Hi- I have a 1989 Pontiac 6000 LE station wagon with a 2.8L V6 with multi port fuel injection. Recently the car died while driving. When I attempt to restart, the engine starts, but instantly dies. Every once in awhile I can get it to run very roughly for a few seconds and then it dies.
There was a trouble code stored (code 45 for rich exhaust) but it's possible it could have been stored a long time ago, long before this current problem.
Eventually I started doing basic troubleshooting for spark and fuel delivery. A spark tester indicated that I had good spark on all cylinders and so I started looking at fuel issues. I connected a fuel pressure gauge to the Schraeder valve on the fuel rail and turned the ignition on to energize the fuel pump. The pressure went right up to proper specs (46 psi), but as soon as the pump stopped running after the 2 second period that ignition on energizes it for, the pressure rapidly dropped down to about 24 psi.
I have a factory shop manual with troubleshooting trees that explains how you can isolate the problem by connecting some special fuel line adapters with shut-offs to the pressure and return lines. I was able to obtain these and did some further testing. In the first step, you energize the fuel pump and then shut off the pressure line before the pressure has a chance to drop. If the pressure then holds, the manual says that this is an indication of a bad check valve in the fuel pump or a leaking coupling hose or pulsator at the fuel pump. If the pressure still doesn't hold the manual indicates that you either have a leaking injector or fuel pressure regulator and tells you a further test to do shutting off the return line to isolate which of those two issues is causing the problem.
When I did the first test (shutting off the pressure line after energizing the pump) the pressure held so this appeared to indicate that my problem was with the pump or it's coupling hose and not with an injector or the fuel pressure regulator. I don't believe that this system has a pulsator.
So I was basically all set to pull the fuel tank and check these things out, but as this is a fairly big job (I've dropped it before for other problems and this time I'd just filled it up with gas to make it even more of a hassle), I started going over in my mind again exactly what the troubleshooting I had done was telling me as I didn't want to drop the tank if the fuel pump wasn't actually the problem.
I started questioning this because although I lose pressure when the pump isn't running, when it is running the pressure is up to specs so I would think that the car should run OK. If the check valve or some other internal leak is causing the pressure to drop only when the fuel pump is not running then this would seem to cause hard starting or increased starting times, but not keep the engine from running OK as long as the fuel pump was running.
To further diagnose things I tried to see if I could get the engine running even briefly to check the fuel pressure while running. I was able to do this once for maybe 5-10 seconds, albeit very roughly, before it died again. The pressure held at just over 40 psi as long as it was running. As soon as the engine stopped, it dropped down again to 24 psi. I can hear what I assume is the fuel pump relay making a click at the exact same time that the pressure begins to drop. This always happens whenever I do the pressure test by just energizing the fuel pump with the engine off as well.
So at any rate, I'm a little hesitant to drop the fuel tank at this point. Shutting off the fuel pressure line appears to hold pressure which is supposedly telling me that I definitely have a problem with or in the vicinity of the fuel pump, but it doesn't seem like it should cause the car to basically not run. There are no signs of any external leakage in the pressure lines anywhere from the fuel tank to the engine including the fuel filter. The fact that the pressure pretty much holds after turning the shut-off closed on the pressure line also supposedly eliminates a leaking fuel injector or pressure regulator as well, so I'm in a bit of a quandry as to what to do next and as I mentioned, a bit hesitant to drop the fuel tank if the problem turns out to be elsewhere even though my tests seem to indicate a fuel pump issue.
I might also mention that to further eliminate the likelihood of a leaking fuel injector, I pulled all six spark plugs, energizing the fuel pump a few times prior to pulling each one and didn't seem to find any sign of raw fuel on the plugs. I also pulled the vacuum hose off the pressure regulator to check for fuel. Apparently there would likely be fuel in that line if the regulator diaphragm were leaking. Again no sign of fuel. But again, I did this testing just as a backup as the fuel rail held pressure when I shutoff the pressure line with the fitting on the adapter hose immediately after pressurizing the system and before the pressure started to drop.
Any advice that you could give me on this issue would be greatly appreciated.
Thanks,
Ted
WoW...There is an intelligent person out there!!! Most posts say "My car...its broke" Looks like you have really done some comprehensive testing. As you observed though, The problem you revealed may have more in common with a delay in starting. My first thought is that though you have pressure etc. Have you tested the fuel for contamination? Water (from condensation etc) can give you similar problems.You mentioned that you just filled the tank I wonder with what? Also how good is the spark, and is it firing when it should? Though I'm inclined to think less of that idea.
