Electrical SUDDENLY FUEL PUMP SHUTTING OFF RANDOMLY

[PonyGTrider]

Hi all,
Today I was driving normally when suddenly my engine came to a stop. I pulled over end tried to restart it but it will start and die again. Having the ignition switch open I noticed that the pump came on and off back and forth priming without touching the key. I closed the ignition switch OFF and ON again and I heard the pump priming normally, I tried to start it but some times it will start but die again.

As I was not too far from home I left the ride and went to pick up some tools a couple of fuel pump relays, a circuit tester, a volt meter, and a spare computer. when I got back I replaced the relay under the driver seat and started it, it started right up. I came home as fast as I could but when I got here it stopped the same way as before, switch open and the pump cstarted priming ON and OFF a few times and could not re start it agai. I disconnected the battery and waited 15 - 20 min. an starter it up normally.

Before all this happened I was running a reman. computer working fine but I had to measure voltage on the O2 sensors. I finished that and decided to exchange that computer with a known good one A9L. It was running fine and suddenly started having the issues described above. So I know the relays I'm using are ok, so I really need your ideas. I didnt have any shorts while doing the O2 sensors testing nor when I exchanged the computers so I have no idea why it started with this BS!

As for now I disconnected the battery, replaced the A9L with the remanufactured ona which has no labels at all so I'm guessing it is an A9L... All I know I went to Autozone and ask for a computer for a '90 mustang GT with a 5 spd transmission. I will wait 'till tomorrow to connect the battery and start this thing and see how it behaves.

I've never had any issues like this with none of these compuetrs I have so I feel lost.
Could the problem be the fuel pump getting hot shutting itself randomly, and when I left the car it cooled down allowing me to get home ??
But then why if I have the ignition switch open the pump comes ON and OFF priming and cuttin and priming and cuttin on its own?

Please share any ideas!!!

By the way I havent run any scan to check for any error codes in the ECM because I've been disconnecting and connecting the battery too many time so I'm gessing and asuming there is not enough data saved in the memory at this point.

Thank you for the help ! ! !

 

[Noobz347]

The fuel pump is bad.

You should also check for fuel leaks or a regulator that smells like gas at the vacuum port.

It sounds like your pump is shot though. You can verify with a fuel pressure gauge. You probably have bubbles in the fuel lines too.

 

[PonyGTrider]

The fuel pump is bad.

You should also check for fuel leaks or a regulator that smells like gas at the vacuum port.

It sounds like your pump is shot though. You can verify with a fuel pressure gauge. You probably have bubbles in the fuel lines too.

Thank you for your intervention. I will check a d verify the fuel pressure The FPR is good with no leaks. If the pump is what failed, it is not that bad compared with a shot ECM. If it starts acting again I was planning to run the pump manually using the jumper in the diagnostic connector.

I’ve never experienced this symptoms with a faulty fuel pump. One time one of them failed but it did that completely, never used to come ON and OFF repeatedly with the switch ON like this time.

Thanks again

 

[AeroCoupe]

You need to verify that the pump is getting 12V+ when it is acting up. You can check for voltage at the inertia switch in the trunk as its the easiest to get to. There have been issues in the past with the connection to the fuel pump and fuel tank sending unit that is just above the fuel tank just to the passenger side of the tank at the rear bumper.

I would take that connector apart and see if there is any water, dirt, corrosion, etc. and clean it up. Use some dielectric grease and put it back together. If the pump is getting 12V+ but it is cycling then its the pump. If the voltage is intermittent then I would make sure the computer is providing a constant signal to energize the fuel pump relay. If that checks out then its between the relay and the pump.

Outside chance the pump suction is getting blocked but I would think the pump would still run but you would not get any or enough fuel.

