Fuel system mystery

My 1990 cabrio has been a bit tempermental of late with faltering when cold and occasional hard starting when warm. I had put injector cleaner in to try and help but with less and less success the past few months. I finally decided to put in a fresh set of clean injectors thinking that would solve the problem like it had some time ago when I had this come up before. Well it did not make any difference. Still falters on acceleration off idle the first two or three minutes in the morning and is hard to start when warm more often now. I have been reading the Bently trying to see what to do next. It suggests fuel pressure, temp sensor and that air measureing flapper thing.

Any suggestions as to the most likely culprit? My ohmmeter thinks the temp sensor is about right but I need to make some measurement when cold to be sure. I will have to invent something to check the fuel pressure. From what I can remember of the air meter, there is not a lot to be done to verify it works correctly. That should not be a starting problem though.
Any suggestions?

Bob in Lake Havasu

From what you wrtoe, you had checked the injectors, the temperature sensor and injectors get fuel.

Have you checked the ignition? Spark plugs, cables, the works.

Isn't there a fuel subsystem that enriches the mixture when cold? There is a coolant temp sensor which feeds info to the ecu enabling the system?

alan

I changed the injectors to a set that had been through the cleaning shop. Checked the temp sensor when the enging was warm and again when hot. The resistance appears normal.
Today I checked the fuel pressure regulator pressure. It is 45 psi.

The spark plugs are recently changed. The distributer cap and rotor were all cleaned 18 months ago when I had them off for timing belt replacement. I suppose that is another potential for problems. Today it was better as it was a warmer day.

Bob in Lake Havasu

Jeeeez Bob, I wish I could point you in the right direction, because mine has the same hard starting issue when hot, when cold, starts no problem, once she is hot and stood a while, she battles to start, like she is flooding, I have to hold my foot flat on the accelerator and just jeep cranking until she fires up.
What I can do is eliminate the airflow meter, mine was stolen ff my car a few weeks ago, I have since replaced it with a good second hand one from a breakers yard, the hard starting is still there, so rest assured knowing it is not your AFM.
Your fuel pressure seems to be spot on.
No Alan, there is no enrichment jet on the Bosch system, the cold start is controlled by the temp sensor, which feeds a voltage to the ECU, telling the ECU, the car is still cold, the ECU then adjusts the injector dwell time to allow more fuel to be squirted into the motor, basicaqlly the mixture strength is controlled by the amount of time the injectors are held open, when the motor is cold, the injector is just held open longer.
Bob, I have also wondered if it is not my injectors, but you say they have been cleaned, maybe cleaning is not enough, they may be worn and leaking.
First prize would be to have them checked for leak down under pressure, I guess a simple way to check if they are leaking would be to drive the car a fair distance to bring it up to temp, then when you get home, disconnect the fuel delivery pipe to the injector rail, thus relaesing the pressure in the system, reattach pipe.
Then wait, ten minutes or so and start the car, if it fires right up, your injectors are leaking down, flooding your intake with fuel as the car stands, if it still battles to start, then injectors are not the problem.
Hope that makes some sort of sense.

----------------------------------------------------------------------------------------------------------------------------------
E30'S AREN'T BUILT, THEY'RE CAUGHT IN THE WILD!!!



When in doubt, use full throttle,
it may not improve the situation, but it will end the suspence.

In the past, a clean set of injectors cured this problem for a long time. Even adding some injector cleaner to the fuel helped a lot. This time however the fuel injector cleaner did not make much difference and changing the injectors that had been to the shop for cleaning and testing did not change the problem. That is why I am looking at all the other potential problems. I checked the leak down when I checked the pressure on the regulator, no apparent leakage. So, if the AFM is not the issue and the temp sensor is not the issue ( I did check the resistance of it) and the injectors are clean, what else could it be? I already removed the idle control valve and gave it a spray of cleaner even though it appeared to function correctly.

I have run out of ideas.

Had two bouts of hard starting when warm today.

Bob in Lake Havasu

Mmmmmm, it is frustrating, I know, as I said, I have the same problem, I am sure it is temp related somehow, just wondering, although you said you have checked the values on the temp sender unit, (200 ohms cold, 3000 ohms hot are the values you should be getting) maybe for sh1ts and giggles, next time you have a hot start coming up, disconect the plug off the temp sender then try and start and see what happens, the ECU should use a default value if it does not recieve a signal from the temp sensor, thus eliminating it for sure.
If it fires right up, you know it is giving the ECU duff gen.
Not sure how easy yours is to get to, may take some planning, I would remove the spring clip prior to the trip, so it is easy to pull off and maybe carry a long nose pliers in the car, or whatever.

