Chevy Bolt EV Forum banner
1 - 14 of 14 Posts

·
Premium Member
Joined
·
2,568 Posts
Discussion Starter · #1 · (Edited)
So last week I took a relatively easy 300 mile round-trip road trip in my Bolt with the family in tow. Did not end up using any fast chargers on the trip, just a ~30 minute L2 charging stop at a winery and overnight L1 charging at my destination.

During the first leg of the trip, I decided to record raw and displayed battery % numbers via TorquePro. I logged the data in 5% increments, and also included 1 value when I was near depletion as well since I had it. I started with a 100% reading and 96.4% raw SOC value, but I did not include it in the chart as I found anything above ~95.3% SOC registers as 100% according to the Batt DIC % PID, and it would have skewed the data slightly.

Most interesting thing is the calculated raw SOC % value of about 7.2% when the Batt DIC % value is projected to be 0. In the past, I had previously observed the raw SOC % value between 4-5% when the Bolt’s battery was almost depleted (this was all prior to the battery recall update for the low cell voltage issue). If my data is correct, my Bolt’s post-recall raw SOC % at depletion is now 2-3% higher than before the update was applied. 2-3% of the battery equates to around 1.2-1.8 kWh of capacity, so it’s very possible the update did indeed lock out more of the bottom of the battery. I still need to do a test to confirm that ~7% is the new real world floor for raw SOC %.
 

Attachments

·
Registered
2018 Bolt EV Premier
Joined
·
1,186 Posts
My collection of Raw SoC %-Displayed % data has been made entirely after the battery recall update because I got my OBD-II module about a month after doing the update.

As you can see, the projected Raw SoC value at the 0% displayed is around 4.8%. So no, I don’t think the update locked the bottom capacity in any significant ways. The data spans more than three months, and the trend line values have been more or less stable.
 

Attachments

·
Premium Member
Joined
·
2,568 Posts
Discussion Starter · #4 ·
My collection of Raw SoC %-Displayed % data has been made entirely after the battery recall update because I got my OBD-II module about a month after doing the update.

As you can see, the projected Raw SoC value at the 0% displayed is around 4.8%. So no, I don’t think the update locked the bottom capacity in any significant ways. The data spans more than three months, and the trend line values have been more or less stable.
My whole "maybe GM reduced capacity slightly on the low end" theory is based off notes from the recall update.
https://www.chevybolt.org/forum/406602-post171.html

The notes say that the update "adjusts customer usable minimum energy". I took that to mean a small chunk of the low end of the battery was blocked off, but who knows if that's actually true or not.
From your real world experience, it's possible nothing was changed. Though I do remember a time that I drained the battery raw SOC down to under 3% (ran the heater till the Bolt died). Need to do another depletion test and see how low it will go post-recall update.
 

·
Registered
Joined
·
2,129 Posts
This old thread was just relinked by @GJETSON in this thread, so I figured I'd go back over some old data of mine and contribute.

Here's a graph of some of my own data (the blue dots) and an attempt to fit some different lines to it.

32185


The lines are:
  • raw = 5 + 0.91 × displayed (shown in red)
  • raw = 4.66667 + displayed − 0.0009 × displayed^2 (shown in green)
  • raw = 4.2 + 1.015 × displayed − 0.001 × displayed^2 (shown in goldenrod)
This shows that there are a variety of lines you can draw that seem reasonable, but non-linear fits (the second two) do seem to work better.
 

·
Registered
2018 Bolt EV Premier
Joined
·
1,186 Posts
I have a question - why are my SoC Raw and Batt % numbers so different?

15% seems like a lot:

View attachment 32200

When charged to full hilltop, they were only about 3% difference.
That looks seriously out of place. As the graphs that me and Vertiformed posted show, the two aren't supposed to deviate by more than about 5%, and that's at the extremes. You should collect the data like we did and see how the numbers change over the range.
 

·
Registered
Joined
·
119 Posts
That looks seriously out of place. As the graphs that me and Vertiformed posted show, the two aren't supposed to deviate by more than about 5%, and that's at the extremes. You should collect the data like we did and see how the numbers change over the range.
Its always been that way, although at high states of charge, not so bad. But even from my screenshots of torque pro from during the summer, they are consistently off by 10% or more at battery states below 80%.
 