To test for fuel contamination remove the fuel line and attach a hose to it. Turn the key to run, and the pump will pump fuel. Let the hose run into a decent sized clean glass jar. if there is water in there it will appear as a layer at the bottom (looks like a thermocline if you have ever scuba dived) Reason I'm not "buying" the pressure drop thing is that It likely would result in delayed start but if there is sufficient pressure at the injectors they really don't care what happens to the rest of the fuel. One thing you could do is unplug the pump and try starting it by spraying carb cleaner into the throttle body...that will give you an idea if the ignition system is working correctly, then you can go back to messing with the pump.
I think you are getting too much fuel in the engine or insufficient spark to ignite it once its in the cylinder. Black plugs can only happen for that reason. Since air flow is pretty much a given (no choke) I would check the TPS (aside from factory test, I also test as you would any potentiometer (pretty much that is what it is) )Also check idle air control valve, egr, and purge system for evap. If canister is full of fuel guess where it goes when running? I'm not going to re-read this but if O2 sensor is showing rich condition ignore it for now...It's likely seeing so much fuel in the exhaust it dosent know what to do with it. Also, don't forget there is an engine under all those sensors. Make sure the timing chain has not stretched and altered cam timing. Late opening valves can do all kinds of stuff.
×
Does your engine have an oil pressure switch? Sometimes these go bad. It is a sending unit which sends a signal to the ECM to allow the fuel pump to remain on. It is for both engine protection and safety in the event of a crash.
However, when the sender goes bad, the pressure built in the initial cranking, which is around 2 seconds when you turn the key on is used up quickly. The engine starts but then immediately dies when the pressure goes away and the ECM thinks there is no oil pressure. Try replacing the sender or bypassing it by consulting an electrical schematic.
SOURCE: air conditioner compressor clutch
be sure the car has a full charge of freon the clutch will not cycle if it is low on pressure please rate -jeff
SOURCE: 94 dakota 5.2 - no spark - auto shut down relay not energized
Hello this is a feature to stop the engine and exhaust system from filling up with gas when spark is lost! years ago fuel would be delivered as usual and you Eng and converter would be full of gas real hazard back then!!!!now they built into computer when spark fails fuel automatically shuts off safety feature??? or not? so i would start to test coil this is a good place to start and if you have engine light on plug in OBD 2 scan tool to retreve the fault hope this helps you out and if you do not have a proper spark tester tool i would invest in one most auto parts stores usually carry them!!! $15-$30 dollars usually.John
1,880 views
Usually answered in minutes!
The thought of contaminated fuel did cross my mind early on, but when I discovered the issue of considerable pressure drop off in the pressure line when the pump isn't running, I assumed that that was what was causing my problem. During all the pressure testing that I've done, I've bled the fuel pressure off of the valve on the pressure gauge a number of times and all I can say is that it looks and smells like gas. I don't really know what I'd be looking for relative to contamination. Would it be noticeable to the eye or nose? And again, even if it is contaminated, what about the pressure drop issue?
Regarding the spark, I used an in line spark tester between each connector and spark plug and the light pulses were bright and regular on all six as I turned over the engine. I also hooked up a test light to each of two wires coming from the ECM to the injectors (another test in one of the troubleshooting trees in the shop manual) and got regular flashes on both. The flashes were not bright, but the manual states specifically that brightness on this test is not important.
Thanks for responding to my post. I don't know if my answers to your questions give you any more ideas. Oh and you better be careful about referring to me as an 'intelligent person'. I might let that go to my head and go tell my wife or something!
The battle goes on. Still no smoking gun....Aaargh!-
I disconnected the fuel line and pumped some gas into a clean jar as you suggested. It seems to look OK. Yellowish tint, very clear and homogenous. No sign of anything that I could identify as a contaminant or any separation or discoloration. Again, I'm not sure if there could be some sort of contamination that wouldn't necessarily be visible to the eye. I did a little research on the internet and found some info stating that because there is an ethanol content in most gas now and because ethanol is water soluble and not lighter than water, it could absorb the water and not necessarily produce a separate layer on the bottom that you could see. Is there any sort of way that you can test the fuel to see if it has water or any other contaminant in it? If it is contaminated, what would you suggest doing. Do I have to drain that full tank of gas or are there additives that I can put in to deal with it. Oh, and when I ran the pump to collect some gas it filled up the container which was probably over a pint in just a few seconds. The Haynes manual that I have says it should put out at least a half pint in no more than 15 seconds or so. It seemed to be putting out at a much higher rate than that. I'm guessing that that is a good thing!
I haven't tried the test with the carburetor cleaner yet. Should I spray a constant stream of it into the throttle body while trying to crank the engine? Is that OK even though the throttle body is just designed to carry air?
