: Fuel cuts off



gsim
08-19-11, 08:20 AM
Have an '84 Sedan DeVille with EFI/throttle body injection.

Will not start, and fuel line dry at throttle body. Filter good, pump working. Trouble is that the 12 volts to the pump only stays on for one second each time the key is turned ON. The fuel pump relay is good, but I swapped it with the horn relay (same type and # relay) anyway, and the trouble persisted. (Horn relay was a good one and horn works fine no matter which relay is plugged into it's socket)

Whatever the source of the signal to that relay, it is only staying ON for one second, then shuts off. I only have a Haynes manual (I know, they suck). So I do not know the source of the signal to the fuel pump relay.

Any ideas??

Gerry

mjs182004
08-19-11, 05:56 PM
Maybe a clog on the fuel line?

csbuckn
08-19-11, 08:09 PM
So no power to the pump while cranking? Fuel come out the filter when cranking?

sven914
08-20-11, 01:11 AM
There might be an issue with the computer not registering a tach signal. The pump primes the lines like it should when the ignition first turns on, but it isn't receiving the signal from the engine cranking.

If your computer is not getting a tach signal, it will register DTC 12. You can retrieve DTC's by turning the ignition to RUN and holding down the OFF and WARMER buttons on the ECC. The computer goes into diagnostic mode when "88" is displayed on the ECC. After that, any stored DTC's will start scrolling on the display.

Code 12 means the computer is not registering the engine is turning over

If it turns out to be no tach signal, you will probably just have to replace the computer wire to the distributor (they come lose, break, or the contacts get all cruded up). Some other things that may cause the fuel pump to not work are high resistance in the pump motor or a faulty ground connection. The car's computer may be able to tell you if the pump is the cause, but the computer doesn't register every fault as it should.

Code 19 means there is high resistance in the fuel pump circuit

Code 20 means that the fuel pump is disconnected/ not grounded

brougham
08-21-11, 02:57 AM
Even with no distributor signal it should still get fuel to the throttle body, it just wont fire the injectors.

jayoldschool
08-21-11, 10:15 AM
Do the 4100 cars have a low oil pressure fuel cut off?

gsim
08-21-11, 10:54 AM
Fuel line and filter tests clear.

gsim
08-21-11, 10:58 AM
Power to pump for about 1 second and cuts off. (One second does not give time for fuel to get as far as the filter. If I wire 12v to the pump it pumps fuel fine thru the filter. Am certain can make car run this way, but not safely.) Fuel pump relay clicks two times within one second, and I think that is the power on-power off situation I read with VM at the fuel pump wiring harness.

gsim
08-21-11, 11:10 AM
Sven 914,

Thank you for input. I will copy down all info you gave me. I was unaware of the possibility of any contacts in play between the computer and the distributor. Of course the Haynes manual has little to nothing to offer in the way of info in that regard. (Will look thru it anyway in case it is in the ignition wiring part of that book.)

I am guessing that feature is the safety feature also which shuts off the fuel in the event of a wreck? Whatever feeds that fuel pump relay is only operating it for about one second. At this point, I have been grateful not to have to remove the fuel tank for replacement of the pump. LOL!

I need to find out what that lead looks like and where it attaches to the distributor. It is a Coil-In-Cap distributor. Do you know if the contacts you mentioned are inside of the distributor or elsewhere? Also, do you know the location of the computer itself? Do you know the place where I would attach a code reader? I assume in the vicinity of driver's-side fender well?

Thanks for all your help,

Gerry

gsim
08-21-11, 11:26 AM
I do not know if the 4100 cars have low oil pressure cutoff. Mileage is around 117,000 and oil level good. Have not seen oil idiot light come on.

csbuckn
08-21-11, 12:24 PM
Do the 4100 cars have a low oil pressure fuel cut off?

