Hi all,

I swapped in a (used with 12,000 miles) 2020 PB8A 10-speed transmission in place of my failed 2018 PB8A unit that had 244,212 miles on it. I just finished yesterday except for the bottom skid-plate stuff, correcting fluid levels after the engine has been run, etc.

Here are some establishing details:
Apparently the Transmission Control Module and the rest of the vehicle need to be "in-sync" before the car will start. More specifically, it says this in the service manual:
"If you replace or substitute the TCM, the vehicle cannot be turned to the ON mode until the keyless access system registration procedure is complete."

The old transmission has some metal and a few small parts floating around in it so I don't want to use anything from it. (I was having valve body issues as it failed) The manual also says that the valve body and TCM are mated together. If you change the valve body, you have to change the TCM with it.

So, I'm using the 2020 transmission with the 2020 Torque Convertor with the 2020 valve body and 2020 TCM in my 03/2018 Accord. I did have to use my existing 2018 TCM wire harness since the junkyard cut the 2020 harness all up. They seem to have the same wires at the TCM connector and the Countershaft Speed Sensor.

This is the issue:
I think it's fine so far, but I can't tell much of what is happening yet. I can't even get the car in the "ON mode" as mentioned by the manual.
I get in, the START button is slowly flashing and the white ring around the Instrument Cluster is lit up. As soon as I press the START button (without pressing the brake, but it does the same with the brake pressed) I hear a click in the engine area, and the START button goes dark, the white ring continues to be lit and nothing else happens.
I can hear a slight electrical whine coming from the PCM or TCM too. If I remove some TCM fuses, you can tell that the TCM is moving valves as it gets power.
So, when they say that "the vehicle cannot be turned to the ON mode" it looks like this is what I'm seeing.

To perform the keyless access system registration procedure I gather that the dealer uses their HDS devices plugged into the OBDII port. I really don't like my local dealer much (so much so that I will get parts from other dealers 50-90 miles away when I'm traveling or have them shipped from even farther) and would like to have my own equipment anyway. Today I ordered a Foxwell NT710 for Honda/Acura but I'm not positive that it will do what I need at this moment. (I still can use it for other stuff I'd guess) I already have a OBDLink+ but it doesn't see the PCM currently. Possibly because it's powered by OBDII.

Does anyone have any tips or confirmations that what I'm seeing is expected with a new TCM? I'm all ears.

 

"So, when they say that "the vehicle cannot be turned to the ON mode" it looks like this is what I'm seeing."

FWIW, I have an answer to this part I believe. I wanted to be sure that I could pinpoint the likely cause of the car just clicking and the START button going dark. I rolled the old transmission so that it was sort of under the car (up on jacks) and disconnected the TCM cable from the new unit and plugged it in to the old unit. I connected an extra ground as well in case it makes a difference.
With the old TCM connected, the car lights up like normal and shows all the unhappy error screens that accompany a battery being disconnected for a while. I'm pretty certain now that the TCM just needs the keyless access system registration procedure done now. Hoping that either the Foxwell NT710 can do it or someone else has a good option for me.

 

A short video of the transmission lower and raise events.

 

Sorry, I haven't updated yet. Here it is.
The Foxwell NT710 is nice enough, but it can't seem to communicate with the new 2020 TCM directly. I'm guessing that the Honda devices can communicate to the various modules to a greater degree than the NT710 can. There does seem to be a procedure where the TMC can be swapped from old to new in the NT710, but it looks like you have to start with the old one first. I didn't want to get into that in case it didn't work right.

I discussed the concept of using the same TCM that comes with the Valve Body with a Honda Technician. He said that they typically use the existing modules when they can to avoid issues. I didn't wholeheartedly believe in his approach, but I had a thought that made it an easier concept to try.

I re-installed and reset the existing 2018 TCM before ever starting the engine. With the transmission having only 12,000 miles, I figured it might be new enough that a reset of the TCM would suffice.
At first there was a little bit of shaking/noise so I loosened the flywheel-to-torque-convertor bolts a little bit and let the two parts find their best, smoothest, natural alignment with each other with the engine running and then re-tightened the bolts. That seemed to make a noticeable difference. I ran the engine a bit and let it warm up and then checked the coolant level and ATF 2.0 levels and filled them up a bit more. Due to the next issue, I was not able to yet do a complete ATF level checking procedure because I was not able to get into Sport Mode.

One thing I struggled with for a while was clearing all the VSA/TPMS/BRAKE/ACC/etc. codes. I could get nearly all of them to go away but then when I drove again for a very short distance (+/- 100 feet) they would all come back. 11 Honda safety feature codes is what I would find each time. Sometimes a DTC would line up with one of those but I could not trust the DTC info since one of them was saying that I had a hot brake pad when I hadn't actually driven anywhere yet.