----------------------------------------------------------------------------------------------------------------------------------
E30'S AREN'T BUILT, THEY'RE CAUGHT IN THE WILD!!!



When in doubt, use full throttle,
it may not improve the situation, but it will end the suspence.

I have considered doing what you suggest but was not sure what value the computer would use. The stumbling when cold is also annoying as there is a stop sign 40 yards from my drive way and I usually have a time getting it going without dying there. I could just disconnect it for a day and see what it would do.
The resistance when hot (180F) 240 ohms, and cooler (129F) 500 ohms, are correct. I have not checked it completely cold, that should be about 3000 ohms.

Bob in Lake Havasu

Quote
Bob in Everett
My 1990 cabrio has been a bit tempermental of late with faltering when cold and occasional hard starting when warm. I had put injector cleaner in to try and help but with less and less success the past few months. I finally decided to put in a fresh set of clean injectors thinking that would solve the problem like it had some time ago when I had this come up before. Well it did not make any difference. Still falters on acceleration off idle the first two or three minutes in the morning and is hard to start when warm more often now. I have been reading the Bently trying to see what to do next. It suggests fuel pressure, temp sensor and that air measureing flapper thing.

Any suggestions as to the most likely culprit? My ohmmeter thinks the temp sensor is about right but I need to make some measurement when cold to be sure. I will have to invent something to check the fuel pressure. From what I can remember of the air meter, there is not a lot to be done to verify it works correctly. That should not be a starting problem though.
Any suggestions?

Hey Bob, It sounds like you have a history with this car as to its fuel issues. From the faults you list It does sound like something in the fuel system.

One word of caution, you might be chasing your tail. Is there a competent motronic mechanic around you. It might pay to have a pro sit down with your car for an hour or so.

I've had a few cars here that I decided to hand off and it was money well spent. Of course that's if you have competent people you can go to....

A few questions on your condition;

1) Did this start all of a suden or did these conditions gradually happen over time
2) Mileage?
3) What happens when it's cold and it starts to die, and if it does die is it an easy start
4) On hot starts does the motor seem flooded, and does it take a little while to clear itself before she starts to run even.

Cheers, Rick

Just a thought, what are your compression numbers like?
Mine are on the low side, and I am wondering if it has not got something to do with it.
1. 975Kpa
2. 950Kpa
3. 925Kpa
4. 925Kpa

----------------------------------------------------------------------------------------------------------------------------------
E30'S AREN'T BUILT, THEY'RE CAUGHT IN THE WILD!!!



When in doubt, use full throttle,
it may not improve the situation, but it will end the suspence.

1) Did this start all of a suden or did these conditions gradually happen over time

It has been getting gradually worse for about a year. Injector cleaner helped at first but now seems to make no difference.
2) Mileage?
The car has 224,000 miles now. I replaced the idle control at about 120,000 thinking it was not working but it was. Changed the injectors then and it ran well for quite a while. When I changed the timing belt the last two times, I washed and cleaned the dist cap and rotor very well. so those have about 100,000 miles on them now. I changed the igntion wires but that made no difference when I had this problem in the past. So I put the new set aside to reinstall the old set once the problem was solved by cleaning the injectors.

3) What happens when it's cold and it starts to die, and if it does die is it an easy start
Most often it falters and I manage to keep it running by opening the throttle a bit. When it does die, it seems to be flooded.

4) On hot starts does the motor seem flooded, and does it take a little while to clear itself before she starts to run even.
Yes, it acts like it is flooded. Cranks and cranks and eventually catches weaky then will accelerate rapidly. Still will falter a little until it runs for a minute.

Once started cold it will idle fine and if allowed to warm up a bit it will drive normally but takes a couple minutes from dead cold. Gives me time to scrape the frost off the windows. If only wet, I want to drive away immediately but have to keep the revs up or it will die for the first minute or so.

Bob in Lake Havasu

Quote
Flyboy
Just a thought, what are your compression numbers like?
Mine are on the low side, and I am wondering if it has not got something to do with it.
1. 975Kpa
2. 950Kpa
3. 925Kpa
4. 925Kpa

The compression checks last were pretty consistent but have not done one in a while. Generally around 170 psi. The head has probably never been off the block. The valves were adjusted a few months ago. Only a couple needed to be adjusted. Most of my driving is short trips with little steady speed although on Saturday I did take it for a good run on the highway for about 100 miles each way. Did not appear to be helpful though.

