Chevy Bolt EV Forum banner

1 - 9 of 9 Posts

·
Registered
Joined
·
27 Posts
Discussion Starter #1
I have not fully charged or checked my capacity in months (mostly between 50% and 75%), and a couple days ago I did a round trip of 180km. Used the trip meter and average efficiency and car comes out at 57 kWh. Car is a 2019 with 11000 km on it. I have never had a reading anywhere near this low. Checked at the halfway mark and at the completion of the trip and the results were the same both times.

Some internet searching shows that the car loses accuracy regarding capacity and SOC over time if never allowed to approach full charge. Has anyone else had any experience with this? I will be charging to full overnight and letting it sit for a few hours in case that is part of the problem.

I'll update after full charge and subsequent capacity test, ideally to a lower SOC, but in the meantime I would be interested to hear if anyone else has had experience with unusual measurements after long periods below 100% SOC.
 

·
Registered
Joined
·
3,600 Posts
unusual measurements after long periods below 100% SOC
thinking this is pretty standard for most all lithium-ion devices. BMS is constantly needing to calibrate based on charge discharge cycles. My wife's notebook computer actually requests to do a battery calibration once in awhile so it's battery meter is more accurate.
 

·
Registered
Joined
·
1,766 Posts
** Cell balancing is making sure that the voltages across all cells are equal. The Bolt does that when the battery reaches the 'full' point that YOU specified ("hilltop reserve" for the older models, or you set the % for 'full', or it is 100% because you never set that option). You don't need to charge to 100% (but it probably wouldn't hurt once every 3-6 months, for sh1ts and giggles).

** "guestimates" on how much a Li-Ion battery will hold can be off by a bit. "fully" charging (to hilltop reserve) and then driving do to 10% or so may increase the estimate of your battery's capacity. Maybe do that once every 6 months?
 

·
Registered
Joined
·
5,334 Posts
** Cell balancing is making sure that the voltages across all cells are equal. The Bolt does that when the battery reaches the 'full' point that YOU specified ("hilltop reserve" for the older models, or you set the % for 'full', or it is 100% because you never set that option). You don't need to charge to 100% (but it probably wouldn't hurt once every 3-6 months, for sh1ts and giggles).
Can you provide a link to where that came from? I have seen nothing to indicate this is the case.
 

·
Registered
Joined
·
27 Posts
Discussion Starter #5
I'll note at this point that after a full charge the car was reporting seemingly erroneous data. For one I unplugged and plugged in again and charging restarted with an estimated 30 minutes until full. I couldn't wait that long so I drove to work, and then checked the battery percentage, which still read "full" in the app after dropping what should have equated to 2.3% provided the capacity is 60kWh. I will need to drop more percentage to get an accurate measurement as plugging in numbers at this point gives a false result around 74kWh. My sense is that 100% is required to balance and I left it lower for long enough that the calibration was off. I think I will discharge more, then do another 100% charge to see how it evolves.
 

·
Registered
Joined
·
163 Posts
I should also mention that I have yet to ever see any evidence of cell balancing with several dozen logs and different experiments. The service manual implies that 0.03V is the maximum out-of-balance allowed, and my current battery has never gone that far out of balance.

However, even when I had a failing battery, where the cell was up to 0.2V out, there was never any evidence of attempting to balance it.

I'm not convinced that it actually does. I have a friend who has a third of a Bolt battery as grid storage for his house, and no BMS. He regularly monitors the cell levels, and they're perfectly in sync. Granted, different use case than driving, but still. It may just be that these cells rarely ever need balancing.
 

·
Registered
Joined
·
5,334 Posts
It may just be that these cells rarely ever need balancing.
Balancing lithium ion cells at any point other than the hockey stick knee at either end of charge is a pointless waste of time for several reasons. It would entail extra expense, and complication. Extra complication would increase the likelihood of failures. The benefit would be essentially zero. Since the voltage curve is so flat for most of the charge, you are not going to get the cell SOC very close anyway. Last, and most significant, is the fact that healthy lithium ion cells maintain their balance extremely well without any intervention. As Hippocrates said, "first, do no harm."
 

·
Registered
Joined
·
163 Posts
Balancing lithium ion cells at any point other than the hockey stick knee at either end of charge is a pointless waste of time for several reasons. It would entail extra expense, and complication. Extra complication would increase the likelihood of failures. The benefit would be essentially zero. Since the voltage curve is so flat for most of the charge, you are not going to get the cell SOC very close anyway. Last, and most significant, is the fact that healthy lithium ion cells maintain their balance extremely well without any intervention. As Hippocrates said, "first, do no harm."
But why?

It's not as flat as you think, it's actually quite easily discernible to 0.1% SoC. It's flatter than other chemistries, but very easy to track. It only looks flat when you have a 0..5V graph, but when you only consider the operational range, and note that even the most basic ADCs have 10-bit resolution (1024 values or 1-3mV difference depending on how you set it up), and based on what we've seen with the Bolt they most likely use a 12-bit ADC (< 1mV resolution), then it's not flat at all.

30304


If you have a system that can balance cells, what's wrong with noticing that you have two cells a little far apart, and getting them closer, regardless of the SoC?

For example, you might be at 50% with your cells hovering around 3.65V. The BMS happens to notice one cell at 3.63 and another at 3.67. The car is idle - and you have an active redistribution system. So why not balance it? The more you push cells out of balance, the worse it gets and the more damage can be done. Waiting until you are fully charged doesn't seem prudent.

Also, balancing only when fully charged is difficult as you can only charge to the highest cell, which means your other cells are technically undercharged. Then you have to stop charging, balance, then start charging again.

I've read several research papers, and very few actually talk about restrictions on when you can balance, most of them include various equations that allow for any voltage level. There is, however, restrictions based on the balancing method used. This site here describes it easiest:


Active Redistribution
Active redistribution compensates for the imbalance by moving charge from higher-charged cells to the adjacent lower-charged cells. In essence, you are equalizing the amount of cell energy across the array. This can happen during every functional period of the batteries' life; charging, at idle and during discharge. There are two ways that active redistribution works: capacitive and inductive topology.

With capacitive topology, a switch capacitor is used to measure the voltage and move energy from cell to cell. The capacitive method however, is limited because it only works when there is peak voltage. Inductive topology, on the other hand, allows for the storage of energy from high-level cells and then delivers the voltage when needed. The two biggest obstacles for inductive topology are cost and a higher part count.
Emphasis there mine. You can easily balance at any point, depending on the technology used.

Having said all that, like I said, I've left my car idle, charging, not charging, with constant cell monitoring, at least a dozen times. I have yet to catch when it's balancing the cells, but all of my cells are within 0.03V of each other (the limit as per the service manual). But even when I had a failing pack that was 0.2V out, even at full charge there was no evidence of cell balancing.
 

·
Registered
Joined
·
5,334 Posts
The car is idle - and you have an active redistribution system. So why not balance it?
But it doesn't, as far as I know. It has a simple bleeder resistor system, like any cheap e-bike, so it would be wasting battery energy to create heat. It's cheap, and simple, and good enough to get the job done.
 
1 - 9 of 9 Posts
Top