Off-Road Forums & Discussion Groups

Off-Road Forums & Discussion Groups (
-   Jeep-Short Wheelbase (
-   -   weird fuel injection /headlight porblem, need help (

**DONOTDELETE** 11-07-2002 08:58 PM

weird fuel injection /headlight porblem, need help
alright this is the weirdest thing I have ever seen.
First, I have a 78 CJ7 with a Howell FI kit on the 258 and have been having problems for the last 3 months or so. Runs bad, surges, exhaust smells funny (think rich). Overall no power. I have replaced the FI wiring harness, all the sensors, the fuel pump, the fuel filter, coil, ignition module, plugs, rotor, checked cap, FI computer, chip in the computer. But it still runs, I can drive it everyday.

Heres the weird part. when i first get in the Jeep and turn the key to "on" and then turn the FI on with the rocker switch I have on the dash "switched power" the fuel pump primes like it should (jeep not running). But when I turn on the headlights the fuel pump primes again, and will continue to prime everytime you turn the lights on. Also when I am at idle with the jeep running and turn the lights on the very slight "huM' from the tank (fuel pump working) changes.

I dont see how the headlights and the FI are connected. Anyone have ANY ideas, I am at a total loss. There is no- where the FI and Jeep are connected other than the rocker switch, the coil, the solenoid and grounds. Help me out here guys. Thanks

RRich 11-07-2002 10:09 PM

Re: weird fuel injection /headlight porblem, need help
Believe it or not, the pump and headlights are connected.
I'm not being facitious either.

Think about it - both are connected to ground. And if you
have a fuel pump relay, it too is connected to ground, as
well as most everything else.

1. Make sure your battery ground strap goes to the engine,
and is in good shape.

2. Make sure you have a good ground strap, preferably more
than one, between the body and the engine. Preferably the
braided flexible types. Regular stranded wire won't last,
it won't flex enough, breaks. The braided straps are made
to flex.

3. And - make sure you also have a ground strap between the
body and frame, at least one.

You may even find it good to run a ground strap from the grill,
where the headlights are, to the main body. Sometimes the
bolts connecting it get corroded.

What you are experiencing is called a "ground loop."

LEVE 11-07-2002 11:15 PM

Re: weird fuel injection /headlight porblem, need help
I'll second RRich's thoughts.

80CJ7 11-08-2002 09:49 AM

Re: weird fuel injection /headlight porblem, need

I have been researching TBI conversions recently and have an uneducated guess at what the problem might be. I have seen several mentions of the requirement for a good electrical supply to the ECM for TBI systems. What could be happening is that you have barely adequate voltage when no accessories (such as the headlights) are on, but when you turn the lights on, the momentary voltage drop causes power to the ECM to fall below the threshold needed for operation. If this is the case, then any high current accessory should produce the same effect - maybe try the blower motor, other light, or winch if you have one.

The low voltage to the ECM has also been implicated in poor running TBI setups. Check the voltage at the battery and at the ECM and if the difference is more than a volt or so, find a better way to get power to the ECM. Maybe as a test, try using a wire straight from the battery to power the ECM. I also have to agree with the above posts about good grounds, this is especially important to look out for in an old CJ - they have been known to have some corrosion problems [img]images/graemlins/tongue.gif[/img]


dave 11-08-2002 10:47 AM

Re: weird fuel injection /headlight porblem, need help
I'm also going to agree with RRich and Leve about the ground (negative wiring). I'm wondering if somehow the power source (positive wire) for the fuel pump is somehow connected to the power source for the headlights. What I'm thinking is somehow the fuel pump relay is connected to the power source for the headlight switch.

I would suggest using / rewiring an engine oil pressure switch to turn on / off the fuel pump. I know RRich and I have posted on another ORC board about using an oil pressure switch on the GM TBI system. Once the engine starts cranking and the engine oil pressure is about 3 PSI the oil pressure closes, and power will be sent to the fuel pump. GM uses a fuel pump relay to prime the fuel pump, but if the GM distributor is not sending reference pulses to the GM ECM (which means the engine is not cranking), the GM ECM will turn off the fuel pump relay after 3 seconds. It's like GM engineered two power sources for the TBI fuel pump.

**DONOTDELETE** 11-08-2002 01:26 PM

Re: weird fuel injection /headlight porblem, need help
Thanks for the ideas guys. THe grounds are pretty good on this jeep but Ill check them all again and make sure they are REALLY good. I understand the voltage loss happening when you turn on a high power accessory but I think it only happens with the lights, Ill have to check that again. All the gounds for the FI are bundled togehter and connected to the TBI. It is a good ground, we have checked for voltage loss .02 voltage drop so thats not the problem. I guess Ill just keep checkin stuff

LEVE 11-08-2002 05:26 PM

Re: weird fuel injection /headlight porblem, need help
1. Voltage drop does not denote a bad ground.
2. A bad ground can be bad under one circumstance and in an istant correct itself.
3. I really think you're assuming the grounds are good.
4. Really take apart and inspect the grounding straps, cables, and termination points.
5. It takes little time, and pay back is great.
6. When you poull on the lights something's causing the voltage to lower... and that's the return path in the ground. The resistance is current sensitive.
7. My Howell TBI is not subjec to this problem because I have checked, replaced, and repaired grounds.
8. Advice it's cheap, but usually on this forum it's free, and likely good. [img]images/graemlins/shocked.gif[/img]

