I posted a few weeks back about the 86 F150 SWB 4X4 I picked up. It had a fuel pump problem that miraculously cured itself long enough to drive home. Since then, it has been intermittent, usually healing itself before a full diagnosis could be made. Today, it failed long enough that I traced the problem down to the fuel pump relay control circuit.

When the circuit failed today, I limped home by jumping the fuel pump test point to ground. All fuel pumps ran and the engine ran. Removing the jumper would stop the pumps and the engine. Logically, the fuel pump relay and the 3 pumps are all OK. I have a Chiltons manual and what info I can get from the Internet about the control circuit. The relay pulls when EEC power is on and the ECU supplies a ground via pin 22. The test point is the same as ECU pin 22. The conclusion is that there is an open wire between the ECU pin 22 and the fuel pump relay. This is where it gets into unknown territory. Wires tend to be pretty reliable and I see no evidence of tampering. I changed the ECU with one from an 88 F150. It ran at first, then the intermittant problem of the fuel pump not running returned. Same fix, ground the test point and it runs. The odds of 2 faulty ECU's is very low. The control wire from ECU pin 22 to the fuel pump relay is supposed to be tan/light green according to the manual. I have to assume there is a splice somewhere in the harness as this circuit connects to 3 points and a single wire only has 2 ends. My question is where might this splice be located? ECU is under the dash in this truck, harness runs through the firewall and up over the engine. Before I cut everything open, does anyone know where the splice may be?

 

Not sure where the splice is, but it's probably not too far away from where the test port enters the harness.

Before you go any further, I would pull the codes. Make sure you do not have any codes related to the ignition circuit. If the computer is not getting any pulses from the distributor module, it thinks the engine is not running and will shut down the fuel pump.

I know it's running when you short the test point to ground, but the ignition module is made to run in a limp mode by itself. I don't know if it's possible for it to actually run with some of the signals missing,, since I believe it's also the way it fires the injectors, but I would double check to make sure you do not have a ignition problem. Check the codes.

 

Very true, codes need to be checked. No MIL/CEL on, but that's not always the entire story.

While I was posting, #1 son was troubleshooting another problem involving turn/brake/4 way. Found #1 fuse blown in fusebox. Replacing this fuse solved the problem he was looking at, but lo and behold, now the truck starts and runs without the test point jumper!

Now, I'm informed that the fuel pump enable consists of a 1 second timer in the ECU that is continuously reset by PIP pulses from a running engine. The predicted behavior is that the pump will be enabled for 1 second immediately when the key is turned on. If the engine is cranking, or running, PIP pulses will be present and the pump will continue running. Without PIP pulses, the fuel pump enable will remove the ground from the fuel pump relay and the pump will stop. I'm going to have to do some looking, but I wonder if some condition may be set with the #1 fuse out that would stop the pump enable from being set? Long shot, and yes, I know, read the codes! #1 son is currently out on a test drive to see if it fails again and I'll be checking for codes as soon as he's back.

Should add that when the engine operates normally, the pump will run for 1 second when the key is turned on without starting the engine. If the no run condition is encountered, the pump will not run for 1 second when the key is initially turned on. Scratching my head as to any effect the #1 fuse has on this?

New Info;

FAIL! and I'm not surprised. Actually kind of relieved that the #1 fuse was not related. Could not see how?

Tried to read codes. I have an OBD1 code reader and I have used it successfully, but I could not read any codes, nor establish communication with the ECU. All this raises my suspicions that I have some harness damage somewhere. Also should note that the latest failure came after an ill advised trip into the mudhole. Mechanical shock may play a part in this intermittent condition.

I should note that this truck has had a hard life. It looks like someone has been here before, lots of unrestrained cables and plumbing. I have a donor truck that may give up it's innards to save this one. Big project to change everything, (including the dash), as the donor is an 88. Gonna sleep on this one.

 

First year of a MIL/CEL connected to a computer was 1988.

IOW you don't have one, hafta pull codes to see if something is amiss that it knows about.

 

Well, still no luck with the code reader, but I was able to prove out the fuel pump control wiring right to pin 22 of the ECU.

So, symptom remains that the ground supplied by the ECU to the fuel pump relay via pin 22 is intermittent. The truck runs fine when the fuel pump test point is grounded, so it is safe to assume that ignition and injector functions are OK. I'll repeat that the short fuel pump run expected when you turn the ignition to on, but before the engine is started does not happen when the problem is present. Occasionally, the pump will run for a second upon turning the key on and then the engine will run fine without grounding the fuel pump test point until something happens to repeat the failure. It will occasionally fail while running and removing the ground from the test point will stop the engine when the problem is present.

Here is what I know is good;

Fuel pumps
Fuel pump relay
Wiring from fuel pump relay to pumps including tank change switch
Power to fuel pumps
Wiring from fuel pump relay to ECU pin 22 and OBD1 test point.
ECU (by substitution)

Here is what I think is good;

PIP signal from ignition because injection still functions and this would not stop the initial pump enable signal upon key on.

Basic power and ground to the EECIV system because the engine will still run.

Here is what I think may be happening;

Something is telling the ECU not to run the fuel pump enable timer. I am reasonably sure this is not related to the PIP signal as explained above. I really cannot find anything else that would produce this symptom.

I suspect my inability to use my code reader may have something to do with the early version of EECIV. I believe there is an alternate method, but that would flash the MIL/CEL lamp which apparently, I do not have?

The problem does seem to have an element of mechanical intermittent, i.e. it will shut down after hitting a bump, but I have not been able to pinpoint this.

Any ideas appreciated.

 

Use a test light or volt meter and count blinkenlights or meter sweeps:

https://web.archive.org/web/20110414...tion.com/?p=13

 

Thanks for the link. Knew there was another process, but had no luck finding it.

Anyways, no codes???? I know that is not right, but I suspect something is amiss in the wire harness. Just found a correct harness and ECU in the local scrap yard. Time to start swapping it out.

 

You should get a code 11 if everything was alright. These very early 85-86 fuel injected trucks are difficult to work on, they were the first ones and they kept doing things different and changing things. They got some of this stuff worked out on the later trucks.

 

May have found the answer.

ECU harness was pretty haggard, previous owner had opened it up all over the place. 85/86 is quite different from 87 and later, but we were lucky enough to find a correct unmolested harness.

Removed the old and installed the new. Found an orange wire disconnected and bared near the fuel injector connector near where the harness passes through the firewall. New harness had this wire grounded to the engine at the intake manifold. Turns out this is the ground for pin 49 on the ECU. There are other grounds on the ECU, but I don't know how the internal structure of the ECU handles this. So far, everything runs fine. I have to assume this ground was essential for providing the fuel pump ground enable via pin 22.

Still no luck with my code reader, but that is a problem for another day. Thanks for all the input.