diy solar

diy solar

LFP overdischarge conflicting voltage limits

Luthj said:
Some other interesting bits from that paper.

"Guo et al. found that the dissolution of SEI occurs within 0 to −10% SoC, severe copper dissolution then occurs below −12% SoC, with severe internal short circuiting occurring at or lower −20% SoC19. For the batteries used in this study, 0.5 V corresponds to −1.9% SoC and as such, in agreement with Guo et al.19, only SEI dissolution and gassing occurs."

The good news from this, is that while its very undesirable to discharge the cell below 2.0V, major damage between 2.0 and 0.5V takes a fair bit of time. If the cell is promptly brought back up, then capacity loss can be mitigated.
Click to expand...
Digging in my notes I found the data of my "short circuit test" mentioned above:

After dischaging my Headway 12 Ah cell down to 2.5 Volts, I connected a 2 Ohm resistor across the terminals. The voltage gradually dropped down to 0 V. I have no record about the timing. I removed the resistor 24 hours after I connected it to the cell. Then I charged the cell with 0.1 A up to 2.5 V. The charging current was then set to 1 Amp, and the cell was charged up to 3.55 Volts.
A capacity test afterwards revealed no loss of capacity.
 
I believe in operating on the side of greatest safety .... Why push the envelop and risk reducing the life of the cells. There is VERY little to be gained by charging a little bit higher or a little bit lower .... but there is the possibility of loosing significant pack life.
 
Luthj said:
If the cell has 100AH, then to get from 2.5V to 0.5 requires an additional 1.9AH of power to be removed from the cell.

Obviously this will vary with each specific cell. In order to get to the rapid damage zone of -20% SOC, 20AH would need to be removed from the cell. This may only be possible if the cells which remain in the pack reverse charge the first cell to reach 0V. At that point the damage gets pretty extreme.

Though at 0.5V capacity loss seems significant at 1% per day.
Click to expand...
One of the big mistakes in all of my previous considerations was to assume, that 0 Volts represents 0% SOC. I learned that 2.3 Volts is equivalent to 0% SOC.
With that in mind all "conflicting" data suddenly make sense to me!!
Thanks again for your help!
 
ArthurEld said:
Bottom balancing would help.
Your expectations are higher than mine.
It will be interesting to hear what you end up doing.
Click to expand...
I believe that balancing has no impact on my issue, which is dealing with the characteristics of single cells.

First off all, I must find out how much Ah does it take a 100 Ah cell to go from 2.7 Volts down to 2.3 Volts, assuming small currents. Then I will divide the Ah by the Amps being consumed by the BMS and I will get the maximum time a 100 Ah Battery can be left unattended. Finally I can scale up or down the data for other capacities.
 
Arthur,
All of us provide comments which are sometimes more and sometimes less helpful.
And there is no way to learn from "no comments" .
There is no reason to apologise!
Regards Hans
 
Not exactly on topic, but kinda related...i externally powered my BMS through it's LVD relay, connected to a SSR, so that if activated the power drain of the BMS is terminated at the same time. At a disconnect of 2.6v/cell, this should still leave plenty in the cells even with their own self discharge rate for months. This was to eliminate all parasitic drains possible, as related to the discussions above.

I simply love all the testing/experimentation folks are doing...real world data. Well done Hans (y)
 
Solarfun4jim said:
Not exactly on topic, but kinda related...i externally powered my BMS through it's LVD relay, connected to a SSR, so that if activated the power drain of the BMS is terminated at the same time. At a disconnect of 2.6v/cell, this should still leave plenty in the cells even with their own self discharge rate for months. This was to eliminate all parasitic drains possible, as related to the discussions above.

I simply love all the testing/experimentation folks are doing...real world data. Well done Hans (y)
Click to expand...
How do you reconnect once the BMS has hit LVD and opened the SSR (and so shut itself off)?

I suppose your concern is protecting your battery over long unattended periods. Those of us using our batteries for solar power generation/storage are more concerned that the system automatically wakes itself up once the sun begins to shine after it’s inadvertently shut down because a weak cell discharged to LVD...
 
fafrd said:
How do you reconnect once the BMS has hit LVD and opened the SSR (and so shut itself off)?

