I've searched around, I'm sure the answer is here but maybe someone can read this and redirect me, I'll go fast.

1994 Honda Accord, has had a history the past few years of not starting when in the sun/heat; would later start after the sun had gone down. I trolled around on the forums and sites and seemed that this was consistent with a failed/failing Main Fuel Relay (MFR). Even up to a week ago, with the old MFR, it would start -- I drove it up into my garage to get it out of the heat/sun in the meantime. So I bought a (new?) one, had it shipped, pulled the old, and replaced it. No change -- I thought for sure that would handle it. So here's the way it stands right now:

With the OLD MFR -- Engine light turns off after 2 seconds at Key position II (on). No noise from the fuel pump starting up -- nothing. Will turn over but not catch fire/run. I actually, very briefly, got it to kick for about a second several times, albeit roughly, but never caught. Since those first few times -- no more kicks -- it just turns over but never fires.

With the NEW MFR -- Engine light STAYS ON -- does NOT turn off. Still no noise from the fuel pump. Will turn over same as with OLD MFR, but no fire.

I've already pulled every single fuse, both under the hood and in the drivers compartment, and they're all good, by the way.

Would the initial symptom (not starting when hot) be consistent with a failed MFR? It certainly seemed dead - on consistent with everything I'd read.

Thoughts? Opinions? Links? Thanks in advance.

 

If the Check Engine Light (CEL) is on, have you checked for any codes. If no codes and CEL on w/ engine running, the ECU is in backup mode.

I have had to replace the ECU on my 94 Ex twice to clear faults that were not indicated by diagnostics; code 9 CYP and code 21 VTec solenoid valve.

Sun/heat cause stress on the ECU same as the MFR board opening circuit board faults.

good luck

 

Since you can't hear the fuel pump, check the circuit with a meter or test light?

• WHT/GRN from underhood 15A fuse #33 to main relay pin 7 (hot all the time)
• BLK at main relay pins 3 & 6 (grounds)
• RED/WHT from underdash 15A fuse #2 to main relay pin 5 (hot when on or start)
• GRN/BLK from ECU pin A7 to main relay pin 1 (I think this should be hot for two seconds when the key is turned to the ON position)
• BLK/YEL from main relay pin 4 to fuel pump (ditto)

 

pull the code before you groundhog around your wiring harness.

 

Okay, that'll be my next step; I'll see if I can get someone to check the codes for me. Might be a hard proposition since I can't drive it. . .I wonder if places rent a reader? At any rate, thanks thus far. If I think of anything else that might give more information, I'll reply. Thanks in advance.

 

Your car can't be read by OBD2 scanners. OBD2 only became available in 1996 Accords, and a few 1995 Accords.

You can read the codes yourself. Check DIY forum for instructions.

good luck

 

Fair enough, I'll give it a shot. Real quick -- if the battery has been disconnected a couple of times AND the engine hasn't fully started since then and run for any length of time, are those codes still going to be in place or will it have been cleared anyway? Thanks so much in advance.

 

It depends. Some codes will emerge w/i secs if the fault is detected; circuit faults. However, some codes require mulitple engine starts before the ECU will set the code, ECR codes, catalytic converter efficiency (not applicable to you), MAP probably. The manual doesn't identify the precise details of what conditions set the fault code.

good luck

 

Any CELs in those past years in conjunction with the no-start condition? It seems like finding out why the fuel pump isn't running would be the priority vs. determining why the CEL behaves differently with the old main relay vs. the new one. Putting a test light on the BLK/YEL wire on the fuel pump connector in the trunk would be where I'd start. It there's no power there for two seconds after the key is turned to the ON position then work backward to troubleshoot the other inputs and grounds on the main relay, starting with the GRN/BLK signal wire from the ECM that energizes the fuel pump side of the relay.

If there is power at the BLK/YEL wire then it's likely a bad fuel pump. Fuel pumps tend to go bad when it's hot out. Not real common, but I seem to remember a recent thread chasing numerous "problems" and finally finding a bad pump.

 

Alright then that is where I'll start. . .fuel pump is in the trunk, you say? I will dig around on here to find the DIY in testing this, unless someone so happens to havr the link handy in the meantime. TIA.