[octagondd]

I am trying to find real world pack balance data from LFP users with no BMS. I want to see how the pack stays in balance over time. There probably isn't much out there, but anything is better than nothing at this point.

Do you have an LFP pack with no BMS?

Do you have any data from the history of your cells, particularly how well they stay in balance?

Did you initially balance them?

Top or Bottom balance?

What discharge rate are you typically drawing from your pack?

Do you balance your pack every so often to bring them back in line?

Do you use cellogs or other device to log all or some of the cells?

Would you share this data via a link or email?

Thanks,
Dave

 

[JRP3]

I'll copy my post from the other thread.
I'm running bottom balanced without a BMS. My SE/CALB cells range in capacity from 110ah-114ah. Four months after my initial bottom balance I checked again and they were all within less than 0.08 volts of each other. I rebalanced again with a more accurate meter so they were all within 0.004 volts of each other and I'll check them again in December. I've pulled 5.5C from the cells occasionally and had one low discharge event to an average of 1.77 volts per cell yet they've remained well balanced at the bottom. I do undercharge the cells to be safe. The PFC voltage trim is a bit less accurate than I had hoped. If I need a full charge I'll flip the charger back on for another 15-30 minutes while I monitor closely, but not very often. I like the safety bottom balancing gives me while driving in the event of low SOC but as long as you stay away from the top and bottom I think you can get away with top, bottom, or middle balancing, without a BMS. I do have an adjustable charger, PFC-30, so I can change current draw and cutoff voltage as temperature changes the resistance of the cells. If I did not have an adjustable charger and wasn't using a BMS I'd probably top balance the pack. I'm using Dimitri's EV Display for amp hour counting which makes life easier being able to see how much charge goes into and out of the pack.
__________________

 

[octagondd]

JRP,
What are your individual cell Voltage readings after a typical charge like? Are they within say 0.02V of each other? This is what I would figure they should be at given a 3% capacity difference and a voltage difference of .6V between bottom balance and end of charge.

 

[JRP3]

Do you mean resting after turning off the charger, or near the end of charge but still charging? It also depends on how much I charge them.

 

[octagondd]

I would say right after the charge. Just curious if my math is the same as the real world. In other words, charging from 3V to 3.6V you may only have 3.58V in your 114 Ah cells since that would be a 3% difference in capacity.

 

[etischer]

I am currently running without BMS. After balancing, my cells are within 0.002v. I balance them when they are more than 0.02 out of balance.

I have 99 ThunderSky 100Ah.

I top balance about once a month by charging the low cells (@ +5 amps) to match the voltage of the highest cell (usually 3.420v). Then I shunt the highest cells (@ -1 amp) for 10 minutes.

I have my cells located in 4 groups. The cells which are mounted under the car and under the hood are generally lower in voltage compared to the cells mounted in the trunk. Sometimes I charge an entire group of around 30 cells before individually charging each one.

My Manzanita shifts to CV when the cells are around 3.400v, it holds this for 35 minutes then shuts off.

Resting voltage is around 3.33v.
Pulling 1C they dip to 3.23
Pulling 2C they dip to 3.13

After 25 miles of driving (-25Ah)
Resting voltage is around 3.30
Pulling 1C they dip to 3.19
Pulling 2C they dip to 3.09

At 2.8C, my max controller current, I've never gone below 3.00V ave/cell.
I typically only use about 30% of my capacity, so I'm far away from the extreme ends of the SOC curve. I have about 10k miles on the car.

Temperature seems to be the biggest factor that affects the normally consistent performance. I usually see 10V sag for every 100A of current. Works out to be 1 mOhm per cell.

 

[DavidDymaxion]

I'd like to add the questions:

What is your max charging current?

What is your max regen current?

I am trying to find real world pack balance data from LFP users with no BMS. I want to see how the pack stays in balance over time. There probably isn't much out there, but anything is better than nothing at this point.

Do you have an LFP pack with no BMS?

Do you have any data from the history of your cells, particularly how well they stay in balance?

Did you initially balance them?

Top or Bottom balance?

What discharge rate are you typically drawing from your pack?

Do you balance your pack every so often to bring them back in line?

Do you use cellogs or other device to log all or some of the cells?

Would you share this data via a link or email?

Thanks,
Dave

 

[JRP3]

I would say right after the charge. Just curious if my math is the same as the real world. In other words, charging from 3V to 3.6V you may only have 3.58V in your 114 Ah cells since that would be a 3% difference in capacity.

No it doesn't work that way. If I charge my smallest cells to 3.6, which I rarely do, the largest cells will only get to around 3.45 or so. Voltage climbs much faster near end of charge so it's not linear at all.
After sitting for a while they all drop down to 3.336-3.338.

 

[etischer]

My max charging current is 20 amps @ 230 volts ac, so battery current is about 14 amps.

I have my regen current limited to 120 amps. Once the battery voltage reaches 336 volts (3.4 v/cell) the regen torque is tapered to zero.

I'd like to add the questions:

What is your max charging current?

What is your max regen current?

 

[JRP3]

I'd like to add the questions:

What is your max charging current?

What is your max regen current?

My max charging current is 38 battery amps at 120 battery volts from a 240V 30 amp circuit though not for long since the PFC 30 goes into overtemp limit and cuts back to around 36 amps. Charging 120VDC from 240VAC is not very efficient. Normally I charge around 25 battery amps.
Max regen I've seen is a little over 200 amps for a couple seconds, normally between 50 and 150 amps.