Bob in Lake Havasu

Bob, what if the ECU is not getting the signal from the temp sensor? The ECU would think the engine is fine and no need for enrichment? Or possibly an issue with the ECU itself? Worth checking continuity and resistance values at the ECU connector?

alan

Have you tried unplugging the O2 sensor to see if the Motronic defaults make a difference? Sometimes a faulty O2 sensor can mask problems or simply induce them...

I don't think the 02 sensor or the temp sensor will have those affects on the way Bobs car is running though, I've had both and from my experince (I hate this computer, spell check does not work on this site) those were not the simptoms.

Bob, A few more questions

Your answers to my questions really does seem like the car is flooding on hot starts and when it stalls cold. There should be a way to prove out the spray pattern on the injectors (Peter, any help here?).

When I test fuel pressures I do a complete run test, even on the street, and also a leak down over night and a few others but the gauge always stays right against the window, hooked in to the wiper through out. I don't believe just a run test (while the car sits there) is enough.

1) when the cars starts dead cold does do anything out of the normal
2) when the car hot starts does it go to a high idle or do anything strange
3) does the car ever run right, like within a days outting
4) have you had the AFM apart, or the throttle body

Rick

That is an interesting possibility. Although the fuel consumption does appear to change with engine temp. When cold it takes more fuel on a particular road I drive than when it is warm I am pretty sure. I could check the wire harness. I better check those wiring diagrams for the pin numbers.

That brings to mind another possibility too. I have an after market chip in the ECU. Maybe it lost part of its mind. I could put the stock one back and do a few sorties.

The one thing I have not checked is the oxygen sensor. Not sure how it would affect starting when warm.

Bob in Lake Havasu

Thanks for all the feed back guys.
When the engine is dead cold it starts right up and the idle is high for a second or so then settles to normal. I had the idle control valve off and it worked freely but I put some ether in it to flush out anything that might be there. So not much unusual there. No change afterward.
On warm starts if it does start, it seems quite normal. If it does not, I hold the throttle wide open and crank for several seconds. It weakly starts to catch and I leave the starter ingaged until it suddenly revs very quickly. have to get out of it or it will really go too far. Then it still idles rough for a few seconds. once past all this trauma, it runs beautifully.
I have never messed with the AFM. I have not even looked in there in months. not since I changed the air filter some time ago.

Bob in Lake Havasu

Quote
rkj
I don't think the 02 sensor or the temp sensor will have those affects on the way Bobs car is running though, I've had both and from my experince (I hate this computer, spell check does not work on this site) those were not the simptoms.

Bob, A few more questions

Your answers to my questions really does seem like the car is flooding on hot starts and when it stalls cold. There should be a way to prove out the spray pattern on the injectors (Peter, any help here?).

When I test fuel pressures I do a complete run test, even on the street, and also a leak down over night and a few others but the gauge always stays right against the window, hooked in to the wiper through out. I don't believe just a run test (while the car sits there) is enough.

1) when the cars starts dead cold does do anything out of the normal
2) when the car hot starts does it go to a high idle or do anything strange
3) does the car ever run right, like within a days outting
4) have you had the AFM apart, or the throttle body

Rick

I've been thinking about this idea that the car is flooding when hot. This is actually counter intuitive when you consider that the way most of us have cleared it is to hold the throttle to the floor while cranking until it catches.
Think about it for a moment. When we crank the engine over; it can't be turning a very many rpm at all; definitely not enough to open the AFM flap more than a slight gap. Now think about what happens when we push the throttle to the floor...it opens the throttle body switch to WOT which tells the ECU to give maximum duration to the injectors for maximum fuel flow.
So, if the engine was already flooded; it should not start because firstly, we are dumping more fuel in it and not actually sucking any more air in to dry things out.

When you think about it this way, you will see that the engine is actually starved of fuel until we give it WOT to dump a big charge in there.

I had the hot start issue a few times on the iX and I always looked in the rear view mirror to see if there was a black cloud of smoke when it finally fired up...never a wisp. If the engine was flooded; it would have produced a nice visible dirty black cloud!

With that in mind; I would like to suggest that the situation is not one of flooding but rather of fuel starvation.

