starts, then stalls

[Blue90]

I was just helped in finding a fix for my BCM and now I'm addressing another problem: the car, 1990 coupe, starts right up, runs for a couple of seconds or minutes, then stalls.

The fact that it starts right up tells me it's probably not a spark problem.  

The next culprit would be the fuel delivery system, then maybe the security system.

Where's a good place to start?

Thanks,

S.

[Guest Corvanti]

the first place to start, for me - would be to make sure you can hear the fuel pump upon turning the key to the on position. if ok, then check the fuel pump pressure:  http://reattaowner.com/roj/component/content/article/67-engine-a-drivetrain/fuel-system/288-fuel-pressure-testing

 

if the pressure is ok - 40 to 45 psi, let all know for advice to go onto other problems that it may be.

 

a fuel pressure tester can be found here: http://thereattastore.com/diagnostic-tools/

[Mr. Anderson]

I would bet on the fuel pressure regulator. Kerry is right about first checking for fuel pump operation and correct pressure first. You can jump 12 volts directly to the pump to verify operation via the green wire under the hood.

[Blue90]

Thanks for the advice.  I do have a fuel pump and vacuum gauge tester that I've never used before.

I'll check fuel pressure and fuel pump load test, then get back.

S.

[Blue90]

Turns out the pressure tester I have only registers to 10 PSI.

I'm buying one from Ronnie's Store along with a fuel pressure regulator (why not?).

Be back in a few days.

S.

[Blue90]

I hooked up my DVM to the green wire on the driver's side of the engine and got 12.42 volts.

With my new fuel pressure tester I got 43 psi.

I am now waiting for the fuel pressure regulator to arrive; should be here Monday or Tuesday.

The bottom nut on the FPR looks like a bear to get to, any advice (besides not cursing out loud?).

Thanks,

S.

[2seater]

Your fuel pressure is spot on if that is with the key on or hotwiring the pump. If there is no fuel present in the vacuum hose connected to it, replacing it will not change anything. Getting to that bottom nut is a real bear unless you at least partially remove the fuel rail. Frankly, unless I read your fuel pressure test wrong, save the new regulator for another time.

Edited July 19, 2015 by 2seater (see edit history)

[Alfre]

check the security system I have the same problem and the security system is at fault.

Al

[Blue90]

The fuel pressure was tested with the key on (with help from not-a-mechanic Karen).

I put everything back together as I was getting ready to replace the fuel pressure regulator.

I started it up and it stalled right out the first couple of times.

Then it ran very rough for a couple of minutes, before stalling.

When I first went to check for codes i got an "error" display on the dash.

I tried again and got b120c (a Riviera code according to the FSM; b552h (probably from disconnecting and connecting the battery); and e026h (quad driver circuit).

The car runs for a couple of minutes now, very rough, tailpipe gases suggest rich mixture.

Also the "lights on" illuminates on the dash, the parking lights are on front and rear, and will not turn off via the switch.

Getting closer but still confused.

Thanks,

S.

[Blue90]

Al, How did you determine the security system was at fault and what have you done about it?

Thanks,

S.

[2seater]

Try disconnecting the MAF sensor before starting the engine. Did you leave the fuel pressure gauge in place to verify the pump continues to run when the engine is running? Fuel pressure should drop by several psi in response to engine vacuum.

[Blue90]

I just disconnected the MAF sensor; car started right up, ran smooth for a few seconds, then stalled.  Car set E034C code.

I noticed on page 8D-45 of the FSM is a flow chart for troubleshooting the MAF circuit.  I do have extra MAF sensors.

I did not leave the fuel pressure gauge in place while running the engine.  I will try to do that soon.

 

"error" now shows in odometer space.  I checked BCM overrides and found B506, which I had saved at 144; and B507, saved at 228, now read 130 and 0 respectively.  I went and changed them back to specs and they immediately went back to 130 and 0.

 

I'll clear all the codes and disconnect the battery to see what that does.

 

Thanks,

S.

[Mr. Anderson]

Make sure that when you are finished setting the option content numbers that you press and hold the HEATER and DEFROST buttons on the ECC for approx. 3 seconds. This will override the existing numbers stored in the BCM.

