There is a fairly simple diagnostic chart for code 43 in shop manual. It will require backprobing A26 and D14 at ECU to measure O2 sensor voltage output signal. This should ID whether wiring/connector fault or faulty O2 sensor.

If fault appears w/o justification, then ECU is suspect.

good luck

the cat tends to eat up any smoke sent through the tailpipe its best to use seafoam imo ...considering ur issues i would strongly suggest u testing

Haven't had trouble with cats and the smoker before, but it's something to bear in mind.

When I get time between work and school, I'll dig into in again and see what comes up. Thanks for the advice, folks.

It's not trouble, it's how cats are designed, to neutralize gas. That's what they're for.

Sure, but what I meant by trouble is I've tested systems before to expose leaks and the smoke made it through the cat on the vehicle.

I thought the catalytic process needed the exhaust heat to truly function efficiently. You're saying you've seen it neutralize the smoke at room temperature (I'm assuming) and so I found that interesting because it's not something I've had an issue with. I'll make sure to keep that on mind as a possibility. Everyone's experience is helpful knowledge.

Okay, now that I'm out of school I finally have time to trouble shoot this thing again. Sorry to resurrect an old thread, not sure if that's a bad thing around here.

I backed probed the wires and had good continuity from the o2 sensor to the ECU on certain pins. I made sure to test through the plugs themselves in case there the wiring break might be right at a plug. There's a snag though. I was using All Data because I don't have a true blue shop manual.

It labeled the three ECU plugs 434, 435, and 436 respectively. Which looks correct, but it labeled them in reverse from what I can tell. The thing is that even though it also indicated A26 and D14 on the wiring diagrams, the way it labeled the ECU plugs makes it unclear whether or not pin 14 is the same as the actual D14 and the same thing for pin 26. Since it doesn't actually have a place showing where A26 or D14 is on the plugs, and the plugs themselves don't have the pins labeled, I had to use the diagrams they had for the plugs themselves. All those really have are the plug holes numbered from left to right, top to bottom. By counting over 26 pins it put me on a wire that was not the correct color. Obviously I wasn't getting continuity. So I just kinda looked for the correct colored wires. By finding the correct colored wire in the plug I could get continuity. I got continuity for the white wire (D14), the yellow/black wire (A26), the orange and black wire (A6), and a green and white wire that appears to be ground.

Since I was reasonably sure I had good wiring I ordered a good used ECU, but they sent me one out of a manual car. I imagine there's too many differences to use it as a test ECU, but since I have it I tried it just because. The car basically drove the same, and eventually the light came back on. No other noticeable differences between the ECUs.

So, should I still try an official Auto trans ECU or is there something else I should do with the wiring tests?

At this point I've tried two 02 sensors (Both Denso)
new Fuel filter
checked fuel pressure
Removing, cleaning, testing fuel injectors
testing for vacuum leaks
checked for exhaust leak around 02
checked timing
checked plugs and wires

Thanks.

To test O2 sensor for voltage output (0.1-0.9V alternating w/ mean of about 0.45V) you only need to backprobe the sensor wire (wht in 1990 manual I have and pin D14 at ECU) preferably at the ECU but could be nearer O2 sensor if you're confident in wiring integrity to ECU. Other pin is a good engine ground at any location. I pretty sure D connector at ECU is the smaller connector (22 pins), based on 1990 hard copy manual I have.

This could be an issue w/ fuel injector balance causing a lean or rich condition which would also set Code 43. A Honda HDS is necessary to examine the LTFT and STFT. See following link for clone HDS system at reasonable price.

Universal Honda Acura Diagnostic Can Bus System HDS Kit New | eBay

Also you might try checking fuel injector resistances for a weak injector. See link:

Part 1 -How to Test for a BAD Fuel Injector (Honda 2.2L, 2.3L)

This will only disclose a fuel injector w/ electrical fault. A partially blocked injector may work fine but release too little fuel. This will cause ECU to change pulse width to compensate causing increased LTFT. If LTFT gets too large, I suspect code 43 is set.

There is a leakdown test for fuel injectors which measures fuel pressure and uses a special electrical tool to rapidly open/close the injector and compare leak down times (time to leak from say 50 psi w/ 5 sec string of open pulses of injector). See youtube (scannerdanner) for an example.

good luck

Okay, I'll start with testing the again first since that's obviously what I'm equipped to do right now.

I testing good continuity in several stages. There's a major wiring harness plug on the passenger fender near the ABS pump. I tested good continuity from the 02 to that plug. Then from the other side of that plug to the ECU. Then, just for thoroughness, I tested from 02 sensor all the way through the whole harness to ECU.

So since that plug harness is easier to access, I'll use that for the next stage of wiring tests unless you really think it's worth getting back at the ECU again.

I did pick up an injector pulse tester a few months back, this one, and it didn't show an imbalance in the fuel flow of each injector. I didn't actually test for ohms because of that. I'd already removed them, cleaned their screens and the fuel rail, and reinstalled them. So I figured the fact that they showed even flow rates was good enough. I'll double check resistance.

After that, I guess it's time to pony up for the HDS system you linked.

Thanks a lot for the help.

looks like you've done anything you possibly could do... i would pick up another ecu from the junkyard at this point

I read your initial post and found point that "car runs fine as long as light (CEL) is off". I don't know but suspect the car goes to "limp mode" when CEL comes on. This may cause the drivability issues that you noted. This pattern suggests an ECU problem, however your repeat of problem w/ manual shift ECU doesn't predict a new ECU will be successful.

No easy path forward suggests itself.

good luck