I posted this on another Honda forum, but I thought I'd check and see what the collective wisdom over here has to offer. I wasn't sure where this should be posted so I put it here. If it belongs elsewhere, please excuse my ignorance.

I went against my gut and listened to the gal at the JDM engine place nearby and put an F23A1 motor in place of the F22B1 in my Accord. On the advice of the JDM place, I stripped the F23A1 down to a long block and used all the goodies from the F22B1:
intake
exhaust
fuel rail
fuel injectors
fuel pressure regulator
sensors
ECU

Everything went pretty well, but I've got some pinging in 3rd and 4th gear between 2,000 - 3,000 rpm at a light to medium throttle. It's the worst at about 2,500 rpm with more throttle rather than less. I assumed the pinging was due to the half point compression difference and the fact that the ECU can't take the knock sensor input. I run premium gas and also put in some plugs that were one range colder. The premium seems to have lessened the knocking and the plugs seemed to have no effect.

Anyway, I took a road trip to Montana and pulled the plugs when I returned. All of the plugs were white, so it appears as though the knocking may be due to a lean condition, so that makes me wonder if the problem I'm having is an ECU calibration issue. If so, I think it will be hard to dig my way out of this without spending a lot more money. I have checked the EGR. It works and the passage was cleaned when I had the intake off. Also, I have found no vacuum leaks. No codes before the swap and no codes after the swap.

I'm wondering if anyone else here has done this swap and if they've had any issues related to pinging. If so, have you done anything to fix it or do you just live with it?

Thanks in advance.

 

Normally you'd want to use the same ecu as the engine you're using unless they are functionally the same. However it would take a lot of work to get an f23a1 ecu to work in your car. It might be best to convert to obd1 and use a chippable ecu with a tune, unless your state has strict emissions testing.

You could just try to get the supplier to sub that out for a f22b sohc unless they don't have any.

 

check out this series on DIY Auto Electrical Mods. the link here is for part 1. there are a total of 13, all listed at the bottom.

this type of mod in some situations (such as yours) can be a simpler, more cost-effective way to set things right than installing and tuning a chipped ECU. of course be sure everything mechanically is right before you begin (mechanical timing, fresh plugs, no CEL, no leaks in or out, good compression...)

 

I hear ya, holmesnmanny. I didn't like the idea but the gal that sold me the engine says that they've sold over 150 for the same swap I did and never heard any complaints from their customers. I trusted what she said against my better judgment. I'm getting an education and this is the price of tuition.

I looked into reverting back to OBD I so I could get the ECU tuned. That's a pricey option and I'm afraid of opening another can of worms. It would cost more than the engine by the time you include parts and labor for the tune.

Shmallow, thanks for the link. That looks like some really good info and the two examples he provided (MAP and EGR) just might do the trick. Normally, I don't like hacking into factory wiring. I'm going to try to build up a wiring harness with some breakouts so I can easily revert back to the original configuration if things go wrong. The only issue I have with that is finding a connector that matches the one on the sensor. I've got plenty of the connectors that mate with the sensor from the JDM harness. I might have to dissect a donor sensor. Any leads on where to find the connector on the sensor, other than the sensor itself?

 

You know, the more I look at your swap I would think it actually should run fine even with the compression differences, for an p13L10 ecu like I have, which is usdm, it runs fine even with the compression difference of my jdm motor.

I'm wondering if maybe the injectors for the b1 aren't big enough for that engine. You should try to find out(this is where u do some of the work yourself lol) if the cc's for your b1 injectors are as much as the a1 injectors put out. I don't think the a1 injectors fit in your rail. I think i seen someone else have issues getting it to fit but you should try 'cause that could be the key.

Also, did you change any of the wiring and did the connector for the crank sensors mate up with the a1 motor or did you have to change anything ?

 

Looking at gaskets it appears the F22B1 and F23A1 use the same intake and exhaust manifold ports. .1 litre/.5 CR difference is not going to cause a ping issue. If the engine is running lean enough to ping I would look at other possibilities.

You stated the EGR plenum is clean and the valve works.
Did you use new gaskets for the intake and exhaust?
Only thing I can think of is either the base ignition timing is not set correctly, the F23 may have had some carbon buildup in the combustion chamber causing an even higher compression ratio, and screwing up the A/F ratio, or you missed a vacuum hose. What is your fuel pressure looking like?

 

Since his car is obd2 base timing is nonadjustable.

 

Already checked out the injectors. The injectors are the same size. The F22B1 has low impedance injectors while the F23A1 are high(er) impedance (no resistor pack).
I didn't change any wiring. The crank sensor mated up just fine.

New gaskets are SOP for me. If there were a leak, it would have to be uniform for each cylinder as the plugs looked identical.

