Hi All, Here's what I know: '87 328 ~35k Miles Over the past 3 months the idle bump at cold start dissapeared. (Still started fine, just idled at 1k instead of higher. Backfired once about 2-3 months ago on a hot start. Checked AAV for operation- works as advertised. Replaced injectors and seals 2 days ago. Removed and cleaned throttle body at the same time. Checked Thermo time switch for operation- works as advertised. Set mixture (Closed Loop) with Duty Cycle- 50%. Mixture adjustment works as advertised when engine is warm. Insured Plug is in adjustment hole when running engine. On cold starts with Open Loop the mixture goes to extreme rich- 22% on duty cycle. Engine only runs if I keep goosing the throttle to try to lean it out. Once warm it runs fine. (Great actually- it is quieter, smoother, and has tons more power than it did. Throttle response is also much better.) Hot starts after at least 15min roughly the same as cold starts, although not as aggravated. Also- the mixture still goes to aroud 22% duty cycle. If I short the Thermo circuit it stalls the engine. I can't hear any difference between air injection on and off. Thus I have no idea if it is working properly at this point. (But I am suspicious...) As far as I can tell the Lamda sensor and system are working properly. What then is causing the profound enrichment when the engine is anything less than warm?? Something changes- to go from 22% up to ~50% is not a subtle change. Thus I am missing something that I hope you can point me towards. Control pressure regulator?? 4th Fuzzit valve?? Many Thanks for your thoughts on this one. Cheers, dce
If your problem was "warm running is good, warm restarts are OK, but only cold starts are bad", I'd say checking the Warm-up Regulator behavior would be at the top of the "to do" list -- but your report of warm restarts also being poor doesn't seem to fit that exactly (but, since the Warm-up Regulator is a "dumb" open-loop device, it wouldn't hurt to verify OK, or not OK, before digging deeper IMO). Can you clarify "If I short the Thermo circuit it stalls the engine" -- is this shorting the coolant temperature switch closed during warm (i.e. closed-loop) running? If so, this makes sense, because shorting that coolant switch closed enables the air injection system which would make the O2 sensor measure very, very lean so the injection ECU tries to compensate by (wrongly) having the frequency valve dump in a bunch of fuel to try to richen things up.
Yesterday when the engine was warm I would disconnect the spades from the switch and "Simulate" a cold engine by shorting the two it would immediately die. Dead, no protest, no nothing, just...done. Tonight when I confirmed that the switch was working it would/did start with the circuit closed, albeit with the complications listed above. But...Closed or open, it doesn't matter- I still see roughly a 22% Duty Cycle when starting the car after anything but a really brief stop.
Please go back and see my edit -- shorting that coolant switch should cause an engine stall when warm-running. I believe it is correct for the FV duty cycle to default to a somewhat richer condition for either a cold or warm restart (unless the warm restart is done very quickly such that the thermocouple in the cat stays hot enough to say to the injection ECU, via the slowdown ECU, "I'm warm -- so run closed-loop"). Can't comment if 22% duty cycle is normal or not though for that default value.
Ahhh.. Got it- I was wondering if there was something in the mix that would preclude simulating a "Cold" engine when it was actually warm. (The Probst manual deosn't really clarify this point.) So that is at least normal operation- although it still doesn't tell me anything about why the FV is trying to reduce a nearly full rich mixture. I think that is the key: The FV isn't pushing the mixture to full rich, it is trying to compensate for something else that is making it run rich. Thus..... What can cause a ridiculously rich running condition that overcomes the FV's ability to compensate upon start/restart??
I would definitely focus upon the WUR. The WUR is really a fuel pressure regulator that affects the AF ratio under all conditions, not just "warming up." On your 328 you'll find there are three inputs controlling the WUR fuel pressure (= AF ratio): a) temp off the block sensed through the WUR housing, b) electrical enrichment input based off engine temp sensor, and c) vacuum enrichment based off engine load. If any of these three goes off due to an internal failure then you've got a messed up AF ratio. Replacements aren't cheap at around $400 but I suspect your problem lies with the WUR.
Before you get too deep, one basic thing that is often overlooked is make sure the air sensor plate is resting in the proper place. If it is too high you will never get everything adjusted where it runs right all the time. I hope this helps with your problem. Good Luck Larry Fletcher
I think you might be right- the question is: How do I adjust the plate downward?? The Probst book shows how to adjust an updraft plate, but not really a downdraft plate. Any suggestions?? Thanks, dce
Yes, there's an intertwinned layer of electro/mechanical/thermal/vacuum actuator gizmology that's model (manufacturer) dependent so the K/KE Jet general texts don't (and really can't) go there. The best you can hope is that there's something in the vehicle manufacturer's WSM that gives a real engineering-type system overview. Otherwise, the OM and deducing system operation from the relevant wiring diagram are the only other pieces of information that I know to use. (In your case, Fig 2 of the '89 328 wiring diagram would do OK.) IME cold-running on K/KE is a fairly "stupid" mode, and you've already quasi-verifed that your coolant temp switch is working as intended (closed cold, open warm). The only other thermal input device to the injection ECU on Fig 2 is the oil temp switch. The FV is a passive output device so the problem would be "something" telling the injection ecu to run the FV overly rich (if the 22% duty cycle is incorrect for open-loop restarts). If you wanted to check the oil temp switch (Item 154 on Fig 2), its logic should be closed cold and open warm (check the OM for the approx actuation temps) -- give a shout if something comes up; otherwise, I'll assume that you've got Fig 2. and can check 154's behavior (i.e., it's open when above something like 5 deg C -- which it might be always if you live in a mild climate) at the electrical assy in the trunk (but this would be the lucky/unlikely fix IMO). A couple of questions for you: 1. When you've got it at warm-idle, running closed-loop, how much is the FV duty cycle changing from the average 50% value that you've set in order to cause the O2 sensor output to oscillate from 0V to +1V? (And have you verified the O2 sensor signal is changing reasonably from 0V to +1V?) 2. When you've got it at warm idle, running closed-loop, and then unplug the O2 sensor what duty cycle does the FV default to, and what is the voltage coming from the (unplugged) O2 sensor wire? TIA
Good questions. 1. The FV duty cycle is changing +/- 2% or so. And the O2 sensor ranges from .01 to around .09vDC. In short they seem to be working. 2. This is interesting. I'm pretty sure I did this, but I don't think that it really changed that much- I actually cannot remember at the moment. This is a legitimate question though since if I understand correctly it should default to 50% and pretty much stay there. (Which it might have done- I'm going to have to try it again to verify- but I'd really like to get the cold start problem nailed first- my neighbors are going to disown me if I keep stinking up the arrea!!) Thanks, dce
Did you make a decimal point error in you message? Should it be 0.1 to 0.9 (which would be A-OK)? Yes, you've got the right picture for #2 (I'm just trying to understand what the default duty cycle value is versus the duty cycle value you used in the "closed-loop" adjustment -- my guess, too, is that they should both be 50% -- but that's just a guess). This is where you'd really like to have that extra "working" 328 K-Jet with Lambda so you could check things like "is a 22% FV duty cycle restart normal?" If it isn't the "unlikely" oil temp switch fault, you've got to get back to the WUR first on any K-Jet (with or without Lambda) having cold-start/cold-running issues.
