Hey Guest, welcome to ChevyBolt.org. We encourage you to register to engage in conversations about your Bolt.

Question for battery experts: effect of torn anode tab or folded separator?

10424 Views 35 Replies 20 Participants Last post by boltage, 11 mo ago

boltage

Discussion Starter · Aug 22, 2021

GM says that the battery cell defects are a torn anode tab and folded separator, which increase the risk of a fire when both are present in the same cell.

What are the effects of each one individually? Does either one individually tend to cause a cell with lower voltage than the others, or other observable (without disassembly) effects? Obviously, we now know that any such observable effect is not 100% reliable in determining whether a cell has the defects.

1 - 20 of 36 Posts

Vertiformed

· #2 · Aug 22, 2021

Disclaimer: I'm not an expert. (But I don't think we have very many li-ion battery-industry folks here, so you'll mostly get armchair “experts”.)

A torn anode tab would seem like it would restrict current flow and potentially get hot if the battery were discharged at high current for an extended period. So it might be detectable with a thermal camera in a high-discharge test.

Reactions: 1

GJETSON

· #3 · Aug 22, 2021

The main function of a separator is to keep the two electrodes apart, to prevent an electrical short. A folded separator would also allow easy access for dendrites to reach the opposite electrode.

MisterJJ

· #4 · Aug 22, 2021

Vertiformed said:
Disclaimer: I'm not an expert. (But I don't think we have very many li-ion battery-industry folks here, so you'll mostly get armchair “experts”.)

A torn anode tab would seem like it would restrict current flow and potentially get hot if the battery were discharged at high current for an extended period. So it might be detectable with a thermal camera in a high-discharge test.

Disclaimer: I'm not an expert. But I am an engineer that has previously worked for many years in the Li-Ion battery pack assembly field (small electronics, not cars).

That seems reasonable for testing for a torn tab, but of course it would depend on the size of the tear and unlikely to be done on existing pack.

The fact that they've claimed that it is the combination of a torn tab and folded separator would lead me to believe that;

A: A torn tab by itself has no effect and any additional heat generated would be negligible. High risk packs, like for medical devices, have an impedance check done after assembly to verify connections to cells and a significantly torn tab would show up in this test. I have no idea if GM does this but they certain should.

B: A folded separator by itself has no effect on the safety of the cell. There may be some change to the capacity of the cell, but it would be very minimal, unless it was severe but if that was the case, it should have easily been found in production earlier.

C: Therefore, you need both defects, as GM has stated. This leads me to believe that the torn tab is, or can be, in proximity to the folded separator. There's all sorts of combinations of possible failure modes with such a situation. The torn tab could damage the separator at the fold during assembly. The torn tab could have exposed edges that are more likely to form dendrites, that are not prevented by the fold in the separator. Lots of possibilities.

Yet, without knowing if these defects actually existed in the burned Bolts, it's nothing more than speculation on our part AND on their part. You'd have to build millions of new cells, verifying these defects don't exist, install them in cars and wait a few years before knowing if you've really fixed it. I doubt GM will go down that road.

See less See more

Reactions: 3

GJETSON

· #5 · Aug 22, 2021

MisterJJ said:
High risk packs, like for medical devices, have an impedance check done after assembly to verify connections to cells and a significantly torn tab would show up in this test. I have no idea if GM does this but they certain should.

They do check impedance at the pack level, in the car. I doubt a torn tab on one cell, out of 288, would show up, as significant.

Rectangle Font Technology Pattern Electric blue

See less See more

GregBrew

· #6 · Aug 22, 2021 (Edited)

MisterJJ said:
Disclaimer: I'm not an expert. But I am an engineer that has previously worked for many years in the Li-Ion battery pack assembly field (small electronics, not cars).

That seems reasonable for testing for a torn tab, but of course it would depend on the size of the tear and unlikely to be done on existing pack.

The fact that they've claimed that it is the combination of a torn tab and folded separator would lead me to believe that;

A: A torn tab by itself has no effect and any additional heat generated would be negligible. High risk packs, like for medical devices, have an impedance check done after assembly to verify connections to cells and a significantly torn tab would show up in this test. I have no idea if GM does this but they certain should.