I think you should probe the wires on the back of the oil pressure switch to see if you have power going in and out. If the sensor is bad or reading low pressure, it will tell the computer not to start the car.

brougham
08-21-11, 03:34 PM
Low oil pressure switch is unlikely to be the problem. There should already be fuel pressure before the engine even starts to turn over.

csbuckn
08-21-11, 05:52 PM
I think he says it primes when he turns it on.

sven914
08-21-11, 06:28 PM
I have heard that the 4100 used an oil pressure switch to disengage the pump (but I doubt it). There is no DTC for the oil pressure switch, which makes me think it is not computer controlled, and therefore is not used in disengaging the fuel pump. There are codes (17 & 18), for a crank shaft position sensor, which would definitely be usable in determining if the engine is running and in disengaging the pump.

Looking back, it does make more sense to have a crank sensor or an oil pressure switch to tell the computer to turn on the fuel pump, and to use the tach pulse for the injectors (like Brougham said).

jayoldschool
08-21-11, 07:25 PM
I know that on the later cars, if the oil pressure switch is the problem, the car will start, but then die about three seconds later. Not the case here, it seems.

csbuckn
08-21-11, 07:40 PM
I don't think it would have a code for oil pressure switch. It would be in line on the power wire to the pump...? Jay, I wanna say my 90 didn't start when it was unplugged.

brougham
08-21-11, 09:58 PM
I think he says it primes when he turns it on.

It sounds like it tries to but he said the fuel line is dry at the throttle body. So something is stopping it from getting that far. If the problem is really that the injectors aren't firing a bad ignition control module can cause that. On a lot of GMs from that time that's what tells the computer when to fire the injectors. I don't know if it works that way on the 4.1 tho. And it won't necessarily have a code either.

I always thought that aything that forces the engine to shut down like oil pressure kills the ignition, not starves it of fuel. I don't think they had that back then either, or even now?

jayoldschool
08-21-11, 11:40 PM
That's the way it works on the LT1 cars. No oil pressure after start, the power to the fuel pump is cut.

gsim
08-22-11, 08:41 AM
Oil pressure would be zero to very low initially, so I cannot see why or how they could make that to turn pump off after only one second of cranking. It is literally on-off duration of one second or less.

I will be reading the codes stored in the computer after I charge the battery this AM. It is dead, having sat in the garage for over 2 weeks. Will post the codes I see when I do that today. Found all low-voltage wiring going to distributor. 3 wires go from cap to distributor body and are connected to one polarized plug. There is one fat red one that goes from cap to somewhere else. All contacts inside of the connector caps on these low voltage wires appear to be very clean. I applied a light dab of vaseline petroleum jelly to each one and wiped the plug clean so as to insure that there would be no short circuit. I reconnected those and will try the car again when the battery has recharged.

Thanks to all of you so far for input and ideas. Have been letting this car sit because have had an ongoing job installing a 44 ft culvert across driveway near house to handle excess rainwater. So will soon be able to stay on my Cadillac trouble. OH, I do have a brake warning light that has been on for 8 months. I do not have ABS and I can find no pressure switch on Mast cylinder or on brake lines anywhere. Only see a hand brake switch and brake light stays on when I disconnect that. I do not see how it could be relevant when brakes work so well.

Gerry

gsim
08-22-11, 08:48 AM
It sounds like it tries to but he said the fuel line is dry at the throttle body. So something is stopping it from getting that far. If the problem is really that the injectors aren't firing a bad ignition control module can cause that. On a lot of GMs from that time that's what tells the computer when to fire the injectors. I don't know if it works that way on the 4.1 tho. And it won't necessarily have a code either.

I always thought that aything that forces the engine to shut down like oil pressure kills the ignition, not starves it of fuel. I don't think they had that back then either, or even now?

Only oil pr fail shut offs I have ever seen were the type that killed ignition on a gasoline engine. Trouble is that power is only sent to pump for one second and then turned off, whether trying to crank or not. Can gerry-rig it to run, but then would be unsafe to drive. No way to shut fuel off if it was in a wreck and I was unconscious.

