probably not done.

Sunday, September 16, 2007

I sanded the wood surround and then hit it with paint stripper. That peeled off whatever nasty non-factory finish had been slopped over the top, leaving the smooth original factory finish wherever it hadn't started chipping off. I cleaned it, sanded it lightly, and set it aside to dry. I then found a ruined bolt of about the right diameter and ground it into a smooth bullet shape. I glued (Shoe Goo, naturally!) this in place of the broken-off plastic retaining pin.

This evening I enlisted Jill's help to put the lid on, the weather's turning. (Yesterday was magnificent but today was iffy, much colder and windier. Rain is predicted over the next few days.) It went easier this time, we're getting used to the trick twist required to get all the pegs in the holes. (Something's tweaked.) I checked the windows and they seem to be aligned correctly. Good.

The console wood was dry, so I sprayed on a coat of clear enamel. Shines right up!

Monday, September 17, 2007

...After work I put on a couple more light coats, same procedure.

Tuesday, September 18, 2007

Once it was done I installed it into the car. It went smoothly, and the extra clacker even worked. I used a drill to re-open the hogged-out ashtray retention screw holes that I'd filled with Shoe Goo much earlier. The ashtray frame then tightened down properly. Looks good! It rained last night so I even washed the car and squeegeed off the glass. I checked the trunk and it didn't seem wet, though because of all the junk in it I wasn't able to dig down to the bottoms of things to be absolutely sure.

It was kind of chilly this morning, and the car really didn't want to start. It kept dying until I manually held it at high idle for a short while. I know the cold-start valve works, the problem is probably the warm-up regulator, the timing, or a vacuum leak. (Things I haven't tested yet.)

On the way to work there was water dripping down onto the console wood. Gee, I wonder how that got water-damaged in the first place? Obviously the front hardtop seal is sub-optimal, I'm going to have to deal with that.

...After work we were driving home a bit late and I got to use the headlights again. I decided the left high beam was still pointed a bit outwards, so I cranked it in some more after we got home. Someday I'll get there!

Wednesday, September 19, 2007

Sunday, September 23, 2007

I pulled the exterior trim and started trying to work the glass loose using the heat gun, a putty knife, and wooden wedges. I need to remove the glass, clean up the butyl seal material as best I can, repair the defroster wiring, and reinstall everything correctly. Big job! Before I had to break for breakfast I was able to get one side edge freed. (If I next do the other edge I should then be able to lever the glass off the bottom edge. Carefully!)

After breakfast I finished removing the glass. It wasn't all that hard, just heat the outside of the glass with the heat gun where the butyl made contact, evenly. After a couple of minutes of heating the putty knife could be forced between the glass and the butyl, from the inside. Once enough was released the glass could be gently pried from its place and wooden wedges slipped in. Repeat until done.

Once the glass was completely loose I could pull the cable through the hardtop enough to expose the wiring. The wires that are soldered to the glass's contacts are bare, the fool that had installed it last time didn't use sufficient electrical tape to keep them from touching. (Both wires come off the lower right-hand corner of the glass and were twisted against each other.) With the short cured I was able to hook up the battery charger and measure 4.3 A of current. That may or may not be the full amount, but it's certainly enough to indicate that at least some of the glass will clear.

While the glass was out I could easily reach the surface rust that was starting in the corners beneath the molding, presumably due to water leaking in past the glass. It wasn't bad, I scraped and wiped it off, then brush-painted black over it.

I then used Xylene on a rag to clean and soften the butyl rubber a bit, then used the heat gun to warm things up. I then pressed the glass back into place, pulling the cable back into the channel in the hardtop as I went. I pressed down and used heat all around the glass, but it didn't really seem to seal all that well. We may end up using a little RTV on it later, we'll see. I then put all the trim back, inside and outside. I greased up the turn locks since they were very stiff.

Flush with success I then tackled the soft top. With the top halfway up I was able to flex the soft top mechanism sufficiently to gain access to the broken-off end of the bolt by bringing it forward past the vertical weatherstrip. That put a lot of stress on the bows, but nothing broke. I got out the Harbor Freight Pittman arm puller I'd bought earlier and cut a chunk of 3/4" square steel to span across its opening, making a miniature machine press. I then got a socket that fit over the bolt head, center-punched the broken off bolt so the point of the press's screw wouldn't slip out of place, assembled the puller, bar, and socket over the broken bolt, sluiced it with Kroil, and cranked the screw down on the center of the broken-off bolt, pushing it into the hollow of the socket. CRACK! The rust let loose and I was able to push the bolt out a little bit. I had to take it back apart and find a small nut to use as a spacer in order to push some more as the puller's bolt wouldn't fit into the bolt hole in the arm. With that in there I was able to push the broken bolt head out all the way. Hallelujah! (No way that thing would have come out any other way. I could have ground off the head and drilled out the remains, but that would have been very difficult to do without messing something up.)

I was worried that the thread end lodged in the other arm was going to be jammed, but when I was messing with it it just screwed out with my fingers. More good luck (and Kroil). I looked around for a suitable bolt to hold things together temporarily, but didn't have any luck. A proper new bolt is available, but is $10 and I'd have to order it, and wait for it. (That had been part of the plan, but so was finding a temporary bolt.) I wonder...

I used the wire brush to clean off the rusted shoulder of the head end. I got out the drill set and found that an "X" drill size was just about perfect to ream out the hole in the arm. With everything cleaned and reamed the head went smoothly back into the arm. I then ground off the threaded end of the broken bolt at a 'V' angle and welded the threaded shank back onto the head. The grinder and the cutoff saw (used as a narrow grinder) were used to put the rough profile back onto the shoulder and neck areas, then I used the M8×1 die to repair the threads. I greased the bolt and put it back. (This was working out nicely.) The soft top then operated properly and didn't hang up anymore. I tightened down the tension wire with the nut on the back of the repaired bolt. After that the soft top could be put up and down normally. The right side now was flat across the top of the window opening, no more drooping center, and the tension wire was taut so that the canvas over the window wouldn't flap. Even better!

Hey, I just saved $10! (Way more than that, since nobody else would probably attempt to repair the bow mechanism.)

In celebration my son and I went out for ice cream. In the convertible. While it had been beautiful all day it was getting cold and windy, and there were threatening rain clouds. We went anyway. It sprinkled a bit, but we had the windows up and I had borrowed Jill's rear windscreen, so it was actually quite nice. When we got back I put the hardtop back on. I did some touch-up painting on the chips along the leading edge and re-adjusted the rear turn locks that I had removed. Cleaned, greased, and adjusted, they worked a lot better than before.

Monday, September 24, 2007

As suspected it doesn't all work, less than half seems to be clearing. However, if I can't have it all what does work is not badly distributed; there are big 3" and 4" swaths right down the middle that work, and that's where you most need it. Of the rest, most of it is on the passenger's side which is where you need it more than on the driver's side.

The steering wheel squeaks as you turn it, there's a lip on the back that sticks out and rubs on the face of the dash. It's very irritating, and has worn a big stripe into the dash material. (Not a problem, as it's hidden behind the wheel.) I took off the steering wheel and used the grinder to shave down the plastic where it protruded. Plastic dust got everywhere, but it cleaned up all right. I then put the wheel back. We'll see if that takes care of the problem.

On the way to work the squeak was entirely gone. Good.

...Band started again today so I was (finally!) out after dark, and this time I had a screwdriver. On the way home I stopped and did a final dialing in of the headlights. The rest of the way home the lights seemed nearly perfect. No hot spots, no lit trees, no wonky off-and-up-to-the-side. Just a nice even lighting field, in front of the car. I think I can finally reinstall the light bezels.

Also on the way home, after a nice long freeway drive, the 3–4 shifts were very smooth. They're often a bit rough when driving to and from work. It may be that it needs 'running out', or it may be that there's a thermal switch (usually responsible for softening cold shifts) that isn't right. Anyway, it was shifting perfectly once I got off the freeway.

The temperature regulation of the ACC is not really very good. Always overshooting and ending up too hot or too cold, it seems. I wonder if there is good airflow over the sensor? Perhaps I need to resolder the ACC brain box? I'll have to check into that.

Tuesday, September 25, 2007

BOSCH
TEMPERATURREGLER
(B) 1 147 328 025
12V
made in Germany
(M) 000 822 10 03

With the coating stripped I resoldered the board. I took out the big electrolytic, which had a suspect peeling-loose plastic cover, and tested it on the bridge. It checked out OK. I put it back together, and back into the car.

One should always check the obvious first. I'd checked the foam end of the ACC hose and it was OK, but while lying underneath the dash putting the ACC brain back it seemed to me that there was an awful lot of hose hanging down anyway. When I looked I found that the other end had fallen off of its fitting, so of course there would have been little/no airflow across the sensor. What a fool. I put it back. With any luck I didn't damage the ACC brain by operating on it unnecessarily.

...On the way to work, and again afterwards on the way home, the ACC temperature regulation was correct. I may have gotten there the hard way, but get there I did.

Wednesday, September 26, 2007

MODEL: PI 375
INPUT: 11–15 VDC
OUTPUT: 115 VAC +/– 10%
MADE IN CHINA
DATE CODE: 0213
SCHUMACHER ELECTRIC CORP.
MOUNT PROSPECT, IL 60056 USA

I need to dig up another defogger power connector, and complete the modifications to the defogger relay to allow untimed operation.

Thursday, September 27, 2007

The seat jack mechanism was filthy, full of gritty dirt and hair, but the reason it wouldn't move was that there were a couple of empty .223 shell casings jammed under it. I removed all the crap and vacuumed it out, then I disassembled the jack mechanism to clean it. I put it back together with fresh grease. I oiled the lift bar's pivots too, but I didn't remove it from the car to do a better cleaning and greasing. I've fought that heavy spring before, and no thanks!

There was a little surface rust on the sidewall where the seat's past up and down motion had scuffed the paint off. I scraped and wiped that off and put on a little black paint.

The seat tracks came off with no problems, though they were full of crud. None of the sliding feet were broken or missing, so I just scraped out the tracks and wiped things down, then re-lubricated with vintage Lubriplate. After that the tracks moved easily.

I ran out of time to address the seatback adjustment, but I think that can be done while the seat is still installed. Reinstalling the seat was awkward, getting the seat slotted onto the lift bar is always difficult for me. With it finally on I then used the Lubriplate on the sliding guides at the rear and reinstalled all the bolts.

The seat now works much better.

Friday, September 28, 2007

Sunday, September 30, 2007

Anyway, with that stitched back together I put on a little black paint for rust protection, lubed the adjusters, and reassembled the seat. A little Simple Green was used to lube the seat back panel, that really helps in putting it back. The seat works great now.

Monday, October 1, 2007

The front hardtop seal is still leaking, it drips in the rain. There are no obvious problems that can be seen.

Tuesday, October 2, 2007

The black sweatshirt plugging the passenger's defrost vent has gotten wet. That's not good, I hope it's just drips from the known leak on top. The towel on the driver's side is dry. I have hopes that inner-tube rubber shims behind the top seal will help there.

Wednesday, October 3, 2007

...No, that didn't do it. I looked closer, and it's apparent that the window does need to go up a little further, but the top rubber seal's outer lip may need to be shaved in order to not interfere with the glass. More thought required!

Thursday, October 4, 2007

Tuesday, October 9, 2007

Wednesday, October 10, 2007

None of this was relevant to my purpose there, I got the cigar lighter and a male two-pin plug to use in making this car's rear bucket auxiliary power adapter. I also got its push-pull rear window defroster switch and timer, for miscellaneous fabrication purposes.

The 126 was a Euro model that had been imported. (Manual seats, no sunroof, dual-zone manual HVAC.) It didn't have the EHA valve I was hoping to snag, it was too early for that, nor was the ignition module the same. (I should probably have a spare.) However it did have the vacuum switch for adjusting headlights, so I got that and its bezel. The headlights themselves were USDOT, and of no interest. The purple vacuum lines from the switch were capped off, and too deeply embedded in the car to nab in the time that I had.

Thursday, October 11, 2007

I filled up the car again today. Again, just shy of 16 MPG. Seems to be pretty repeatable, and about right for the model. The folding solar calculator I'd procured for the car was non-functional. I don't know what happened to it, the display was insane and wouldn't react to the keyboard. Good thing I had a backup or I'd have had to wait to get home to figure out the mileage. Horrors!

Friday, October 12, 2007

I stopped at Harbor Freight and bought their on-sale 12 V liquids pump. Toad parts for the transmission? $30. If it's not full of toady goodness I can always use it for something, someday. 50 PSI, 240 GPH. It's not rated for continuous duty, but if it works at all for the proposed purpose I suspect I'd be running it at reduced power, which will help immensely with the longevity.

Sunday, October 14, 2007

Keeping the "Tear Mender" fabric glue bottle in a sealed glass jar (peaches) was a good idea. It's not that cheap, and my last bottle dried out. When I opened the jar it reeked of ammonia, so plenty had tried to get out of the glue but didn't get far. It worked just fine, no sign of drying out. (It's an emulsified rubber glue that goes on like thick milk.)

