Tesla Model 3 Battery Cell Failure - Experience Reqd!

[nkiernan]

I'm hoping to get some advice/suggestions from anyone more experienced with these batteries.

I have a 55kWh Tesla Model 3 pack from a 2020 car, and I've had it on a bench for the past two years or so...slow project progress! Being newish to HV batteries and keeping my distance, I didn't run them down to any storage voltage, all cells about 3.96V. All four modules are disconnected from each other and raised off the bench, and taped up.

I periodically measure the voltage of the modules and they've always stayed steady and similar across modules. Back a couple of months ago, I had some tap wires soldered to the BMB slave board pads for use with an Orion 2 BMS. Checked all cell voltages before and after and all remained even with no sign of issues. I never connected these tap wires to anything since, while I continued working on other parts of the project.

Recently I was doing some testing of all 96 cell voltages and now find that one cell on module 4 has dropped to approx 0.4V! Not good. No visual signs of any issues around the module or on the BMB slave board. I've read other threads about TM3 battery cells damaged where people were considering splitting the module for smaller voltage packs etc.

So I expect this is unrepeatable as I wouldn't be on for isolating the cell (trying the split it from the neighboring cells) to try resuscitate it on its own and bring it back to the voltage of the other cells, and so I'll have to see about finding a spare module somewhere.

• My query is, does this make sense? Is this something that can randomly happen on quality packs or would there have to be some external issue that caused this?

• Might there be a way to safely try resuscitate that cell?

[nkiernan]

Coupe of photos of the BMB

[johu]

In an e-scooter pack I had one cell go high self discharge and taking all cells with it that were in parallel. It happened after just one charge cycle though and then it was a matter of weeks to drop to 0V.
On a used pack that seems unlikely I tend to think. Only way to find out: disconnect the BMS (pretty much the only other possible culprit), charge the cell cluster back up and monitor it. It might have taken damage from deep discharge so it is partly a post mortem analysis...

[jrbe]

There are a few things to consider.

The low voltage cell is very likely damaged. At best it will have reduced capacity. At worst it's a big fire risk if you try to recover it.

So what's connected to the battery that is causing a draw there? Or is it a fault in one cell in that group?
If the bms was connected was it responsible? Did it try to warn you of a low cell somehow?

It might be challenging but it's worth figuring out what happened before you get a new module and it happens there or in another spot in the pack. Especially if you plan to reuse some of the wiring / bms.

[nkiernan]

Thanks for the replies Johy and jrbe.

Its an interesting one to me becuase the modele has been isolated and not connected to the other modules or anything that draws power for a year or so. The only work on it was adding the BMS tap wires as shown in the photos above. So those wires go to a 24-way plug ready to connect to the BMS but it has never been plugged in and the Orion BMS was never connected.

After soldering the tap wires, the individual cell voltages were checked (before and after) and showed no sign of shorts or issue, so looked like all was good. Don't think the Tesla BMB board on the module does anything in this isolated state so I'll need to go back and recheck the 24-way plug and unwrap the wires to see if anything shows up there now that some time has passed.

For 24 cells in series, I would have thought at least the cells either side of this low voltage cell would start dropping also, but it doesn't appear to be. Would that be normal?

[jrbe]

Its an interesting one to me becuase the modele has been isolated and not connected to the other modules or anything that draws power for a year or so. The only work on it was adding the BMS tap wires as shown in the photos above. So those wires go to a 24-way plug ready to connect to the BMS but it has never been plugged in and the Orion BMS was never connected.

Do you have a way to disconnect the ribbon cable from the battery and check resistance across the 2 wires to that cell group? I'd guess you see a very small amount of resistance between those wires like 10k that you don't see between other cell group pairs.

After soldering the tap wires, the individual cell voltages were checked (before and after) and showed no sign of shorts or issue, so looked like all was good. Don't think the Tesla BMB board on the module does anything in this isolated state so I'll need to go back and recheck the 24-way plug and unwrap the wires to see if anything shows up there now that some time has passed.

Definitely worth checking. Check for flux residue between cell groups too.

For 24 cells in series, I would have thought at least the cells either side of this low voltage cell would start dropping also, but it doesn't appear to be. Would that be normal?

Resistance between that cell's connections would only draw power at that cell group. Think of it as if you had a tiny light bulb between that cells connections. The bulb (the short) will pull power from that cell group only. It's not connected to other cells unless you skipped a couple series wires for the bulb / resistance connections.
These are the expected results if there's resistance between the problem cells connection. It won't spill over to the cell groups on either side.

