Tech help needed with battery temp sensor [SOLVED]

Mitsubishi i-MiEV Forum

Help Support Mitsubishi i-MiEV Forum:

This site may earn a commission from merchant affiliate links, including eBay, Amazon, and others.


Well-known member
Jul 10, 2014
Chemainus, BC
Hi All,

I am looking for some assistance with an issue. Last week I dropped the pack in my 2012 to replace some weak cells. The job went smoothly (i am always nervous that I'm going to screw something up) and the car seemed fine until it shut down during a short test drive ( RBS light and car with exclamation point light came on). After 10 minutes the error lights went out and I drove home.

Using a borrowed Autel MAxipro scanner I sifted through the many DTCs and found P1A71 - CMU08 Battery temperature
I then went for a drive again with Canion open to the battery temperature page and saw that temp sensor 40 was started increasing as soon as I drove up my road. In 200m it had gone from 2C to 30C.

When t40 reached 60C it shut the car down and it wouldn't start for 10 minutes. I am assuming that because the T reading increases and decreases very quickly that the cell itself is not heating up and cooling down.

The odd thing is that when accelerating or regenerative braking, the temperature climbs, when maintaining speed on flat road, the temperature is stable and when stopped the T drops quickly.

So I assumed that I screwed up one of the CMU boards when I was changing cells, but CMU08 is on the block containing cells 53-60 and I didn't even remove that one from the tray.

according to the chart on this thread T40 corresponds to cell 53.

I removed CMU02, CMU03, CMU09, CMU10, CMU12 when I changed out cells.

I would like to try and fix this in one go. I have to ask a neighbour for his lift and his time when I do this work and I know he enjoys helping but I feel it is a lot to ask of him. So I am hoping for some assistance to drill down to the problem.


If I replace the board on CMU08 should that fix the problem?

Could there be another cause of this issue?

I been through a long thread by Martin Winlow on speakev regarding problems with a CMU board, so I now have a little knowledge, but I am not knowledgeable about circuits and electronics.

I would greatly appreciate any assistance with this issue. This is the car I drive daily and without it life is going to be quite challenging.

Cheers and Thank You

It's too bad that you found that thread with the WRONG pack layout information, it really needs to be deleted or marked as not to be used.

Here is the pack layout.

There is a thread about the CMU boards with pictures and identification of the board components,
That's the pack layout I have used (the correct one you just posted). Is the temperature sensor map in that thread correct?
I just had a thought. Cell 53 is on the end of the cell block and the bus is bolted to it. Lets assume that the temp sensor is working, if the busbar was loose on the terminal would the terminal heat up quickly? Could that explain the behaviour of the sensor? I did have to loosen the nut in order to remove the adjacent block.


i was wondering that exact thing--that the screws or nut might not be fully tightened.

The temperature sensors are near the screws that connect the CMU board to the cell (+) terminals.

If the nuts were loose then it would create a high resistance junction that would definitely heat up as current passed thru the pack.

p.s. the CMU table of voltage and temperature sensors is correct using canion numbering; MUT "numbering" uses letters e.g. A, B, C to identify the cell position in a module.

Thanks for chiming in. I spoke with an EE neighbour and he concurs that a loose nut would cause the sensor behaviour.

Taking the pack out to tighten a nut! What a pain. I will be thrilled if that is the only problem.

As a follow up, it was a loose nut on the terminal of cell 53. Problem solved. Thanks for you assistance Kiev.
As a follow up, it was a loose nut on the terminal of cell 53. Problem solved.

It's standard practice amongst many industries (particularly heavy diesel) to paint mark nuts/bolts after torquing to spec in order to avoid missing critical steps

This is particularly important if working on a device is a multi-day process as it's easy to overlook stuff (human memory is terrible)

Something advising this should be added to the FAQs.

I'd also suggest photographing everything before/during disassembly as this helps a lot if you need to work out how cables need to be dressed during assembly to avoid issues later