Monday, October 15, 2007

I stopped by the upholstery shop and got some more canvas. It's a little lighter than the last stuff, but it's what they had. I forgot to pick up a snippet of black, maybe tomorrow.

Tuesday, October 16, 2007

(B) 89 71 24
10 M 81
(M) 126 540 47 45
12V
10A
Made in W-Germany

Anyway, I brushed it off, resoldered the joints, and burnished the pins. It reassembled easily, and continued to work on the bench. I put it back in the car. I don't expect anything to come of this, unfortunately.

The car started what I would call 'normally', which is to say somewhat reluctantly but not with extreme difficulty. Didn't want to idle worth beans.

...At lunch I picked up a square of black canvas. I also cut a piece of the beige canvas to cover the other big tear at the back. That fabric glue works great.

...After work, in the parking lot, I cut some more patches for the little holes. Easy-squeezy. Good thing, too, as it had started raining by the time I got home. (I wasn't worried about getting wet, but if it's wet I can't glue on patches.)

One oddity was that as I pulled out of the parking lot at work I heard a bunch of relay chatter from over by the fuse box. Could be a lot of possibilities, perhaps I should yank any non-critical relays and see if it goes away? If it was the fuel pump or OVP relay it could be part of my problem.

Wednesday, October 17, 2007

Distributor cap is dirty—soggy from condensation, or the sensor that tells the cold start valve to squirt is bad or has a bad connection?

Btw, even with ether it doesn't roar to life, it's a slow onset of firing. Once running it's a lumpy, low idle that gradually gets better as the car warms up. Once warm, it runs well.

I found a bolt missing (stripped) that holds down the air horn, but plugging the hole with a big sheet metal screw made no difference to starting or idle speed.

Spark plug wires? Often an issue when things are "dewy".

When my M117 started getting hard to start, then progressed to refusing to start—I found what I assume to be the cause to be loose intake manifold bolts (to the head) on one side (finger loose) and ALL the rubber seals, tubes, etc. in the intake system were rock hard and loose. This started the rebuild of another engine with the best parts from both that is now in the car (after 18 months of off-and-on work/play).

An easy way to check the spark plug wires is to go out on a soggy night and have SWMBO crank the car while you look under the hood in the dark. A miniature lightning display says you need new plug wires. In my experience, which is all with pre-1980 gassers, in the no start in soggy weather conditions it has been plug wires probably 95% of the time. The newer MB wires seem to last longer. coil ignition wires seemed to last a max of about 5 years. The 1980-something MB gassers I have had to deal with have had OK plug wires for the time I have had them. I think 3 years is the max so for.

Thursday, October 18, 2007

So, I pulled the coil wire, wiped it down, and put it in the shop oven to dry. While drying, I cleaned the dirt out of the coil cover and put it in the oven too. Corona dope is lacquer, right? So when the wire was warm, and presumably dry, I then dipped the wire part in a can of lacquer, let it drip off a bit, then put it back in the oven. After a few minutes it wasn't tacky anymore but was still flexible, so I put it back in the car. (Bold? Perhaps, but what am I going to do, ruin it? It's crap already.) Where it had been arcing to the valve cover I wound electrical tape around it, then dipped that part again. After a bit of drying, I took it back to the car, cut a length of bicycle inner tube (mountain-bike size so as to fit over the cap end) and slipped the wire through it. (Belt and suspenders!) The two worst offenders from last night, #5 and #7, also got the lacquer treatment. The bundle of spark wires (#'s 5–8) were tubed as well. (That tube had to be slit open and tie-wrapped back together, so I used the smaller size tube that I have more of.) The other four spark wires were routed much better, but I still tubed them too, through the point where they went through a plastic-lined metal bracket on the way to the plugs. I then cut some more rubber and draped it over the idle solenoid, as one of the wires goes close to that. I put the valve cover back on, and cut some rubber to go between it and any spark wiring that wasn't already well-protected. This entire procedure added at least one extra layer of insulation anyplace the wires were touching metal, and as many as three.

For grins I pulled plugs 1–4, and they didn't look too bad. They could use a good cleaning, but I didn't have time this morning. The plug sockets were all Beru, as were the cap ends that I checked.

The car started much easier than on most mornings (although it was dry out), and it idled at a more reasonable RPM. I think I'm on to something here! Now I just have to remember to actually replace the wires. (Out of sight, out of mind...) Time will tell if I've really improved the situation.

It seems reasonable to conclude that it might not have been the ether that helped with starting those two times so much as removing the air cleaner, which cut down on the shorting. (Ether sprayed in the snorkel did no good, but spraying it down past the metering plate helped.)

Friday, October 19, 2007

Saturday, October 20, 2007

I then removed the top window molding and the sunvisors. This will, I hope, expose where the water leak is. I suspect the passenger-side front corner joint.

...Later in the day it was raining again. The leak is definitely through the black butyl (sticky) rubber window glass seal all along the top, no question. I guess it's time to break out the runny RTV window sealant, I'm not going to pull the glass in order to replace the butyl. To sop up the water over the console I draped the black towel over it. That may save the switches an unnecessary shower or two.

Sunday, October 21, 2007

I then hosed out the channels to remove the glop. I'll need to wait for dry conditions before I can use the sealer. I washed off the grilles and put them in the oven to dry. They don't need to wait for drier weather, at least. When dry I painted the screens black, and sprayed the red touch-up paint upwards from below to catch the sides of the grille slits on the aluminum cover pieces that had started to lose their paint. This refreshed paint will help the appearance a lot.

When I moved the car to the lawn for washing it really had trouble staying running. It started fairly easily, but once it had died it didn't want to re-start. I think I should spring for the 'tune-up' kit, rotor through plugs. I left it idling to recharge while I washed off the muck.

...After church things had dried enough that I was able to wipe down and rattle-can some black on the air intakes underneath where the screens sit. I then wiped down the top of the windshield glass area and blew it off/out with compressed air. (There was a definite breeze through the seal about where the rear view mirror is located.) I put what I had of the runny windshield sealer RTV all along the top edge of the glass to spill over and run down the failing seal, puddling against the metal, and I wiped a small bead on from below as well. I didn't have quite enough RTV to do what I wanted (saturation overkill), but I think I got it anyway. We'll see. I had a little problem with dripping on the underside, but I believe I got it all wiped up in time.

Monday, October 22, 2007

Tuesday, October 23, 2007

It promised to be a beautiful day, so I put the top down. It was still pretty chilly, and as I was pushing the top down into place I heard a surprisingly loud 'POP'. Yep, one of the rear quarter windows split completely across. Crap.

After work I stopped by the upholstery shop. They sold me a scrap of vinyl side-curtain material for $5 (the material's priced at about $18 a yard), it should be enough to do two or three of those windows. My current idea is to cut out the old material leaving about a 1/4" rim, and glue a new piece to that. Not beautiful, but it might not be too bad, either. I guess we'll see. Or not. I do know that a tape job really looks bad, and doesn't hold up all that long.

Thursday, October 25, 2007

Still undecided as to the glue to try for a window patch. A fellow lister suggested using magnets to clamp the vinyl together. Not a bad idea!

Friday, October 26, 2007

We got boxes of parts delivered today. Looks like the ignition parts, and some of the parts for the 190D. I got the vacuum pod too, but not the dash cover. The sunvisors came, they're nice looking, and soft. I wonder how they'll hold up?

Saturday, October 27, 2007

I found a too-short bolt holding the air horn down, it stripped out the one thread it had 'bitten', so I found a longer screw to replace it with. (I was checking to see that it wasn't loose.)

Though cold, the car started with less than a second of cranking. Serious improvement! During a few errands today it started a bit more normally, that is, after a few seconds of cranking, so it's still cold-blooded. We'll see over the next few days if its extreme starting difficulty has been cured or not.

Sunday, October 28, 2007

Tuesday, October 30, 2007

The U-Pull has a 116 there that still has its front bumper. The front bumper end pieces, those black rubber wraparound things, look to be the same on the 116 and 107, and the ones in the yard are a lot nicer than this car's. I wonder if I should go get them?

Wednesday, October 31, 2007

I got some of the dash impedimenta pulled off, the A-pillar covers, the support struts, and two of the three retaining bolts are off. I checked the big box that came yesterday, and yes, it is the dash cap.

Since I was there, I pulled apart the old corroded flasher switch and brushed off all the internal and external corrosion. I installed it in place of the real switch so that I wouldn't have to disrupt the auxiliary clacker wires, and it works fine. A spare!

I put the window switches and the ACC panel back in place so that the car can be driven.

All this took about an hour. Somehow it seemed longer.

I next need to remove the steering wheel and instrument cluster, the remaining bolt, and then disconnect the heating ducts and cables. Then the dashboard should be able to be removed. I believe the car can continue to be driven even with it out, though it will not look so good. It doesn't look so good now!

Regarding the kickdown switch, I now believe that it does work, but it doesn't respond instantly, nor at all at higher RPM's when trying to force a 2–1 downshift. I'm guessing it's supposed to be this way.

Thursday, November 1, 2007

With the dash out I then rigged the car for running again. The instrument cluster is resting on the towel that formerly plugged the defroster vents. (The seasoned traveler should never go anywhere without his towel!) Ugly as sin, but it ought to work just fine.

Ah yes, the defroster vents. The ring on the flap that the actuator hooks to is broken off. The flap pivots easily enough, I was able to close it by hand so I won't need that towel anymore. I will have to fabricate a new attachment for the actuator, no doubt using sheet metal and Shoe Goo.

All this took about another hour. Somehow it seemed longer. Have I said that lately? :-)

...Somehow during the dash removal the grounding for the instrument cluster has gone bad. There are some very weird interactions of the lights and gauges, especially with the turn signals. Normally I might suspect the cluster itself, but using the turn signals causes the seat belt light timer to cycle again. That indicates that wherever it's picking up ground is probably the problem spot. I'll look into that, the major ground clump on the car frame above the fuse box (G102) is the most likely suspect.

There is a loud rattle on bumps, which turns out to be the bottom center of the windshield glass slapping against its dead butyl seal. I'll need to get some more of the RTV sealer to put along there. With the dash out it can now be accessed. (It's also a heck of a lot louder out in the open, and I think the dash itself was restricting the glass's motion before.)

Friday, November 2, 2007

Some of the rag padding that's glued to the firewall had come loose, so I got out the contact cement and glued it back up. That neatened the area somewhat.

...At lunch I picked up some more runny RTV. Tomorrow?

After work I filled it up, and got my worst-ever mileage: 14.4 MPG, I wonder why it's so low?

Regarding the O₂-Sensor warning light, the little light in my head finally lit. This car drives that light based on mileage, and not feedback from the engine computer. That explains the weird connection to the speedometer, and why the light was stuck on, and why there was no bulb in the socket. I'd heard about this sort of thing once upon a time, but it never occurred to me that that was the situation here. The official servicing for this warning is to replace the sensor, and to remove the light bulb. I found a weird Hall-effect (?) sensor at the end of the odometer wheel stick. That's probably it. It can probably be moved so as to turn off the light, but what's the point? It'd just turn back on again some random number of miles from now. Far better is to measure the duty cycle of the sensor itself, or to check the average current at the feedback actuator. (Which, of course, is how all modern systems drive the light.) I'll be removing this 'problem' from the list.

Saturday, November 3, 2007

Now that I was there I decided to have a closer look at the O₂-Sensor warning light driver output that is clearly marked on the PCB. I powered up the speedometer and verified that it reacted to being fed an input, my signal generator would drive it nicely. So it's OK. The lamp output was being driven high at all times. What was very interesting was that there was a 100 mA fuse (shaped like a small radial electrolytic capacitor) plugged into the rear of the board, with a plastic access plug in the shell covering it. This fuse was blown. Don't tell me...

Yes, the fuse is part of the O₂-Sensor system. I replaced the fuse with an LED/resistor pair so that I could see when it was sinking current. When I spun the little magnet wheel on the end of the odometer stick I could make the light go on and off. I turned it so that the light was off and replaced the fuse with a short piece of wire. Then the lamp output was off. (It stayed on with the LED in place, the voltage drop was too high to turn off the light.) How sophisticated! When enough miles had clocked by the sensor caused the fuse to blow, which turns the lamp on, more-or-less permanently. One could spin the little wheel to the right place and replace the fuse to restore the system to its original state, but it seems to me it's a hell of a lot easier to just forget the light, note the mileage, and replace the sensor as needed. Better yet, only replace it when it fails. I removed the wire and put the dead fuse back before reassembling the instrument cluster.

Moving on, I cleaned the window seal off with alcohol and a clean rag, I used a butter knife to work the cloth into the gap. I then used the runny RTV on the bottom edge, and worked some up the sides too for good measure. I made sure to push up on the glass to let a good bead get down into there, and I ran my finger down the bead to spread the excess over the entire butyl seal down to the metal frame. That used up most of a new tube of the stuff.