·
Registered
2018 Bolt EV Premier
Joined
·
1,186 Posts
Its always been that way, although at high states of charge, not so bad. But even from my screenshots of torque pro from during the summer, they are consistently off by 10% or more at battery states below 80%.
Looking at the other thread where you posted this, maybe this could be a symptom of failing battery cells.
 

·
Registered
Joined
·
6,492 Posts
The 0% displayed crosses at 4.92% raw instead of 4.8%.
Last week I unknowingly hit a strong headwind and barely made it (had to slow to the minimum toll road speed of 60 mph) to my next charging spot showing 1% remaining as reported on the charge station. Last checked my raw percent using my OBD2 reader at 7%. I'm assuming it hit 6% raw. I've been seeing this roughly 5% disparity for awhile at the low end of the battery. So I'm wondering if the car quits at 0% or does the car have a hidden 5% battery capacity based on the raw value?
 

·
Registered
12/16 build, 2017, white LT
Joined
·
13,532 Posts
Last week I unknowingly hit a strong headwind and barely made it (had to slow to the minimum toll road speed of 60 mph) to my next charging spot showing 1% remaining as reported on the charge station. Last checked my raw percent using my OBD2 reader at 7%. I'm assuming it hit 6% raw. I've been seeing this roughly 5% disparity for awhile at the low end of the battery. So I'm wondering if the car quits at 0% or does the car have a hidden 5% battery capacity based on the raw value?
 

·
Registered
2018 Bolt EV Premier
Joined
·
1,186 Posts
Last week I unknowingly hit a strong headwind and barely made it (had to slow to the minimum toll road speed of 60 mph) to my next charging spot showing 1% remaining as reported on the charge station. Last checked my raw percent using my OBD2 reader at 7%. I'm assuming it hit 6% raw. I've been seeing this roughly 5% disparity for awhile at the low end of the battery. So I'm wondering if the car quits at 0% or does the car have a hidden 5% battery capacity based on the raw value?
Wow, a thread from 2 years ago! And this time I actually have something to add because I did hit a displayed battery charge of 0% a few months ago. Okay, where to start...

This was after a very tight drive, and I had just barely made it to my home. The drivable range showed up as 0 km. However, the OBD-II reading showed 0.392%, which seems to be about as low as it can get before truly hitting 0% and becoming completely immobile. The raw number was 5.098% - slightly higher than what I've gotten on the trend graph - and the car could still move, so I decided to drive to a nearby DCFC 3 km away.

I was able to expend additional 0.4 kWh to arrive there, and while the displayed SoC was still at 0.392% the raw SoC was now at 4.314%. I wasn't going to risk moving further, so I started charging the battery right away.

This seems a bit reminiscent of the top "buffer" as far as displayed SoC goes. Displayed SoC will be at 100% when the raw SoC is around 95.29%. However, you can still continue to charge if the charger doesn't automatically cut itself off when it sees the displayed SoC is 100% like some DCFC will. Eventually, the raw SoC will continue to go up - sometimes into 97% territory - until the car's BMS thinks it's fully charged, when the number will be set to 96.08% and the charging session is terminated. You can use up to roughly 0.6-0.7 kWh of energy from the battery and the displayed SoC will remain at 100% because the raw number hasn't fallen below 95.29%.

Likewise, it seems that the car's BMS was allowing me to eke out a bit of energy - at least 0.4 kWh in this case - even though the displayed SoC was already as low as it could get before flatlining.

So in my observations there is indeed some sort of small buffer as far as displayed SoC is concerned. The raw SoC will have about 1% leeway after hitting displayed 100% or 0%. However, I think the BMS will not allow you to actually go beyond that under normal circumstances. In other words, the 0-4% and 96-100% regions are not "usable" areas and should not be considered as buffers.
 

·
Registered
Joined
·
6,492 Posts
In other words, the 0-4% and 96-100% regions are not "usable" areas and should not be considered as buffers.
Thanks. I suspected that. This was my first time getting this low. I had perfect weather at my previous stop and didn't realize the conditions had deteriorated so quickly while I was driving to the next stop. Must have been a cold front as I ended up hitting snow. Even knowing the conditions I hit 3% at the next stop that I added to make sure I got to my final destination. 880 mile trip was a success, but was a little exciting at the beginning and end of the trip. My first leg at the beginning of the trip I hit 4% after driving through the mountains at 19F and my battery was conditioning.
 
1 - 14 of 14 Posts
Top