I'm buying a non-runner MIEV here in Japan!

[Rational]

You could still swap it, your original DTCs should be replaced with VIN mismatch errors if the original contoller board is also defective

Okay. Well in that case, I have a second MCU inbound still as I ordered two. I guess I could just swap that in as is, unopened and see what codes show up.

Car won't move though right if the chassis numbers don't match? So I'll have to reprogram it if so before it will move?

Man.. I hope my FULL CHIP arrives soon...

 

[MickeyS70]

Car won't move though right if the chassis numbers don't match? So I'll have to reprogram it if so before it will move?

Probably not, but see what happens.
In the meantime did you re-check the DTCs; have they changed after replacing (all?) IGBTs?

 

[Rational]

In the meantime did you re-check the DTCs; have they changed after replacing (all?) IGBTs?

Yes, here above is what I got during the first scan after replacing the MCU.



Then after clearing the codes and trying to move the car in reverse and drive.



Then this is the result after clearing the codes again. But it still throws an engine exclamation point light and the P1A2A errors again.

Then for comparison below is what I got on my very first code scan with car scanner back on 27 Sept, 2024 before I touched anything on the car:


Page 2:

 

[Rational]

What do you mean by ‘good’?


OBD data has already determined that it’s a 88 cell 16kWh battery, what would be interesting to know is when it was built.

I gather you couldn’t find the label mentioned in post #14 but based on indicated range and current capacity it could well be the original pack, manufactured in 2010/11 with 1st gen LEV50 cells.

Oh, I hadn't noticed prior screenshots had said it was 16kWh. I only noticed the 88 cells up until today.

What was leading me to believe it was only a 10 kWh battery was the OBDZero page showing it as only 10.5 here:



Is this a default I need to change or something for it to show accurately in OBDZero?

 

[MickeyS70]

If you look at the schematic below you will see that there are actually 6 IGBTs / rectifier diodes. They seem to be physically arranged in 3 blocks inside the inverter.

What it also shows is that the output is directly connected to the motor unit, therefore if you disconnect G-07/G08/G09 (beware of potential HV)
and get the same DTCs when trying to move, the problem is most likely inverter related.

But if P1A2A is replaced with something else, suspect a faulty motor..

Is this a default I need to change or something for it to show accurately in OBDZero?

No need, this is accurate, it shows current max capacity based on SoH

 

[Rational]

If you look at the schematic below you will see that there are actually 6 IGBTs / rectifier diodes. They seem to be physically arranged in 3 blocks inside the inverter.

What it also shows is that the output is directly connected to the motor unit, therefore if you disconnect G-07/G08/G09 (beware of potential HV)
and get the same DTCs when trying to move, the problem is most likely inverter related.

But if P1A2A is replaced with something else, suspect a faulty motor..
View attachment 1219


No need, this is accurate, it shows current max capacity based on SoH

Funny you should mention this page. I was just looking at it in the online manual.

But I wasn't realizing I could disconnect the three motor wires and see if the codes remain or not.

So that's a good test then if so.

I could disconnect the main fuse and make sure there's no voltage inside, then disconnect those three motor wires.

If the codes go away then the problem is with the motor/reduction gear?

Very concise way to be sure it sounds like. I'll try it. Thank you. I wasn't quite catching that when I read it 15 minutes ago on my own.

This is why I hang out with you guys.

 

[Rational]

One curious thing I found in Car Scanner under the all sensors page is that only two of the IGBT temperature sensors read full temp.

The other one reads a very low temp. Do you guys also see this when you run it on your cars?



IGBT 1 temperature reads -1.67 C.

Yet IGBT 2 temperature reads 29 C and

IGBT estimated temperature reads 29 C.




But hobdrive looks fine. So maybe no issue with the temp sensor after all on my car?

 

[kiev]

i suspect there is some wiring harness issue or connector issue because of the low igbt temperature reading, also the impossible high motor currents. Also the CAN buss error codes.

Does the car start to READY mode?

If so was the HV error (!) lamp ON or not?

Did the HV error lamp come ON after you tried to shift to R or D?

If so, then try to clear the DTCs again, turn key OFF, then just turn the key to ON position and read any DTCs and note if any MIL. If DTCs show up, record then clear them, and post up what you find.

