Any Honda techs on the board? My ECU fuse pops on long drives

  • Thread starter CyborgGT
  • 12 comments
  • 4,523 views
CyborgGT

CyborgGT

2,833
United States
Colorado
GTP_Cyborg
I've been trying to diagnose this issue for the past couple of weeks, and it's really starting to suck not being able to drive. Hopefully someone here can help me out?

1992 Accord LX
JDM H22A w/P13 ECU
Accord automatic transmission w/P19 TCU

My problem has been that the car will drive for a while, then the engine will shut off and leave me with a blown ECU fuse (under-hood 10A). Otherwise the engine swap has ran seemingly perfect in the year that it's been in.

The first time this happened the main relay and fuel pump were unplugged (thought maybe it was a simple failure of one of them), then after the car sat for a bit it started up and I was able to make it home.

The second time it died on me I tried doing the same things but with no luck, so the car was towed home. I tested the main relay (which was re-soldered for the second time in its life; I know these were problematic so I figured 'why not?'), fuel pump, ECU, and fuel injector resistor box. Couldn't find any problems, and yet with all this unplugging of things and the car sitting, it eventually fired back up. Though wierdly it was the injector resistor box that was looked at before the fuse stopped popping and the engine was able to start. I had been using the one that came with the H22, but I still had my old Accord unit laying around and decided to swap it out just in case.

Then it died a third time. The car can make short trips, but going for long drives seems to do it in. Maybe something's warming up to 'optimal failure temperature' or something.

Using a DMM, everything in the ECU fuse's circuit checked out fine except for the plug going into the IACV. Main relay, injector resistors, injectors, O2 sensor... all good (I might be forgetting something at the moment but it was all looked over). Battery voltage is supposed to be dropped across that harness plug (the valve itself tested fine according to what Haynes said to test for). The plug was dropping 1v or something like that. No continuity with ground was seen on either side of the plug (OL on the DMM). Am I wrong for thinking the ECU looked to be the culprit here?

So I just replaced the ECU and a few things have changed. That IACV plug has gone from dropping just the 1 volt to now 7 volts. Trouble codes have gone from no CEL codes and four TCU codes, to a solid CEL and code 14 from the D4 light. I don't remember the old TCU codes off hand, but I'm pretty sure 14 was one of them; it's for a line pressure control solenoid. I haven't driven the car yet, I'm afraid this mysterious short might steadily fry this newer ECU I paid dearly for (were OBD1 H22A1s just not available in America with automatics or what?). But I did let it idle in the driveway until the engine warmed up. I've got solid engine mounts, so it's hard to tell if the engine's running rough or not - vibrations are very normal for me.

To do tomorrow: It looks like my aim now is to find this line pressure control solenoid and test it. But I have no idea what it even looks like. Researching around the internet for a bit, I'll also check to confirm I'm getting battery voltage at pins A25 and B1 (power to IACV), 02 sensor, and the injector resistor box. I do have a new fuel pump I'll install, the one in there now is sounding like an old radio with the way its whine fluctuates.
 
I hate when you over think things and it turns out to be something simple.^^^
 
Pulled these off Autozone's site. They're not the best, but I don't have a scanner. Fuse 28, at the top-center of this first image, is the one that pops when the ignition is turned on (key on, engine off).

While the engine is that of a Prelude's, the swap involved re-using the Accord's engine harness and modifying it. For the VTEC system (VTEC solenoid, VTEC pressure switch, knock sensor, and IAB (secondary butterflies in intake manifold), I bought a subharness online - I know it's in perfect shape and doesn't touch the rest of the engine harness anyway; it pins these components directly into the ECU. I converted from an external coil to internal coil distributor, which involved splicing together the YEL and BLK/YEL wires at the plug going into my old external coil (top-left sector of first image below). Otherwise I just had to swap plugs to fit the new engine's components: EGR and oil pressure sensors, and extended the wire for the 02 sensor (inspected, looks/tests good).

And finally, some wiring adjusted for the automatic transmission and shifter to work right; I'll try to get a good diagram or something to show these mods. I don't see it in these diagrams, but I spliced together the GRN/WHT and WHT/GRN wires that would normally go into the TCU (spliced on the harness side; the TCU pins were removed). For the shift lock to function, I tapped into the GRN/WHT wire at the shifter position switch and ran it to pin A7 of the TCU.

