Hey guys,

I've got a 350 in my jeep YJ and just got an unusual problem this morning. It's been running fine, and was idling fine for a half hour while I dug out the driveway and scraped the windows (ice storm in Halifax).

Driving in once I got rolling it just suddenly died and smelled of fuel. I popped the hood and liquid fuel was just pouring out of the top of the TBI. I popped the air cleaner off and it was a pool of fuel there.

Any ideas where to start looking?

Kevin

 

Well, I think it was the fuel return line was kinked. Straightened it a bit and it's no longer pouring gas out but it won't start.

I'm thinking it's just flooded and should pull the plugs and crank it for a bit.

Any other ideas?

Thanks,

Kevin

 

May need to replace the plugs, likely carbon fouled. As for the return line being kinked causing this, not too sure. It may have been the cause of a blown out gasket or FPR diaphragm, but that issue will still exist.

Within the upper part of the TBI unit is: o-rings to seal injectors, upper pod to lower pod gasket, and the FPR diaphragm. Any of which will cause flooding.

RBob.

 

Weird. Turns out wiggling the return line only seemed to fix it because it was bumping wires in the area. The plug on the back of the distributor (one nearest the firewall) when wiggled makes the injectors spray fuel like crazy regardless of whether the key is on/off, cranking or anything. If it isn't spraying full blast, it doesn't spray at all, even if the engine is cranking.

Also, the fuel pump seems to be running anytime the battery is connected.

Time to brush up on my wiring diagrams to see what could possibly be causing this...

Kevin

 

Yes, do need to get to the wiring diagrams. That connector on the distributor pigtail is also likely to be bad. The terminals are either dirty or have been sprung (not enough tension for good contact).

Some GM set ups did run power to the injectors all of the time. But it is better if they are ING+ switched.

Need to do the same with the ignition distributor/coil and ECM set up. Sounds like they too have power all of the time. NOte that the ECM is supposed to have battery power on the proper pins.

With an IGN+ switched power to tell the ECM when to power up and down.

When these items get wired to the ignition switch, need to be sure to pull the it from the proper feed. If using the accessory feed it will lose power during cranking, leading to no start conditions.

Need to have it from an IGN+ feed that is also hot during cranking.

RBob.

 

I attached the relevant wiring diagram page. (I think, I've been a lurker here for years learning lots, but this is my first time posting and trying attachments. Great site by the way, I've never had to post because I've always found the answer already here somewhere.)

The engine is from a 91 K-code 3/4 ton suburban, and I carried all the wiring over from it.
Note, the diagram is for a 4L80E equipped vehicle, but I actually have a manual, SM465 behind the engine. All of the tranny wiring is simply incomplete. It's been working like that for years, with obviously a CEL.

Anyways, at the top C15 and C16 are the injector control. As I understand it, the 7.5A fuse shown is wired directly to battery hot as others specify IGN if they are switched. The PCM on the right grounds the injectors to get them to open in very short bursts to allow a metered amount of fuel out, right?

As far as that, either I have a short, or the PCM is continuously grounding them for some reason.

At the bottom is the distributor with it's connections. The EST plug (D11-D14) is the one that wiggles and switches it from no fuel ever to all fuel all the time. It seems bizarre to me that Electronic Spark Timing has any effect in this situation, and I don't understand why it's even getting power.

Finally, I get a clicking sound like a relay when I wiggle the plug, I suspect it is the fuel pump relay because that's the only one I can think of on top of the engine but I haven't been able to confirm (it's bloody cold out there).

 

The EST system sends distributor reference pulses (DRPs) to the PCM. The PCM uses them to know when to fire the injectors and to enable the fuel pump relay. Along with using the time between DRPs to calculate the engine RPM.

Wiggling the plug is causing false DRPs to be sent to the PCM.

RBob.

 

Thanks so much for your help RBob, I really appreciate you taking the time to help out a new member to the site.

Unfortunately, now I'm really confused.

As background, I checked all fuses and vacuum lines, nothing blown or changed.

Went back to the Jeep and tried a bunch of diagnostics.