B: A folded separator by itself has no effect on the safety of the cell. There may be some change to the capacity of the cell, but it would be very minimal, unless it was severe but if that was the case, it should have easily been found in production earlier.

C: Therefore, you need both defects, as GM has stated. This leads me to believe that the torn tab is, or can be, in proximity to the folded separator. There's all sorts of combinations of possible failure modes with such a situation. The torn tab could damage the separator at the fold during assembly. The torn tab could have exposed edges that are more likely to form dendrites, that are not prevented by the fold in the separator. Lots of possibilities.

Yet, without knowing if these defects actually existed in the burned Bolts, it's nothing more than speculation on our part AND on their part. You'd have to build millions of new cells, verifying these defects don't exist, install them in cars and wait a few years before knowing if you've really fixed it. I doubt GM will go down that road.

Which raises the question...how would a "visual inspection" after assembly do any good whatsoever?

Reactions: 1

GJETSON

· #7 · Aug 22, 2021

GregBrew said:
Which begs the question...how would a "visual inspection" after assembly do any good whatsoever?

Yeah. I think that is total BS. Are we expected to believe that Hyundai would have figured out how to QC these LG cells, and LG, GM, and VW would have missed this?

www.chevybolt.org

Goodbye Bolt “Nebula”

So, I sold my 2020 Bolt EV premiere yesterday. I owed $27k and they offered $25.5k. I couldn’t refuse. Especially, after driving the 2021 Kona EV Ultimate. Spoiler alert, I ended up getting the Kona EV. things I will miss about the Bolt: 360 view, true one pedal driving and the rear view...

www.chevybolt.org

www.chevybolt.org

Reactions: 1

Vertiformed

· #8 · Aug 23, 2021

GregBrew said:
Which begs the question...how would a "visual inspection" after assembly do any good whatsoever?

[Pedantry alert: I wouldn't correct most people, but @GregBrew is a self-proclaimed nerd, so ought to care about such things…]

No, it raises the question. Begging the question involves circular reasoning, like saying “a visual inspection after assembly would do lots of good, because visual inspections are beneficial, particularly when done after assembly”.

Reactions: 2

sparks

· #9 · Aug 23, 2021

I do not like my anodes get torn an folded
Carry on Y'ALL

Eriamjh1138

· #10 · Aug 23, 2021 (Edited)

Required I am not a battery engineer.

Edit: post removed. Carry on.

Dark-Fx

· #11 · Aug 23, 2021

Eriamjh1138 said:
Required I am not a battery engineer.

The separator isn’t between the anode and cathode. It’s between cells welded together in parallel to from a group or between groups.
The separator is there to provide padding, spacing, and a form of pressure between cells (or cell groups).

This is wrong. The separator is indeed in the cell to keep the anode and cathode sides apart. It sits in the electrolyte. They are absolutely needed because lithium dendrites will easily form in the electrolyte, and the separator stops the growth of them from reaching the other side.

The Four Components of a Li-ion Battery

<p> </p><h3 class="view-title" style="margin: 0px; padding: 0px; font-size: 28px; font-weight: normal; line-height: 36px; color: rgb(34, 34, 34); text-align: center; font-family: Arial;"><br></h3...

www.samsungsdi.com

Reactions: 4

AZBILL

· #12 · Aug 23, 2021

I posted this on another thread, I am not a battery expert but am an EE and have performed safety analysis. The argument on the other thread was that "exposure time" does not matter.

I have a EE degree and have worked in aerospace my entire career, during that time I performed several system level safety analyses. Exposure time does have an effect on the likelihood of an unsafe event happening, especially if it is a latent fault that is not detectable.

We know GM has identified the faults as torn anode and folded separator. These are the latent faults that could be in your battery. We know that punctures of the separator will lead to shorts, and thus fires. We can also discount external punctures do to running over objects or collisions.

So in a fault tree for the fires, the cell defect is one leg. There are at least two other possible legs to the fault tree.

1) physical tear due to anode rubbing against separator
2) puncture due to dendrite growth

Case 1 hypothesis: The torn anode rubs against the folded separator during the swelling that occurs in the cell during discharge and charging due to heating and then the reduction of swelling as the battery cools. This rubbing will not stop after a certain number of years or charge/discharge cycles, it will continue for the life of the car.