[muehlpower]

My M3 LR battery sat separated on the shelf for 1.5 years. I then installed and connected the modules into my case, but in a different order. The two short modules are rotated and swapped. That was 2.5 years ago now. I've never charged it! The cell voltage has dropped from approx. 4.1V to 3.9V with a 0.014V min/max difference 4 years ago and now. I also did various tests on the drive unit and the DC-DC converter, which consumed power. The self-discharge is almost zero.

[Bratitude]

i can think of 2 mechanical possibilities:
- residue flux from solder cell taps/ left over solder blobs
- broken/ interfered cell fuses on the bottom of the battery module. (as you know) these m3 modules are very exposed on the bottom and moving them around is challenging without damaging any of the cell connections

how shaft have you been with keeping the bottom clean?

[nkiernan]

The self-discharge is almost zero.

Thanks for the comment. That's what I would hope, so points to something external and I'll do some more checks

[nkiernan]

i can think of 2 mechanical possibilities:
- residue flux from solder cell taps/ left over solder blobs
- broken/ interfered cell fuses on the bottom of the battery module. (as you know) these m3 modules are very exposed on the bottom and moving them around is challenging without damaging any of the cell connections

My first TM3 pack was delivered in the back of a van with no protection on the underside, and found cells with 0V readings because of those damaged cell fuses. Was lucky that they changed the pack for me and we both learned from that.

So I have them raised on a plywood frame that keeps the underside safe, and they've been like this for a long time. So flux residue or solder blobs or something around the pads sounds most likely at this stage. Will get the magnifying glass at it tomorrow, cheers

[bielec]

Just for information. The Model3 pack which I have been using for solar storage for the past two years has just developed a cell imbalance.
This happened in the last two weeks and I noticed it on the Tesla CAN data which I continuously display. A cell difference of 20mV to 80mV is reported on cell number 39. The cell is low by this amount and it varies depending on various factors such as State of charge, Charging current and discharge current. It is not temperature dependent. I have always kept the state of charge between 20% and 90% and it usually sits around 60%. There are automated cut-offs to disconnect outside these limits. I have a small immersion heater plumbed into the coolant loop which keeps the temperature from falling below 10 deg.C.
I am preparing for the worst and will probably have to strip the pack down. In the first instance, I will try charging cell 39 independently and then see how long it stays in balance. Prior to this the cells were always within 10mV , thanks to the Tesla BMS. This cell has now obviously slipped outwith the capability of the BMS to correct it.

[nkiernan]

When you mention Tesla BMS, had you balancing working with the M3 BMS?

[bielec]

I have always used the battery with the internal Tesla BMS. It is a complete sealed pack. The faulty cell 39 continues to lose voltage and is presently 138mV lower than the rest.

I am considering cutting a hole in the bottom plate in the vicinity of cell 39 to attach charging wires/probes. This would save the big job of opening the whole pack which includes stripping all the Penthouse components. Does anyone know the exact location of cell 39? And how much clearance is there between the bottom plate and the terminals of the cells?

[Bratitude]

I have always used the battery with the internal Tesla BMS. It is a complete sealed pack. The faulty cell 39 continues to lose voltage and is presently 138mV lower than the rest.

I am considering cutting a hole in the bottom plate in the vicinity of cell 39 to attach charging wires/probes. This would save the big job of opening the whole pack which includes stripping all the Penthouse components. Does anyone know the exact location of cell 39? And how much clearance is there between the bottom plate and the terminals of the cells?

Don’t even attempt this.

You’ll get aluminum shaving on the cell taps if u mange to not touch them.
This is a very bad idea

[johu]

Thought I should repost here should you need replacement modules https://www.bruktdel.no/nbf/Bildeler/te ... 02-698542/

[bielec]

Don’t even attempt this.

You’ll get aluminum shaving on the cell taps if u mange to not touch them.
This is a very bad idea

I agree.
I had considered using nibbler type shears, which do not produce swarf but there is probably not enough clearance between the housing and the battery cells to operate such a tool. Removing the Pyrofuse and the contactors would isolate the modules and reduce the voltage at any point to 96v.

It is just a bad situation with only bad solutions! I learn from other forums that whole module replacement doesn't work , for balance reasons.

[Person1]

You could pluck the wire bonds on the cells in the bad brick, using ceramic tweezers. Then, find the problematic cells in the brick. Then reconnect all other cells in that brick. You'd need to solder a wire in.