Sunday, November 4, 2007

The heater box was cracking where the center vent pod attaches, this is very common and if left unchecked can result in the pod pulling loose from the box and thus failure of the center vent to operate. I cut a piece of sheet metal to go over the weak area, then I scratched it up, cleaned it (and the heater box) with acetone, then Shoe-Gooed it into place over the cracks. With any luck this will grip strong enough to stop further damage.

Monday, November 5, 2007

It was icy out, I put the little space heater in the car to thaw the glass the easy way.

...Out shopping at lunch I found an interesting piece of craft foam sheet, thin stuff like some kids' toys (masks, etc.) are made of, about as rigid as manilla folder material. It might be a suitable replacement for the foam seals of the defrost vent flap that are decomposing.

I also got some speed nuts at the hardware store and installed them in place of the duct tape. Now that it's all hooked up it's kind of fun to watch the vent flaps do their little dance as you drive.

...On the way home after band I kept hearing an intermittent clicking from behind the left side of the dash. I've heard it before, but this time it was a lot easier to track as the dash is out. There's not much there on the left side, not when compared to the right. Turns out that it was the keyless entry unit. It kept clicking after I shut off the car. Alarm function? That wasn't even hooked up, so it must have gone insane. Now it's unplugged!

Tuesday, November 6, 2007

Friday, November 9, 2007

Tuesday, November 13, 2007

Monday, November 26, 2007

Tuesday, November 27, 2007

Monday, December 3, 2007

Tuesday, December 11, 2007

Crap. This sort of behavior is getting very old.

Wednesday, December 12, 2007

Saturday, December 15, 2007

We're having a party today, the 20^th (or so) annual Wicked Wodka. I moved the car off the pad to make room for guests. It started instantly, in less than half a second of cranking. It didn't want to run much thereafter, so it's got a choke (equivalent) problem. Warm-up regulator?

Sunday, December 16, 2007

Tuesday, March 18, 2008

The car drove to and from work with no problems. A lot to do, though!

Wednesday, March 19, 2008

Thursday, March 27, 2008

Friday, March 28, 2008

Tonight, after work, the snow started again. Getting old!

Saturday, March 29, 2008

Sunday, March 30, 2008

Monday, March 31, 2008

Saturday, April 5, 2008

Thursday, April 10, 2008

Saturday, April 12, 2008

It was so nice that I took the lid off of Jill's 560 SL. I could use the hoist for hers, but I also wanted to go topless today. She, however, was getting ready for a concert and couldn't help me. That's OK. I got the top off by myself, it's not that hard if you crouch inside under the center and then stand up. Then you just walk very carefully out of the car while balancing it on your head and neck. Putting the lid down on the ground without dinging it is much harder than getting it off the car (it's so short), but I was able to do it. I don't think one would be able to put it back on the car this way without hurting anything, but I don't need to do that. I don't think I'd try this on a car with nicer paint!

Sunday, April 13, 2008

Monday, April 14, 2008

In fact it rained pretty heavily today, but the top did its job. I had a lot of errands to do and filled up the car a fair amount doing them. Not much like yesterday, that's for sure. A tiny bit of water slopped onto me on a couple of corners, out of the top lock holes. The window glass was still dry along the top edge, so this is likely the main top seals at (not) work.

Tuesday, April 15, 2008

Wednesday, April 16, 2008

Sunday, April 20, 2008

Saturday, April 26, 2008

Daniel actually helped, he manned the hose. I had to leave the doors cracked afterwards to let the puddles dry. (He's pretty aggressive around the windows, which weren't quite all the way up.)

Sunday, April 27, 2008

Tonight I noticed that the brake warning light wasn't illuminating at all, 'cause it let me drive off with the parking brake on. (Not far, no damage done.)

Monday, April 28, 2008

Lying in bed, after a really nice day driving around with the top down, the weather guys mentioned that it was raining already. So I had to get up and go out to put the top up. (Hey, it is sprinkling!) Not sure how wet it would have gotten, the rain was very light.

Wednesday, April 30, 2008

Jill called mid-day to tell me that it was snowing heavily at our house, she wanted to make sure I had my top up. I looked out the window at the sunny parking lot, and marveled at the fun weather we've been having this year.

Thursday, May 1, 2008

Coming home after work it was lovely, so I put the top down. It was dry by then.

Sunday, May 4, 2008

Tuesday, May 6, 2008

Thursday, May 8, 2008

Friday, May 9, 2008

Saturday, May 10, 2008

Monday, May 12, 2008

Wednesday, May 28, 2008

Thursday, May 29, 2008

Friday, May 30, 2008

The oil drain plug came out easily, and is still in good shape. I found not one, but two copper crush washers on it. Dumb. I dug one of my OEM Knecht/Mahle filters out of storage, they come with new washers and O-rings. I cleaned the plug off and put a new crush washer on and put it back. Then I moved the pan over to the filter housing and dropped that. What a mess! The housing was pretty dirty, so I put it in the solvent tank and cleaned it out/off. (The housing was made by Knecht too.) I pulled the bolt out and replaced its sealing washer with a new one from the filter kit. I scraped out the bottom of the housing, there was dirt collected in it. I then put it all back together with the new filter and O-ring, and put in 1.5 gallons of Delo and started it. Once the pressure came up I stopped it and lowered the car, then checked the level. I added oil until it came up to level. The manual claims 8 liters, so 2 gallons is just about right.

The stinking hood pad was falling down again, so I got a rubber glove and the gallon of non-solvent contact cement I got at the liquidator's and slathered it all over the place. What a mess! That stuff is very runny and dries slowly, it wasn't sticking even by the time I was ready to go to work. I just dropped the hood carefully so's not to smear glue all over the place.

...After I dropped Daniel off at school I tried pressing the pad up again. With the engine heat it had dried enough that it stuck this time. Mostly.

At work I checked Rusty's prices, it looks like around $52 in parts, and a grinder to take out the old rivets. (The new joints affix with bolts.) The official repair is to shell out about $280 for parts because the dealer only offers new upper control arms (with ball joints included). Nah!

...Because of Jill's road trip today I also had to pick up Daniel after school, and bring him back to work with me. I was nearly late, but I wanted to drop the top as it had gotten quite nice outside and was in fact uncomfortably warm inside with the top up. In too much of a hurry, I neglected to manually push back that bow that has elastic bands (shot, on this car as on so many) to pull it back, and as a result I put too much stress on it and it started to tear loose from the aging soft top. I'm not sure just how much damage was done, I won't know that until I next put up the top. Chances are I can just glue it back again. Carefully.

...After work I pushed it up again to catch the remaining loose spots.

Saturday, May 31, 2008

Later in the evening the sky looked threatening and I heard thunder, so I put up the top. The work on the turn lock seals helped a lot. The tearing of the bow from the canvas isn't that bad, I think it can be ignored for now.

Tuesday, June 10, 2008

Sunday, June 15, 2008

Monday, June 16, 2008

...After work, since Jill and Daniel are away for the week I parked in Jill's garage spot and put away the picnic canopy. Might as well enjoy the amenities of an empty house!

Tuesday, July 1, 2008

Wednesday, July 2, 2008

I decided to check the spark plugs as it's been running rough lately. Six looked good, with mild dry white ashy deposits, but #2 was wet, and #7 was oil-fouled. I cleaned those two in the Handy-Sandy, I guess I'll have to keep an eye on them. I checked the manual, and it's rather interesting that the firing order is 1-5-4-8-6-3-7-2, which makes the two dirty plugs electrical neighbors. I checked the plug wires, and there was nothing wrong there. The plug gap is rated 0.8 mm, which makes it .031". The plug gaps were a little large, around 0.9 mm, so I brought them in a little. The plugs were all out for awhile, and after I put it back together the engine started instantly. Perhaps I've got leaky injectors that are dribbling and causing a touch of cold flooding? Something to think about.

...Yes, the car runs fine now. Will keep an eye on this. The rain shrank all the torn seat leather so that now there are large gaps at the tears, and pokey bits to jab you where edges curled up. Crap.

Thursday, July 3, 2008

I then used contact cement to reattach some of the front firewall sound insulation. I also replaced the foam tube for the ACC cabin temperature sensor: the first replacement was falling apart, and it was actually an inch too short anyway.

Someday I'd really like to put this car back together.

Thursday, July 10, 2008

After work I dug around in the piles of crap and found the partial valve stem seal kit. (I'd used one set of the four on Jill's 450 SL eight years ago.) A minor miracle! (I also found a complete good set of rear Textar brake pads, and a bunch of dead monovalve inserts.) I put the seals in the trunk of the car, with all the other parts in waiting.

Thursday, July 24, 2008

Friday, July 25, 2008

Out on errands today it was quite warm. The car was running better than usual lately, I wonder what's up with that? The plug fouling problem seems to be variable.

Wednesday, July 30, 2008

They fixed it before lunch, no charge.

Thursday, July 31, 2008

Sunday, August 17, 2008

Monday, August 25, 2008

Tuesday, August 26, 2008

I noticed that the coolant level is down substantially. Not a good sign: fouling plugs and consumption of both oil and coolant! I checked the tightness of some of the exposed head bolts and didn't notice anything wrong there, but that doesn't mean much. (I sure wish I knew what top-end work was done, and by whom.)

Thursday, August 28, 2008

So, what small thing can I do on this car that won't screw anything up? I suppose I should look at the foam flap tips on the defroster vents, since that's really the only thing besides the lack of time keeping the dash off. (OK, I also want to do the rear fog light modification at this point, and it is easier to glue the dash cap on when the dash is out. But still...)

I refilled the reservoir with some of the 560's used green coolant, at least the color matches! That'll do until I service the system properly. I also tightened the two exhaust manifold bolts on #4, they had worked completely loose. I put the battery charger on, it was still thirsty.

But is the foam flap the most pressing and worrisome thing? No! Instead I removed plug #7, again, to find it, again, very oily. I cleaned it, and checked #'s 6 & 8: they were fine. I checked #2 and it was a bit wet, just as before. I then removed the valve cover, which required removing the fuel lines in the area and the brake booster vacuum line. I unplugged the ignition module and the fuel pump relay, then cranked the engine over 'til #7 was roughly near TDC. I removed the cam oiler tube to make room for the spring compressor. I compressed the springs a bit and removed the cam followers, that was pretty easy. I tried to remove the valve springs, but the KD spring compresser (that I'd bought for using on Jill's now-sold 450 SL) was itself too springy and the forks kept deforming under the heavy stress required. (I had this problem the other time I used it.)

KDT-3087 K-D TOOLS VALVE SPRING COMPRESSOR—$56.06 Tool compresses valve springs on most overhead valve engines including GM, Ford, Chrysler, Nissan and Toyota. Jaws adjust to fit over spring retaining washer. An adjustable handle allows easy clearance in tight areas. GM adapter included. use no. 3269 Fulcrum Rail if overhead cam is not available.

The stem seals are not visibly damaged, nor are the valve guides sloppy or loose. I'll replace these seals anyway, but I suspect that there's either a head gasket or ring problem causing the fouling. The coolant loss suggests head gasket. Damn.

I was out of time this morning, but I don't need to drive the car until tomorrow. I'm sure I can get it back together before I need it.

Friday, August 29, 2008

Once the springs were back in the rest of the engine reassembly went quickly. It started instantly and didn't leak gas, so that's good. I drove it to work (unfortunately rather late), it ran fine. Especially so since I'd cleaned off the spark plug again.

Head gasket? Rings? I guess something's wrong.

Saturday, August 30, 2008

I'm told that it still could have been the seals, but that if they weren't leaking before they will be now, and that I shouldn't deduce anything until I actually replace the suspect seals properly. Well, doing so was on my list...

Friday, September 5, 2008

Thursday, September 11, 2008

Thursday, September 18, 2008

Sunday, September 21, 2008

...Jill went off to rehearse for the Opera tonight, so I took advantage of her newly-freed hardtop hoist and the temporarily open bay. I he-manned the top into the bay and set it down under the lift, then lifted it. I used compressed air to blow out all the grit that had settled over the summer on the interior horizontal surfaces. I then moved the car inside out of the rain; its top is thoroughly wet, much too wet to furl so I got out the big fan and pointed it at the top. Maybe a few hours will suffice to dry it enough to put away, then I can drop the top onto it and get back out of Jill's way. We'll see. In the worst case I leave the top hanging there until I can make the switch.

...Before bedtime it was dry enough to put on the hard top. Done. There was some rain water in the storage well, but I mopped that up first.

Monday, September 22, 2008

Tuesday, September 23, 2008

89 5065
12 Volt 10 81
001 545 52 32
Made in W-Germany

I also retightened the ground cluster over by the fuse box. That is also part of the warning circuitry. (Ground is always part of everything, and a loose ground can trigger timers, etc.)

