To Bolt or Not that is the Question—We Have a Failing Traction Battery

By Paul Gipe

We have been driving electric for more than seven years and haven’t had any problems–until now. We’re currently driving a 2020 Chevy Bolt EV, but we don’t know for how much longer: It has a failing traction battery.

Our Bolt has lost more than 30% of its range since November 2021. We’re now working with 43-44 kWh of usable capacity out of the original 65-66 kWh when the car was new. Worse, the battery’s capacity continues to decline as does our range.

The LG Chem battery is under warranty and GM is obligated to fix the problem. Therein is the problem. What will GM do and how does that fit with what we want? First some background.

My Kingdom for a Battery

GM is in a battery pickle. The traction battery on some 2017 and 2019 Bolts have exploded and caught fire. Enough have done so that the NHTSA (National Highway Transportation Safety Administration) and GM have issued a battery fire safety recall for the entire production of Bolts through at least 2022.

The plan is to replace batteries in “all” Chevy Bolts with a new 65-66 kWh battery. That’s a lot of batteries to replace, about 150,000 or so. (This is the same battery that was part of the Hyundai Kona EV recall as well.)

LG Chem’s plant in Holland, Michigan will have to build batteries both for the recall and for any new builds that GM is planning. GM’s Orion assembly plant has shut down periodically as the company prioritized getting replacement batteries to current Bolt owners. Even so, the recall will take years to complete at the current pace. In short, there are not enough batteries to go around.

GM has notified us that our Bolt is part of the recall. However, our position is at the end of the queue behind 2017, 2018, and 2019 Bolts. Officially, there is “no remedy” yet for our Bolt. That’s bureaucratic speak for “we haven’t even placed you in line yet.” GM won’t say when we might get a battery under the recall.

To complicate matters, our failing traction battery doesn’t have anything to do with the fire safety recall. It has another, more pressing problem.

Torque Pro and Sean Graham’s PIDs

How do we know our battery is failing? The answer to this question requires a bit of explaining.

I am a nerd and I record data on the Bolt every time we charge it. I record about a dozen parameters. Some are found on the Bolt’s dash display, such as the odometer reading, others are displayed on my smart phone.

I use Torque Pro on my Android phone to record several measurements from the Bolt’s brain through a dongle plugged into the On-Board Diagnostic (OBD) port. Torque Pro takes the signals from the Bolt and translates them into something meaningful using PIDs (Parameter IDs if you want to know) developed by Bolt driving techies like Sean Graham.

One measurement I get from Torque Pro that the Bolt doesn’t provide on the dash display is state-of-charge. Knowing the state-of-charge (SOC) and the kWh consumed on a trip, you can calculate the number of kWh available when you started the trip. (I’ve explained how that’s done in Slight Chevy Bolt Traction Battery Degradation after Nearly 30,000 Miles.) This calculation was what alerted me that something was rotten in the Bolt. 

There is also a PID specifically for the Bolt’s traction battery capacity in kWh. This has remained steady at 58 kWh during this period—another indicator that something is amiss. The measurement from Torque Pro’s PID and that from the calculation using SOC should be approximately the same. They are not. They are off by as much as 15 kWh, sometimes more.

When we charge the car to 100%, we have only 43-44 kWh available to drive the car while there should be 58 kWh available. And there’s the rub.

Chevy Dealer Clueless

So, I took the car into the local Chevy dealer. They told me it would take 24-hours to run diagnostics on the battery. No sooner had I got home than the dealer called and said the car was ready. Hmmm, that was odd. When I got there to pick up the car, Chevy’s service “consultant” told me that the battery had the charge limiting software put on it “over-the-air.” I told them I didn’t think GM did that and in fact was told by GM’s concierge they wouldn’t do that without my permission. “Oh, yes, we do over-the-air updates,” she replied. And that’s what was in the official service report.

I dutifully wrote this up and posted it to fully expecting howls of derision from the forum’s denizens and they didn’t disappoint. (See Has GM Used Over-the-Air Updates on Chevy Bolt Safety Recall?) This suggested I needed even more data on the Bolt’s battery.

Nerds to the Rescue

I dived back into Sean Graham’s PIDs that record the cell voltage from the Bolt’s battery. There are 288 cells in 96 cell groups, three cells per group. These in turn are grouped into modules. I’d seen posts by some of the nerds on the forum showing the cell voltage for all 96 cell groups in their batteries. That was more than I wanted to key in on my smart phone.

Fortunately, there are PIDs for average cell voltage and maximum and minimum cell voltage. I chose to start there.

I pulled up the numbers. The average cell voltage was 4.11 V, the maximum was 4.12 V, and the minimum in cell #33 was 3.82 V. They looked reasonable to me.


I posted the voltages up on the forum and the electrons started flying. I had a bad cell I was told and I’d better take the car into the dealer right away.

Turns out that NHTSA’s fire recall bulletins specifically mention the difference between the minimum cell voltage and the average voltage. On page four of the ten-page bulletin it says a difference greater than 0.08 V is cause for alarm.

The difference for our Bolt was 0.3! Yowser.

One forum participant even pointed to where in the car the bad cell was located. It was found in Module 4 under the rear seat.

Lost in Limbo

I called GM’s concierge, got a case number and was told that a California recall specialist would contact us with an offer based on California’s lemon law. GM indeed did call, told me my options, said they’d look for a host dealer and that was that. It’s been more than two weeks and not a peep out of GM.

We’re stuck in Limbo. I can’t get a response out of GM corporate and the local dealer doesn’t call me back. We’re still using the car in the meantime.

At my last charge, the car reported that I had only 39 kWh to work with. I hope that was an outlier and it bumps up to at least 43 kWh on the next charge. If not I may have to take more drastic action like park the car in front of the Chevy dealer. . .

Progress of a sort, I finally reached a human at GM. She looked up my case—and best of all they in fact did have me in their system—and found that yes someone had dropped the ball. So she was going to try to kick up a notch.

Nothing happened.

I called again. And yes, I was still in the system to replace a “2018 Bolt.”

We don’t have a 2018 Bolt. We’ve never had a 2018 Bolt. We did have a 2017  Bolt but that was turned into GM almost two years ago.

They also had us down for a “buyback” and not a swap or “trade” in their preferred jargon. Wrong again. We want a swap or a trade. We’d prefer a new battery but if they are not going to get us one promptly, we’ll take a trade.

GM’s not getting high marks for competence here.

We’re still in Limbo.