How would that happen? Well, many years ago there was a Datsun 510 model that suffered a similar situation only we called it vapour lock back then. Basically the Datsun fuel line ran too close to the block and when the car was hot and turned off; the heat cause an air bubble to form in the fuel line right in front of the fuel filter which sat near the same area as the FPR on the M20.
The 510 has a mechanical fuel pump that was not in a closed loop and was feeble compared to the modern electric fuel pumps; so it was not able to clear the vapour lock.

I am guessing something similar happens to the M20 fuel lines when the car is turned off hot. Having the closed system with high capacity fuel pumps and Fuel Pressure Regulators; it is likely that any vapour locks or air bubbles would quickly be cleared.

I say this because we know that holding the throttle to WOT(instructing the ECU to dump fuel in) does eventually get bring the engine to life and quickly.

It is also my experience that it's pretty difficult to flood a hot engine.

Maybe we have been looking at this problem the wrong way?

BTW; I don't believe to cold start/idle issue is related to the hot start one...unless the FPR is faulty.

You bring up some good points Peter. The WOT is what I am not sure of. Would that tell the computer to turn on the injectors if the engine was not seen as running?

I put a pressure gage on the fuel return line while I was at the regulator test effort. When the fuel pump is running, there is 15 psi at the regulator on the return line. This means that fuel is flowing back to the tank. As soon as the pump is off, the pressure falls to zero. The high pressure side of the regulator stays up at pressure and had to be released into a container so I could get the gage out of the line.
I have not noticed a cloud of black smoke either.

I suppose I could clean my distributer and rotor to see if that makes any difference. I have not checked the oxygen sensor but that should not affect the starting since it would be cold anyway. (I think).

In the past, when I have had this problem, putting clean injectors in cured it for quite a while. This time not.

Another suggestion was to check the wire harness for broken conductors in the temp sensor circuit. Have not done that yet either. I need to find the pinout diagram for the ECU.

Bob in Lake Havasu

The ECU uses input from several sources...engine RPM(slow at cranking speeds); temp sensor input; #1 plug wire sensor; the AFM; the TPS and the O2 sensor.
With all of these sensor inputs being in a closed loop; any one of them can cause problems.

The reason I'm suggesting the fuel starvation rather than flooding is because the ECU will ignore most of the other inputs when it sees the WOT signal. It is designed to marginally increase the fuel/air ration for a richer mix. So; if there isn't a puff or cloud of black smoke, it indicates there was a lack of fuel rather than too much. The ECU can open the injectors but if there is no fuel to pass thru them; not much will happen.

I no longer have my Bentley or Probst manuals but it is all very nicely described in those two publications.

I will have to look again in the Bentley. Thought I had understood it. In the past, a new or freshened set of injectors completely cured the problem. This time it made almost no difference.

Bob in Lake Havasu

Any possibility the new injectors are deficient?

alan

You know, I have considered that but have not figured out how to tell. My thought is that they would not be identically bad. The new ones had been to the Witchhunter cleanning shop and came back with a print out of performance that said they were within specs. Not that it could not happen.

Bob in Lake Havasu

Quote
Archeo-peteriX

Quote
rkj
I don't think the 02 sensor or the temp sensor will have those affects on the way Bobs car is running though, I've had both and from my experince (I hate this computer, spell check does not work on this site) those were not the simptoms.

Bob, A few more questions

Your answers to my questions really does seem like the car is flooding on hot starts and when it stalls cold. There should be a way to prove out the spray pattern on the injectors (Peter, any help here?).

When I test fuel pressures I do a complete run test, even on the street, and also a leak down over night and a few others but the gauge always stays right against the window, hooked in to the wiper through out. I don't believe just a run test (while the car sits there) is enough.

1) when the cars starts dead cold does do anything out of the normal
2) when the car hot starts does it go to a high idle or do anything strange
3) does the car ever run right, like within a days outting
4) have you had the AFM apart, or the throttle body

Rick

I've been thinking about this idea that the car is flooding when hot. This is actually counter intuitive when you consider that the way most of us have cleared it is to hold the throttle to the floor while cranking until it catches.
Think about it for a moment. When we crank the engine over; it can't be turning a very many rpm at all; definitely not enough to open the AFM flap more than a slight gap. Now think about what happens when we push the throttle to the floor...it opens the throttle body switch to WOT which tells the ECU to give maximum duration to the injectors for maximum fuel flow.
So, if the engine was already flooded; it should not start because firstly, we are dumping more fuel in it and not actually sucking any more air in to dry things out.