I've been looking into a number of other things here and there. I mentioned earlier that I'd pulled all the plugs once after finding the original fuel pressure drop to try to further verify that I didn't have a leaking injector. I didn't mention that I believe all of the plugs on the front bank looked quite black. I'd say the deposits were fairly thin and they were dry, but quite black. I'd replaced the plugs fairly recently and except for the deposits they still look to be in good condition. It didn't cross my mind at the time that they might not be firing. As I mentioned initially when this first happened there was a trouble code (code 45) stored in the ECM which is for 'rich' on the oxygen sensor circuit. Again, for all I know that code could have been stored a year ago. I don't always pull the codes every time the engine light comes on. For the last couple of years the ECM frequently and intermittently stores the code (code 43) for the knock sensor circuit. The shop manual seems to indicate that this is usually relative to a wiring issue. I haven't been able to find one and the engine doesn't seem to be knocking so I haven't worried about it too much, but this is why I don't always go pull the codes every time the engine light comes on. So at any rate, I suppose the dark plugs and the code 45 are an indication that the engine has been running rich, but I'm not sure why. As I mentioned, I hooked up an in-line firing indicator to each plug connector and when I cranked the engine, it seemed to put out bright consistent flashes. I just tried the same test with a pencil type induction firing indicator and got similar results, but I'm not really sure if that actually tells you that the plug is firing. Could the bright consistent flashes be there even if the plug itself isn't actually firing. In other words, I could be getting good output from the ignition coils and through the wires even if the plug wasn't firing. Over the last seven years I've been through two ignition modules on the car, but when they died I had misfiring issues, but the car still would idle fairly well. I happen to have one of the old ignition modules and so as another further test, I decided that I would try substituting it with the current one just to see if the car would idle or if it would make any performance difference. Although the substitute module was basically bad, I remembered that the car would at least idle somewhat decently with it so I thought it was worth a try if for no other reason than to just eliminate one more possibility. I didn't really expect there to be any change and sure enough there wasn't.
One more thing in the possible ignition problem department. Or fuel I suppose. For a few months leading up to this I had been noticing a problem that I thought was transmission related. Even with the age and high mileage of this car, it's always had a lot of power. The transmission (automatic overdrive) has had a sporadic problem with slipping out of gear at a stop or occasionally when I back off the gas while driving for years, but it just happens maybe a few times every several months and hats it. But recently, I'd noticed thatat high speeds on the highway, especially when the engine is under load and usually going uphill, when I push on the gas the power just isn't there. No misfiring, but just lack of power. I finally decided that maybe the transmission just wasn't shifting up into the highest gear as I seem to be having other transmission issues (engine goes up to high speed and clearly transmission isn't shifting as soon as it should. Also sometimes lets out a big clunk when it does) But now I'm beginning to wonder if that high end under load lack of power could be related to the current engine problem? (plugs? fuel delivery?)
So at any rate, I'm getting pretty frustrated and still trying to come up with another list of things to check. I don't think that I mentioned that during some of the few times that I've been able to get the engine to idle at all ( if you could call it that. It feels like it's running on about one cylinder at best and any attempt to push the gas pedal kills it immediately. Even so it won't go for more than a few seconds. Maybe 10-15 at most.) it has stored another code (code 33) for the MAP sensor, but when I read the sections in the shop manual relative to that, it says that a rough, unstable idle can result in this code being stored and I definitely have that. There are other sensors that I wonder about, but in most cases the factory shop manual says that when there is a problem in a particular sensor circuit, the ECM will usually operate the circuit on some sort of default values and so again, as with the fuel pump pressure issue, I'm inclined to wonder whether or not problems in most of the various sensor circuits would produce the severity (basically no start) of symptoms that I'm having. Along with the MAP sensor, some of the other ones that I wonder about are the Throttle Position Sensor, the coolant temperature sensor and the oxygen sensor.
I'm currently wondering a bit more about the throttle position sensor. The manual seems to indicate that it is one of the most important sensors relative to fuel delivery. I had discounted it fairly early on as the shop manual indicates that any problems with it should set one of two codes (Code 21 if the signal voltage is too high or Code 22 if the signal voltage is too low.) However, the shop manual has a section for intermittent problems that also don't necessarily set a code that under possible problems relative to hard starting lists a binding or sticking TPS as a possible cause. Reading this again it crossed my mind that it the TPS were stuck or binding, maybe it might be putting out an incorrect voltage for the actual throttle position, but a voltage that nevertheless was inside the allowable voltage parameters that would thus prevent a trouble code from being stored. Just a thought. I'm looking for any possibility at this point!
