1998 QX4 wont start please help!

[boatflipper]

Hello all, I've been hanging out on this forum for a while and have learned a whole lot but this is my first post so thanks for having such an informative site. Here's the situation: My wife's QX4 with 100k on her has run flawlessly since the day we bought her used 2 years ago. A few days ago she drove home and parked her and the next time we went to start her she cranked with no start. Figured it was fuel so I did the step-by-step diagnosis and found that the fuel pump wasn't running. I discovered that the ECM/ECU wasn't sending a ground signal to terminal 1 on the fuel pump relay so the relay wasn't operating (the relay itself is good). The relay is getting 12v at terminals 2 and 3 when the key is on and when I apply 12v to terminal 5 with the relay removed the pump will run and I could get her to start. Figured I could MacGuyver it so I removed the relay and made a jumper wire from terminal 3 to terminal 5 and she started and ran fine - for a while. Went to start her later and she sputters and dies. Now she'll sometimes start and sometimes won't so we have to rent a car for my wife and kids 'till I can solve this issue. What would cause the ECU to not send the fuel pump ground signal? I'm thinking bad ECU or some sensor? I'm getting a code reader after work but your wisdom/advice is much appreciated. Thanks!

[Slumpert]

Did you test the actual pin on the ECU. A damaged cable/connector might be going on.

I think I read somewhere that the system has two modes for the fuel pump.. a short startup time when key is first turned on, and then a continuous time if running.

[boatflipper]

Thanks for the reply. I will test the pin for continuity to the relay plug as well as a ground signal. You are right about the two modes. The ECU sends the signal for 5 seconds (to prime the fuel pressure) when the key is first switched on and then shuts off unless it "sees" that the engine is rotating via the camshaft or crankshaft position sensor. I get no signal in either mode so it could be a harness issue that just popped up. However the fact that I could jump the fuel pump and it worked for a while (bypassed the possibly bad harness) but now has intermittent issues causes me to suspect something else is going on.

[boatflipper]

Ok so I used the code reader - no codes :-( Jumped 3 and 5 and she started and ran fine again. Took her around the block and everything was good until I tried hard acceleration. Then she decided to hesitate and stall. She wouldn't start again so we towed her back to the house. I'm gonna head to the local u-pick to see if they have a compatible qx4 or pathfinder so I have a parts source before I start throwing money at this issue. I know somebody on this forum has to have some further insight.

[Slumpert]

Looking at schematics for my 1997 Pathfind which should be very similar to yours, I see two harness connectors between the relay pin 5 and the fuel pump.

I think the most likely issues would be a damaged/loose harness (even if voltage was okay, is it still able to supply the current the fuel pump needs?)

Right at E1 - M1 if there was a issue it would explain both issues.. not getting signal grounded to turn on the relay and not getting good power from the relay as those seem linked in the diagram .

Otherwise not sure what is going on besides compound issue

[boatflipper]

The only issue at this point is ground signal from ecu to relay. I checked continuity from harness at ecu to the relay terminal 1 and i got the beep. However when I grounded the ecu harness end, I did not get ground at terminal 1. Very strange. Also maddening is the fact that if I give the pump power she will sometimes start and sometimes not I found a 97 pathfinder at u-pull-it today and tomorrow I'm going back and puling her ECU since it's only $17 - if thats not the problem I can always sell it on fleabay. But you could be right - I've been giving the pump power by jumping the relay socket. If the harness has some issue at the e1-m1 then the pump might not be getting a full 12v or enough amps. Will take another look. Thanks for the input.

[Slumpert]

ECU are tricky to test ... For that pin the ECU sinks the voltage to ground when on..

I think I read you should see +12 on the ecu pin when it is not functioning and just over 0 when active

turn key on and and see if voltage goes from >0 to approx 12 after the 5 secs of MODE 1 expires.

Do not run a seperate power wire to test this... putting power to a pin when the rest of the ECU is "off" is really risky

[boatflipper]

Slumpert, thanks a lot for helping me with this. Of course when I went to the junkyard someone had taken the ecu :-( I was really hoping to tule a faulty computer out. Anyway, what I really need to know is if there is a sensor check before the ecu applies the ground signal to the fuel pump relay. We know that when the key is first turned on there should be ground at the relay for 5 secs but there is not. Once the car is running there should always be ground at the relay bc the camshaft position sensor is telling the ecu that the motor is turning. There is never ground at the relay terminal #1 bc the ecu is never sending the signal (at startup or when the motor is running). For some reason the computer is not sending the ground signal even when the key is first switched to "on". Either the computer is bad or there is some sort of start up check where a sensor is telling the computer not to give the ground signal to run the pump. When I hotwire the pump she will sometimes start and run but when I accelerate or put her under load (like with the AC on in park) and open the throttle she dies like she's not getting enough fuel. Fuel pressure regulator? Fuel injector controller? Whatever it is it's fuel related and its telling the computer not to run the pump or its faulty and the computer isn't running the pump bc of this.