 

[tomofreno]

Do you have an LFP pack with no BMS?
No. I have a minibms with no shunts.

Do you have any data from the history of your cells, particularly how well they stay in balance?
Yes, I've posted data several times on my build thread SwiftE. Generally, 32 of the 36 have remained balanced to within +/-4mV or less for almost 1 year and around 5k miles. That is both after full charge, or at around 35% SOC. Four cells were replaced, and these tend to slowly drift lower in voltage compared to the others, around 0.2V (edit, that should be 0.02V!) in over about 2-3 months, as measured at 35% SOC. I usually charge them individually with a power supply at that point to re-balance, so don't know if they would continue to drift or reach some stable voltage.

Did you initially balance them?
No.

Top or Bottom balance?
Sort of bottom, about 3.1V if I recall correctly, and that is only the 4 replacement cells.

What discharge rate are you typically drawing from your pack?
Around 0.1C to 0.3C at lower speeds, 0.4 to 0.8C or so on the highway, about 0.7C to 1.4C up an about 4.5% average grade (varies quite a bit) for 19 miles, I drive maybe once per week. Up to about 2.2C fleetingly when I accelerate more aggressively, usually more like 1.5C.

Do you balance your pack every so often to bring them back in line?
Just the 4 replacement cells. The other 32 have never been balanced.

Do you use cellogs or other device to log all or some of the cells?
I did log around 4 cells at a time a number of times but it is pretty boring as they are always in balance and all show about the same voltage changes under load. Some of the plots are posted on the thread.

What is your max charging current?
~0.2C. Usually charge at 0.14C to 0.17C depending on temperature in the garage (charger overheats, Manzanita PFC30)

What is your max regen current?
It has been as high as a bit over 1C fleetingly in a quick stop on a highway. Usually is 0.8C or less.

I think the larger fraction of C current you draw, the more you will have to balance your cells, and I expect their life will also be shorter. I expect racers have real problems with keeping cells balanced, and that is why they think you are nuts if you don't use a balancing bms.

 

[tomofreno]

I just checked my cell voltages since it had been several weeks, and I can't use the car (charger died). Voltages ranged from 3.298 to 3.300V, including the 4 replacement cells. SOC is 55%. I think I last individually charged the 4 replacement cells to balance them back in June.

 

[dtbaker]

excellent info Tom.... I'll try to plug my running experience in this thread after i upgrade to Li.... probably in Nov.

 

[tomofreno]

Dan, note the soc was 55%. The change in voltage with Ah is very small at 55% soc. At 20% soc I expect there would be greater imbalance. But that doesn't matter to me, as I never go below 30%, rarely below 35% soc, and there the balance is typically almost as good as I reported at 55% soc. I find life is very simple if you just operate on the flatter part of the V versus Ah curve, 30% soc to about 95% soc - where the exponential voltage rise starts. But like I said above, I think the higher C currents you regularly drive through cells, the more frequently you will need to balance. No data, just a guess that more stress accentuates cell differences. A comparison of JRP3's imbalance data and mine seems as if it might support this, but scant data so hard to say. I consider a 0.01V difference big, and usually re-balance the cell at that point. His cells are 100Ah, mine are 180Ah, both using the same controller and similar motors, so he likely pulls higher fraction C currents than me. For several months I measured all cell voltages before and after each charge, and I swear they became better balanced over time, without me doing anything. I quit doing it because the voltages were always almost the same. I still do it about once every few weeks or so, with the same result.

 

[dtbaker]

I consider a 0.01V difference big, and usually re-balance the cell at that point.

measured when? right after end of charge, or 'at rest'?

I am not sure I would consider .01 v within my multimeter accuracy. I think its rated at +/- 3%... but I dunno if that is absolute accuracy, or the repeatability.

I was thinking that balancing wouldn't really be required unless voltage differences were maybe .02 or more just after end of charge, when differences would be magnified as some might start climbing the voltage knee while others not.

 

[dtbaker]

Do you have an LFP pack with no BMS?
No. I have a minibms with no shunts.

please clarify what the minibms does if it does not do any shunting at end of charge.... does it just shut down the charger as soon as the first cell hits the target high voltage?

 

[tomofreno]

please clarify what the minibms does if it does not do any shunting at end of charge.... does it just shut down the charger as soon as the first cell hits the target high voltage?

You can get all the info you want on the minibms in the thread on it in this forum. You have a choice of ordering the minibms with or without shunts. I ordered it without them. Its other functions are HVC and LVC, similar to other bms. So yes, it shuts down the charger when a cell hits a high voltage of 3.6V, and it has a relay on the main board which can be used to trigger an alarm and/or cut the throttle if a cell hits a low voltage of I think 2.5V - but not sure on the actual voltage level. The low level has some temperature compensation to account for more voltage sag at lower cell temperatures.

 

[etischer]

After reading some comments in other threads about the troubles of balancing mid-soc, I turned my charger up to see what would happen. I increased the voltage from 336 to 337, just one volt, about 1/8" of a turn.

I had always been charging to 3.37, and the cells would finish 3.40, they generally stay within 20mv.

With the charger turned up ever so slightly, I noticed my highest cells started shooting though the roof, about 1mv every few seconds. I let it get to 3.5volts, and while it was doing so, my low cells were starting to drop in voltage due to the current tapering off. I now had over 120mv spread when I almost never had more than 20mv before.

It's pretty easy to have cells within 10mv when I don't charge above 3.37v. I'm starting to realize how out of balance things can be when you get to the knee.