CJ7Taz 11-08-2002 06:13 PM

Re: weird fuel injection /headlight porblem, need help
I was busy with the brakes on the neighbor's Explorer and didn't have time to respond earlier but I see you got it. That power drop you experience when you turn the headlights on is called inrush. They draw more current during the first instant when the filament is cold. When the filament heats up, it presents a greater resistance. Ever notice that most bulbs fail with a pop when you first turn them on, they hardly ever fail while they are burning.

Yes, that was my opinion too that you have a bad common connection somewhere that is dropping the voltage to the ECM below a threshold for an instant. May be on the ground or positive side. Unless you have a dual post battery with the ECM connected to one set of posts and the rest of the Jeep connected to the other posts, you have common connections.

Just as a side note, a ground loop is a high, radio, frequency thing. It's along the order of VSWR a.k.a. SWR, but rather than being reflected, it follows the circular loop back to the source and round and round.

</font><blockquote><font class="small">In reply to:</font><hr />
4. Really take apart and inspect the grounding straps, cables, and termination points.

[/ QUOTE ]I found an equipment ground strip at the hardware store, I've installed one on the frame and will put the other inside the tub under the dash. I've run ground wires to everything, nothing depends on frame grounds. Even the tail lights and fuel tank grounds run in wire back into the tub rather than tying to the frame. The signal ground for the ECM ties directly to the block and the headlight/parking light ground ties to the big aluminum alternator bracket (no sense in going alternator to battery and then back to the headlights).

LEVE 11-08-2002 06:47 PM

Re: weird fuel injection /headlight porblem, need help
Taz, a ground loop is not frequency dependant... remember that the voltage in a CJ's DC, direct curent. Ground loops are the situation where there is more than one electrical path to ground for electrical/electronic equipment. The interfering path is not always placed there by choice. In a Jeep Electrical theroy it's placed there due to corrosion and rust.

A ground loop forms when a return current flow cannot follow the engineered ground path. So, those pesky little electrons start looking for another way home. Just like water, the electrons seek out the path of least resistance and start flowing through it. The path is often constricted and all them electrons are trying to still get through the path, and the path heats up. It's a cyclical thing... most people wouldn't understand! [img]images/graemlins/grin.gif[/img]

That's why lots of funny things happen when a Jeep has a ground loop. You just never know just where that path is going to spring up. Try handling 165amps starting current though your nice new radio... you can imagine just what will happen tot he radio.

In your example of RF, the energy is transfered from the final amp to the antenna lead cable, and then to the antenna. However, the antenna resistance is not matched to the frequency of the wave being fed into it. Not all of the energy will be radiated out of the antenna. That energy which is not radiated starts to seek it's way back to Mommy Earth and takes the path of least resistance.. guess where? Yep, back the way it came... Reflected RF can, and often does burn out the final amplifier in transmitters because of this problem.

CJ7Taz 11-08-2002 07:51 PM

Re: weird fuel injection /headlight porblem, need help
You are misusing the term "ground loop" to describe multiple ground paths. Ground loops cause an interference hum in RF circuits, even in computers. I'm not as learned on radio frequency as I am in some other areas so I'll not debate it here. Just plug "ground loop" into a Google search and see what you get. In obtaining my EE degree I did study what they were and what they do. I even accidentally created on in the lab once.

DC travels from minus to plus (yeah, I know conventional current flow is + to - but it's the electrons flowing so it's - to +) so it only flows in one direction through the multiple grounds. RF will chase it's tail, so to speak, through the loop. I guess because there is always a different potential only a few feet away. Like I said, I'm no expert on RF, I did the digital option and was only exposed to the simple RF and power properties and theories.

And no, I'm not buying the Fourier Transform theory. that the DC unit step is just a bunch of sine waves added together. In the first place, if they were, why would we need crystals? We could just filter out the frequency we wanted. In the second place, the sine waves continue from - to + infinity. So why would you need a headlight switch? If the waveforms started at minus infinity, the headlights would just come on when it was time, no need for a switch.

All times are GMT -5. The time now is 03:47 PM.

Powered by vBulletin® Copyright ©2000 - 2019, Jelsoft Enterprises Ltd.

User Alert System provided by Advanced User Tagging (Pro) - vBulletin Mods & Addons Copyright © 2019 DragonByte Technologies Ltd.
vBulletin Security provided by vBSecurity v2.2.2 (Pro) - vBulletin Mods & Addons Copyright © 2019 DragonByte Technologies Ltd.
Copyright 1994-2009, VerticalScope Inc. // Off Road forums & discussion groups sitemap
side by side | atv | dirtbike | snowmobile | sandsport | competition | land use | Jeep | Toyota | Ford | GM

For the best viewing experience please update your browser to Google Chrome