928mac

Did you try disconnecting the temp sensor?
If it failed it will show -40 deg and the computer will run the injectors really rich.
If it makes no difference with it off then check the sensor with a meter cold and then hot. Maybe do that first.
(not at home so I can't tell you what resistance)

The reason I think it corrects it's self after you clear the memory is because it has a code set by a sensor (MAF, Temp, ~) out of range and it takes a while to fail again.
To bad you could not scan for the code, and then you would know the problem.
My 2 cents

BTW Bang on the MAF with your hand while it is running, if it stumbles and wants to die then that is your problem. Not a correct test but if it stumbles, replace it.

Brett Jenkins

I've checked it with the temp ii plugged in and unplugged with no change. I tested the resistance for it at the LH plug and at the sensor itself. Got 1.5 at the LH and about the same from the LH pin to ground and the same from the EZK pin to ground at the sensor itself.

Checking MAF specs now. Everything at the LH/EZK and fuse panel seem ok. Checking the hot wire circuit at pins 3 and 5 on the MAF gives me 00.9 and the manuals says it should be 3.6 to 4.1. Going through other testing now.

John Speake

Unplugging temp 2 is he best way to test it. This is because an open or short circuit will trigger the LH to revert to normal 80deg C setings for the fuelling. As Brad says, a faulty temp 2 wih highish resistance (say 10k ohms) will make it run very rich.

I'm not in the office, but the hot wire is around 1 ohm IIRC

Hilton

Sounds to me like the O2 loop adaption circuit is screwed and its consistently adapting off the end of the scale. This would be consistent with it being ok for a while after battery disconnect, and then happening on cold/warm starts after the problem arises.

Time to try your LH in a local's car - the commonly failing part in the LH includes the O2 loop.

John Speake

Good thought Hilton...

White Lightnin'

So, the unfiring cylinders would be dumping raw gasoline into the cats/exhaust system? Would this have caused the cats to glow red as MrMerlin inquired about?

Brett, did you have a strange-odored smoke coming out from under your car/back of engine area?

Mine did 99% of what you describe plus the smoke... today. I shut it down and let it cool and after about 30 minutes, it returned to normal -no stumbling RPMs, no smoke. It did scorch my paint at the exhaust tips though...

Brett Jenkins

The cats fire was one of the first things I checked. They were not glowing red.

I have another MAF coming Monday. Hoping that fixes it. Swapped out LH relay today with no luck.

If MAF doesn't fix it, it's probably the LH, which is what I have been afraid of.

Has anyone ever done a Bosch parts cross-reference on the LH? There are several I see on Ebay that are identical in appearance and only slightly off on the Bosch part #, and are significantly cheaper. I don't know too much about the functions of the LH, so it might be a moot point if the internals are unique.

Hilton

There were some old threads about other LH versions from Volvo and Saab's. However LH2.3 is pretty much unique to the 928 - so those other LH's won't work without much modification and figuring out how to change the rom's (IIRC Volvo and Saab both went from LH2.2 to LH2.4)

(hey Ken - there's a job for ya.. figure out how to equip a turbo '86.5 with Volvo's LH2.4 system from their turbo engines )

Brett - if you mean other 928 LH versions, any 928 LH from 87-95 is functionally the same and can be used in your car.

The only functional difference is found on the ECU chips which can be swapped. However your best bet is just to get yours rebuilt by JDS or similar - once you've made sure its faulty by putting it in another local's healthy S4.

Landseer

Despite newer wires be sure to scrutinize for a nick on a coil wire. Doesn't take much.

Brett Jenkins

I may have found the problem, or created another one. I pulled the LH and cracked it open to take a look. The board in one corner, under a protective plastic cover, appeared to be cracked. I removed the cover and sure enough, it had several cracks in it. I don't think my removing the LH and opening it up would have done this. So, looks like a replacement or rebuild is needed.

I know The Nine Source has a used one available, spoke to him earlier about it, but I'm not sure I want to spend that much on a used one, since its failure could be right around the corner. Unless someone has a super cheap one, I guess I will send this off to get rebuilt ASAP.

Brett Jenkins

I assumed the plugs would be fouled horribly from the rich condition, so I pulled them and they are black as night. Is there anything else I should do because of this, in addition to new plugs and changing the oil? Should I run anything through the fuel or cylinders to clean up the soot, and then change the oil.

MAF , LH, plugs and plug wires should be here Wednesday, so hopefully I'll be back on the road by Thursday.

Mrmerlin

I would put new plugs in and not worry about changing the oil unless its time,
the engine hasnt been running long enough to foul up the oil.
New wires and other ignition parts are always a good idea

daveo90s4

If you have cats don't you also have the 'lovely' ignition monitoring system too, along with the red and green diodes to check for illumination?

Brett Jenkins

I believe the ignition monitoring system was introduced in 89.

SQLGuy

I believe LH in an '88 is 2.2, not 2.3.

Re, "Sounds to me like the O2 loop adaption circuit is screwed and its consistently adapting off the end of the scale.", does Bosch really allow that much EGO authority from the fuel map? I would have expected them to limit it 15 or 20 percent, since the map should be pretty close and only need Lambda for fine tuning and/or compensating for fuel pressure, battery voltage, etc.

As a recap, though: very rich, same behavior with O2 disconnected, swapping relays makes no difference (this is expected, as relays are just relays, they're pretty much going to make things work or not, but not make them work badly), intermittent.

Things that would cause the system to send too much fuel:

1. MAF voltage too high, stuck high, or reading high, or maybe MAF losing signal ground.
2. Fuel pressure too high - bad FPR or vacuum leak at FPR.
3. LH thinks engine is really cold. Sounds like you already eliminated this possibility.
4. Maybe (but I'm not convinced), Lambda overcompensation. Since my '88 runs passably well with the O2 disconnected, I don't think this is the issue. In fact, mine had the O2 disconnected for the first year or so I owned the car, and, other than being a bit rich at startup, it really ran well and plugs looked fine.
5. General LH going bad.

I think that 1, 2, or 5 are most likely here. Have you checked the condition of the MAF plug and wiring? Have you checked for fuel in the vacuum lines to the FPR and dampers? Does the car run any worse with the vacuum line at the FPR disconnected?

Can we get a picture of the cracks you found in the LH?