This is a very long, drawn out thing to discuss. Basically short story -


Main relay went out completely a few days ago, i acquired another one from a local junkyard, and everything works, except the fuel pump.

So starts the diaging... Removed the relay and checked the joints, fairly good condition. Tested the connectors on the plug, and everything has power when it should. Now, i've checked all fuses and grounds, everything checks out, I've found the problem, but i'm trying to dig deeper.

I tested ping a7 from the ecu to the main relay, it sees itself, HOWEVER.. the ecu does NOT produce a ground, I spliced open the wire a bit and ran a ground to it, and it turned the fuel pump on.
I'm assuming when the relay went bad, it took ping a7 with it. I know pin a8 runs on the same circuit, is it possible to move the pin to a8 and see if it works? The ECU will not send the ground signal to the main relay, so i'm assuming the ecu is now bad..


Anyone think otherwise? I've spent 5 hours tested every, possible, thing. Grounds, power wires, everything. And yes, for the time being, i'm jumping 5-7 on the plugs of the relay to allow me to drive my car...

 

If the ecu is nt sending the output ground to the relay. Make sure all your inputs are going into the ecu first. Im not sure what inputs it needs to see but im sure it needs the keyswitch voltage and a couple other things. If you have all the inputs and that ground is a output then putting a permanent grou.d to the coil of the relay should be fine as long as the 12v positive isnt constant. Than your battery will drain. Let me check my helms

 

Alright I checked the helms looks like the ecu would be the problem. If the first set of tips is closing and you have power to the black\white and voltage coming out at the black yellow wires. Than that supplies voltage to the ipg1 and ipg2. .and voltage to the second coil in the relay. If you splice in a ground The problem would be is the fuel pump turning on when the key is in the radio position. Not acc. Have you checked the resistance of that wire from ecu to the relay?? I've seen wires look fine but broken inside the insulator.

 

Check IPG1 and IPG2 to the ECU/ECM, both must have power for ECU/ECM to supply a ground to the PGM-FI Main Relay.

Power for the IPG1 and IPG2 is supplied by the fuel injector relay in the PGM-FI Main Relay. 94

 

He said all other functions were working. If he wasn't getting those inputs his injectors wouldn't have voltage. Unless there's a broken wire going to those inputs. But he said he checked all that wiring.
I was wondering why does that relay need a hot when cracking voltage from one of the fuses? Does it need more voltage to that coil to keep the tips closed?

 

A7 is not getting the ground signal from RM7 , to send to the relay, when i ground A7 itself, the relay turns the pump on. In a jist, ECU parts RM7 and Q10 are bad, for right now, i'm running a ground wire to the "a7" wire of the relay, providing my own ground, everything else works as it should, this will hold me over until i can either fix these parts, and/or buy a new ecu.

 

The ECU/ECM does not supply the injectors with "voltage", [12V+] that is supplied by the fuel injection relay in the PGM-FI Main Relay, [independent of ECU/ECM].

The ECU/ECM supplies the ground to pulse the injectors.

OP stated engine runs if fuel pump relay is supplied a ground, [direct chassis ground.

That would indicate ECU/ECM is working properly except for fuel pump relay control, so all wiring, powers and grounds to ECU/ECM must be good, with the possible no power to either IPG1 or IPG2, again, both are needed for fuel pump control output.

Sorry OP, I do not have a wiring diagram for a 92 Civic so I can not give you pin outs or colors, however, on most Honda/Acura IPG1 and IPG2 are the same color as the fuel injector relays output and the 12V+ lead at the injectors and will be the only 2 leads at the ECU/ECM that are that color.

If one of them does not have 12V+ when ign. switch is on the ECU/ECM will not supply the ground for the fuel pump relay, the most common problem is a corroded Junction Connector, that is repairable with some contact cleaner.

The other possibility is a connection problem at the point the 2nd IPG lead is added to the ECU/ECM harness, repairing it would require you "untape" a section of harness to expose and repair the connection, it is a lot easier to cut the IPG lead with no power and jump it to the one that does have power.


