Does anyone have an idea? My 1992 Acura fuel pump will not come on for the intitial 2 seconds to charge the system. The problem seems to be the computer will not let relay coil close. It has voltage on one side of relay coil, but the green wire back to the computer will not ground it and close the circuit. I can jumper the green wire to ground and the pump will come on so it is not a pump or relay issue. Does anyone know why the computer might be holding it out?

 

Have you tried to repair or replace the main relay?

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

 

Yeah... I repaired it then on top bought a new relay to be sure. The problem seems to be the green wire back to the computer is not grounding and changing the relay state. I can jumper that wire to the frame and it will operate. Not sure why the computer will not allow the circuit to close.

 

Try another ECU. Not uncommon for an ECU to go bad. Not to mention it's a DA so it's under the pass side where DAs like to collect water.

 

As mentiond, it is most likely the ECU/ECM, but confirm you have power at IGP1 and IGP2 at the ECU/ECM when ign. switch is on, they are the only two (2) yellow/black leads at the ECU/ECM. 94

 

I don't have another ecm on hand. The LED on the ecm seems will indicate some trouble codes, but it was doing that while the ignition key was on without the jumper. If i am reading it correctly they looked like a 12, 7, and 8.
The first one is an emmisions problem, but the other two are TDC sensor of throttle position sensor. I am wondering if any of those would not allow the computer to close the fuel pump circuit through the computer.

 

Is your car automatic?

 

Yes it is auto.

 

"As mentiond, it is most likely the ECU/ECM, but confirm you have power at IGP1 and IGP2 at the ECU/ECM when ign. switch is on, they are the only two (2) yellow/black leads at the ECU/ECM. 94"

Is there anything that would make it obvious the ecm is bad? If i check for continuity between the green wire input from the main relay and the frame (ground), the input never changes states. I will check the IGP1 and 2 when i get the chance and when i locate them on the computer.

 

So I checked voltage to ecu and removed cover to see if any evident problems with board. Had a small rust spot on board at one of the pins. Removed rust and blew board off with canned air. Plugged back in. Turned switch on and fuel pump came on for its 2 second charging of fuel system. What do guys think...find another ECM or trust this one?

 

Repair the cowl leak above the ECU/ECM and you should not have any more problems. 94

 

I repaired a leak i had there a while back. I did unplug the ecm and let it sit. After plugging it back up, if i turn the key to the 2nd position after a few seconds the pump will charge up the fuel system. It seems as though the more i start it the shorter amount of time it takes for the pump to come on. I was wondering if the ecm might have a problem or if it is something else?

 

If it got wet because of the leak, it is definatly possible the ECU/ECM was damaged.

Remove and open it up and spray down the CB with contact cleaner allow it to dry, then when you plug it in again, spray a little WD40 into the plug(s) ans socket(s) to remove any corrosion from the terminals and see what happens. 94

 

The problem with fuel pump coming on was better after unplugging and plugging back up, but it would take about 15 seconds for pump to come on when i turned key to the second position. I removed covers from ecm, and removed the board from the ecm housing. I used 97% alcohol and a soft camel hair brush and scrubbed both sides of circuit board. I then blew the board off with canned air. I reassembled the ecm and cleaned the plugs. I also used di-electric grease on plugs to keep out moisture. Now when i turn the key on the fuel pump immediately comes on and charges system. I am gonna call that repaired. Thank you guys for the help.