Improve battery with cell replacement?

[dusanhu]

I haven't found a similar topic on this forum, apologies if it's there and I have missed it...

I have a MY13 with 223 000 km on clock, SoH 25.4 Ah. According to PhevWatchdog statistics, this significantly below the expected SoH at this milage. So I am wondering if there is a reasonable way to improve capacity with replacing the weakest cells in the package. The BMU was reset 400 days ago by the previous owner.

According to PhevWatchdog the maximum cell voltage difference ever measured is 88mV, but from these overall statistics it is still not clear to me which cells might worth to be replaced, if any.

I have already ordered a miniVCI, and plan to use it with MUT 3, so I was wondering if I should perform a DBCAM, and check the cell voltage at the low end of the DBCAM process, and also check which cells stop charging first during the charging phase. But how could I log the cell voltages during the DBCAM? As far as I know PhevWatchdog only logs the min and max voltages, but not all the cell voltages separately.

-what might be the best way to identify cells to be replaced?
-does this plan make sense at all? (replace the weakest cells -> BMU reset -> smoothing -> DBCAM)

View attachment phev1.png
View attachment phev2.png

thank you

 

[Sunder]

Not familiar with your software, but there's two fairly standard ways to "eyeball" a weak cell.

1. Testing remaining capacity. A cell that has reduced capacity will be first to fill as well as first to empty. Most reliable way is to get the battery perfectly top balanced, and watch it as it discharges. The ones that drop in voltage fastest are the candidates for being most degraded. However, allow it to rest before making judgement, as internal resistance can drag some cell voltages lower than its capacity might suggest. A few seconds of no load will fix that.

2. Speaking of internal resistance, get the battery to an even state, and put a high load on it. The ones that drop in voltage more are likely to be more worn.

Both of these are "eyeball" methodologies, and without cycle testing, it's still just a best guess.

I dropped off this forum for a while, but came back as I am thinking of adding in a secondary battery. Mine is a MY14.5, and this (Australian) winter, is the first one I have noticed a significant loss in range. So I will probably be doing the same diagnoses as I am recommending you as soon as I either find or get a new OBD-II connector. Also trying to find a cheap MUT3 box too, as I understand you can use it to force a balance.

 

[dusanhu]

thank you very much for your reply Sunder!

Currently I am using PhevWatchdog, which is an absolutely amazing tool, but during discharge it can only register the voltage and number of the highest and lowest voltage cell, not the voltage of all the cells. I have ordered a miniVCI, with that I should be able to use MUT 3, but I don't know whether MUT3 is capable of registering all cell voltages during discharging.

What tool would you use to register individual cell voltages during discharge?

For test #1, do I understand correctly that first the car should be charged fully(until EVSE shows zero Amp load, so balancing has also completed), then discharge it for some time, then stop the load, leave it rest for a couple of minutes, then compare the cell voltages?

For test #2, does "getting the battery to an even state" means the same as with #1, full charge, until balancing completes, right? Then high load can be some intense acceleration, DURING which the voltages should be registered, right?

 

[Sunder]

Oh... I was hoping something like a ODB-II tool would be able to tell every single cell voltage, and if that couldn't the MUT-3 could. Has kind of put a spanner in my planned works. Not that I've had much time lately to do much.

Using graphs from the charger side, I've notice that it doesn't balance every time. There's a distinct drop from ~2200w to nearly nothing for 30-45 minutes on about 10% or fewer of the charges, before it returns to charging at the maximum rate. So interestingly, I don't think it does a top balance, and I couldn't be certain it does a balance every charge.

Unfortunately, I think I'm blind without the usual tools I have access to for eVehicles I've built myself (So far an eBike, and a mostly completed eMotorcycle). I need to do more research before I can offer any Outlander specific advice. It'll be a fun journey, but things tend to take a long time with me, as I can work long hours in bursts, then have a fair bit of time to work on hobbies.

Hopefully someone more experienced can offer you better advice than I have so far.

 

[dusanhu]

The car on the OBD interface itself can most probably provide the per cell data, I am just not aware of any software which would register it during discharge.

During charge, PhevWatchdog does register the voltage of each cell. I have checked with the author of the Watchdog, and if I understand correctly, during discharge the app register lots of driving specific parameters, and it would consume to much resources to register the voltage of each cells as well.

I wonder if many people ask for it, then it might be possible to add a manual switch in Watchdog to change it to read/register "charge mode parameters" during discharge.

I can't tell if MUT3 can do it as I still didn't receive my miniVCI cable, or if there is any other capable tool.

Here are the values of one charging which I've registered in case you are curious.

https://1drv.ms/x/s!AvmU7dKfmqHbgcYkfmdyX4N7nIojuA?e=MvYGpP