OK, this is weird. I've wondered why opening the right-hand door causes the warning buzzer to sound on this car, but not on our later SL's nor on our now-sold 450 SL, and I can see from the car's schematic (page 117) that although they wired the two door switches separately to the warning module they also tied the two switches together, negating the diode that's in the warning module to isolate them! Later SL's have them separated. I wonder...

I pulled the two door switches, and the wiring found there matched the schematic, as did the wiring to the courtesy lights themselves. (It's always good to check.) It looked like one of the two brown/blue wires that comes off each switch is just a tie between the two switches, one that isn't necessary and that was eliminated for the 1983 model. (It's probably a transitional step in the design between a single door circuit and the two circuits of later years.) Gambling, I cut one loose from the driver's switch and beat the odds! It was the inter-switch tie and after that the warning module only buzzed when the driver's door was opened, yet the courtesy lights still functioned as they should. I taped up the loose wire and put the car back together.

...The drive to and from work proved that the flakey light behavior was cured.

Wednesday, September 24, 2008

...After Daniel went to bed I pulled the car into Jill's empty bay (lights!) and loosened the hardtop. I followed the directions in 68-570, minimal though they were, and got the outer trim removed. (For the record first you peel out the weatherstrip, then remove the nine (9) screws, some of which hide under a flat strip. The trim face then pulls outwards from clips in the rear of the hardtop, slowly as the butyl seals let loose.) With the trim removed the glass was sitting loosely in its frame. All the butyl (?) rubber had come loose from the glass, there was no need for further disassembly. I cleaned the glass and scraped the surface dirt off the butyl, then ran a thin bead of runny windshield RTV (leftover from the windhield sealing job) around the butyl, then laid the glass back on. I clamped it against this new bead as best I could, using spring clamps and weighted boards leaned against the glass. I left it alone to set up.

Thursday, September 25, 2008

...The rattling is gone and the wind noise is much reduced. Sehr gut.

Monday, September 29, 2008

Tuesday, September 30, 2008

Wednesday, October 1, 2008

Thursday, October 2, 2008

Friday, October 3, 2008

Oops. Now that the steering knuckle is loose, however, it becomes much harder to remove bolts! The brake hose was in the way of using a pusher on the ball joint's post, and the steering arm had to be removed to get the hose brace loose. I got the steering arm off and I think I can get the pusher in there without removing the brake caliper, but next time I'll just take off the caliper and the steering arm first.

I can't find the 'pusher' (Pittman arm remover) that I modified by narrowing the jaws using the torch, but I do have the unmodified one that I used on the soft top bow. Perhaps I can get that one to do the job, though its jaws are too wide to grip the top of the steering knuckle.

I notice that the fucking dog has at some time done a cat-chasing dance all over the hood of the car, the paint is all gouged up. (Will that dog survive much longer?)

Saturday, October 4, 2008

The new joint, TRW, looked good. It has a 5 mm Allen socket in the end for easing installation, unlike the original. I put it in the steering knuckle first, then bolted the UCA to it. Very clever, except that I'd trapped the brake caliper on the wrong side! Crap. I had to remove the joint from the steering knuckle and reinstall the brake caliper. Fortunately the socket hadn't fully gripped the ball joint's post yet, and some mild tapping with the hammer released it.

While I was there I found that the insulation on the brake wear sensor cable was decomposing, so I cleaned it off with a wet rag. Hungry, and waiting for the cable to dry, I broke for breakfast.

...Back on the job I taped up the wire using about half a roll of electrical tape, once down and then back up. I did everything exposed in the wheel well. I then reinstalled the ball joint, and then the steering arm. The hub was then nice and solid, so I put on the wheel, using wax on the lug bolts. Even with the greater leverage the wheel felt nice and solid, not like before. (No more 'chunka-chunka'.) I let the car down, I'm ready to do the other side now. It should go much faster, this side took about four hours total.

Sunday, October 5, 2008

I had found that the cable to the brake wear sensor was ripped loose at the sensor block, and several inches were missing. While the paint was drying I cut into the tiny 1/4" wire stub that remained on the sensor block and got the bare wires exposed. I scraped them with a razor blade, then soldered new 18-gauge wire onto the nubs. I used heat-shrink tubing to protect the joints, then potted the end with a bit of Shoe Goo. A large piece of heat-shrink then went over the entire thing, bedding into the blob of glue. I heated the whole thing with the heat gun and tightened the outer sheath over the glue. I then smeared the expressed glue around the end of the joint and the block itself, hopefully providing some mechanical strength as well as waterproofing. This was set aside to dry for the moment.

With the paint now dried enough I installed the new ball joint in the UCA, then put on another coat of paint to cover the scuffs. I put the brake caliper back on loosely, the ridge-ectomy helped considerably, then installed the ball joint into the steering knuckle. It tightened down easily, I didn't need the Allen wrench for the center. There is a chunk out of one of the brake pads, but I don't think it's particularly significant. I then tightened down the brake caliper.

During all this it became apparent that the boot for the lower ball joint had a small crack in it, but was otherwise intact and in good shape. With the steering knuckle mounted again I wiped off the expressed grease (clean looking) and used brake cleaner to clean off the boot. A bit of Shoe Goo served to pot the small tear, I think it had been torn during this job by the weight of the half-supported steering knuckle pinching the boot at that point.

To give the glue a bit of time to set up I came in and updated the log up to this point.

I then went back out and finished the job in about 45 minutes, much of that was spent repairing the brake pad sensor cable. The insulation on what was left was decomposing, so I wiped it down like I did the other side, then trial-fit the wire to get the length of the new piece right and cut off the excess. I soldered it together and used heat-shrink to cover the solder joints while the cable dried. I then wrapped electrical tape from the top down and then back up (just like the other side), giving it a nice solid covering, and installed it. I then put the wheel back on, using wax on the lug bolts, lowered the car, and drove off in it taking Daniel to meet Jill and some friends for lunch. Just about two and a half hours were spent this morning on this job.

If nothing else was wrong, and having done them before, I bet that you could get down to about an hour a side for this job. Not that bad!

Monday, October 6, 2008

Wednesday, October 8, 2008

Thursday, October 9, 2008

Friday, October 10, 2008

I have the Tex seat covers on sale too!...$415.20 plus shipping for a set of 2 MB Tex seat covers. Head rest covers $63.00 per set (2). Armrest cover $42.60.

Quite cold outside this morning, and the car hasn't been driven in several days. It started instantly. Seems to like sitting for awhile, same-day restarts can be a bit difficult.

Thursday, October 16, 2008

Rusty sez:

It is a great price. GAHH is dying right now, no one is buying anything, so they are giving it away. These prices are 25% off the regular prices.

Tuesday, October 21, 2008

Thursday, October 23, 2008

Tuesday, November 4, 2008

Wednesday, November 5, 2008

...Definitely colder, tonight after being open most of the day the trunk isn't yet dry.

Thursday, November 6, 2008

Friday, November 7, 2008

Wednesday, November 19, 2008

Thursday, November 20, 2008

Saturday, November 22, 2008

As it was idling rather lumpily I started pulling spark plug wires. They all made it worse except #6 and #7! Pulling either of them had no significant effect on idle. I double-checked the wiring order to the distributor cap and it was correct according to the firing order stamped on the valve cover. Well, there are only a limited number of other things that it could be. I decided to think about it for a bit, and came in for breakfast. The list of potential culprits is rather short:

• no spark
• no fuel
• no compression

When I got back to it I checked the spark on #7 by putting in one of the old spark plugs and laying the newer one on the valve cover, connected. Started, the engine idled as usual with a nice bright spark. That's not it. I then pulled half the plugs, the ignition module's input, and the fuel pump relay and tested the compression, six strokes per reading. I repeated it twice:

Plug	Bar	(PSI)
#5	10	(150)
#6	8	(120)
#7	9.5	(140)
#8	8	(120)

No joy, but #6 and #8 are kind of low. Still, it's #7 that's been trouble, so what does that mean? I don't know. Low compression on this side, but very oily rings in #7 boost it? I put all the plugs back. I looked at getting the injectors out, and they're really buried on this side. It's not going to be easy to get them out. Maybe later. Some suggestions from the mailing list:

Are the plugs in #6 and #7 normal (that is, do they have evidence of combustion)? If not, and you have gasoline fouling, it's time to replace the boot between the air horn of the fuel distributor and the throttle body, the throttle body to intake seal, and the seals between the two halves of the intake manifold (especially the latter). A leak there will give you idle fits while working pretty much as normal at speed—the mixture on those cylinders will be too lean with a leak there.

Would the manifold seals fail like that, so quickly? I'd expect more trouble from the rubber boot, I'll try to have a look at that... This stuff hard to get to?

Except for the boot between the air meter horn and the throttle body, yes. Naturally, you have to remove the fuel distributor to replace the boot.

You have to pull the manifold to get to the throttle body seal and the manifold seals. Sadly, there is no reason to think that whoever pulled the heads took the manifold apart to replace the seals (it's in two parts, bolted together, with a rubber ring between the halves at all 8 runners).

Also check for a leaking brake booster line or bad booster or other serious vacuum leak on that line—IIRC, it screws into the rear of the manifold, and if there is a major leak there, it will cause the back two cylinders to run very lean.

My 450SLC went from running very nicely to idling very rough, then to won't start in about two weeks. The only things I could find wrong with the engine were the dried out and leaking rubber bits in the intake manifold.

Monday, November 24, 2008

Upon reflection, I think I'll put the car together and run it for a few days in order to see if the oil fouling is improved. There's little advantage to tearing into the intake manifold right now, there's almost nothing 'saved' by doing so except removal of the air cleaner. I have other, more pressing projects.

Saturday, December 13, 2008

Given the lack of real utility of the smog pump system (it only runs for maybe 30 seconds when the car is first started) I just left its belt off. That'll make the car quieter at startup. I put the belt in the trunk, it'd be fairly easy to put back (it is the frontmost belt in the stack). The air conditioner belt is also starting to fray, so I pulled and cut off the loose bits. It could use replacement, but it's not critical nor do I have a replacement on hand.

The cruise control cable's rubber sheath was decomposing at the actuator end, lots was gone. The wires under it were still in good shape, but they wouldn't last long without more protection than that. I used window cleaner to clean off the cable, and once it dried I wrapped the full length with electrical tape. Two thicknesses where there was still rubber, four at the one end where it had all fallen off. Not great, but better than it was.

The washer tank was leaky at the pump grommet. It needs a new one, but I didn't have one on hand. I cut a square of bicycle inner tube rubber, nipped a too-small hole in the middle, and then jammed the pump into it. It slipped into the hole with some force, finally, but I don't know if the hole opened up enough or not. Doesn't really matter, right now. I'll get a proper new grommet at some time.

I then reassembled the car. I didn't even forget to plug the ignition module and fuel pump relays back in. It started easily. I checked manifold area with brake cleaner, no real signs of leakage, so I put the air cleaner back on and shut the hood. I moved it outside into the snow so Jill can park the Chicken Wagon, her winter wheels, inside. Maybe that'll shut her up? I don't know when I'll get a few days of weather suitable for test-drives to see if the oil fouling is cured. Maybe in the Spring!

I forgot to re-tighten the smog pump mounting hardware; I must not forget to tighten it later before driving so that the hardware doesn't work loose and get lost.

Wednesday, March 4, 2009

Tuesday, April 7, 2009

Today Jill and I had lunch, and did a bit of shopping. The car behaved itself, but Jill was not pleased with its state of discombobulation. After work I romped on it on the way home a few times and blew out a bunch of smoke. Once cleaned out it seemed to stay that way. It's not running materially worse than it did this morning, the next several days should prove out whether or not the valve stem seal replacement did any lasting good.

Wednesday, April 8, 2009

Thursday, April 9, 2009

Tuesday, April 14, 2009

...I brought some more paper boxes home. I was able to carry three (perhaps I could have managed five), as contrasted to the twenty-seven I was able to put in the 560 SEL!

Tuesday, April 21, 2009

Wednesday, April 22, 2009

Thursday, April 23, 2009

...It snowed again today (but it didn't stick). This is something of a drag. The car does seem to be running and shifting better than it did before, so I think I'm on the right track.

Friday, April 24, 2009

There was a heavy frost last night, it was cold, Jill said 26 °F. I had to put the space heater in the car to thaw the glass.

...Sure does run better on all 8 cylinders!

Thursday, April 30, 2009

Monday, May 4, 2009

Friday, May 15, 2009

Saturday, May 16, 2009

Friday, May 22, 2009

Saturday, May 23, 2009

While working in the area I managed to snap off the vacuum control line (purple) from the charcoal cannister vacuum valve. Again, just like on August 29, 2007.