[boatflipper]

Talked to a certified Nissan/Infinity mechanic today and he said it was most likely a bad computer and that it was a common problem with this vehicle. Planning to search for a used computer today. I'll post results.

[Towncivilian]

In my experience, ECU issues are incredibly rare.

Are both sides of the battery terminals clean and free of corrosion? Fuel pump relays and fuse also free of corrosion and scorch marks? Fuel filter changed within the past two years?

Loosen and retighten engine ground screws:

[boatflipper]

Towncivilian I agree 100% on the ecu probability but I don't know where else to look. New fuel pump, in tank filter and in line filter. Relay and fuse and their sockets free of corrosion. Battery terminals ok. Continuity between pin 11 on ecu and relay. The computer isn't sending the ground signal to start the pump. The computer is also not controlling the injector/fuel pressure system to regulate the amount of fuel when under load. Unless there is a sensor I'm missing it has to be the computer. It's supposed to send the signal when the key is first turned on and it doesn't. I will clean and re-tighten engine ground screws. Thanks for the input.

[Slumpert]

Maybe check the power wires going to the relay..

I don't mean see if it is 12v, check that there is solid connections all the way to the fusebox, change the fuse, and then follow the cable to the battery.

You might just have a bad supply issue here..

Warning----------- Not for faint of heart------------------------ Something I as a total hack redneck home garage mechanic would do..
Remove the fuel pump relay and ground Terminal 2 or 3 then turn the key on for 2 sec.
Should pop the 15amp fuse #17.. If it doesn't then you know there is a bad connection. It is possible that will complely fried out the weak wire, and while not usually enough to cause any fires, having a fire extinusher handy is recommended as that wire may well indeed be covered in oil and other flamable stuff.

[fastpakr]

I'm also curious about the quality of the ground to the ECU itself.

[boatflipper]

Ok I paid $100 for a junkyard ECU and plugged her in. Now I get ground to the relay for 5 sec with key turned to on and constant ground when cranking or running. After priming the fuel system she started and ran again with the relay in place so the old computer was bad. However she still stalls and doesn't want to restart if you push the pedal down quickly under load. I'm wondering if the new aftermarket fuel pump I put in isn't putting out enough pressure :-/

[boatflipper]

Frustrating thing is that the bad computer caused her to suddenly no work but replacing the computer hasn't completely solved the problem. The other thing may be that she's getting too much fuel for the old plugs to handle?

[Towncivilian]

How many miles do you have on this QX4?

[boatflipper]

100k

[Towncivilian]

Plugs are due at 105k, so I would change them out with OEM specified plugs now. Timing belt is also due at 105k.

[boatflipper]

Timing belt was done around 80k plugs are worn gonna change them out. Thanks.

[boatflipper]

Ok the mechanic says I need to reset/retrain/relearn the computer because I replaced it and that I may need to take it to the dealer to do this. I have found a couple of posts detailing how to do this with the key and gas pedal but they are for newer vehicles. Mine is an early (up to 2/98) 1998 model built on 7/97. Any help is appreciated.

[Towncivilian]

There are no computer relearn procedures, only manual adjustment of the fast idle cam, idle speed, ignition timing, and idle mixture ratio. It is described beginning on page EC-34 of the factory service manual. I don't think this will help your no-start issue though.

Check under the distributor cap for excessive dust and dirt build up, blow it out with compressed air if necessary. Also check the rotor for excessive scorching.

[boatflipper]

Ok here's what I did: It seemed to me that replacing the fuel pump and filters would allow the injectors to spray more fuel and the stalling never occurred during idle only when pushing the throttle. The more I replicated the scenario the more I became convinced of a flooding/rich situation rather than a lean one. I pulled the plugs and noticed they were pretty worn with a large gap. So my theory was the spark was weak but it was ok with the prior lower fuel pressure and the old computer had "learned" to run correctly. By increasing the fuel pressure I was now flooding out the motor. So I replaced the plugs, cap and rotor. She now started and idled like a sewing machine! I still had some stumbling at idle (which as low) but no issues under load. Adjusted the TPS a little to up the idle and she now runs better than she ever has. The problem WAS a faulty computer! Thanks for everyone's input!