Case 2 hypothesis: The torn anode and/or folded separator contribute to accelerated and localized dendrite growth, that could puncture the separator. Again, dendrite growth does not stop after a certain time period or number of charge/discharge cycles.

Case 1 is the most likely root cause in my mind.

See less See more

Reactions: 2

sparks

· #13 · Aug 23, 2021

sparks said:
I do not like my anodes get torn an folded
Carry on Y'ALL

Don't care for dendritetes neither

Eriamjh1138

· #14 · Aug 23, 2021

Dark-Fx said:
This is wrong. The separator is indeed in the cell to keep the anode and cathode sides apart. It sits in the electrolyte. They are absolutely needed because lithium dendrites will easily form in the electrolyte, and the separator stops the growth of them from reaching the other side.

The Four Components of a Li-ion Battery

<p> </p><h3 class="view-title" style="margin: 0px; padding: 0px; font-size: 28px; font-weight: normal; line-height: 36px; color: rgb(34, 34, 34); text-align: center; font-family: Arial;"><br></h3...

www.samsungsdi.com

Thanks for the correction. Post edited.

Martin D

· #15 · Aug 23, 2021

I don’t think visually inspecting the exteriors of cells is going to tell them anything unless the defects can be seen on the exterior.

This issue is happening as the cell is assembled. Somewhere in the line stacking proceedings, an occasional error is occurring which creates the possibility of a defective cell. An occurrence of the initial defect may or may not lead to the lesser occurrence of the other defect. Then the cell is folded and sealed, removing the ability to visually determine an error has occurred.

GM needs to go over the assembly line with a fine tooth comb, getting a camera on every procedure to find where the anomaly happens. Then the line can either be adjusted to stop the problem, or an occurrence of the misalignment , as seen by the camera, kicks the defective cell out of the assembly line.

Once they have a way to either correct or remove the defects from the “good” bin, they can assemble batteries that won’t spontaneously combust.

See less See more

AZBILL

· #16 · Aug 23, 2021

They could check the cells after making them via x-ray or CT scanning equipment, before they put them in the modules. Make this part of the manufacturing process going forward.

Reactions: 2

bolt-ish

· #17 · Aug 23, 2021

AZBILL said:
Case 1 hypothesis: The torn anode rubs against the folded separator during the swelling that occurs in the cell during discharge and charging due to heating and then the reduction of swelling as the battery cools. This rubbing will not stop after a certain number of years or charge/discharge cycles, it will continue for the life of the car.

I thought the tab was torn, meaning outside the cell

Gimme_More_Amps

· #18 · Aug 23, 2021

MisterJJ said:
B: A folded separator by itself has no effect on the safety of the cell.

What if one corner of the separator is folded over, like a dog eared page in a book. That would leave a small area between anode and cathode plates with only air between them, would it not? And be prone to arcing? Or bridging by dendrites?

Reactions: 1

AZBILL

· #19 · Aug 23, 2021

bolt-ish said:
I thought the tab was torn, meaning outside the cell

There statement stated in the cell, not external:

On Friday, the company specified that the defects are a torn anode tab and a folded separator in the cell.

Plus, a torn anode on the outside of the cell would be clearly visible during an inspection and before the next assembly step. Both the anode and cathode tabs have to run from the outside of the cell to the inside of the cell to make the electrical connection. Those tabs are at the edge of the cell, and the separator fold would be on the edge of the cell as well. So all very close together.

bolt-ish

· #20 · Aug 23, 2021 (Edited)

AZBILL said:
They could check the cells after making them via x-ray or CT scanning equipment, before they put them in the modules. Make this part of the manufacturing process going forward.

Even if possible, I doubt it is practical. GM talked about two cell production factories with partner LG that would manufacture 70 GWh a year. Each pouch cell is 0.2 kWh, = 350 million cells a year. Presuming best case of 24/7/365 manufacturing, that is 350 million cells produced in 31.54 million seconds, so about 11 cells per second

Just how many scanning devices and readers are you planning to put in the factory ?
How long do you think the full step would take ?

If the full step only took a minute (outrageously unlikely), you have slowed down production 660x

Disclaimer: Not a battery expert, period. But with the help of a calculator, arithmetic is within my grasp.

1 - 20 of 36 Posts

This is an older thread, you may not receive a response, and could be reviving an old thread. Please consider creating a new thread.