 

[General karthief]

Now I'm not a mechanic, that said I would put a test light on the power input side of the relay and drive it. When it acts up you can tell if it is losing power before or after the relay

 

[PonyGTrider]

You need to verify that the pump is getting 12V+ when it is acting up. You can check for voltage at the inertia switch in the trunk as its the easiest to get to. There have been issues in the past with the connection to the fuel pump and fuel tank sending unit that is just above the fuel tank just to the passenger side of the tank at the rear bumper.

I would take that connector apart and see if there is any water, dirt, corrosion, etc. and clean it up. Use some dielectric grease and put it back together. If the pump is getting 12V+ but it is cycling then its the pump. If the voltage is intermittent then I would make sure the computer is providing a constant signal to energize the fuel pump relay. If that checks out then its between the relay and the pump.

Outside chance the pump suction is getting blocked but I would think the pump would still run but you would not get any or enough fuel.

I appreciate your suggestions and ideas, right at this moment I will start taking some measurements. As I mentioned swapped computers with the one I was running before this ordeal. I'm almost 100% sure both computers are good but just want more verification of that. In the past I did have issues with the connector you suggest, cleaning it was the solution but it was not acting the way it is doing now.
One question that comes to mi mind is: Assuming that the pump is bad, when the pump stops will the relay click as if it gets de-energized??? Because that's what it's doing. If I hold the relay in my hand it clicks when the pump starts and it clicks when the pump stops. couldn't the pump just because the relay stopped providing power and pump is good? Of course I will find out by checking the voltage at the inertia switch, I'm just thinking loud this time I guess.

Thanks a lot for your suggestions

 

[PonyGTrider]

Now I'm not a mechanic, that said I would put a test light on the power input side of the relay and drive it. When it acts up you can tell if it is losing power before or after the relay

Yep that is a good idea, I'm just concern at getting stranded again on the road, but it is worthed to take that risk.

Thank you

 

[PonyGTrider]

Just a quick UPDATE,
I run few tests and all came normal: I measured the voltage at the inertia switch and I got 14.5 Volts I activated the inertia switch and it opened the circuit properly. I then restablished the switch and it worked fine. While running I shooked the hell out of the fuel pump connector under the rear bumper and it is solidly connected not even a "blink" on the voltage.

I don't hear any abnormal noise on the fuel pump, it simply sounds normal. Have to remind all that I'm using the other computer, not the one I had connected when this whole thing started. Could a problem in the computer is creating these failures?

One thing I noticed is the exhaust fumes smell nicer with this other computer (NO LABELS ON IT) with the Labeled A9L the exhaust smell more rothen and stronger, I also noted a rolling idle on that same computer.... Does that tells anybody anything or it is just my imagination???

ANY WAYS, is there a way to identify the computer without that has no labels, If not from the outside, maybe internally? The one I got remanufactured from Autozone???

 

[AeroCoupe]

If the fuel pump relay clicks it means the coil has either energized (pump running) or de-energized (pump not running) and the only thing that controls that is the ECU. So if the relay clicks and the car dies when it should be running then I would be 99% sure the ECU has an issue.

There were several different ECU's in the MAF cars and they all had a little different strategy so one could run richer than another very easily.

 

[PonyGTrider]

When the failure occurred I pull over, the ignition switch was still open and strangely I could hear both solenoids (The one under the driver seat and the one in front of the passenger side strut tower, now relocated inside the fender) opening and closing rapidly and the pump priming ON and OFF, ON and OFF, I’ve never experienced that before.

I was watching this video about replacing the three bad capacitors and also included a verification of the 5 volts signal, and few more functions. The guy talks about the fuel pump bad conditions and the transistor that manages it. I followed his suggestions of testing the fuel pump transistor and pin 22 and it seems to be working fine. I proceeded to re-solder the three capacitors as I noticed that a leg on one of them had a questionable contact. Those are new capacitors I installed them, replacing all of them on both computers.
I then installed the computer back, connected the battery and started up. I took a few miles ride beating on it hard waiting for posible issues… Nothing!!! No problems at all!! So I just don’t know if the issue was fixed by re-soldering the capacitors or if the problems are just on hold temporarily.
I just ran a scan and found these error codes KOEO (o) 81, 82 and code 15 in the Continuous Memory (c)
I know the first two are irrelevant to my issues (Smog system deleted) and the last one probably the ECU needs more time to re-learn all the parmeters.