 

[kiev]

In the meantime IF you want to check the motor windings:
shift to N and set the brake, disconnect the 12V and pull the cover off the MCU,
check for no HV on the leads from the pack,

[side note: inspection of low voltage signal wires can be done later while cover is off]

back to motor windings:
and disconnect the big cap and remove the U,V,W motor leads from the igbt connection such that the motor leads are free and not contacting anything, then measure the resistance from phase to phase, then check from phase to chassis (motor housing).

Block off the wheels so the car won't roll away and release the brake, then
Carefully jack up the right rear just enough to rotate the tire, and set some safety stands,

Now rotate the rear wheel and listen for grinding or any noises. Connect your voltmeter to the U V motor leads and rotate the rear wheel, and notice if the voltage is swinging, then check the U-W leads, then V-W leads. The voltage level is related to speed, faster it spins the higher the voltage. If you want to see the sine wave on a scope, then create a virtual ground point using 3 small capacitors connected in wye. If the signal has too hash noise on the scope, then add 3 large value resistors in wye (R parallel to the C) to make a low pass filter.

If the motor has a shorted winding then spinning the wheel may be difficult as if the brake is holding. You can get a feel for this by connecting 2 motor leads together and try to spin the wheel, then connect all 3 leads together and try to spin the wheel. A shorted winding will also have a distorted or lower amplitude sine wave, or no voltage output at all.

 

[kiev]

i also suspect that the "old" or original MCU control board may be the source of the issue. Didn't realize that you were gonna be swapping the old board into the "new" mcu. You certainly like to do extra work, maybe slow down a bit and work smarter not harder. The old codes were stored in the old control board, and maybe the control board is damaged or defective due to the high current load.

Right now you are in a diagnostic and test phase, so don't seal everything up until you know it's fixed and working, leave the covers off so you can take measurements, etc.

Maybe the current sensors or their wiring or connectors are damaged? That all should be checked to verify is good. For example, The current sense signals go to the control board, then sent to the EV-ECU. so wring out the harnesses from the upper to the lower plenum of the MCU.

 

[MickeyS70]

Things to consider before swapping MCU/Motor

 

[Rational]

Does the car start to READY mode?

Yes. The car starts to ready mode without issue.

If so was the HV error (!) lamp ON or not?

It is not on during the initial start.

Did the HV error lamp come ON after you tried to shift to R or D?

Yes. The yellow car with exclamation mark comes on only immediately after I shift it into R or D.

If so, then try to clear the DTCs again, turn key OFF, then just turn the key to ON position and read any DTCs and note if any MIL. If DTCs show up, record then clear them, and post up what you find.

I have cleared codes many times after the error lamp shows up and the DTCs are identical to this:


Same results taken at the same time in hobdrive. E.g. after clearing codes, restarting, trying to shift to R or D, error lamp lights and then these codes appear again:


I have a video of it I will find and post so you can see it all in real time.



Here's the video.

This is the first start up of the i-miev upon arrival. The yellow exclamation mark check engine light appears immediately after shifting into R or D.

 

[Rational]

i also suspect that the "old" or original MCU control board may be the source of the issue. Didn't realize that you were gonna be swapping the old board into the "new" mcu. You certainly like to do extra work, maybe slow down a bit and work smarter not harder. The old codes were stored in the old control board, and maybe the control board is damaged or defective due to the high current load.

Right now you are in a diagnostic and test phase, so don't seal everything up until you know it's fixed and working, leave the covers off so you can take measurements, etc.

Maybe the current sensors or their wiring or connectors are damaged? That all should be checked to verify is good. For example, The current sense signals go to the control board, then sent to the EV-ECU. so wring out the harnesses from the upper to the lower plenum of the MCU.

Yes, I'm also suspecting the old/original MCU control board at this point. I also thought last night that perhaps the control board got damaged during the surge that damaged the IGBT.

It's not intentional nor desired on doing extra work. I'm just trying to make progress during the daylight I have based on my current understanding. So I am only doing extra work each time for lack of knowledge and experience on next steps. Thankfully with your input here guys I can work a bit smarter now with these recent specific points. Thank you.

Yes, I was hoping for a first fix with the IGBT swap like Neo5 had so risked sealing it up the first time. But also because I thought I had to have coolant running. But I thought about it last night and realized it won't overheat immediately during testing. So I will keep the MCU loose and uncovered so I can test and won't button things up again until all is confirmed to be working and the car moving.