The one thing I could not get rid of was the steering alignment error. I have come to believe that it was the only true issue, and that all the other errors would pop up since the systems are dependent on each other. I want to stress that I was very careful to mark the position of the steering linkage in relation to each other and to put it back exactly the way it was. I also was careful to mark the subframe alignment in relation to the body of the Accord. I can't quite figure out why the car thought that the steering column alignment had changed other than just having the battery out for an extended amount of time had reset something. Unless I missed something, I can't see how it would have even known what I did while the battery was disconnected.

In any case, I started with what seemed to me to be a reasonable amount of gas in the tank (somewhere around 1/4th of a tank or a bit more but dwindling as I ran the engine and went for short test drives) and after checking over a bunch of stuff, I recalled that I read somewhere that calibrations should be done with a certain level of fuel. (Maybe 1/4 or 1/2?) I don't remember what the threshold was, but I decided after a couple of days of frustration to go fill up the tank and try the steering calibration again. That's what worked! I probably could have saved 4-10 hours of work had I had more gas in the tank to begin with. Once I cleared the codes and performed a steering calibration, everything went green and have stayed that way since.

I've been dealing with one bent wheel 0.102" that I had straightened (I still have two less-bent 0.064" wheels from potholes, etc. that I will have straightened later) and now I've found that two of my tires themselves are a bit out of round too (internal belt damage, I thought it was that bent wheel all of this time) so I'm going back to a different set of tires that I had some really good luck with before these. Those will be put on this Saturday the 25th.

I've put about 250 miles on the 2020 transmission by now and it's running really smoothly it seems. I accidentally spun the front wheels a few days ago and had to remind myself to be careful until I was sure everything was working well.

I will keep the 2020 TCM and another sensor and then recycle the old transmission I think.


Now I have a pretty clear idea that I'm seeing indications of the beginning of a head gasket leak. I'm guessing that started around 220,xxx miles from what I recall. That was the first time I heard air bubbling around in the heater core on start up and I've been keeping it topped up and watching it since. It has never overheated or even looked like it might overheat. Reasonably often the coolant expands into the overflow bottle and never contracts back into the radiator at cooldown. I've bled the air out of the system many times and it's still ending up with air in the radiator. Also, I'm noticing some rough startups sometimes too. Looks like a head gasket will be my next big project.

 

I suspect your issues could be solved with an SM2 Pro clone from AliExpress and a copy of Honda Diagnostic System. The functions within it should enable you to perform the necessary calibrations to clear the error you're experiencing.

EDIT: I should've read further in the paragraph before responding. Glad you got it sorted.

I would also perform a 'block test' with the combustion gas detecting fluid to determine if your head gasket is indeed compromised. You can also pull the plugs and take a look at the electrodes for any inconsistencies.

 

I have pulled the plugs a couple of times quite a few months ago. I also used a borescope to look in the cylinders but didn't see anything unusual on the piston surfaces. I've got a newer borescope arriving this weekend that I'll try. perhaps I can see the gasket areas better with it. I have also considered using the combustion gas detection method. So far, I have not noticed the smell in the coolant, but my nose might be out of calibration.

 

When you're bleeding the cooling system are you setting the heater to full blast?

 

Yep, I am. 🙂 Also I've done it on level ground and on my 6° angled driveway so that the top of the radiator is the highest point.
I rinsed out and changed the coolant the first time at the interval specified by the Maintenance Minder (which was at 110,000 miles in 2020 according to my list) and it was fine for around another 110,000 it looks like.

In late January of 2023 around 220,000 was when I noticed the water noise in the heater core on startup and I started experiencing rough startups sometimes in April 2023 while on a road trip.
I'm guessing at cylinder 3 and maybe a little of 2 as well based on the stored misfire codes.
Spark plugs were finally changed around 190,000 miles even though it was called for in the Maintenance Minder around 95,000 to 100,000. I checked them back then and they still looked good and I had no misfire history.
No misfire codes until April 2023 to my knowledge.
​​​

 

Yeah 190k on a set of plugs will do that. I suppose keep an eye on the level and go from there. Maybe the block test kit as well.

 

Oh, no, I didn't explain very well. I didn't mean that the misfires occured at 190K because of the age of the original plugs. I just changed them because I was reminded again to do it and I skipped it the first time around.
I just checked and I actually changed them July 30, 2022 at 208,139 miles.
I didn't notice any misfires until April 2023 around 232,800 or so and that was a couple of months after I noticed the lower coolant level and started topping it off some mornings. It doesn't need topping off every time so I suspect that it's a minor leak currently and sometimes it leaks more than others.

I'm firmly of the opinion that the head gasket issue is fouling a plug or two for some startups. Sometimes it's fine, other times it misses for 15-20 seconds on startup. Might be related to where the piston stops the night before. After they warm up I don't experience any further misfires at all until the next morning.

 

You could absolutely be right. Time for further testing.

 

New info on the head gasket issue:
https://honda-tech.com/forums/honda-.../#post52776232