Cold start/drivability issues

[JGood]

Picked up a $600 automatic 89 325i for a 24 Hours of Lemons car.

The car takes a good 10 seconds of cranking before very slowly sputtering to life when cold. Once running, it runs with a pretty decent idle, doesn't lope or sound like it's missing too bad, it will idle all day without really any hiccups. But if you try to give it throttle, it will usually stall, or try very hard to. If you very, very slowly get into the throttle, it will usually sputter its way up to 2k or so and rev pretty good at that point, but under any load (while driving), it will cut out and miss pretty bad, to the point where it's not really drivable.

This doesn't go away until about 30+ minutes of running. Then, it runs like it's got a freshly rebuilt engine. Pulls very hard, idles great, no hesitation, no smoke. So it appears to be a full engine temp symptom, not just a coolant temp symptom. With coolant up to temp, say 10 minutes of running, the issue is still there, until the entire engine is actually hot, after 30+ minutes of running.

Things new on the car before we bought it: FPR, fuel filter, plugs, cap, rotor, wires. The head was also 'rebuilt' 1 year ago. Apparently the rockers and cam were replaced, and the PO said the cam had significant scoring on it when he removed it, hence the replacement. I take anything he said with a grain of salt, as although he was familiar with e30's, he was not an expert by any means. He claims he smoke tested it and found no leaks.

Things I've checked so far:
TPS - Idle switch works (cont. between 2 & 18 with throttle closed, no cont. with throttle open), WOT switch does not (not continuity at WOT).
CPS - 560 ohms
ICV - 20 ohms between each coil, 40 ohms between both coils, moves freely)
Temp sensor (DME) - Only checked this after ~15 minutes of running, temp was probably 160ish degrees, got 400 ohms, which is about right.


I'm thinking a few things:
1. Valves adjusted way too tight. But, it idles pretty good, even when cold, so I'm hesitant with this theory.
2. Compression issue. Again, with the good cold idle, this doesn't sound like the issue.
3. AFM
4. Vacuum leak
5. Injector issue(s)

Any other ideas, or suggestions of what to start with? We only get so many chances to work on it, as once we play with it for 30 minutes or so, it's up to temp and will run great until cold again. Just trying to get this sorted with minimal time/labor.

Thanks!

[meengreene30]

Have you checked fuel pressure? Sounds a little what mine was doing. I was getting almost 90psi, unhooked return line from regulator and was a much better 46psi. We found that my in tank return line was clogged about 10 inches in. I also have an 89is with the 63 liter tank that the return line enters at the bottom behind the fuel filter, not on top of the fuel pump. Maybe the o2 sensor could be bad too, try unplugging it and see if it makes a difference.

[JGood]

Have you checked fuel pressure? Sounds a little what mine was doing. I was getting almost 90psi, unhooked return line from regulator and was a much better 46psi. We found that my in tank return line was clogged about 10 inches in. I also have an 89is with the 63 liter tank that the return line enters at the bottom behind the fuel filter, not on top of the fuel pump. Maybe the o2 sensor could be bad too, try unplugging it and see if it makes a difference.

Nope, haven't checked pressure yet, that's actually my next step for the next time I fire it up cold. Although I wouldn't have expected high pressure to cause the symptoms I'm experiencing.

Also considered an o2 sensor issue, and will try unplugging it. The symptoms I'm seeing seem a bit more dramatic then what I'd think an o2 could cause as well, but it's easy enough to try!

Thanks for the input!

I'm also going to put my Peake code reader on it. Since the TPS WOT switch isn't working, I can't do a stomp test.

One other thing I found interesting, according to realoem, this automatic car should have the variable TPS with the round plug. Mine just has the standard 3 pin switch. Which is fine by me, makes it easier to deal with, just strange that it appears realoem is wrong here.

[JGood]

Did some more work on the car.

Compression test, cold:
160-170-170-135-170-170

Compression test, warm:
170-170-170-160-170-170

Cyl 4 is a little low, and the plug was a bit wet with fuel. My issue isn't a single cylinder misfire though, the entire engine is cutting out at once.


Fuel pressure was only ~33psi at idle, but it was idling great at that pressure. I made the car start to stall by giving it throttle, and as I did that, pressure came up to about 40psi, and stayed there as it sputtered and stalled. Not sure what to make of that. 33 is low, but it doesn't seem to be affecting the idle, or causing the issue I'm having.

Plugs looked a bit white from possibly running lean, but nothing terrible.

No codes in the DME.

No vacuum leaks found thus far with the carb cleaner method, will be smoke testing though.

Swapped AFM, ICV, and CPS, no change. Verified TPS is working properly, idle and WOT switched closing/opening as they should.

Unplugged o2 sensor, no change.



Again, the issue is only existent when engine is not up to full temp. The car cranks for a while, slowing sputters to life, then holds a decent idle. Giving it throttle will cause it to sputter/stall, unless you get on it very slowly. Any load will cause it to miss and break up pretty bad, it's not really drivable until fully warmed up. Coolant temp doesn't seem to matter, it's actual engine temp. Coolant gets up to temp in 5-10 minutes, it takes 30+ minutes for engine to get to temp and clear up. While idling when cold, it also seems to randomly stutter and sometimes die without touching the throttle. Maybe once every 2-3 minutes it will do that.

Next step is to check valves, cap/rotor, and maybe swap the main & fuel relays. Any other ideas?

[UserError]

I'd suspect maybe a coolant temperature sensor, with a drivability issue like that. Check the resistance of the sensor, it should be between 2000 and 3000 ohms, depending on the ambient temperature. It should also drop as the engine warms up, though the only way to check that is to shut off and measure the sensor at various engine temperatures.

[JGood]

I'd suspect maybe a coolant temperature sensor, with a drivability issue like that. Check the resistance of the sensor, it should be between 2000 and 3000 ohms, depending on the ambient temperature. It should also drop as the engine warms up, though the only way to check that is to shut off and measure the sensor at various engine temperatures.

Yeah, mentioned that in my first post, it tested good.

And again, this issue does not appear to be related to coolant temp. Coolant is up to temp in 5-10 minutes, and the car still runs like shit. It takes a good 30+ minutes of running until the car runs well.

[JGood]

Turned out to be a bad distributor cap/rotor. They were burnt/melted and had some very strange patterns in the contacts, almost as if the rotor was crooked. Either way, those parts are in the trash and she runs great!