Yes. It seems I may even have to dig down and somehow check the wiring harnesses at some point here if the initial checks of the control board and motor pass their individual tests. Would this mean running continuity checks on each wire end-to-end with a multi-meter? Checking resistance as well somehow?

 

[Rational]

In the meantime IF you want to check the motor windings:
shift to N and set the brake, disconnect the 12V and pull the cover off the MCU,
check for no HV on the leads from the pack,

Yes, I do want to make further progress today so will also run these tests so as to give more information on the motor condition. These are excellent recommendations and should give some insight into those components at least.

back to motor windings:
and disconnect the big cap and remove the U,V,W motor leads from the igbt connection such that the motor leads are free and not contacting anything, then measure the resistance from phase to phase, then check from phase to chassis (motor housing).

One question on this though. Why disconnect the big cap? Will things still work without it for testing or wouldn't it be required to remain in place to maintain the system design in an operating condition? Or not really for testing and just adds risk during diagnosis testing?

Carefully jack up the right rear just enough to rotate the tire, and set some safety stands,

Will do. But curious on why only the right rear? And is there no difference for my JDM RHD car on which wheel is chosen? Can I jack up both rear wheels for the test?

Now rotate the rear wheel and listen for grinding or any noises. Connect your voltmeter to the U V motor leads and rotate the rear wheel, and notice if the voltage is swinging, then check the U-W leads, then V-W leads. The voltage level is related to speed, faster it spins the higher the voltage. If you want to see the sine wave on a scope, then create a virtual ground point using 3 small capacitors connected in wye. If the signal has too hash noise on the scope, then add 3 large value resistors in wye (R parallel to the C) to make a low pass filter.

Fantastic test idea here. I don't have an o-scope so will have to use my multi-meter for the short-term. I can get an o-scope next month if I'm still in need of it to fix the car though.

If the motor has a shorted winding then spinning the wheel may be difficult as if the brake is holding. You can get a feel for this by connecting 2 motor leads together and try to spin the wheel, then connect all 3 leads together and try to spin the wheel. A shorted winding will also have a distorted or lower amplitude sine wave, or no voltage output at all.

Genius idea. This will be insightful for sure. Thank you.

 

[Rational]

Things to consider before swapping MCU/Motor
View attachment 1225

Oh, interesting. On a used motor that paper factory tag containing the initial phase angle is long gone I'm guessing though right?

So I wonder how to obtain it?

Looks like I can get the current phase angle from my current motor somewhere in the "service data" when doing an OBD reading with MUT3.

Big thank you for pointing this detail out.

Well, it's 8am here so I'll dig in now. Looks like a fun day of learning at least in performing these tests.

Thanks guys. I'm grateful for you both.

 

[Rational]

Just sharing some initial findings this morning.

I've disconnected the three motor wires from the MCU: G-07/G08/G09.

G-07/G08/G09

At first it was still throwing the same check engine light and P1A2A error codes when shifting from P to R. On 2-3 attempts and clears.

BUT, when I put it into N and cleared codes, then attempted again to go into D it did not. Now I can go to any position without the error light coming on nor any codes being thrown.

Note again that this is with the UVW leads disconnected. G7/G8/G9.

So I wonder what this really tells us? Were the first few attempts residual error readings and it's now in the clear because the motor is disconnected?

Does that seem to indicate the fault is caused by the motor? I guess reconnecting UVW again and retesting will tell us if it does or doesn't throw error codes again and if it does it's the motor?

Anyway, in the meantime, I've just disconnected the aux and removed the main traction fuse in preparation to test the motor spinning output with the multimeter between each of the three UVW leads.

As for the physical rotating of the rear tires, they do spin. Right rear (driver's side) is about twice to three times more difficult to turn than the left rear during the resistance part of the cycle.. But they both spin. There is a point in the rotation of both where the resistance is higher than the rest of the cycle. Perhaps this is the motor resistance against the magnet? It's only perhaps 1/4 to 1/3rd of the total cycle that has this resistance.

Now rotate the rear wheel and listen for grinding or any noises.

No grinding noises heard or felt on either side.

 

[Rational]

Here's a short video of me testing the rear tire turning ability in order to test the motor and reduction gear for damage.

 

[Rational]

Here are the results of the tests between the UVW leads.


Between G-07 and G-08.


Between G-08 and G-09.


Lastly between G-07 and G-09.