I've ordered the factory 'Shop/Service Information' and 'Electrical Troubleshooting' manuals from HelmInc.com, just waiting for them to get here.

0900c1528005fa23.gif


0900c1528005fa24.gif


0900c1528005fa25.gif
 
Have you tried a fuse with higher amperage? 15 or 20?

And try the wires you spliced, see if anything is going on there. Usually it's something you modified yourself, instead of the factory wires.
 
Dennisch
Have you tried a fuse with higher amperage? 15 or 20?

And try the wires you spliced, see if anything is going on there. Usually it's something you modified yourself, instead of the factory wires.
Click to expand...

With the stuff he has done, I agree.
I also know that japanese cars up til the mid 90s do have wire hick ups here and there. As old as his car is, it's possible the wires have become brittle and broke at one of the kinks under the dash or where the wires come out through the firewall and into the engine bay.
When you swap an engine and move all the harnesses around to get them out of the way, that possibility increases tremendously.
Again, with all the stuff done, there are a lot of splices and connections you have to go through. Unless all the connections you made were properly soldered, sealed/shrink wrapped, you have to go over all of them before you look at the stock wires.
 
Looking at the system, it looks as if the feed from Fuse 2 powers the coil in the main relay, which in turn enables closes the contact from Fuse 28, which in turn allows voltage to be present downstream. You should not have power at those pins you mentioned unless the car is starting, or running from what I can tell. Those pins aren't the source of the power, the power is coming from the fuse, and then feeds the IACV, O2 sensor, the injectors via the resistor block and then the engine management system.

Weird then, that the Fuse 28 should blow when just in the On position. Maybe the wording was wrong in the description and it should have said "hot when on or start" instead of "hot when run or start" but I'd check that.

Sorry I didn't have a lot of time to look at it, and my trade is in industrial electrical systems, not automotive and the two are quite dissimilar.

Might I also suggest that if the voltage drop across the IACV has increased to 7V, that is a sign that the previous ECU may have had an internal error, allowing the resistance at the ECU to be lower, hence the higher voltage. I suggest this was a problem. Is the fuse still blowing?

Good luck!
 
I haven't driven the car since installing the new ECU, I'm afraid that if the short is elsewhere that it might have been what fried the old ECU. But I did let the car idle for a bit with the new computer in, and that's when I got a solid check engine light (doesn't flash codes) and a code 14 from the D4 (TCU) light. So for now I'm just going to try and diagnose this code 14, for a "line pressure control solenoid."

And no, Dennisch, I'm not about to use a higher-than-rated fuse in the ECU circuit.
 
Last edited:
Punknoodle
Looking at the system, it looks as if the feed from Fuse 2 powers the coil in the main relay, which in turn enables closes the contact from Fuse 28, which in turn allows voltage to be present downstream. You should not have power at those pins you mentioned unless the car is starting, or running from what I can tell. Those pins aren't the source of the power, the power is coming from the fuse, and then feeds the IACV, O2 sensor, the injectors via the resistor block and then the engine management system.

Weird then, that the Fuse 28 should blow when just in the On position. Maybe the wording was wrong in the description and it should have said "hot when on or start" instead of "hot when run or start" but I'd check that.

Sorry I didn't have a lot of time to look at it, and my trade is in industrial electrical systems, not automotive and the two are quite dissimilar.

Might I also suggest that if the voltage drop across the IACV has increased to 7V, that is a sign that the previous ECU may have had an internal error, allowing the resistance at the ECU to be lower, hence the higher voltage. I suggest this was a problem. Is the fuse still blowing?

Good luck!
Click to expand...

You read my mind.
 
When it rains, it f*king poors. My bike was stolen while I was at work last night, and was seen still locked up not 20 minutes before I left. It's a minimum 30 minute walk to anywhere I could possibly need to go.
 
Ah, **** man. I'm sorry to hear that. Theiving scum. If I were you I'd check out how that Fuse 2 is powered, and see why it is energised in the on position.
 

Latest Posts

Back