Gerry

gsim
08-22-11, 09:36 PM
Sven 914 and others,

I just read code on computer as per Sven advice. It reads as follows: -1.8.8 52 and it repeats those two twice more, then it reads 7.0 repeatedly and no other numbers. ( Turned ign off and repeated and it was same. ) Anyone know what that means?

Gerry

sven914
08-22-11, 11:48 PM
The only thing coming up is code 52, which is ECM Memory reset. When the battery died, the computer's memory got erased and all of the codes were deleted.

Try starting it a few times while the battery is fresh, and then see if any codes sets.

gsim
08-23-11, 07:22 AM
Oil pressure is always zero before starting car, so I fail to see how a bad connection there could cause a no-start problem.

gsim
08-23-11, 07:31 AM
The only thing coming up is code 52, which is ECM Memory reset. When the battery died, the computer's memory got erased and all of the codes were deleted.

Try starting it a few times while the battery is fresh, and then see if any codes sets.
Thanks, Sven. Will try that. All I did was turn on sw and read codes. Who knows, maybe removing low voltage wires to distributor and re-connecting them with a tiny dab of petroleum jelly on each one may even have cleared it. (really tho, I am usually not quite THAT lucky)

Gerry

gsim
08-23-11, 11:57 AM
OK, I just did try to crank/start it several times. Same thing. Fuel pump runs for one second after key is turned to run whether I try to start it or not. Read codes and are same. I see -1.8.8 52 three times. Then I see 7.0 repeated as long as I hold buttons down. Seems that the codes deleted via the dead battery took whatever codes that were stored there, and that this trouble does not generate a code.

Gerry

sven914
08-23-11, 02:34 PM
So the computer is not registering anything issues with the fuel pump, the distributor, or crank sensor.

My next step would be to test the oil pressure switch. Yes there is no oil pressure before the engine is cranking, but the oil pressure builds as the engine starts to turn over. Some '90's GM cars use the oil pressure as a backup to the crank sensor (for when the crank sensor fails). With those systems, once enough pressure is built up, a signal is sent out by the pressure switch, and the fuel pump is allowed to turn on. It takes less than a second for a cranking engine to build oil pressure.

brougham
08-23-11, 08:42 PM
Oil pressure is always zero before starting car, so I fail to see how a bad connection there could cause a no-start problem.
Exactly. It doesn't. The problem is something else. Who knows maybe it's the pump itself. At this just point hit up ebay for a service manual and follow whatever it says.

csbuckn
08-23-11, 10:15 PM
Does fuel make it to the throttle body when you put power to the pump?
Here's the wires and stuff. Dont really know, looks like there are a few fuses involved.
http://i935.photobucket.com/albums/ad196/csbuckn/100_9302.jpg
http://i935.photobucket.com/albums/ad196/csbuckn/100_9301-1.jpg
http://i935.photobucket.com/albums/ad196/csbuckn/100_9304.jpg

cadillac_al
08-24-11, 09:25 AM
I don't see any mention of testing fuel pressure. Isn't that the first step? Will it start if you dump a little fuel in the throttle body?

csbuckn
08-24-11, 10:43 AM
It does say to test fuel pressure and also to take the return line off the throttle body and test pressure again to make sure its not the fuel pressure regulator.

gsim
08-25-11, 09:02 AM
Thanks Sven,

I noticed that when I first turn on key I see red oil alarm light. In a second or two it goes out IF I TRY TO START IT, so I would assume that it sees oil pressure building. Thing is, the pump turns off in just one second whether I am cranking or just sitting there with the key switched to "run". There is a return line for excess fuel and a vent too, so it seems the pump would run constantly if the key is in run. Or maybe it gets to a certain pressure and then shuts off if not running like my Ford truck does. Am likely going to try to find a really good shop manual for just this car, and not a 'one size fits all' book like I have.

gsim
08-25-11, 09:06 AM
Cadillac Al and C Buckskin,

I opened fuel line and found it DRY at the throttle body, so fuel pressure is ZERO there at all times. Since the pump turns on and then off in just one second, there will never be any fuel pressure there. I am wondering where the fuel pressure regulator is so that I can just buy one and swap it out to eliminate that part.