Other final thoughts- I'm not sure with this MPFI system exactly how to know for sure that fuel is being delivered to the intake ports. In a carbereted engine I could just pull off the air cleaner, look down the throat of the carburetor, work the throttle a few times, hopefully see the fuel spraying down into the intake manifold and feel fairly confident that fuel was getting into the intake ports of the engine. That doesn't seem quite so easy to figure out here. As I mentioned in my initial query, I did a test mentioned in the shop manual where you hook up a test light to each of two wires that come out of the ECM and go to the injector harness and crank the engine to see if you get flashes. I did get flashes and so that supposedly tells me that I'm getting the necessary electrical signals to the injectors. Now I'm wondering how to be sure that the fuel is actually getting there. Everything in my factory shop manual seems to focus on whether or not the injectors are leaking (stuck open), but is it possible that they could be stuck shut. The manual has a section on testing for injector balance, but it seems to involve having a testing device that you use in conjunction with a pressure gauge. I have the pressure gauge, but do not currently anyways have whatever the other testing device is. And once again, is it likely that if I had a problem with the injectors that I would have more than one faulty injector and if not would that be enough to basically keep the car from starting?
Last, but not least (well maybe least, but I guess it's hard to resist the temptation of using that expression anyways!), I thought that I might check the resistance of the plug wires, although I imagine that if they were bad I probably wouldn't be getting the nice bright flashes on the firing indicators. (Gee, I seem to be becoming an almost cynical pessimist about all of this!
Well, if you've made it this far you're probably wondering why I haven't added a Table of Contents and an index along with extensive footnotes to all of this, but I guess I've got to stop somewhere. So if all this hasn't put you completely to sleep, I'd welcome any further thoughts that you might have. Thanks for your indulgence. Ted
The battle goes on. Still no smoking gun....Aaargh!-
I disconnected the fuel line and pumped some gas into a clean jar as you suggested. It seems to look OK. Yellowish tint, very clear and homogenous. No sign of anything that I could identify as a contaminant or any separation or discoloration. Again, I'm not sure if there could be some sort of contamination that wouldn't necessarily be visible to the eye. I did a little research on the internet and found some info stating that because there is an ethanol content in most gas now and because ethanol is water soluble and not lighter than water, it could absorb the water and not necessarily produce a separate layer on the bottom that you could see. Is there any sort of way that you can test the fuel to see if it has water or any other contaminant in it? If it is contaminated, what would you suggest doing. Do I have to drain that full tank of gas or are there additives that I can put in to deal with it. Oh, and when I ran the pump to collect some gas it filled up the container which was probably over a pint in just a few seconds. The Haynes manual that I have says it should put out at least a half pint in no more than 15 seconds or so. It seemed to be putting out at a much higher rate than that. I'm guessing that that is a good thing!
I haven't tried the test with the carburetor cleaner yet. Should I spray a constant stream of it into the throttle body while trying to crank the engine? Is that OK even though the throttle body is just designed to carry air?
I've been looking into a number of other things here and there. I mentioned earlier that I'd pulled all the plugs once after finding the original fuel pressure drop to try to further verify that I didn't have a leaking injector. I didn't mention that I believe all of the plugs on the front bank looked quite black. I'd say the deposits were fairly thin and they were dry, but quite black. I'd replaced the plugs fairly recently and except for the deposits they still look to be in good condition. It didn't cross my mind at the time that they might not be firing. As I mentioned initially when this first happened there was a trouble code (code 45) stored in the ECM which is for 'rich' on the oxygen sensor circuit. Again, for all I know that code could have been stored a year ago. I don't always pull the codes every time the engine light comes on. For the last couple of years the ECM frequently and intermittently stores the code (code 43) for the knock sensor circuit. The shop manual seems to indicate that this is usually relative to a wiring issue. I haven't been able to find one and the engine doesn't seem to be knocking so I haven't worried about it too much, but this is why I don't always go pull the codes every time the engine light comes on. So at any rate, I suppose the dark plugs and the code 45 are an indication that the engine has been running rich, but I'm not sure why. As I mentioned, I hooked up an in-line firing indicator to each plug connector and when I cranked the engine, it seemed to put out bright consistent flashes. I just tried the same test with a pencil type induction firing indicator and got similar results, but I'm not really sure if that actually tells you that the plug is firing. Could the bright consistent flashes be there even if the plug itself isn't actually firing. In other words, I could be getting good output from the ignition coils and through the wires even if the plug wasn't firing. Over the last seven years I've been through two ignition modules on the car, but when they died I had misfiring issues, but the car still would idle fairly well. I happen to have one of the old ignition modules and so as another further test, I decided that I would try substituting it with the current one just to see if the car would idle or if it would make any performance difference. Although the substitute module was basically bad, I remembered that the car would at least idle somewhat decently with it so I thought it was worth a try if for no other reason than to just eliminate one more possibility. I didn't really expect there to be any change and sure enough there wasn't.