Edited July 20, 2015 by Mr. Anderson (see edit history)

[Alfre]

My security lite is on and the starter does not operate so I ran a wire from the starter solenoid to the battery plus terminal to use  briefly and the engine will start and run about 10 seconds and stop. Apparently the ignition switch does not read the key resistor ,at this time I have not repaired it.

Al    

[KDirk]

It sounds like either his new values didn't save or the EEPROM has other problems preventing the data from being retained. The fact the odometer is now reading improperly leads me to think it is a bad EEPROM or bad BCM. That it held the values (it did work) for a day or two and is now screwed up again has me thinking something else is wrong with the module or the chip.

KDirk

Edited July 21, 2015 by KDirk (see edit history)

[Blue90]

Al, I don't think that's my problem, but I will keep it in mind.

It does seem like an EEPROM or a BCM problem.  I can't believe I'm beginning to recognize that now.

Tomorrow I will replace it with a used/tested unit I recently bought.

Thanks,

S.

[Ronnie]

My security lite is on and the starter does not operate so I ran a wire from the starter solenoid to the battery plus terminal to use  briefly and the engine will start and run about 10 seconds and stop. Apparently the ignition switch does not read the key resistor ,at this time I have not repaired it.

Al    

 

It runs for a few seconds because the fuel pump runs for a short period each time the key is turned to the run position.  You are overriding the starter lockout by just jumping to the solenoid but the fuel pump is still locked out by the security system.

 

You should start a new thread about your problem to avoid confusion with the original posters problem. It would be helpful to know the model of your Reatta.

 

EDIT: I might be wrong about the theft deterrent system on the Reatta keeping the fuel pump from running. I've been working on a similar problem with my '99 Blazer and that is how it works but that might not be the case with the Reattas.

Edited July 21, 2015 by Ronnie (see edit history)

[Blue90]

Make sure that when you are finished setting the option content numbers that you press and hold the HEATER and DEFROST buttons on the ECC for approx. 3 seconds. This will override the existing numbers stored in the BCM.

 

I have been doing that.  I notice the number flashes once when holding down the buttons.

Thanks,

S.

[KDirk]

You're doing it right then. If it isn't holding the new values (or overwriting them) after saving then it almost has to be a bad BCM and/or bad EEPROM. The EEPROM is a standard 28C16 and can be bought new (blank) and programmed but it would be difficult to do without a good existing chip to pull the vehicle specific data from. If you really get into a jam and determine the chip (not BCM) is bad, PM me and I'll see if I can program a working chip for you. As of right now, do you get an odometer readout? If that is corrupted you have bigger problems. As well, can you display the VIN in diagnostics? It is also stored in this chip and if it is bogus, no telling what it will show there. Actually, scratch that, yours is a 90 so no way to show the VIN without a scan tool.

KDirk

[Blue90]

The odometer readout shows ERROR.

I pulled out my BCM and replaced it with the used/tested one I have, with its own proms, and all functions tested normal; odometer readout, security light, interior and exterior lights, HVAC system.  I did not have a chance to play with the PROMS to exactly determine what is bad (EEPROM or BCM).  I will try to get to that tomorrow.

With new BCM, car started right up, ran for a few seconds, then stalled.

S.

[Blue90]

Yesterday I got a chance to switch around BCMs and PROMs.

Whenever I used my EEPROM I got ERROR in the odometer readout.

When I used the EEPROM from the used/tested BCM I bought all was normal.

So I've determined my BCM is good but my EEPROM is bad.

I'm hoping Kevin can help me reprogram this chip.

Even with the functioning BCM and PROMs, the car starts right up, runs smooth for a few seconds to a couple of minutes, then stalls.

Any suggestions greatly appreciated.

Thanks,

S.

[KDirk]

Is the working EEPROM out of a 1990 as well? If it is I can probably work from that. I don't own a 90, I have 88's and 91's. 88 is close, but not exact and 91 is quite a bit different. I would probably need both the bad and the good one and would then make a new (third) chip and send all three back rather than trying to save the original for reuse as I suspect the chip itself may be bad and don't want to risk the good chip. I'll PM you with details later today as I'm on my way to work.