Wednesday, June 3, 2009

I then unplugged the ignition module (having learned my lesson before) got out a small tack hammer, and turned on the ignition. I tapped on the warning module, they key lock, and the dash ground cluster. (Having the dashboard out has some advantages!) Nothing made the seat belt warning light cycle back on, which it has been doing on bumps for some time now. Time to hit the schematic! ...Near as I can tell, however, the only thing besides parts inside the warning module itself (which has been resoldered) that could retrigger the timer is the power (via Fuse F7), or the ground cluster G102. (Given the original symptoms [where the turn signal also caused the light to cycle] it's most likely G102 at fault.) Near as I can tell, if there is a problem with the ignition switch (or other upstream power) that causes a short interruption in power to F7, then this will also disrupt power to F10 as well, which should cause the rear-window defogger circuit to drop out. So, since the heated glass isn't even there these days I can just turn 'on' the defogger and see if it drops out at the same time the warning light comes back on. (It has an indicator light in the switch that'll be easy to monitor.) If it does then I know the problem is upstream. If not, then it's F7 or the aforementioned bits that have to be the problem. (I can't really definitively deduce quite so much, since I don't know the relative sensitivities to power loss of the two timers, but you've got to make some assumptions in order to get anywhere.) I'll try that test next, after checking F7 itself.

Failing that, I have an old logic probe (Global Specialties' LP-2) which I can use to trap glitches on the warning module's power or ground lines. It'll take some wiring up, however, so I don't really want to do that unless I have to.

...The test drive today showed that the cycling seatbelt light is now 'cured'. Perhaps I banged too hard with the hammer, hard enough to jar whatever was wrong into a better state rather than just provoking its misbehavior as I'd intended? If so such a 'cure' is, no doubt, temporary, but I'm not going to wait around for it to come back—I'll just deal with it later if/when it comes back.

I did pick up an auto power adapter at Goodwill for feeding +5 V to the logic probe should I need to pursue that later. (It'll need some work to get it connected to the probe.) I threw it on the project pile for safekeeping. :-)

Thursday, June 4, 2009

I then decided to take another stab at G102 so I removed the two bolts and pulled out all seven ring terminals. I used the Dremel wire brush to clean up most of them, but one was pretty gunky with some sort of varnish. I used a penknife to scrape it off, and the bulkhead where the Dremel just smeared the varnish (?) into opacity. That gunk can't have been good. The ring terminal solder joints themselves all looked good. Once cleaned I put it all back together. (While this is possible to do with the dash out it sure was a lot easier to do it this way. Better light, too.)

...More test driving over bumps, and no sign of the seatbelt light coming on spuriously anymore. I'm going to call it fixed.

Friday, June 5, 2009

Saturday, June 6, 2009

Sunday, June 7, 2009

Tuesday, June 9, 2009

Wednesday, June 10, 2009

Thursday, June 11, 2009

The steering wheel had been making some flapping noises when turning, so I pulled the combination switch back out and re-routed the horn wires away from the center column. That should do it. [It did.]

I swapped spark plugs again. It's been running fairly well, but #7 is still oily. [It runs amazingly better with that one plug working right. Quieter, smoother, and much more powerful and responsive.]

Friday, June 12, 2009

On the way to school (for his last day of first grade) Daniel requested Loathing and Popular. No problem, happy to comply!

...At lunch I went downtown (to 'Felony Flats') to see Darrell, a friend of a friend of a friend that paints in his garage when he feels like it. He said to call him 2 weeks from today, he could patch up the paint for less than $300. (I also offered a case of beer as a gratuity.) He thought it would take a week or less, and I need to get the official paint code for him so that he can get (cross-referenced) matching paint. If this works out it should be an excellent deal. Most prospective customers just wouldn't be able to get past the patchy paint even if the car was otherwise pleasing. If they then wanted it to look really nice, paint it properly later.

Monday, June 15, 2009

The right-side speaker grille had come loose at the top, so I removed it and put it back on. One of the two clips is missing, but that doesn't really show. I used my usual thin butter knife trick to help slip the clip on top into place, then screwed it down on the bottom.

I then cleaned up a bit and glued the metal insert back into the broken parking brake release knob. That isn't turning out so well, but it'll probably do.

Tuesday, June 16, 2009

      70 1715205 71
 107045 12 08148  2
      172G 172G M -
531 587 592 639 802
    440 467 494 519

531:	automatic antenna with radio
587:	581 & 583
581:	automatic climate control
583:	electric window lifts for front doors
592:	heat-insulating glass with windshield band filter and rear window pane laminated glass
639:	634 & 636
634:	elimination of first aid box
636:	elimination of warning triangle
802:	change of modelyear last figure [2] shows new modelyear
440:	bumper guards OR tempomat/cruise control
467:
494:	California version
519:

This sort of checks out, but there are missing codes and this car has the first-aid box mount on the rear deck. The disturbing thing is that the 'paint code' also matches options the car has. But if all these are options, where are the paint and leather color codes?

Running the codes on the Russian site gives:

531:	automatic antenna (from 01.01.1963)
587:	automatic climate control and electric window lifters (front doors) (from 01.01.1977)
592:	heat-insulating glass, all-around, heated rear window pane (laminated glass), band filter
639:	elimination of first-aid box and warning triangle (from 01.08.1977)
802:	Change of year of model, last figure [2] shows new model year
440:	tempomat (cruise control) (from 01.09.1975)
467:	central locking system and instruments with english lettering (from 01.04.1970 up to 31.10.1993)
494:	California version
519:	Becker radio Grand Prix electronic cassette - USA (from 01.08.1980 up to 30.11.1990)

Those also match the car pretty well. So where are the color codes? 172 as a paint code is anthracite grey, even if the G could stand for Glasurit. No part of the car is/was ever grey that I can tell. As an interior color it's blue MB-Tex. Also not a match.

Another paint-code site offers up this car as a sample of 587 pajettrot, the only halfway-close red metallic paint in the list. Looks pretty close, but supposedly wasn't available until 1984.

Checking Glasurit Color Online however, shows that Pajettrot (their code #3587) was available from 1981–1995. They say they only ever offered 9 metallic reds to Mercedes, but all of them were available in 1981. Pajettrot was available in their Line 90 and Line 55 paints. (Not in 22 or 68, whatever that means.)

Black leather has a code of 201 (101 for MB-Tex), but I don't see one of those codes anywhere either.

Too bad the MB database isn't accessible by the chassis number, which I do have!

The mailing list guys all say it's 172G (Glasurit #172, Anthracite Grey Metallic), but that's not what the car actually is. (I'd thought 172G was the paint code too, originally.) A repaint then? If so it's very thorough, I've seen no sign of gray, and I've been all over this car. The bad overpainting in red I'd always though was cheap repairs on top of the original paint, but maybe it's cheap repairs on top of an expensive color-changing repaint?

If this isn't the factory original paint, then all bets are off as to what it actually is, so no code-matched paint is likely trustworthy. Probably gotta do it the hard way.

Anyway, disgusted with all this frustration I went out and vacuumed out the car and put back the carpets. I used the heat gun to soften and bend back the driver's footwell firewall coating that had peeled loose and flopped down over itself. It cracked into pieces during the process, but seems to be in position now. I may need to glue it down later if it doesn't stay put. I also located the under-dash panels and the extra soft top seals I had in reserve. I put a good rubber floor mat down too, but that won't go with the car, especially since Costco no longer sells them.

...In the evening I took a piece of pop can and glued it around the torn-open snap-clip hole of the passenger-side cardboard (forward) under-dash kick panel and clamped it into place. (Shoe Goo, of course.) I also dug up three of the metal button snaps that hold the cardboard to the main plastic panel. (I know I need two on the passenger side, I'm just not sure if there's any cardboard that needs it on this car on the driver's side. If there isn't, the extra can go back in the spares bin. It probably should have a cardboard piece, but it's missing.)

Wednesday, June 17, 2009

I mucked out the trunk (billions of old receipts, and some U-Pull booty), and uncovered the old sunvisor clips I'd saved for this car. I'll be able to do that next.

I took a closer look at the 'primer' exposed where the paint is failing. Hmm, this is the first time I've ever seen metal-flake primer! OK, the car is just a crappy repaint. (Decent paint, poor preparation.) The original color is much lighter than I'd thought it would be, hence my thinking it was just primer all along.

...After work I pulled out the shifter surround again and tightened up the wire lassos for the auxiliary clicker to make it work again. (It had stopped recently.) I dug up another screw for the surround to replace the missing one.

Thursday, June 18, 2009

I then installed the driver's-side footwell panel, and I also had to locate replacement screws for it. Like yesterday, I washed the very dirty courtesy light to remove a lot of dimming grime. This only took 15 minutes, much better than the other side! I then removed the steering wheel and instrument cluster to reattach the key-in buzzer connector that came dislodged during replacement of the dashboard. This, however, didn't provide me sufficient access so I removed the steering lock from the car (Job 46-640), which allowed me to see what I was doing. (The switch connections had gotten bent somehow so I was screwed from the start and definitely needed to take this out to fix it.) While I had the steering lock out I lubed it, but that seems to be a mistake since it no longer defers locking until the key is removed, it now locks immediately once the key is turned fully off; I think there's a broken spring inside. (And the buzzer doesn't sound if the door is open with the key off, which is not normal unless you've removed the key once to release the lock.) Lubing it probably loosened things up enough that something fell out of place. Oh well. It can always be fixed with a new locking mechanism, but those aren't cheap and you'd need a key in order to get one intact from the U-Pull, which is unusal to find there. I brushed and painted the ignition switch collar so it looks better. I took apart the instrument cluster and cured the ringing buzz it had on bumps that was caused by a loose 7mm nut at the top. I taped over the melting face of its black styrofoam plug for the unused gear selection window and painted it black, blew the dust off the dials with compressed air, and then I put it all back together with a new bulb for the missing brake pad lamp.

This pretty much completes the reassembly of the car. (I left out the steering wheel emblem until I'm sure that I'm done pulling the instrument cluster.) All that's left now is to install the new upholstery and ragtop. And to deal with the paint. Then, I think, perhaps we put the for-sale signs on. The car's running a lot better since all the vacuum lines were reconnected and the frame ground for all the electronics was dealt with, we'll see if this has any effect on the plug fouling problem. (Bad fuel/air mixture was suggested as a possible contributing factor that could keep rings from sealing properly, and both those problems could cause bad mixture.)

...The new bulb for the brake pad lamp was perhaps a mistake: it's always lit. I'll chase the under-hood wiring first, since there are some suspicious taped-up knots.

Friday, June 19, 2009

An interesting sideline I found in the car's schematic is that the dome light switch shows an independent ground, and that it can be switched on directly or go on with the doors, yet that's not how it acts. (It acts a basic ON/OFF switch only.) I checked the surrounding model year schematics and found no differences until the three-way switch and timed relay of 1983, which this car definitely doesn't have. Supposedly the switch has a BR/GY wire and another one, unspecified but possibly also BR/GY, as well as a ground through the switch body. Well, that's not even how these cars are generally designed, everything has its own ground wire going to a well-defined ground point, which here would be the big G102 cluster behind the dash. And in reality it is not as drawn. The switch in the car has two wires: a RD/GY one that's hot, and a RD one that goes to the lamps (and thence presumably to ground). There is no continuity to the switch's mounting clips, the only potential source of the ground shown on the schematic, nor would that do much good anyway as they don't necessarily contact a good ground in the car, not to mention this is switched power and not switched ground, so a ground at the switch is worse than useless. In other words, nothing like any of the supposedly relevant schematics. In fact, it looks like the car has the wiring shown in (only) the 1978 schematic! Oh well.

I then tested the brake wear sensors and found them working, as near as I could tell. (No shorts, anyway.) I taped up the one damaged sheath and then put the connections back together, making sure the brown wires (ground) mated in the connectors. The warning lamp behaved as it should with only the key-on lamp-test function lighting it, so I reinstalled the instrument cluster.

Not much left of reassembly, really, so I got the glove box out of the trunk and reinstalled it, and hooked up its light. (That'll be nice to have again.) I sorted through what was in it and put away extra parts and threw away the trash. I left the steering wheel emblem in there, it'll be the last thing to do.

Saturday, June 20, 2009

I moved the metal reinforcement rod from the old cover to the new one. The new cover went on without too much trouble, but the hog rings were tough to deal with using regular pliers. I only put on the minimum number of rings necessary, the ones around the rod at the rear. I didn't tie the shroud (around the sides and front) to the pad with the rest of the rings. Not yet, anyway. I used an office stapler to replace the four staples on the rear sides where the cover material wraps around the frame and staples to itself through slots, and a gasket cutter set to cut the four round holes in the rear sides for the mounts.

I then moved on to the back; its cover came off pretty easily. There was a bit of glue to peel loose from the frame, I'm not sure if this was stock or not but I'm not intending to glue the new cover. The back pad was also in good shape. The new back cover slipped on easily enough, but there were no holes for the backrest. Removing the plastic grommets from the backrest holes in the old cover was tough, those things are tight. I was scared to just mark and cut, but I didn't see any real alternative.

Procrastinating, I pulled apart the old headrest, that was surprisingly easy. The new cover slipped on without too much difficulty, and after fighting it for awhile I got the slip channels closed on the bottom. Thus was finished the first of the six pieces, and after only three hours! (This is going to be slow.)

