Car runs fine AFTER CEL comes on!!!
 

Hey, all, 1991 Accord CB9

Recently, my car will shudder and misfire and lose almost all power a few minutes after warming up. But then, as soon as my check engine light comes on, the wagon runs perfect!!!

I'm getting code 43, which, as far as I've read, points to bad o2 sensor.

Everything else is perfect, as far as how it runs, it's just that minute or two after it warms up. And again, once the CEL comes on, it instantly snaps back to full power and drives fine...

The fact that it runs perfect after it throws the code makes me think electrical issue, rather than mechanical.

(Fuel Pump, spark plugs, distributor, EGR, PCV all recently replaced)

Code 43 doesn't necessarily mean that you have an O2 sensor issue. Replacing the O2 sensor fixes the issue in most code 43 cases. I recommend getting a Denso O2 sensor. Paying the extra money where it has the proper connector is worth not having to solder in the original connector.

I think the code 43 in OBD1 world means the engine computer can't get the air/fuel mixture correct. The O2 sensor is the easiest "fix", but you also risk throwing money at the problem and not fixing it. I had this code on my 95 accord that was also obd1. I did the O2 sensor swap and it didn't fix my issue. When I had my accord, I didn't have the knowledge that I have now. The challenge I would have faced and you will face on an OBD1 vehicle is that you can't easily read live data from your engine computer.

Hopefully the O2 sensor is the fix and please let us know if this works. If not, I may have some suggestions for possible testing based on what tools/experience you have in troubleshooting.

Code 45 is fuel "too rich or too lean". Code 43 is just "fuel system problem" which isn't very helpful.

There's an old-school test for the O2 sensor:
Engine warmed up & idling.
Backprobe the signal wire from the O2 sensor with a reasonably responsive voltmeter. It should swing back & forth, centered around 1 volt, maybe a couple swings per second.
Sluggish sensors have a slower voltage swing and that's your clue to a common failure of those sensors.
If the voltage is stuck high or low, then it might be the sensor is OK but some other reason the system cannot get the mixture right. Maybe a vacuum leak sends it lean outside the control-authority of the sensor? Maybe a bad FPR or some other reason for it to go rich?