KDirk

[padgett]

Just a wild thought but there is an interlock of the fuel pump with the oil pressure sender. This is designed to protect the engine if it loses oil pressure.

 

You said you had 12v on the green fuel pump connector. I'd monitor it and see if it goes to zero just before the engine dies.

[Blue90]

Padgett,

I'll give that a try but right now I'm EEPROM-less.

I sent Kevin my bad chip and the good used one from a '90 BCM I recently acquired.

The only other functioning BCM I have is from a '90 Riviera; I don't even want to go there.

Thanks,

S.

[Blue90]

It's been a long couple of months without being able to drive my car but I believe I'm getting closer to a fix.

To make a long story short:  I sent my eeprom, a good used one I got out of a good used BCM from Jim F. to Kevin.  Along with a new eeprom he reprogramed them to my original chip's specifications (mileage, VIN, options 1 & 2) and sent them back to me.  Almost 3 weeks later Kevin gets the package returned; smashed.  He had his address inside the package or it would never have been returned.  He retested the chips and sent them back to me.

With the good used BCM and reprogramed eeprom the car is no longer doing strange things; however there is still a fuel delivery problem; car starts right up, runs for about a minute, then dies.

I hooked up my DVM to the fuel pump green wire under the hood; with the key on it reads 12.09 volts.  When the car starts it drops down to around 9 volts.  So I'm pretty sure it's not the fuel pump.

I'm trying to replace the oil pressure sender.  The metal part that the wrench fits on is larger than 1" but smaller then 1 1/16"; it's bigger than a 24mm wrench and smaller than a 27mm.  My FLAPS store didn't have any 25mm or 26mm wrenches for me to try.  I thought maybe I could use a deep-well socket but the sender body is too wide, and I still couldn't find a fit.

How is this oil pressure sender replaced?

Thanks,

S.

[Blue90]

I just went to a specialty tool shop and learned some interesting things.  The oil pressure sender is indeed a 1 1/16" using a six-point deep-well socket; with a twelve-point it just spins.  Problem is the six-point socket is 5" deep and won't fit in the space, and it cost $45, non-returnable.

Open-end 1 1/16" wrench is too sloppy; box wrench in that size only comes as twelve-point, also too sloppy.  Story may well be the same with a 26mm but the shop had none to try.

[padgett]

Think I've used a socket from the top and a pair of water pump pliers through the fenderwell.

[Blue90]

I just reread Padgett's post #23, so I went and tested the car again.

With the key on my DVM showed 11.90 volts.

With the car running it showed about 9.0 volts.

When the car stalled, the DVM read 12.45 volts.

If the oil pressure sender was good, the fuel pump volts would have gone to zero, right?

If the OPS is bad, it wouldn't make the car stall, right?

Sometimes the universe works in strange ways.

[padgett]

At this point I'd also monitor the fuel pressure on the rail. Spec is around 40 psi but anything over 20 should idle.That said it could be an ICM heating up and failing. I've seen that before. I always keep a couple of spark plug testers around to check. (embedded HTML seems broke had to embed manually).

There is nothing like having the right tools around particularly if you like oddballs and orphans. Now that I am becoming auld and frail I have enough to outfit a three bay garage including a lft ('vert needs freeze plugs, is there a way to take things loose, remove the rad and just move the engine enough to get to the backside plugs ?)

[Ronnie]

Stanley,

Try this test to determine if the fuel pump is causing your engine to stall.

 

Below are instructions for powering the fuel pump directly from the battery to make it run continuously as long as the wire is connected. Powering the pump directly from the battery will eliminate all the fuel pump controls, (OPS, BCM, ECM, fuel pump relay) from being the cause of the engine stalling

 

If you have a fuel pressure gauge connect it so you will be able to monitor the fuel pressure while the pump is running.

 

With the jumper wire connected, start the engine and watch the fuel pressure gauge to see if the pressure drops when the engine stalls.

 

- If the fuel pressure drops when the engine stalls replace the fuel pump.

- If the engine stalls and the fuel pump maintains pressure the fuel pump is not the problem.