That headrest went so well I did the driver's side too. That only took 15 minutes. If only the rest of the job was so easy!

Refreshed, I tackled the back again. I slipped the cover onto the frame and eyeballed where the holes for the headrest needed to go, also referring to the old cover. Throwing caution to the winds, I grabbed a Sharpie and one of the nylon bushing clamp rings and traced ovals onto the seat cover where I thought they best went. I used the utility knife to cut out the MB-Tex, and down through the foam layer, removing both from the hole, then I flipped the cover over and pushed one of the plastic bushings into the cut and pinched it against my leg with my fingers. I could then feel its outline through the padding, and used the utility knife to cut down to the bushing and chop away the padding. With the bushing pushed through the resultant hole I pushed on the nylon clamping ring and it was done! Then I did it again for the other hole. With bushings in place I slipped the cover onto the frame again and started putting the cardboard edges into the channels. Once I had the cover fully in place I slit and cut out the round and oval holes on the sides, and installed the plastic bushings for the oval slots. I then installed the hinges, making sure I didn't shift their gears so that the seat back would end up straight. The old seat covers make nice ground cover to protect the new upholstery from getting scuffed. Installing the back panel was tricky as it's very gummy from tape, and tight because of the new upholstery sharing the channel. I lubed everything with Simple Green and wedged and hammered it (with hands) back into position, and put in the screw. Running out of time, I got the two seat sections bolted together again on the hinges and got the track adjustment lever installed, except for the springs. I threw the tracks into the solvent tank to soak and then cleaned up the work area. This is taking forever, but so far it's coming out well. More than five hours into it and I'm not even halfway done. I'm hoping the other side goes quicker now that I (sort of) know what I'm doing.

Sunday, June 21, 2009

...Nourished and refreshed, I removed the other seat. The clock's started now, as I can't drive the car without it! Turns out the seat jack has to be all the way up to remove the seat from the car, I wasted a few minutes figuring that out. With the seat out I vacuumed under it, it was filthy again. Lots of leaves. I removed the tracks, and found the wide part of the two bottom black slides pointed towards the inside of the seat, and that they were on the inside of the car, away from the seatback adjustment handle. I'm not sure if this is significant. I threw the tracks and the feet into the solvent tank.

I removed the seat bottom cover, I didn't bother removing the hog rings that held the seat cover shroud in place as the shroud was rotting so I just tore it off. Faster than removing hog rings! (Though I did have to remove the ones at the rear of the cover where the reinforcement rod is.) With the cover removed I found that there was a split down the middle of the underlying pad. This was due to stress allowed by the split cover, the pad material itself was still in pretty good shape other than this. I poured contact cement down into the crack from both sides and massaged the edges back together. I put a pice of heavy gasket paper on the underside, then I wedged the crack together with some concrete chunks and set it aside to dry. All this and the new cover should help hold it together pretty well.

While waiting for the glue I took apart the seat back. It came apart easily enough, practice helps, and I noticed that the two outside panels (face and side) had been replaced with regular automotive foam-backed vinyl. That explains a lot of the earlier seat butchery! One of the oval plastic grommets, the one in the replacement panel in fact, is missing, but I think we can do without it. (I'm certainly not going to wait for a parts order for something like this!) Total disassembly accomplished, at about one hour into the project.

I then slipped the back cover on and, like before, marked the cutouts with a Sharpie. The knife made short work of cutting, as before. (It really helps to go through the vinyl, foam, and jute web from the top before you turn it over and start chopping through the felt padding.) Seat reassembly went just as before, and at about two and one half hours in the seat back was fully assembled, including the back panel. It was time to see how the glue was setting up on the bottom pad.

The glue was fine, so I started putting the seat cover on. The slowest part was putting the hog rings back on the reinforcement rod. I also had some trouble fitting the stapler in there to staple the two flaps around the support. Then I cut the holes and put the hardware back. The solvent tank loosened the dirt and goo on the rails, so I washed them out and then put the slide feet back on in the same pattern they came off. A little lithium lube and the track was ready to go, so I put the hinge covers on. This is tricky, the alignment is crucial or you'll break your new cover. I glued back the chunk of torn-off backing foam on the rear carpet using contact cement. I then put the seat into the car, the tunnel-side forward bolt is the first one you need to put back if you want the job to go quickly. I then put on the headrest. Voila! Done, and at just about four hours, a notable improvement from yesterday. The bottom-right chrome cover on the driver's seat is running into the hump and getting pushed out of place when you drop the seat height, I'm going to have to look into that so the new cover doesn't get ruined.

I then used contact cement to glue back all the felt strips in the soft top storage bay, the glue on them was all powdery and failing. After I cleaned up the work area I called it good for the day, just as it started to rain. Nap time!

I'm going to wait to do the new soft top until we have a long stretch of good weather, because I have no idea just how long it'll take me.

Monday, June 22, 2009

Thursday, June 25, 2009

Friday, June 26, 2009

Saturday, June 27, 2009

Tuesday, June 30, 2009

Wednesday, July 1, 2009

1. Open the rear soft top latch by rotating both handles rearwards.
2. Flip both sunvisors down out of harm's way, and lower the windows.
3. Open the doors of the car and flip both seats forwards, making sure the seat belts won't trip you as you step in. Clean out the rear seating area so that you have maximum maneuvering room, both walking and sitting.
4. Put a soft board under the rear of the hardtop glass by the peg so that when you tip it up the chrome trim will rest on the board.
5. Tip up the top vertically onto the board, and back into it.
6. Crouch down and walk the top onto your back, grasping the handles to keep it from falling.
7. Stand up, balancing the top on your shoulders. You'll be leaning forwards and craning your neck. Get comfortable, and get the top level with the ground before proceeding. Make sure you're not off-balance. (If you like, you can pretend you are Atlas carrying the world.)
8. Walk into the rear of the car, being very careful not to trip or bang anything on anything else. This is paint-killing time if you're not careful. If you get in the driver's side the defroster cable will naturally drape to the inside of the car and not get pinched under the top.
9. Get centered on the car and slowly crouch down, watching to make sure the rear peg goes into its hole first. This is the critical step to avoid damage, and the most difficult.
10. With the rear peg socketed wiggle the top around as required to get the other four locks engaged.
11. Tighten the top down as per usual.

Removal is pretty much the converse of above: get all the car's other bits out of the way as best you can and loosen the latches, then while facing forwards put one knee on the back deck for better control, grab the handles, and then stand up with the top on your shoulders. Walk out of the car and slide the top down your back onto the board.

Tuesday, July 7, 2009

The steering's been groaning a bit lately so I topped off the low fluid. Leaking? It was filled only last October.

Friday, July 17, 2009

Saturday, July 18, 2009

Just a guess, but if you are satisfied there is no leak along the stem then I'd guess a head gasket next for doing flaky things, like leaking less when the ambient temp is warmer. Could be a crack in the head too, but not likely.

Cracked head? I hope not! Wouldn't be too expensive to replace via the U-Pull, but there's never one around when you want it.

Wednesday, August 5, 2009

Thursday, August 6, 2009

Friday, August 7, 2009

Friday, August 14, 2009

Sunday, August 16, 2009

Tuesday, August 18, 2009

Well, OK. I can try that.

Wednesday, August 19, 2009

Friday, August 21, 2009

Tuesday, August 25, 2009

Thursday, August 27, 2009

Friday, August 28, 2009

Saturday, August 29, 2009

Thursday, September 3, 2009

Friday, September 4, 2009

Wednesday, September 9, 2009

Sunday, September 13, 2009

The trunk is smelling fine now.

Monday, September 14, 2009

Wednesday, September 23, 2009

Tuesday, September 29, 2009

I called the dealership today and replacement code-cut keys are $25, versus $29 through Rusty. So I'll be supporting the local dealer on this one! (They moved, I need to find their new location anyway.)

Finally disgusted with the power-steering leak, and a bit panicked by the fact that my dad might be taking this car to Arizona within the week, I ran the car up on ramps after work and had a look-see. I couldn't spot where it was coming from, and none of the usual suspects looked bad. The bottom of the car was covered with oil on that side. I started the car and rolled underneath, and got a big drip in the eye. It was coming off the back/bottom of the pump, a healthy drip every couple of seconds. I pulled the pump and found that there's a big 2" round plug on the back held in with a circular spring wire. I pulled the spring and worked out the plug, exposing a rather dead O-ring that seals the plug. I need a new one, and it's not something that can be found in my generic kit. The pump's numbers:

Type	7672900428	ZF
Fabr.	1506302	1081
MB	A126 460 00 80		65 bar

Wednesday, September 30, 2009

I also cleaned the hardtop off with compressed air and put it back on the car. I'm getting better at doing this alone.

I ordered a spare key (from the dealer) today: P/N 201 760 03 06, $23. I also ordered a seal (kit) for the ZF pump, P/N 000 586 84 46, $17. Won't get it until next Monday, though, which is inconvenient. I got a grommet too, which they had in stock. P/N 123 997 36 81, about 3.5 dollars—a bit dear if you ask me.

Thursday, October 1, 2009

Sunday, October 4, 2009

Tuesday, October 6, 2009

Wednesday, October 7, 2009

Saturday, October 10, 2009

Sunday, October 11, 2009

Saturday, October 17, 2009

Tuesday, October 20, 2009

Tuesday, October 27, 2009

Sunday, November 29, 2009

Monday, December 7, 2009

Friday, December 25, 2009

Sunday, January 10, 2010

Testing of Oxygen Sensors (from a Bosch pdf)

The information here is a simple guide only and does not replace any technical service procedures quoted by a vehicle manufacturer.

Testing of the oxygen sensor should be made after all basic system checks have been carried out, including testing of the fuel system pressure and performing all minor service adjustments and checks as recommended by the vehicle manufacturer.

As previously mentioned the oxygen sensor will only operate correctly once its temperature is above ~360 °C, so the engine should be fully warmed up prior to testing.

Preferred method of testing is to utilise an automotive oscilloscope, however a good quality Digital Multimeter (DMM) with an analog bar graph function can indicate basic operation. Testing should be done in conjunction with a good quality gas analyser capable of lambda or air/fuel ratio measurement to accurately determine sensor calibration as well as function.

Simplified Test Procedure (Digital Multimeter)

1. Locate the oxygen sensor and determine the wiring layout. On heated sensors check for heater element continuity, adequate 12 volt supply to heater and correct ground when the engine is running.

   NB: Many late model vehicles have the ground circuit of the oxygen sensor heating element controlled by the engine management ECU for sensor temperature control purposes. Do not supply direct voltage or external ground to these circuits.

2. With engine speed at ~2000 RPM the sensor voltage should be seen to cycle smoothly between 100–900 mV (0.1–0.9 Volts) approximately eight times every ten seconds [as the engine cycles from lean to rich (and back again)].

3. A contaminated or tired sensor will be slow to cycle between the sensor voltage limits and may not generate the full potential of 900 mV. It may also be noticed that the lean swing may drop to zero voltage.

This will definitely be worth doing, I think. The same source states that the service life of a sensor is between 60,000 km and 160,000 km, depending on type. We're well over that and I bet it's never been replaced.

Saturday, March 6, 2010

Monday, March 8, 2010

Tuesday, March 9, 2010

Wednesday, March 10, 2010

According to the book a grounded sensor is supposed to result in a fixed 60% duty cycle for cold-start purposes. I saw no sign of that, even though I started probing just after cold start. More research is required. I would think that an analog computer, which I think this might be, would have a smoother output than this seems to. Anyway, it ran well on the way to work this morning. Was it the laying on of hands? I may need to go through the engine computer, checking solder joints and capacitor quality.

Thursday, March 11, 2010

I looked close, and found not the two of yesterday but five different temperature sensors screwed into the top of the engine! The two single-pinners from yesterday, a two-pinner near the idle valve, and two more one-pinners at the back of the engine, one on each side. Which one is my sensor? I consulted the ETM and found (on page 108) that the appropriate wire is green, goes to the 'oil temperature sensor' and is shared with the WOT/Idle switch. The Component Location Guide puts the sensor 'beside the oil filter', and the WOT/Idle switch's C144 connector (for diagnostic access) 'on washer reservoir'. This is yet another sensor, on the bottom of the engine. I'll need to check into this, as that sensor is really down in harm's way. I had a quick peek, and C144 wasn't immediately visible.

Friday, March 12, 2010

I rigged a wire from Pin 3 of the engine diagnostic connector into the passenger compartment so that I can watch the lambda valve duty cycle as I drive. (This is the BR/GN wire [ground switched] from the valve, not the BK/RD/WT [positive feed from the fuel pump relay].) I put the Fluke 87 in the car, as it has a backlight.

On the way to work the Fluke 87 said 70% during warmup. While driving it was 65–82%, mostly hovered around 70–75%. Idling was 75–79%, I saw it get to 85% at the end. Idle RPM was lower, down around where it ought to be. This is pretty rare on this car, in my experience so far. I wonder if messing with the WOT/Idle switch connector affected this?