Gerry

gsim
08-25-11, 09:28 AM
C Buckskin ,

Wow! Thank you so much for diagram. I can work with that and pics you sent. Will copy them to paper today. The fuel press thing always was suspect in my mind, but I do not know where it is. Sounds like it is within the throttle body itself. I am thinking that since it is bone dry where the fuel line enters the throttle body, that the fuel pressure reg has no chance to work at all being as there is zero pressure. But if it is an electrical gadget, it could be shutting the pump down immediately after it starts. The pages you sent indicated that normal op would only be 2 sec of 12 volt power at pump wiring harness after key switched to "Run" if all normal. That seems weird. If not pr. regulator, I am leaning towards Engine Control Unit itself, or the pump maybe drawing too much current and doing an auto shut down.

The car will start and run briefly if I do dump fuel in it. I know that I can rig it to run by running permanent 12 v lead to pump. In fact, I can likely run a 12 v line that is hot with Ign sw "On" to the pump relay and let the relay contacts supply the pump as long as the key is "ON". Dangerous to drive that way tho.

gsim
08-25-11, 09:39 AM
Forgot to say that I did test all fuses and did test fuel pump relay at the very beginning, even before reading the one second spurt of 12 V power at pump harness. So although I still think it is an electrical trouble, it is not a fuse or relay.

Had to pour concrete yest around culvert project and have more work to do regarding that job today. So will not yet get to spend a day just working on this trouble. Will be able to stay on it some after a couple more days. Thanks again to all of you.

Gerry

csbuckn
08-25-11, 01:53 PM
What about priming the gas lines. Power the fuel pump till you get gas to the throttle body. Makes me think that the line is leaking somewhere and thats why all the gas emptied out the line and the pump cant push the gas up front.

jayoldschool
08-25-11, 06:14 PM
I agree. If it is actually priming, the line at the TBI would have fuel in it, even if the pump didn't run after starting. I wonder if the line is disconnected from the pump in the tank.

There should be a 12V fuel pump test connection to power the pump. On the later cars, it is found over by the passenger side hood hinge. Put power to that, and the pump should run.

gsim
08-26-11, 09:01 AM
I agree. If it is actually priming, the line at the TBI would have fuel in it, even if the pump didn't run after starting. I wonder if the line is disconnected from the pump in the tank.

There should be a 12V fuel pump test connection to power the pump. On the later cars, it is found over by the passenger side hood hinge. Put power to that, and the pump should run.

Thanks Jay old school. Will check that out. No way that any fuel is going to travel through all of that line when the power to the pump stays on for only one second, no matter what else. Have finally decided that press regulator is mechanical, likely a small ball bearing and spring. Therefore it could not cause this problem because there is no fuel at the throttle body to cause a pressure spike. Am going to run a tempy hot lead to pump wiring harness and just have my wife hold it to the + bat term. Then I will listen to the thing to see if it sounds steady or not. That could tell me if it is failing and causing maybe an excess current draw causing the ECM to take it our of service. In regards to the comment that the line is maybe leaking, that is impossible because it is in my garage and would stink to high heaven if leaking

Gerry

gsim
09-21-11, 10:47 AM
Hey Jay old school,

I fixed my trouble some time ago, maybe a couple of days after your last post. I determined that it was trouble in the tank somehow. Then as I prepped to remove tank, I remembered THAT I HAD TRIED TO SIPHON GAS FROM THE CADDY ABOUT 2 WEEKS BEFORE THE TROUBLE STARTED. The car had not been used since, so I did not make the connection between my attempt to siphon gas from it and the no-start trouble. It was disconnected inside the tank and was a fast and easy fix. Just used a floor jack and scrap or plywood to lower and raise the tank. Stupid is as stupid does, and my older friends tell me that it does not get better at our age either.

gerry

jayoldschool
09-21-11, 10:42 PM
Glad you got it fixed!