1997 Accord EX OBD1 engine swap ECU / CKF questions
 

Hi, I am helping someone with their 97 Accord EX 5 speed; the original engine died and they installed a replacement F22B1 engine. Unfortunately it appeared the replacement was an OBD1 engine instead of an OBD2 engine. I'm not clear on if any / what all parts may have been changed on the OBD1 engine to OBD2 parts (distributor, maybe?) but the car seems to run and drive perfectly fine. It does however throw a few DTCs for P0301, P0303, P0304, and a pending P1399 - misfires on cylinders 1, 3, 4, and it looks like P1399 is a "random misfire". The car does not seem to be misfiring at all... it runs quite smoothly so it seems like it might be a problem reading a sensor that's causing the errors. I've read people talking about having to swap oil pumps or even easier to do a CKF (crank fluctuation sensor) bypass (.:FFS TechNet : CKF Bypass Trick :.) but the 37820-P0H-L02 ECU in the car does not seem to have wires for a CKF sensor. Pins C1 & C11 are empty on the wiring harness. Do the Accord F22 OBD2 engines have this CKF sensor or is this something on the D/B series on Civics, etc. only?

If they don't have a CKF, is there another sensor that may cause the DTCs listed? It really does seem to run great so it doesn't seem like it's actually having any misfires. Thanks!

The OBD2 F22B1 engine has the crank position sensor on the engine block (and maybe the top dead center sensor). The OBD1 F22B1 had all 3 sensors in the distributor.

Look by the timing belt cover on the firewall side of the engine block near the crankshaft for an engine wire harness that has nothing plugged into it.

Also let us know how many wires are going to the distributor on the current setup as well.

My 97 Accord with F22B2 (non-vtec) has both sensors at the crank. I believe the vtec version has 1 in the distributor and 1 at the crank. Either that, or both are in the distributor.

I've found that a harness extension was made that connects the 4 pin ODB2 CKP/TDC connector near the timing belt cover and runs over to the OBD1 distributor. The distributor has 2 plugs: a 2 pin (black/yellow wire and blue wire) that I believe is the ignition and RPM output. There is also an 8 pin (7 wires populated) that looks like it is the ignition input (yellow/green wire) and then 2 wires each for CKP, TDC, and CYP (4 of the wires for CKP and TDC are what run over to the timing belt cover connector)

I hooked up an oscilloscope to ECU pins C2 & C12 (CKP), C3 & C13 (TDC), and C4 & C14 (CYP) and all have signals that seem to check out from what I've seen online. As an example, CKP was about 8100 per minute / 12 teeth = 675 (idle was about 700 rpm so this looks OK) the TDC and CYP also seemed correct other than when I was looking at both, it seemed like the CYP was slightly out of sync with the TDC. I've attached a picture of the TDC and CYP waveforms to show what I mean. It seems like the CYP signal is "late" compared to the TDC signal (CYP signal occurs every 4th TDC, so that part seems OK) The "out of sync" may be OK as well, I've seen a few waveforms online that also look like this (maybe something to do with timing advance?)

The car seems to run great, it's just that the CEL is on and it won't pass inspection that way. Any ideas of other things that might be a problem?

ETA: The CEL doesn't come on sitting there idling so I'd think the signals would be OK? In fact, you can drive it around and it's fine for a while but it eventually will throw the codes listed above. It doesn't seem like light / hard driving makes a difference - even taking it very easy it will eventually throw the errors.

Your PCM is for the 97 EX manual transmission coupe, so it is OBD2.

Which sensors are sending signal to the PCM? Trying to figure out if the distributor sensors or the sensors under the timing belt cover are sending signal.

Since you have an oscilloscope, is there any chance you have a scanner that can read live engine data?

Can you find out if the replacement engine actually fixed the problem, or if the problem remained after the replacement from the previous owner?

The CKP, CYP, and TDC sensors are all being fed from the OBD1 distributor. This replacement engine is an OBD1 engine that doesn't have the sensors under the timing cover (hence the extension that connects the distributor connector over to the OBD2 wiring harness connector on that side of the engine) Are the signals different from OBD1 to OBD2? It doesn't throw errors straight away, it's only after driving a while. This issue is with the new engine only, the original engine did not have any of these issues; it was fine until it died from a mechanical problem (I don't recall what at the moment)

I do have an ELM327 USB OBD2 reader. Are there specific things to look at that might give a clue what the problem is? Right now there are no pending or current DTCs but I could capture other sensor measurements if they'd help.

I would line up the crankshaft to TDC and verify the cam sprockets are also at TDC. The problem with the OBD1 signal setup is the camshaft sends the signal for the cam and crankshaft to the PCM. Maybe the timing belt is slightly off? I'd also look at every ground in the engine bay using that shop manual to ensure they were all reattached.

If this were my car, I'd consider eventually getting the crank and cyp sensors attached to under the timing belt cover. This would essentially be done when you replace the timing belt job. You'd need an oil pump housing from and OBD2 accord, the obd2 crank sprocket, and the OBD2 distributor. You would need the sensors too.

And IF you still have the original engine, there are all of your donor parts.

The original engine is long gone. The car was driven for a year or more with this new engine (in a different state with more lax/non-existent inspection) but the car has been mostly sitting more recently since it can't pass inspection in the current state with the CEL on. Eventually swapping the oil pump, etc. to OBD2 parts when it needs a new timing belt wouldn't be bad but I didn't want to have to preemptively do all that work when it seems the sensors in the OBD1 distributor are sending their signals to the ECU. Especially since it is not throwing any codes related to the TDC or CKP sensors like intermittent, no signal, or malfunction DTCs. I'm wondering if something else might be a problem between the OBD2 car and OBD1 engine but am not sure what else might be causing the false misfire codes? I don't know if there's a list of grounding locations but all the ones I've seen are connected.

Maybe take a step back and connect a scanner that can read some live data. See if the fuel trims remain close to zero. Look for any obvious sensor data that seems way off.