On the way home I again saw high duty cycles. (Assuming that these are to be measured against ground and not across the valve, which would invert these numbers. [i.e. Wednesday's numbers.]) I checked the other vital statistics, and the RMS voltage (referenced to ground) was around 6.5–7.5 V, probably OK, but the frequency was very suspect: usually around 4,400 Hz, but ranging from 2,500–5,500 Hz! The waveform shown on http://www.systemsc.com/waveforms.htm indicates that the 70 Hz range is more what it ought to be, but the figure shows a fine chop that is gated on and off at a slower rate, and I don't know which frequency was the given number. (Probably the large-scale pattern.) I suspect that the Fluke would react to the fine component, so that high frequency I'm reading may be just fine. I definitely think that it's time to haul that engine computer out of there and have a look at it.

Saturday, March 13, 2010

Because of the conformal coating the solder joints look to be in good shape, but also because of the conformal coating they could be bad and not look like it. I did not have any more time today to pursue diagnosis or repairs.

On the way to Jill's concert I saw about 75% during warmup. While driving it was mostly hovering around 75% as well. I saw higher and lower spikes at times. I had tried a real Dwell meter per Probst's book, but I didn't get anything good out of it, however it always was a POS so maybe that means little. I need to try a better Dwell meter, and perhaps an oscilloscope.

Sunday, March 14, 2010

Pin	Function
2	Oxygen Sensor
4	Oxygen Sensor Shield
5	Ground
6	Idle switch (grounded at idle)
7	Oil temperature switch, grounded when cold; Also throttle switch, grounded at WOT; (Labeled "Fixed 60/40 Ratio Input Signal")
8	Power
9	—
11	—
12	—
14	Air injection relay control (grounds to enable)
15	Lambda valve (switched to ground)
16	Ground
17	—   (Duty Cycle Test Connector?)
18	—

(Note that Pin 16 ground, obviously required from looking at the PCB, came from the '83–'84 schematic. [I think the '82 schematic has a fair number of errors and omissions in it; I noticed that its alternator wiring just has to be wrong.] Pin 18, though present in the connector, isn't hooked to anything on the PCB. Pins 9, 11, 12, & 17 are connected on the PCB, but to what I don't know. Only the nine labeled pins in the table have connections in the car's mating plug, so the rest are obviously not used in normal operation.)

Some time ago a coworker gave me an old Tektronix 561A oscilloscope (with 3A1 dual-trace and 3B4 sweep plug-ins) that was otherwise on its way to the dump. I rescued it and put it in the garage, mostly because I couldn't bear to see such a fine old piece of equipment suffer such an ignominious fate. I justified it by saying to myself that I could use a garage oscilloscope, at least until I can find a Fluke Scopemeter at a decent price. (They're awfully proud of those, even used. I've been looking halfheartedly for some time.) Well, if this isn't a good time for a garage oscilloscope I don't know what is. I got it out and fired it up. (An apt term, since it is full of firebottles.) It had some trouble getting going, all the switches and potentiometers are obviously rather dirty. Unfortunately I couldn't get the horizontal sweep to work well enough to use so I tried using spray cleaner on the switch contacts in the 3B4. Thereafter it stopped working altogether. Oops.

On to Plan B. I stripped the varnish off the back side of the Jetronic. The oxygen sensor input goes through an L-C filter and into the hybrid circuit. I removed and measured the Frako electrolytic, and it had about half the rated capacitance and a very poor Dissipation factor, so I replaced it. I resoldered a lot of the board while I was there, focusing on the larger leads that are usually the troublesome ones, and bought it back out to the car. I had left the 'scope on and cooking, and I pulled the 3B4 back out and looked it over, and cycled the main timebase switch many times. I put it back in and it started working, more-or-less properly. Good. I really wanted to use a scope on this, but digging the Tektronix 2336 out of its nest in the workbench seemed like too much trouble.

Time to get brave. I hooked up all the instrumentation and fired up the car. It acted as usual. I could see on the scope that the waveform to the valve is complex, like the picture. There is a high-frequency chop over one half the 70 Hz major cycle, so the Fluke's weird high-frequency reading is probably fine. (The other Dwell meter was also useless, that was a disappointment. The scope, though, was useful, and well worth the price I paid for it!) I hooked the Fluke 83 up to the oxygen sensor input to the Jetronic, and most of the time it hovered right at about 0.5 V. When I blipped the throttle I could see it go up to the 0.8–0.9 V range, and when it came down it went to maybe 0.3–0.4 V, before recovering. That the oxygen sensor can reach those voltages indicated that it probably isn't bad. I could watch the chopped versus flatline ratio of the valve waveform breathe back and forth a couple of times a second as it adjusted the fuel mixture. The by-eye duty cycle seemed much more like 50% than the Fluke's reading, I think that the high-frequency chop is really confusing the Fluke. Offhand it all seems to be working. But I was out of time for today.

My next trick will probably be to connect test wires to Pins 2, 5, 6, 7, 8, 15, and 17 of the Jetronic for monitoring while I drive. I also need to take a close listen to the valve, it's supposed to buzz in operation. I haven't noticed this, but it's not exactly quiet under the hood! I could also build a diode-capacitor-resistor filter to remove the high-frequency chop from the lambda valve drive so that I can use a regular dwell meter or the Fluke's own duty cycle and frequency functions and get numbers I can trust. This might also permit a Dwell meter to work right. I need to investigate Pin 17 (which one source identified as a duty-cycle testing connector) to see if it's got anything interesting on it. I should also probably complete the resoldering of the Jetronic.

Monday, March 15, 2010

PIN	COLOR	FUNCTION
2	Orange	Oxygen Sensor
5	Orange/White	Ground
6	Green	Idle Switch
7	Green/White	Oil Temp/WOT Switch
8	Blue	Power
15	Blue/White	Lambda Valve

At the free end of the cable I cut and stripped each wire to a substantially different length, to help avoid shorts. I didn't have time to resolder the rest of the unit (stupid Daylight Savings Time!), so I installed it back in the car as-is. (I'll finish resoldering later.)

...On my way to work the system behaved as I think it ought: the initial cold-start O₂ sensor voltage was 525 mV and ranged up to somewhere around 750–800 mV before closed-loop operation began. Thereafter the voltage swam back and forth across 500 mV a couple of times per second, as it ought to. (The bargraph feature of the Fluke meters is invaluable there. There's quite a swing, a good 500 mV on the freeway.) I saw voltage extremes on throttle transients from around 800 mV down to near 0 V. At idle it was very good at holding a pretty steady 500 mV reading, it didn't need to 'swim'.

On the way to band and thence home the behavior was the same. On deceleration I think I saw about 830 mV as the highest voltage, 15 mV as the lowest. The car is running well right now, little sign of the logy feeling. Still a bit lumpy though, but if this is indeed a fix it'll take some time to blow the gunk out of the engine.

I'm stoked!

Wednesday, March 17, 2010

PIN	COLOR	FUNCTION
2	Orange	Oxygen Sensor
5	Orange/White	Ground
6	Green	Idle Switch
7	Green/White	Oil Temp/WOT Switch
15	Blue	Lambda Valve
17	Blue/White	Duty Cycle

...and then hooked up the meters and went for a drive. I put the Fluke 83 on the Duty Cycle wire (since the 87's analog bargraph is better suited to the O₂ sensor voltage) and measured about 3.5 VAC, which varied from 69–72 Hz somewhat irregularly. That's probably fine. The Duty Cycle signal was hovering around 50% at idle before the drive began. I had to invert the meter plugs to get 60% at WOT (rather than 40%), but that was no matter. (The duty cycle is obviously referenced to the other side of the valve, which is at B+ and not ground. Reversing the meter leads takes care of that difference.) While driving with a lot of throttle extremes the duty cycle ranged widely, in the 30–82% range, but under more steady-state conditions it looked like it was going to stick to the 45–55% range where it belongs. Some further test driving is called for. Grounding Pin 7 (WOT switch) results in an instant jump to 60% duty cycle, exactly as specified. When parked and idling, after the run back up the hill to the house, the duty cycle was about 65%, though the O₂ sensor voltage was circling the 500 mV point normally. Interesting, but likely insignificant.

Still looking good!

Thursday, March 18, 2010

I made a mistake, though. While reinstalling the unit and hooking the metering back up I accidentally clipped the Fluke 87 onto the O₂ sensor wire while it was still set to Ohms. That is forbidden to do, I hope I didn't fry it. It, at least, was connected to the Jetronic at the time, so maybe that saved it? We shall see soon. (It measured about 293 kΩ, for what it's worth. And it really oughtn't fry it, the Jetronic itself puts 500 mV on it when it's cold.)

...On the way to work the duty cycle looked good, 60% at first. At idle would be anywhere from 65% down to 25%, depending on circumstances, but usually closer to 50%. Steady-state on road conditions looked like the 45–55% that it ought to be, but could go higher and lower.

...At lunch I had a few moments, so I popped out #7 spark plug and found it still oily. I replaced it with a clean one from the bag of used but clean plugs in the trunk, we'll see if the new regimen will cure that or not. I expect it to take awhile, if indeed it's a ring seating problem. I also unplugged the idle actuator as a test, and the RPM's shot up. So that system is working. To give the new plug a shot at cleaning things out I drove downtown (freeway) to Dick's drive-in for lunch, I had a burger and a malt. I saw 83% duty cycle at the beginning of a long deceleration getting off the freeway, and 15% when getting back on the throttle, those were the extremes I noted. All the numbers are looking pretty good, usually right in the center of the nominal ranges. When restarted with the engine warm (but the O₂ sensor cold), the duty cycle was right on 50% before the O₂ sensor started generating voltage.

Saturday, March 20, 2010

Monday, March 22, 2010

Tuesday, March 23, 2010

...When I picked up Daniel from school I had a recurrence of the nasty 'clicking' noise from the fusebox area, just off idle. The air pump (?) relay was going apeshit. I also noticed that the frequency of the lambda valve drive was wildly erratic, and I was getting abnormally high (150 Hz) and low frequencies too, though I believe the duty cycle sample AC voltage was normal. The frequency could jump from one end of the range to the other from one reading to the next. Very weird. After it warmed up a bit the clicking went away, as usual, and the lambda readings stabilized.

...On the way home after work it seemed to be running a touch rough so I pulled #7 plug out again. Very sooty (except for the tip of the electrode which was burned clear), but not oily. Interesting. (I pulled #6 for comparison, and it was fine.) I swapped in a cleaned plug from the trunk stash as it's 'date night' and I want it on its best behavior. I filled it up again too today, 12.6 MPG. Not great! I've noticed some blue smoke on acceleration, especially after a long deceleration, such as at the bottom of our hill in the mornings. Sigh.

Wednesday, March 24, 2010

...After work I pulled the plug again, and it was again extremely sooty. Only about 15 miles on it. I put more PB Blaster in the cylinder, and unplugged the cold-start injector, which feeds the manifold in that area. (I really should block it off mechanically, to be certain that it's not dribbling.)

Thursday, March 25, 2010

Friday, March 26, 2010

Saturday, March 27, 2010

Tuesday, March 30, 2010

Friday, April 2, 2010

Anyway, to that end I pulled the troublesome plug and found it a bit oily, about the same as the other recent checks. I cleaned it and put it back. I topped off the power steering fluid, it was a touch low. Engine oil was fine, as was the coolant. Brake fluid is dirty, but otherwise OK. Belts look OK. Seems to be ready!

The front speakers haven't been working lately, so I pulled out the stereo and the glovebox. Poking around they came back to life, not quite sure what was up with that. I put it all back together, parked the car in the lawn (such as it is), and washed the car.

Ready!

Saturday, April 3, 2010

Today the weather was better, though not perfect. Some more snow at the outset, but not too much. We had a tailwind today on the return trip, and got 20.6 MPG. At home it idled better than it did after the first leg.

The rear window glass is starting to come loose again, and was rattling much of the journey. It seems to be OK for awhile at first, but continued vibration works it loose and then it rattles easily. Let to rest it seals up again, until the next long trip. Or so it seems, anyway.

Tuesday, April 6, 2010

You change it, of course.

This doesn't seem to be getting any better, but it is otherwise running pretty well now. This isn't debilitating anymore—just very annoying.

(I was having some fun with Lilypond, the 160-byte source code for the above music snippet is here.)

Thursday, April 15, 2010

Friday, April 16, 2010

Sunday, April 18, 2010

The glove box light kept popping out, so I glued a little strip of plastic (a length of vinyl phone cable insulation, slitted) into the hole to narrow it so that the light's tab would grip better. That should keep it up.

Today was beautiful, and warm. We had to go to Cd'A, so off came the top, and off we went! It was nice.

Monday, April 19, 2010

Wednesday, April 28, 2010

Thursday, April 29, 2010

Monday, May 3, 2010

He also suggested a hotter plug with a extra-large gap, so I guess I'll probably do that. We'll see how it goes now.