Here is the link of the video I mentioned, very interesting.

View: https://www.Youtube.com/watch?v=p73aWm90XbU

Thank you all

 

[AeroCoupe]

Code 15

OBD 1 code 15.

Hey gents. It’s my first time posting in here. I have a 93 Cobra that will run fine then just shut off and the only way it will start up is if I disconnect the battery and connect it again. I went through the process of elimination by replacing the tfi module, ignition coil, fuel pump relay...

stangnet.com

 

[PonyGTrider]

Thanks for sharing that good and interesting thread.
Today I used my scanner to clear all the codes, ran the engine and then ran a new scan. the obvious 81 82 errors showed up again but not code 15, it gave me code 11 so everything seem to be ok. I'll keep checking that for some time hopping it doesn't come back.

At this point no more fuel pump issues have been present. You can read the previous comment about What I have done to the computer, Idon't know if that was the fix or the problem still there. Maybe I just need to get everything nice and hot to start acting again, it is well known that when the fuel pump is at the edge of its life some times they start acting when hot so we will see.

Thank you

 

[PonyGTrider]

UPDATE:
I took my ride fir a spin and for the sake of it I run a scan and to my bad luck there it is, error 15 is present

I’ve red about this and will try to follow the testing steps. I have no chip attached to my A9L and I think Isael the wire on pin No 1 goes to the fuel pump relay and I’m almost positive I have power on the relay. This points to a likely bad compute. If the computer has a bad KAM is there a fix for this???

Thank you

 

[Blown88GT]

Bad ignition switch or EEC power relay could be killing it, too.

 

[PonyGTrider]

Bad ignition switch or EEC power relay could be killing it, too.

Error 15 relates to a losing power at pin number 1. Pin number 1 is hot at all times regardless if the ignition switch is on or off. I checked the voltage in wire of pin 1 and it never loses power, that makes me believe the computer is faulty.

 

[PonyGTrider]

I'm thinking if this error 15 has to do with the problem I've been experiencing like hesitation and explotions in mid pipes while taking off immediately after a stone cold start. Could it be that the KAM is relearning all the settings as the KAM has to re-learn every time the engine is shut down (Because of error 15). This kind of make sense but this error showed up just recently and I have hose symptoms way long before that.
Sorry I'm thinking loud or maybe i'm trying too hard to resolve those issues

 

[General karthief]

Send that computer to https://www.ebay.com/str/ecuexchange
Then you will know it's right.

 

[PonyGTrider]

Yep, at the end that would be the most effective option.
Thank you

 

[Mcmahst]

As a test, maybe you could rig pin 1 wire with a test lamp, clear the codes, go for a test run long enough to see if you get a flick in the light, and retest for the code 15? While it wouldn’t solve the problem, it would lead to locating where you might have a bad power lead. To retest, you could then rig a new direct wire from the battery and do that again, to see if that resolved the code 15 issue. It would help eliminate or confirm the computer as the issue.

 

[PonyGTrider]

Well now that’s a great idea I didn’t think about. Since I just assembled the computer back in its place I will run a lead from the relay under the seat, I already tested that wire from the relay to the computer connector and the connection is solid. Right now the power is on standby as the battery is disconnected, I will set up that test light bulb and see any flickering as I drive.

I went over the circuit board in a more carefully way trying to see any indication of visual damage.
The first component from pin No. 1 is a micro resistor, I wonder if it is faulty… Last night I didn’t want to start testing, I just put everything together.
This time I will run and monitor this test longer doing scans more often looking for some more accurate results, then since I have the other computer that seems to be an A9P which works wonderfully I will replace the A9L and decide what to do with it.

Thank you much for your ideas