I started with:

Why is my fuel pump always on, even when the key is off?
First I unplugged the oil pressure switch, as it's the only thing between Batt+ and the fuel pump in one path. Still whirrs away as soon as battery connected.
Second, I removed the EST connector from the dizzy. Fuel pump still fires up as soon as battery connected.
I pulled the fuel pump relay and put a voltmeter on the signal wire to the relay from the computer. Yup, computer is telling it to be on.
I plugged the fuel pump relay back in, and all of a sudden the fuel pump is never on. The only way to get it to pump is to apply 12v to the test terminal on the relay.

I can't find anything else to do with that, so I start the injector troubleshooting.
First I check for spark, actually the engine sputters a bit when cranking on fuel that was dumped into the manifold on previous tests, so spark is good.
Next I pull the EST connector and try to check it. I put 12v to the purple/white wire on it (EST reference) as I read that should signal the injectors to open. Nothing comes out, so I try grounding that wire, still nothing.

I checked all the voltages against the PCM connector test chart in my manual, and almost every one checked out good.
1. The fuel pump signal wire never goes hot.
2. Tranny temp signal and coolant temp signal both have 9 volts on them, instead of 2.24 or 1.5 respectively.
EDIT: this makes sense as it's not at operating temp, and the resistance is higher when cold, hence higher voltage drop across them.

Also surprising, is the PCM case supposed to be live? I read 12v from the case to ground...

I'm thinking of changing the Ignition Control Module as it seems to be the fault in many of these bizarre injection problems, and it's plug seems to switch things from full fuel to no fuel depending on position. Note, it's not like I wiggle it and get a squirt, and then another squirt. If pushed one way it sprays all fuel all the time, if pushed the other way no fuel comes out no matter what. That doesn't address my fuel pump oddities though...

Thoughts?

Kevin

 

As for the PCM case being hot, no, it is grounded to the engine block via the EFI harness (or at least is supposed to be).

Thinking about it, it may be as simple as the EFI harness grounds to the engine block are loose or the wires broke at the ring terminals. This is where I would start. That would explain a lot of what you are seeing.

Being a retro-fit, make sure that the PCM case isn't grounded locally. It needs to be isolated from the chassis/body.

RBob.

 

So I checked my wiring diagrams, and I C2/C3/C13 in the attached diagram are the only ones I see that go to engine ground directly, or really at all without activating something else along the way.

If I take my multimeter to this, I should see 0 ohms between the case and those two, and between those two and engine ground?

I'll make sure they're gounded and see what happens. If the case is still hot there's no way I can just fire a self tapping screw into the case and run a ground wire from that to the engine eh?

Kevin

 

THANK YOU RBOB!!!!

Wow, I found it, and I can't believe what it was. Without your help I may not have found it. The wire for C3 had a great ground to the engine, almost no resistance, but a huge resistance between it and the case. When I looked closer I saw that the pin was backing out of the connector. I pushed it back in with the probe of my multimeter, hooked up the battery and the jeep fired. It ran like crap because it was full of fuel and one spark plug wire was hooked to a tester, but it quickly cleared up and is running like a top again!

THANKS AGAIN!!

Great site, great people, much appreciated.

Kevin

 

You're welcome, glad it was an easy fix. On the terminal there is a little flag that catches a ledge inside of the connector to prevent it from being pushed out. I've seen them get bent over, might be something to check.

RBob.

 

It turns out there were actually two issues, and the second one is now solved as well. Info here for future reference should others run into this.

My fuel pump was only working intermittently, and the engine would just randomly start and run fine, die randomly, or not start at all.

I was tearing my hair out for a bit as I could check the pin at the computer and it was commanding the fuel pump on, I unplugged the relay and tested conductivity along the signal wire from the computer to the relay socket, that was good. Put power on, got proper voltage at the relay signal, other side of the relay signal coil was grounded fine. Must be the relay. Bench tested the relay, very solid click and perfect readings from it. I plug the relay back in and get nothing out of it.

I finally come to the conclusion that it must be the relay/socket connection, and bent a couple of the pins on the relay to make sure they contacted the metal in the socket, and it now works as advertised. I'm going to run it a few days and make sure everything's peachy now, but I'm thinking I should replace the socket, as I suspect corrosion inside it.

Hope this helps somebody else.

Kevin

EDIT: I also suspect my oil pressure switch has failed as it provides a back-up should the relay fail, powering the fuel pump anytime oil pressure is above 4psi.