Help with 528i start but not run

[turbo-rampage]

OK here is the latest.

Didn't have time today to get an injector tester, but a friend told me a peanut bulb will do in a bind.

So the verdict according to the peanut bulb is the injectors are NOT pulsing.

I first cleaned the obvious connectors, checked the single lead going to the ECM, and did check the ICM for spark to 0 rpm. That one is a tough call, but it has to be sparking very close to zero.

I decided to pull the ICM apart anyway just cuz. I did find something interesting in the ICM.

Right above the terminal strip for the plug there is a fairly heavy duty looking resistor, which was kind of encased in a ceramic type coating (almost like a ballast resistor). I say this because it has running extremely hot, and the coating is burned off, and it as melted a plastic piece on the terminal strip just below it, and has really baked the circuit board underneath it to the point it is deformed. Is this normal? Seems like an odd place to put a resistor of that size. The other odd thing about it was it had full continuity, no resistence whatsoever. And the circuit board is still OK, in spite of it's melted state (The copper "lines" in it still conduct)

Anyway I did resolder all the conections, put it back together, and guess what. Still doing the same thing. Arrrrrr!!!

I guess I will got back and double clean the plugs/connectors, but looks like I still need more suggestions.

Is there any way to confirm the signal from the tach is making it to the ECM via that single wire connection?

[Peter Florance]

OK here is the latest.

Didn't have time today to get an injector tester, but a friend told me a peanut bulb will do in a bind.

So the verdict according to the peanut bulb is the injectors are NOT pulsing.

I first cleaned the obvious connectors, checked the single lead going to the ECM, and did check the ICM for spark to 0 rpm. That one is a tough call, but it has to be sparking very close to zero.

I decided to pull the ICM apart anyway just cuz. I did find something interesting in the ICM.

Right above the terminal strip for the plug there is a fairly heavy duty looking resistor, which was kind of encased in a ceramic type coating (almost like a ballast resistor). I say this because it has running extremely hot, and the coating is burned off, and it as melted a plastic piece on the terminal strip just below it, and has really baked the circuit board underneath it to the point it is deformed. Is this normal? Seems like an odd place to put a resistor of that size. The other odd thing about it was it had full continuity, no resistence whatsoever. And the circuit board is still OK, in spite of it's melted state (The copper "lines" in it still conduct)

Anyway I did resolder all the conections, put it back together, and guess what. Still doing the same thing. Arrrrrr!!!

I guess I will got back and double clean the plugs/connectors, but looks like I still need more suggestions.

Is there any way to confirm the signal from the tach is making it to the ECM via that single wire connection?

Yes, tach wire is can be checked with standard dwell/tach meter. It's the same wire as negative lead of coil

Do you have +12 on the injectors with key on? ECU also powered by same line. If not, check combo relay and female contacts in it's connector

You may have a bad ecu but that is rare. Good news is there are plenty available used.

[turbo-rampage]

I did check for 12v at the injector connectors, but I will re-check it again tonight. It did have 12 volts present.

I am fairly sure the combo relay is doing everything it should. I actually still have the cover off it so I can visually see the relays pulling in.

I think my next trick will be to run a jumper from the - coil terminal directly to the ECM.

Would a faulty AFM cause this? I did pop off the black cover and clean it with contact cleaner, and I know for sure the micro-switch that sends a signal to the combo relay to run the fuel pump is working for sure.

[Peter Florance]

I did check for 12v at the injector connectors, but I will re-check it again tonight. It did have 12 volts present.

I am fairly sure the combo relay is doing everything it should. I actually still have the cover off it so I can visually see the relays pulling in.

I think my next trick will be to run a jumper from the - coil terminal directly to the ECM.

Would a faulty AFM cause this? I did pop off the black cover and clean it with contact cleaner, and I know for sure the micro-switch that sends a signal to the combo relay to run the fuel pump is working for sure.

No, but I wonder about a bad connection on coolant temp sensor? I've never checked to see if I would get injector pulses with it open

[Lenny D.]

Did you read my post at the bottom of Page 1?

And I cringe at your having used contact cleaner inside the AFM. The wiper board is segmented carbon track, contact cleaner can be very aggressive depending on what you used. Is face still black (that the wiper arm traverses)?
I might hazard a guess you need to swap a known working AFM at this point. But you might also remove the air cleaner assembly, but keep the AFM booted to the throttle body with the connector attached, and push on the AFM door slightly while trying to start, they can wear out in the 'spot' that they live in (idle area) most of their lives.
Stupid question 101 -do you have gas in the car? 'Good' gas - (not 3 months old)?

[Peter Florance]

My turn..

As I've stated here many times, the connections in out cars are the likely suspects generally, before component failure. I had your complaint that turned out to be the female pins of the AFM connector. There is a post about that in the archives, but check the connector for cleanliness, and check that the female sockets stay in the connector housing as you connect it to the AFM. My car would do exactly that, start and die, even at operating temp.

This was fixed in 1981 harnesses but all older cars are suspect.

[turbo-rampage]

Did you read my post at the bottom of Page 1?