When you think about it this way, you will see that the engine is actually starved of fuel until we give it WOT to dump a big charge in there.

I had the hot start issue a few times on the iX and I always looked in the rear view mirror to see if there was a black cloud of smoke when it finally fired up...never a wisp. If the engine was flooded; it would have produced a nice visible dirty black cloud!

With that in mind; I would like to suggest that the situation is not one of flooding but rather of fuel starvation.

How would that happen? Well, many years ago there was a Datsun 510 model that suffered a similar situation only we called it vapour lock back then. Basically the Datsun fuel line ran too close to the block and when the car was hot and turned off; the heat cause an air bubble to form in the fuel line right in front of the fuel filter which sat near the same area as the FPR on the M20.
The 510 has a mechanical fuel pump that was not in a closed loop and was feeble compared to the modern electric fuel pumps; so it was not able to clear the vapour lock.

I am guessing something similar happens to the M20 fuel lines when the car is turned off hot. Having the closed system with high capacity fuel pumps and Fuel Pressure Regulators; it is likely that any vapour locks or air bubbles would quickly be cleared.

I say this because we know that holding the throttle to WOT(instructing the ECU to dump fuel in) does eventually get bring the engine to life and quickly.

It is also my experience that it's pretty difficult to flood a hot engine.

Maybe we have been looking at this problem the wrong way?

BTW; I don't believe to cold start/idle issue is related to the hot start one...unless the FPR is faulty.

Oh boy, there's so much to talk about here!

First off, my car (245k) lately has the old hot start flooding because it carries it's original injectors and getting it to start always works better if you hold the gas pedal down but I never see any smoke out the back so what that means is anybody guess... just sayin. Dinner ready, later, Rick

I am so happy that I have given everyone something to think about

Maybe I will have planted a seed that will blossom forth in one of your minds and supply the real answer

Quote
Archeo-peteriX
I am so happy that I have given everyone something to think about

Maybe I will have planted a seed that will blossom forth in one of your minds and supply the real answer

So..... It's been my experince that with all the E30's in my life, both mine and friends cars, none have suffered vapor lock. In fact, anything but. I think there's an answer to Bobs problem, we just haven't found it yet!

Bob, have you ever taken that barrel plug apart under the intake manifold? I did mine twice so far and was amazed what I found in there. Lots of chaulky white stuff, in fact it was quite amazing anything was transfering signals at all- Not only that but there were a few vacuume lines off the canister and valve by that plug. I also decided to replace the throttle postion switch rather than just clean it, with 230k at the time it was ready to get tossed. A new set of wires and a decent cap put my car back to running perfectly.

I have not had any experince with cleaned injectors but I have replaced a few sets in my 2002's and I think after a certain amount of mileage those poor things just wear out. In my world they are expendables....

I'll keep thinking, Rick

Thanks for the thoughts Rick,
I have had that electrical connector apart, I took it apart to get the injector harness out of the way. The pins and scockets seemed to be clean.
Vacuum leaks from hoses in that area might be a good possibility. Have not changed any of those hoses. I will give them a good going over with the engine running to see if there is a bad one. One test (if you can call it that) I did was to remove the oil filler cap with the engine running to see if it made any difference to have a change in the unmetered air. It acted like it has in the past with some change to idle speed and then corrected. When put back on, same correction.

Bob in Lake Havasu

Quote
Bob in Everett
Thanks for the thoughts Rick,
I have had that electrical connector apart, I took it apart to get the injector harness out of the way. The pins and scockets seemed to be clean.
Vacuum leaks from hoses in that area might be a good possibility. Have not changed any of those hoses. I will give them a good going over with the engine running to see if there is a bad one. One test (if you can call it that) I did was to remove the oil filler cap with the engine running to see if it made any difference to have a change in the unmetered air. It acted like it has in the past with some change to idle speed and then corrected. When put back on, same correction.

Those vacuume lines are a bit hidden, I'm pretty sure other things have to be removed to get at them.

A poorly fitting oil dipstick will cause havoc as well (ask me why I know this ) It will let unmetered air into the system.

alan

Yes, I have checked that too. This has been getting worse for quite a few months. The last three days when cold starting, it fires up more quickly than normal. Like it had fuel available in the manifold already. The warm starts were not so bad but then this evening after setting for two hours, it had to be cranked for several seconds to catch.

Still a mystery. I might try another set of injectors just for the practice.

Bob in Lake Havasu