Carbon buildup is a possibility, but that engine had very little time on it. I may do a Seafoam on it. GM topend cleaner is no longer available.

Before I removed the engine, I put a piece of blue tape on all hose connections and labeled them. When I installed the hoses, the blue tape was removed as I made each connection. I’m nearly 100% sure I got them all, but I’ll check again.

Fuel pressure and a look at live sensor data are next.

Thanks for the suggestions. I'll keep plugging away.

 

Did you ever clean the egr ports ?

 

It just dawned on me. Knock sensor. When you replaced all the sensors did you leave the F23 knock sensor on the engine? Or did you install the F22s?

Knock sensors are tuned to an engines bore size, if the wrong sensor is used it may not work or send a signal to the ECM when the engine is experiencing detonation.

 

lol thanks, mike

The B1 ecu doesn't test for knock sensor input so it wouldn't be a factor here.

I'm really baffled as to what your issue could be. Normally low mileage JDM engines come over pretty clean.

 

I was actually wondering where the hell it was on the B1 as I couldn't recall seeing it anywhere. I guess that explains it...

Strange though, no knock sensor on the CBs or CDs?

 

It's weird cause it's not on the b1 vtec motor but it is on the h23a non vtec. lol

I'm not sure if the V6 is called a CD but it does have a knock sensor...or two rather.

None of the usdm CB's have one.

 

Granted the crank sensor on the 96/97 is actually at the crank and tells the PCM. But there still is a distributor used for the secondary side. Or is the distributor physically locked out from being adjusted in its clocking/timing from the cam/head?
If that is the case, then rechecking the camshaft timing belt may be in order.

 

I don't know just exactly how it's able to set it's own ignition timing I just know it's the way it is.

You can still turn the distributor but it doesn't affect the ignition timing since the two sensors needed for timing are in a different spot obviously, a set spot.

 

True, if the PCM knows when the crank is at X° and it signals the igniter to fire it will do it at X°. But if the distributor is cranked too far either way, it may not be able to advance/retard timing enough. It still relies on a mechanical interface of cap and rotor, albeit if it is off enough it may not cause a misfire but poor spark performance.

basket_case forgive me if this was already asked, its 3A.M. here and I'm getting blurry eyed, was the O2 removed or changed recently? If so what brand did you use?

 

Thanks for all the inputs so far. Sorry for going dark. I’ve been busy doing homework. Lots of things to check on this car and everything else in life doesn’t stop just because your car is acting up.

I did, but maybe I missed something. I’m starting to doubt myself. I’m thinking of pulling the EGR plate and taking another look. And, if bad, I’ll remove and give it another cleaning. Could be like carb jets on my bikes - tough to clean the first time. However, with the engine at idle, I did pull a vacuum (not much) on the EGR valve and it stalled the engine. Also, I teed a vacuum gauge into the EGR line and measured vacuum up to about 3 in Hg while driving around town, so it looks like that part of the EGR system is working (assuming 3 in Hg is enough).

This engine was super clean. I pulled the valve cover and the oil pan. The aluminum wasn’t even stained brown as happens with time. Looked like a new motor. It burns no oil and it runs like a champ at full throttle. The intake and exhaust ports were clean. I didn’t think of sticking a scope in the cylinders but I don’t expect the heads/pistons to be loaded with carbon based on what I’ve seen.

The O2 sensor has not been changed. I’m going to place an order for one from Rock Auto tonight or tomorrow night. I plan to go with Denso. My understanding is Denso is OE, but if there’s a reason to NOT go with Denso, let me know. I don’t mind throwing a new O2 sensor at it - with 200k on it plus the fact that it has been exposed to some coolant in the combustion gases from the bad head on the F22B1, I think it deserves to be retired.

I took some data with the engine analyzer and need to comb through it. There weren’t as many parameters as I hoped for, but it’s better than nothing. At first glance, it seems like everything was in spec. However, the specs are pretty wide so I’m not sure how to tell if something is bad in real life even though it still is “in spec”. I don’t have access to an exhaust gas analyzer so I can’t tell how much excess O2 there is.

Timing was right at 15 degrees at idle and I saw as much at 35 degrees while driving.

I found that the TPS had drifted slightly, down to 4% with the throttle closed, so I set it to 10% per the Honda manual. I didn’t expect any difference in operation and there wasn’t any.

I checked the MAP sensor against a vacuum gauge and also against the MAP sensor that came on the JDM engine. They all agreed within about 0.5 in Hg at three different pressure/vacuum points.

Long term and short term fuel trim were greater than one. If I understood the funky wording in the manual, it means more fuel is being injected, which would tend to richen the mixture. The injectors were flow tested and cleaned at the time of the swap.

I’ve got a list of things to check and I’m burning ‘em down. If you’ve got any more suggestions, let me know. Meanwhile, I'll keep at it.