- If the engine continues to run and does not stall, you are one step closer to determining the cause of your problem.

 

Once you have ruled out the fuel pump as causing the engine to stall you can move on to doing other tests to determining what is causing your engine to stall.

 

_______________________________________________________________________

How to power the fuel pump directly from the battery:

 

1. Make certain the ignition key is in the off position.

 

2. Use a length of wire, 14 gauge or larger, to connect the green fuel pump prime/test connector to the positive side of the battery to supply the fuel pump with 12 volts. The pump should begin to run.

 

Having the connectors on the length of wire is is helpful but not necessary as long as you can connect a wire between the green connector and the positive terminal of the battery safely.

 

 

[Blue90]

I was down with a bad (are they ever good?) cold the last few days so I just got around to testing the fuel pump today.

I made a hot wire of 12G speaker wire with an electrical connector at one end for the green connector, and an alligator clip at the other end for the battery connection.  Thank you Ronnie for promoting safety.

I connected the fuel pressure gauge.

I hooked up the wires with the key off:  Fuel pressure read 42 PSI.

Karen-not-a-collector started up the car:  Fuel pressure went to 35, held there steady until car stalled, then went back up to 42 PSI.

So I take the results to mean the fuel pump is not the problem.

From whence do I wander?

S.

[TexasJohn55]

Time for a spark tester and go from there. If spark is good, check to see if fuel injectors are shutting down.

[Ronnie]

Yes, your test proved the fuel pump is capable of maintaining proper fuel pressure for the engine to run without stalling.

 

Just to rule out a fuel pressure problem completely - I would perform the same fuel pressure test again without using the jumper wire.

 If the test results are the same you will have proved the fuel pump and fuel pump controls aren't causing the stalling problem.

 If the fuel pressure drops while doing this test the fuel pump is losing power and we will have to figure out why.

 

Assuming the fuel pressure remains good...

Next, as TexasJohn said, connect a spark tester to one of the plug wires so you can watch it to see if the ignition system loses spark when the engine stalls.

 

Below is a photo if the spark tester I use. Anything similar should work. They are inexpensive at the parts stores.

 

[Blue90]

I performed the same fuel pressure test without using the jumper wire.  Started car up, PSI went to 38.  Car stalled, PSI went to 42.

I checked the spark plug wires with a spark tester.  With the car running the light went on for all 6 wires (some brighter than others).  I held the tester to one wire and when the car stalled the light stayed on.

The wires and plugs have not been changed for a long time.

S.

[Ronnie]

I think you have ruled out fuel pressure as the cause of the engine stalling.

 

If I understand you correctly you have a strong spark at the plugs when the engine stalls. Correct? If you still have a strong spark as the engine stalls that should rule out the ignition system as causing the engine to stall. Usually bad spark plugs and wires won't cause an engine to stop running. They can cause the engine to miss under load but I don't think they would cause the engine to die..

 

The fuel control system is the next thing that should be checked. It was saved for last because it is the most complicated. A few question need to be answered before you start troubleshooting the fuel injection system.

 

1. Are there any ECM codes being displayed in the on-board diagnostics?

 

2. Does the engine start right back up every time after it stalls?

 

3. If you start the engine and hold the gas pedal down to keep it running at 2000 rpm will the engine still stall or will it continue to run as long as you keep it at 2000 rpm?

 

[Blue90]

1)  The on-board diagnostics show; e026h, b552h and no r codes.

 

2)  The engine starts right up every time after it stalls.

 

3)  I started the engine and kept the RPM near 2000.  The engine hesitated a few times but kept running for over 3 minutes; then the RPM gauge went bonkers, bouncing between 500-5000 RPM, went back to 2000 RPM for a few seconds, then the engine died.  This is the longest I've been able to keep the engine running.

Answer to #3 is that it still stalls even with RPM kept at 2000.

S.

[Ronnie]

Wow! That wasn't the result I was expecting from holding the RPM at 2000. I was hoping the engine would run smoothly. Then all you would have to do is clean or replace the Idle Air Controller and it would fix your problem. What you described is more than just a fuel control problem. The fuel injection components and sensors shouldn't effect the tachometer with the exception of the Crankshaft Position Sensor. It could be causing your stalling problem as well as the other problems you describe.