The above "fix" assumes one IPG lead has power and the other does not, if both IPG1 and IPG2 do not have power, [in that case I beleave you do not have spark either] but you would need to jump from the fuel injector output lead to IPG1and IPG2. or untape the harness to find the break. 94

 

I don't think your reading my post right.
If he is getting voltage to his injectors allowing his car to run. Then he obviously doesn't have a problem with his 12v source going to his injectors meaning the main relay's primary coil is induced closing the secondarys tips allowing 12v positive voltage to pass to the secondary coil. And also sending voltage to his injectors and igp1and 2.unless he had a broken wire or connection to an input it would Cause a problem. But he said already he has both voltage signals and is having a problem on the ecu side of the circuit.
I'm sure with the ground jumped to the primary coil his fuel pump turns on on the first click.

 

Ok -

Let me state again, for search information, everything works. All wires have and see each other. When the ecu is plugged in and the key on, the CEL goes off, but the FP does not prime. IF i pop open the gray cover on the ecu plug, and ground A7, the FP comes on. RM7 internally for the ecu is not allowing the ground to be giving to a7. So i simple sliced the ecu wire to the main relay, and grounded it myself. FP comes on and car runs. I was just seeing if anyone has ever had the ECU not supply the proper ground.

The relay joints are fine. Everything has power as it should. When ohming all the wires, all wires see what they are suppose to. All grounds see the body, all wires see each end. The fuel pump comes on when the relay is jumped via a clip on the back of the plug or the A7 wire is grounded. This shows the ECU is bad, in this particular area. The entire ecu operates accordingly.


pin 32 of U7 drives Q10. Q10 switches the internal +5Vref across RM7 to limit current, in pin 7 (?) and out pin 6. Pin 6 of RM7 runs to pin 6 of QM3, which is the actual FP (and AC clutch) driver and 95% what's wrong. Pin 7 of QM3 goes to the main relay, and it sinks to ground on pin 1.

I can activate the FP by jumping pin 7 to pin 1 on QM3. Pin 6 of QM3 should exhibit some amount of voltage during the pump priming event when you cycle the key.

 

http://techauto.awardspace.com/mainrelay.html


Something that has helped me along the way.

 

I may have misread your post, but just because the injectors get power does not mean IPG1 and IPG2 have power and I see nowhere that the OP has said he has confirmed that both IPG1 and IPG2 have power, "But he said already he has both voltage signals" I looked again and could not find where he confirms that, he has said he has check fuses and wiring but not specifically the IPG inputs to the ECU/ECM. 94

 

I think he mentioned it just very vaguely

 

there is a larger electrolytic capacitor next to the plugs inside the ECU, when that starts to die due to age, it causes no fuel pump.
Replace it, repair any damage to the board (if there is any) and you'll be up and running again.

 

exactly what I was thinking, someone who tune, and works with ecus should know better.

 

lol - Shops , tuners, and everyone else under the sun post for ideas, suggestions or just to spread information.

What have you contributed?? Nothing.. i've searched your post, you sound like some ignorant troll who just wants to hate.

 

Still waiting for you to confirm that there is power at both IPG1 and IPG2 at the ECU/ECM, if you already have I missed it. 94

 

Both have power. for confirmation.

 

Then it is a good bet it is the ECU/ECM.

I do know a couple of guys that have supplied a ground to the fuel pump relay directly from chassis ground one is in a Civic the other is a G2 Integra, both have been running like this for a few years. 94

 

Thats how i'm doing it now, it runs great and starts up fine. I do not have a problem making this a semi fix until i can aquire a new ecu or fix this current one, i'm not worried about it though.

 

just for those finding this thread in the future... this should only be used as a very temporary fix.

hardwiring the fuel pump is functionally fine unless you have an accident or something happens, then the pump can/will become the never ending source of liquid fire.

 

my integra had same problem i sat there for 2 hours trbleshooting it and told the guy to try a brand new relay and shure sht it worked . there was oground. i brought 3 junkyard relays with me that were tested in my civic but they did not work in that integra for reason unknown to me.