And I cringe at your having used contact cleaner inside the AFM. The wiper board is segmented carbon track, contact cleaner can be very aggressive depending on what you used. Is face still black (that the wiper arm traverses)?
I might hazard a guess you need to swap a known working AFM at this point. But you might also remove the air cleaner assembly, but keep the AFM booted to the throttle body with the connector attached, and push on the AFM door slightly while trying to start, they can wear out in the 'spot' that they live in (idle area) most of their lives.
Stupid question 101 -do you have gas in the car? 'Good' gas - (not 3 months old)?

Have you read any of my posts? Top of this page I stated I cleaned all the connectors I could find, including the ICM and AFM plugs. (Based on advise from your post)

Yep the wiper face is still black. I didn't get crazy with the stuff. Just a little squirt and a good blow dry.
Since the car is still doing the exact same thing, I doubt I harmed the AFM anymore than it was. At the time I was trying to figure out why the fuel pump relay was not firing when I opened the AFM door slightly. It turned out to be the combo relay, but I wanted to to clean the micro-switch in there anyway.

I am confident the connectors/plugs in question I cleaned are making good contact. I also verified the grounds on the intake manifold are indeed grounded with a test light and checked all the female plugs to ensure the wire hadn't been pushed partially out of the connector.

Although I doubt 3 month old gas would cause a no start, yes it has a full tank, 1/2 of which is fresh as of this week.

Don't get me wrong, I do appreciate the help.

[Peter Florance]

What I have done in the past, it inject a signal into the ECU tach input lead to see if injectors will fire. One reader followed my advice and connected coil - from a running car to the ecu tach input to test (also joining cars chassis together). I tried to do this with a transformer and diode but you need a pretty good sized signal. I think it would be better to check that junction in the glove box with a tach dwell meter or test light (between +12V and that lead)

[jayne58]

check your fuel filter it might be clogged. At first start the dirt or clog might not be compressed enough to block the passage way thus making it run for a while and die off as it gets pressure.

[Lenny D.]

Top of this page I stated I cleaned all the connectors I could find, including the ICM and AFM plugs.

I reread your post and can't find any reference to your having addressed the AFM connector. That's why I posted what I did, but excuse me for not reading your mind.

Can you verify spark after it tries to start but just cranks? Based on your observation of the ICU component that seems excessively hot, I would try a swap of that. If you can verify consistant spark, you can then focus on fuel. Ballast resistors? They are next to the ICU. Thermo-time switch comes to mind, but if the cold start valve sprays fuel during cranking, it is doing (half) of its intended function.

I might then have to look hard at the ECU if the injectors are not pulsing; the ECU is very reliable, rarely fails, but does. And of course, the connections that are associated with it. I've used a motor-tool wire brush to clean the pins in the large connector in the glove box, but also check the ground point at the rear of the manifold (passenger side). It is the ground for the ECU and the coolant temp. sensor, very critical contact point. Check the connector itself, after thirty years the wires in the crimp will corrode, and you must cut it, expose fresh wire and solder a new terminal. Of, course trace your ground path all the way back to the - terminal on the battery for as close to 0 resistance as possible; the large braided strap is always suspect. And check the connector/lead for the coolant temp. sensor.

At this point you have to eliminate all the obvious possibilities, so go through each system/component and verify their integrity. Not quick, but necessary for reliability, and verifying what actually fixed the problem.

[turbo-rampage]

Lenny,

I try not to make excuses, but here I go again.

Since I am extremely keyboard challenged, I try to balance my posts with concise information, but also try to keep my typing to a minimum. I guess I didn't realize you all were mind readers.

I have trouble believing it to be the ECM also.

I can verify it does have spark after it starts, and also on the way down after it stalls. I think I can safely say the ICM is working OK. Still the (overly) hot resistor is interesting

I did check the ground on the manifold with a test light (I found one near the front and one near the back of the intake manifold) but the point you make about re-doing the terminations is a good one as well as checking actual reisistence.

Ballast resistors check OK in that they have resistence (not an open circuit). Not sure what actual values are supposed to be, but if its got spark all the time I have to assume it is also fine.

The coolant temp sensor I have not done yet, so that is on the list.

I also did run a jumper wire from the negative coil terminal to the single connector on the ECM Still the same start, no run......

I may just invest in a proper injector tester. Not the first time I got bad advise and ended up chasing my tail.

Jayne, can you provide any input on the state of the resistor in the ICM box you fixed? Mine looks like if you touched it it would turn to dust.

[Peter Florance]

resistor is supposed to get hot, If you leave the key on for an hour or so, it will flatten the battery

Ask me how I know.

I'm leaning towards bad afm or bad ecu


There's a dwg in the Ljectronic FAQ that should show typical AFM voltages

http://www.firstfives.org/faq/ljet/ljetronic.html

[turbo-rampage]

I do realize reisitors get hot, but this one looks seriously overheated. As I said the ICM does still seem to be functioning, so I'm going to leave it alone for now.

[turbo-rampage]

I do realize reisitors get hot, but this one looks seriously overheated. As I said the ICM does still seem to be functioning, so I'm going to leave it alone for now.

Oh and thanks for the link, and the continuing advise from everyone. If I was on my own on this the car would be at the crusher by now.

[jayne58]

The ICU when I opened it looks okay and I didnt notice anything like you described. What I had was really just a case of loose soldering verified by testing the continuity and as for the visual condition (resistor) there was totally nothing suspicious. There is a very high possibility there is something wrong with your ICU. In my opinion leaving the ignition switch till the battery dies wont toast it believe me, it happened to me and not once.