One more thing in the possible ignition problem department. Or fuel I suppose. For a few months leading up to this I had been noticing a problem that I thought was transmission related. Even with the age and high mileage of this car, it's always had a lot of power. The transmission (automatic overdrive) has had a sporadic problem with slipping out of gear at a stop or occasionally when I back off the gas while driving for years, but it just happens maybe a few times every several months and hats it. But recently, I'd noticed thatat high speeds on the highway, especially when the engine is under load and usually going uphill, when I push on the gas the power just isn't there. No misfiring, but just lack of power. I finally decided that maybe the transmission just wasn't shifting up into the highest gear as I seem to be having other transmission issues (engine goes up to high speed and clearly transmission isn't shifting as soon as it should. Also sometimes lets out a big clunk when it does) But now I'm beginning to wonder if that high end under load lack of power could be related to the current engine problem? (plugs? fuel delivery?)
So at any rate, I'm getting pretty frustrated and still trying to come up with another list of things to check. I don't think that I mentioned that during some of the few times that I've been able to get the engine to idle at all ( if you could call it that. It feels like it's running on about one cylinder at best and any attempt to push the gas pedal kills it immediately. Even so it won't go for more than a few seconds. Maybe 10-15 at most.) it has stored another code (code 33) for the MAP sensor, but when I read the sections in the shop manual relative to that, it says that a rough, unstable idle can result in this code being stored and I definitely have that. There are other sensors that I wonder about, but in most cases the factory shop manual says that when there is a problem in a particular sensor circuit, the ECM will usually operate the circuit on some sort of default values and so again, as with the fuel pump pressure issue, I'm inclined to wonder whether or not problems in most of the various sensor circuits would produce the severity (basically no start) of symptoms that I'm having. Along with the MAP sensor, some of the other ones that I wonder about are the Throttle Position Sensor, the coolant temperature sensor and the oxygen sensor.
I'm currently wondering a bit more about the throttle position sensor. The manual seems to indicate that it is one of the most important sensors relative to fuel delivery. I had discounted it fairly early on as the shop manual indicates that any problems with it should set one of two codes (Code 21 if the signal voltage is too high or Code 22 if the signal voltage is too low.) However, the shop manual has a section for intermittent problems that also don't necessarily set a code that under possible problems relative to hard starting lists a binding or sticking TPS as a possible cause. Reading this again it crossed my mind that it the TPS were stuck or binding, maybe it might be putting out an incorrect voltage for the actual throttle position, but a voltage that nevertheless was inside the allowable voltage parameters that would thus prevent a trouble code from being stored. Just a thought. I'm looking for any possibility at this point!
Other final thoughts- I'm not sure with this MPFI system exactly how to know for sure that fuel is being delivered to the intake ports. In a carbereted engine I could just pull off the air cleaner, look down the throat of the carburetor, work the throttle a few times, hopefully see the fuel spraying down into the intake manifold and feel fairly confident that fuel was getting into the intake ports of the engine. That doesn't seem quite so easy to figure out here. As I mentioned in my initial query, I did a test mentioned in the shop manual where you hook up a test light to each of two wires that come out of the ECM and go to the injector harness and crank the engine to see if you get flashes. I did get flashes and so that supposedly tells me that I'm getting the necessary electrical signals to the injectors. Now I'm wondering how to be sure that the fuel is actually getting there. Everything in my factory shop manual seems to focus on whether or not the injectors are leaking (stuck open), but is it possible that they could be stuck shut. The manual has a section on testing for injector balance, but it seems to involve having a testing device that you use in conjunction with a pressure gauge. I have the pressure gauge, but do not currently anyways have whatever the other testing device is. And once again, is it likely that if I had a problem with the injectors that I would have more than one faulty injector and if not would that be enough to basically keep the car from starting?
Last, but not least (well maybe least, but I guess it's hard to resist the temptation of using that expression anyways!), I thought that I might check the resistance of the plug wires, although I imagine that if they were bad I probably wouldn't be getting the nice bright flashes on the firing indicators. (Gee, I seem to be becoming an almost cynical pessimist about all of this!
Well, if you've made it this far you're probably wondering why I haven't added a Table of Contents and an index along with extensive footnotes to all of this, but I guess I've got to stop somewhere. So if all this hasn't put you completely to sleep, I'd welcome any further thoughts that you might have. Thanks for your indulgence. Ted
×