...The cloud of blue smoke at the bottom of the hill this morning on my way to work was extra-large. Significance?

...At lunch I went to Schuck's to get a hotter spark plug. The car's rated to take Autolite 64's, yet Autolite 66's were what I found in car before I replaced them with the (stock rating) NGK BP6ES's! Already not one but two grades hotter? That's disturbing... I bought an Autolite Platinum 66 which is supposedly marginally hotter than their standard 66, which is the hottest they had on the shelf. We'll try that. $3

Wednesday, May 5, 2010

Anyway, I installed the new AP66 plug in #7, gapped to 0.044". (I've seen some information that opening them up to 0.042–0.044" from the stock 0.032" [0.8 mm] can be helpful.) The plug that came out was already oily/gassy after 70 miles, though far from the worst I've seen it.

While I was there I cleaned the gunk out of the crankcase ventilation tube to the air cleaner. Very caked it was. I also note that the ill-fitting rubber hose from the smog pump to the air cleaner has finally fallen off and been lost. Oh well.

Thursday, May 6, 2010

Saturday, May 8, 2010

Tuesday, May 11, 2010

Wednesday, May 12, 2010

Saturday, May 15, 2010

Friday, May 21, 2010

Sunday, May 23, 2010

Monday, May 24, 2010

Thursday, June 3, 2010

Saturday, June 5, 2010

Wednesday, June 9, 2010

I also topped off the coolant, it's leaking down. The oil was down too, but not quite ready to top off yet.

Sunday, June 13, 2010

Monday, June 14, 2010

I got one spring (of two) on the squeaky hinge unclipped using vise-grips, with both trunk and horse collar up to relieve some spring tension. With that undone (and I'm not sure how I'll get it back!) I still couldn't pound out the hinge pin. I pried with a screwdriver at the head end, and got it to ootch out a bit, but then I had no more leverage with the screwdriver. I fabricated a fork from a piece of 1/8×3/4×3" steel, and ground a wedge shape into the end, essentially making a custom-fit 'pickle fork' for the hinge pin. Pounding it under the pin's head with the BFH got the pin to move a bit more, and then I welded more thickness to the fork tines and re-shaped it again, and drove the pin out a bit more. (It takes quite a bit of pounding.) The hinge pin no longer protrudes from the other end of the hinge, so I'm on the right track. I ran out of time this morning, however, and the fork isn't thick enough to go further as-is.

Tuesday, June 15, 2010

Wednesday, June 16, 2010

Friday, June 18, 2010

Saturday, June 19, 2010

...I put a towel under the trunk and horse collar, so that when they come off the hinge they'll land on it. Then I removed the pin. Easily, now that I'd gotten it moving? Hell, no! It fought me equally hard every step of the way, even with strong tools. I can't believe how tight that thing was in there. My tire spoon prybar started bending at the tip, so I cut it off to get down to a thicker part of the bar and reground the notch. Once the prybars were out of travel I found that I could stack a succession of large nuts on the pin. They fit over the pin's head yet bottomed against the hinge bracket, then I could pry anew against the nut stack. I ended up using my leg to push against one prybar while the other was wedged against the battery well. It was still extremely difficult, and took a couple of hours to get it all the way out. The pin ended up bent pretty badly, I doubt that it can be reused.

The hinge bores are nylon-lined, I'm really surprised at how this thing was acting. The hinge pin didn't really look rusty either, though it could have been scrubbed off. It's possible the hinge pin was defective all along. In spite of my care I picked up a small paint chip or two, but they should repair pretty well.

Halfway there!

Things to deal with while I'm there: There is a bit of rust under the battery box, and a bit more starting up in the spare tire well again, and the cause of all this is (I think) glue failure on the trunk weatherstripping where it goes around the problematic hinge. The brittle plastic battery box (107 843 00 05) also has some chunks off of it that can probably be glued back.

Sunday, June 20, 2010

I then pulled the trunk weatherstripping out of the channel where it was leaking, and used a hose and toothbrush to clean the dirt out of the channel. I gouged out the loose glue and wiped it all dry. I cleaned the weatherstripping and dried it too, then I got out the weatherstrip cement and glued it all back, making sure to completely coat the surface on the trunk side. (This should keep the water from spilling over into the trunk, and in the channel where it belongs and can drain rearwards.)

The other battery box retaining strap was starting to crack like the first one did, so I welded it too. It got ground, brushed, and painted as well.

The battery hatch hinges were rusted solid, and a couple of days soaking with penetrant hadn't really helped. No matter, I know the trick: I used the acetylene torch to put a red heat on them, and they loosened right up. I then brushed and painted them, then oiled them. Lacking any big rivets as used originally, I reattched the hinges to the plastic hatch cover using some large sheet metal screws driven into some aluminum strips (cut from dryer vent pipe) with starter holes drilled in them. It's solid enough.

The plastic battery box itself was cracked and broken, so I started gluing it back together with Shoe Goo. It'll take a few sessions to get it done. (I don't think I have quite all the pieces, but it'll be OK.)

...In the evening I glued some more pieces of the battery box back on. One more session ought to do it. I think there's only one 4-inch piece of the sealing channel that's actually missing, that's not too bad. (It's like doing a dirty, deformed jigsaw puzzle.)

Monday, June 21, 2010

Tuesday, June 22, 2010

Time to do something constructive! I greased the pin and the hinge bores and faces, and assembled the hinge. The two arms moved easily, yet with no noticeable slop. I bent the retaining tab down to hold in the pin, then attached the horse collar and the trunk to their hinge arms. The paint should be safe now! Both panels move easily.

While I was there I touched up some of the black paint inside the trunk and tire well.

To reinstall a non-broken battery box you actually have to remove the ground cable from the car, and that's where I ran into trouble. The bolt was badly rusted and the captive nut broke loose from the frame, and there doesn't seem to be any way to get at it from the bottom. There's a rubber plug directly under it on the bottom of the car, but it appears to be a decoy. The nut is captive inside the frame, and not accessible at all. (Visibility is very poor through the hole.) I sprayed penetrant all over and put everything away, I was out of time anyway.

Wednesday, June 23, 2010

Next was the battery box, which I finally got to put back! It works best to remove the grommet from the negative battery cable and insert it in the box first, then thread the cable through it and attach it. I used anti-sieze compound, that should prevent this problem from happening again. I then painted the bolt attachment area. The positive cable grommet was a real pain to get put back because you have to do it with the cable through it, but I managed. When I put the battery in the box I found the source of part of the problem: the battery is too small, and if the box is intact the cables don't reach the posts! I pulled on the positive cable to break loose (partially) some of the gluing I'd done on the box, enough to get the clamps attached to the battery. I then bolted the hatch cover hinges down, and then finally the box retaining straps. As I was out of time for more I cleaned up the area, but at least the car is driveable again.

As the rains have again stopped and it warmed up 20 degrees or so it was nice to have the car back in service.

Thursday, June 24, 2010

I then painted the chewed-up end of the torsion rod, as well as the general hinge pin area.

There was some suspicion that one of the expansion tank fuel lines was weeping at its connection. It had a bit of slack in it so I cut off 1/2" and reconnected it. The cut-off piece actually looked pretty good.

It was then time to put the trunk wall back, which took a lot of fiddling. I had to remove the battery hatch again in order to fit the panel into place. Do not tighten any bolts down until you have them all started! The worst part was getting the expansion tank screw started, it didn't want to fit. I then put some non-hardening RV window putty between the trunk lid skin and the support channels, which cures the rattle. The trunk now feels and sounds like a Mercedes ought to when you open and close it.

I still have to reattach the carpet I peeled back from the bolt heads on the trunk wall, but this little project is getting close to done, after 21.5 hours so far!

...After work I filled up again. In honor thereof I put in a clean sparkplug.

Friday, June 25, 2010

...At lunch I took a break and walked down to Schuck's (or whatever they're calling it these days) and bought some more weatherstrip cement, a quart of oil (emergency kit), and a proper sparkplug anti-fouler ($5 for two). After work I installed the anti-fouler on #7, we'll see if it helps any. (It has the correct seat and so should seal better than the last one I tried, and the threads are long enough to keep the threads in the head from getting gunked up.) About 18 miles on the plug from yesterday's fill-up and it's already getting oily. Even if this doesn't help much with the fouling, its presence makes the plug a lot easier to change on a hot engine and cuts wear and tear on the head's plug threads. It's funny just how much difference a fresh plug with no anti-fouler (compression reducer) makes in the 'feel' of the engine. Much stronger and smoother—for a few miles!

Wednesday, July 21, 2010

Friday, July 23, 2010

I also noted at lunch that the AC seems to be inoperative, though the compressor was engaging. Probably low on refrigerant again. Seems to last about two years. I dropped by the sporting goods store to stock up again. $5/can for the base material.

Sunday, July 25, 2010

Friday, July 30, 2010

Wednesday, August 11, 2010

Today Craig's List came through again:

Convertible top only!

Complete assembly. Mercedes Benz original, factory convertible top for R107–W107

Fits all 450SL, 380SL, 500SL, 560SL, 350SL from 1972 to 1989. (Also euro version 280SL from 1974+. No, it won't fit 230SL, 250SL, 280SL.)

This is complete with the entire frame assembly and canvas. Just bolt on your SL.

Frame is perfect. Hinges are complete and work perfectly. Canvas is dark gray / black in very, very good [condition] with no holes, no rips, no wear. Window is in very clear with no scratches or yellowing. Tan headliner is excellent. All rubber molding is excellent. Very clean. From a cigarette-smoke-free 1985 380SL.

Easy install—only 4 bolts.

$280.

509-217-XXXX

Sweet! The pictures looked very good. I called, and made arrangements to get it after work, in the car. We dropped by where he had it stored, and it looked very good. The plastic windows were clear, though with the normal amount of scuffing, the rubber seals were intact, though in need of a touch of gluing, and the top was good, both inside and out, though dusty. The frame was not rusty, unlike my original. Sold!

The top was from a parked car that got whacked in the front, enough to ruin the water pump, etc. Nothing good left on the entire front he said.

The guy expressed an interest in my old frame, and Jill was very interested in not bringing home two, so we made a deal: for $75 he got my old one. (She was bouncing up and down and saying "Yes, yes!" right from the first. Honey, we have got to work on your negotiating skills!) Yes, properly marketed I could probably have gotten more, but this is done now and is not itself a project. That's worth something right now. He got out a 13mm swivel socket wrench and we swapped tops, it took about a half hour. It was a bit fiddly getting the frame back in place and all the spacer shims back where they belonged, but nothing too onerous. The front turn latches (a bit rustier than mine, but who cares) cinched down well, but the rear peg wouldn't reach the hole. I've seen this before, the canvas has to be wet to stretch a bit before it'll reach, they get that way in storage. No big deal, I imagine. The bow hinges aren't quite as easy to work as the old ones, but some oil and use will probably take care of that, just as it did on that one.

I'm very happy that the top is now 'dealt with', and in a ready-for-sale condition, for about a $200 net expense. I should wash and waterproof the top, and the guy suggested Meguire's plastic cleaner to take off the fine scratching. He also suggested that I might try Castrol 20W50 oil to see if it helps with the oil consumption.

Saturday, August 14, 2010

Tuesday, August 17, 2010

Saturday, August 21, 2010

Friday, August 27, 2010

Tuesday, August 31, 2010

Thursday, September 9, 2010

Sunday, September 12, 2010

Monday, September 13, 2010

Monday, September 20, 2010

Tuesday, October 12, 2010

Sunday, November 21, 2010

Sunday, January 30, 2011

Another large chunk of clear coat has peeled off of the hood. Crap!

Monday, May 9, 2011

Wednesday, July 6, 2011

Friday, August 19, 2011

Friday, April 13, 2012

Friday, April 20, 2012

Wednesday, May 16, 2012

Sigh. I don't know that I believe him. If the cylinder was all crapped up due to an oily plug not firing just prior to the test I could easily believe the valves did not seal correctly during the test. I saw much better numbers when I tested it.

Wednesday, June 13, 2012

Wednesday, August 8, 2012

Thursday, August 9, 2012

Thursday, August 16, 2012

Wednesday, August 29, 2012

I began preparing the car for sale. I removed the test wires I had connected to the Jetronic control box, and reinstalled the floor panel and carpets. I then vacuumed the interior and tidied up the trunk. Mostly cleaning and photography left to do, now.

Thursday, August 30, 2012

During the day, when the car was parked, I made sure to park it in visible places and with the signs up. It begins!

Monday, September 3, 2012

Tuesday, September 4, 2012

Saturday, September 8, 2012

Sunday, September 9, 2012

That's a big relief off my mind, I just have too many cars and other projects to keep up with. Now there's one less.

When we got home I took the ad. down and fixed up my web site.

Monday, September 10, 2012

Tuesday, September 11, 2012

Saturday, September 19, 2015

Wednesday, October 28, 2015

Thursday, May 12, 2016

Saturday, March 25, 2017

Return to Site Home