 

The engine hesitation, the erratic tachometer and the code B552 (BCM memory reset) while the engine is running leads me to think you may have a problem with the grounds behind the battery or a bad connection somewhere effecting the electronics. The grounds behind the battery could be loose or dirty. They are isolated grounds for the computers, and other electronic components.

 

I hate to send you on a wild goose chase but I would not continue with trouble shooting this problem or throwing expensive parts at it until you know for certain all the grounds around the battery are clean and tight.

 

After you have the grounds clean, clear all codes and try running the engine again at 2000 RPM, see what happens and then let us know..

Edited October 1, 2015 by Ronnie (see edit history)

[Blue90]

I began the parent of this thread on 14 may 2015 under the heading Dash Electrical Problem.

At that time I disconnected, cleaned, sanded, checked for cracks, reconnected tight with electrical grease the battery, the ground junction block, the ground terminals and the fusible links; all were originally clean and tight.  I checked the vacuum lines for cracks or rotting (I found one end of a T that was rotted but even when repaired had no effect on problem). 

Yesterday I again visually checked all of those connections, all fine.

So I clear the e026h and b552h codes; start the engine.  It stalls immediately.  I try again, manage to get the RPM at ~2000.  Engine runs very rough for almost 2 minutes, then it smooths out and runs sweet for about 10 minutes.  I took my foot off the accelerator and the idle went to1100 RPM, then to about 800 RPM, still smooth.  After about 3 minutes the tach starts flipping from ~500 to ~3500 RPM.  It alternated between the bonkers mode and idling at ~800 RPM for the next 5 minutes and never stalled.  Engine ran for another 5 minutes and then I turned it off.  Checked codes, e026h appeared; I cleared it and went to bed.

Today started up car, stalled immediately.  Started again and tried to keep RPM at ~2000.  Engine ran very rough for about 2 minutes, smoothed out, I took my foot off the accelerator, RPM bounced from ~500 to ~3500 for about a minute, then stalled.

S.

[KDirk]

Stanley,

Is the tach reporting correct engine speed? That is to say, is engine speed actually spiking up when the tach does or is the tach not synchronized with the engine speed when it goes way up momentarily and showing a reading much higher than actual? Should be pretty obvious by the sound of the engine. That 3.8 starts "screaming" at about 4000 Rpms.

I ask because I had an ECM fail on my scrub 88 coupe a while back. Engine wanted to stall out all the time and the tach (when on screen, this is a CRT car remember) would spike to 6500 RPMS and back down repeatedly to "actual" engine speed though the engine never actually went up to 6500 (which is the the top end incidentally).

Given the history of electrical malfunction you've had with this car now (BCM/EEPROM failure) I'm starting to wonder if your ECM got smoked also. That said, being a 90 I presume it still has Magnavox iginition on it. These ignition modules are failure prone - more so than the Delco setup used in 91 - so it may be suspect also. Typical failure mode is leaking of the green slime which causes the module to overheat and self destruct the electonics

New replacements for both versions and their associated coil packs are still readily available and if you are near a you-pick yard they are plentiful on many GM's equipped with the 380p from 88-90 (for the Magnavox type). Whether they are any good is the issue, if you can get one cheap with an exchange warranty then buy a used one to try.

This is all just a suggestion. If you happen to have a spare ECM you might try swapping it in to see what you get. Unfortunately, the diagnostics on these cars misses a lot of powertrain faults and then fails to set any useful codes leaving us to do seat of the pants troubleshooting.

KDirk

[Blue90]

Kevin,

The tach is not reporting the true engine speed.  On the low end, ~500 RPM and up to ~2000 RPM it sounds true; over that it's just the gauge going bonkers, not the engine.

Yes, being a '90 it has a Magnavox ignition; which is clean and dry.

I do have a couple of spare ECMs and Magnavox ignitions (I may even have a Delco out of a '91) that I've collected from u-picks over the years (look good but not tested).

I will try replacing the ECM and see if it has any effect.

S.