I suppose your concern is protecting your battery over long unattended periods. Those of us using our batteries for solar power generation/storage are more concerned that the system automatically wakes itself up once the sun begins to shine after it’s inadvertently shut down because a weak cell discharged to LVD...
Click to expand...
I simply slide the switch from external power to being powered through the sense wires.... you get an instant reading of all the cell voltages.

You are correct, i'm talking from the perspective of an RV which gets basically shut down for the winter. If the pack is around 50 - 60% when the main power switch is thrown at the end of a season, their should be no drain other than the BMS, which i would prefer to have monitoring the cells. If the ambient temperature drops below zero, which happens frequently during our winters, then i hope the BMS drain is also cut off. I know it is a miniscule drain, but i just dont want it happening at say -5deg celcius.
 
Solarfun4jim said:
I simply slide the switch from external power to being powered through the sense wires.... you get an instant reading of all the cell voltages.

You are correct, i'm talking from the perspective of an RV which gets basically shut down for the winter. If the pack is around 50 - 60% when the main power switch is thrown at the end of a season, their should be no drain other than the BMS, which i would prefer to have monitoring the cells. If the ambient temperature drops below zero, which happens frequently during our winters, then i hope the BMS drain is also cut off. I know it is a miniscule drain, but i just dont want it happening at say -5deg celcius.
Click to expand...
Understand the concern and for that use-case, makes perfect sense.

I have another 90Ah LFP battery that I disconnect the BMS harness from when being stored for extended periods because I’ve seen that reduces current draw but you’ve motivated me to check how much current is still being drawn by the BMS even when power transistors are shut down / off...
 
It makes a big difference, which kind of system one prefers:
1. solar power recharges the battery after an LVP disconnect, or
2. battery remains disconnected after an LVP, in order to first become aware of a potential problem with one cell, befor putting it online again.

It is not easy to choose the system suitable for your particular application.

In case of version 2. it is most important to know exactly how much currenrt is drawn from the batterie after the disconnect.

There are various loads, which can drain the cell:
a. BMS electronics
b. cell voltage monitor
c. cell modules mounted directly on top of the cells

I am using a version 2 System with the following characteristics:
- BMS electronics: 60 uA,
- cell voltage monitor 30 mA
- cell modules 0.25 mA

Important: I use a switch to disconnect 4 wires of the cell voltage monitor after checking.
As a result, my System draws 0.31 mA after an LVP disconnect. This is less than 0,01 Ah per day.

I still don't know how much Ah I can draw after an LVP disconnect at 2.7 V until I hit the 2.3 Volts (0 % SOC).
Assuming 0.1 Ah is the remaining charge at 2.7 Volts, I can leave my system unattended for 10 days maximum.

I will try to measure how much charge is left at 2.7 Volts, but this will take a while. I will repor my findings......
 
Hans Kroeger said:
Important: I use a switch to disconnect 4 wires of the cell voltage monitor after checking.
As a result, my System draws 0.31 mA after an LVP disconnect. This is less than 0,01 Ah per day.

I still don't know how much Ah I can draw after an LVP disconnect at 2.7 V until I hit the 2.3 Volts (0 % SOC).
Assuming 0.1 Ah is the remaining charge at 2.7 Volts, I can leave my system unattended for 10 days maximum.

I will try to measure how much charge is left at 2.7 Volts, but this will take a while. I will repor my findings......
Click to expand...
The main thing about all your testing and fact finding on your particular cells, is that you get to know exactly when you must check on them due to discharge rate. You set a reminder on your phone or email system to go check and you know if you dont, you are risking 100's $$$

I'm still trying to get to know my pack, (but out of commission just now due to broken stud), but my eventual aim will be to move my LVD to the highest point away from 2.5v but still be functional. With my cells being 280Ah, i'm hoping to have a reasonable buffer left in the cells after LVD. The reason for this, is that i will still have a BMV712 connected, which uses up 0.6Ah per month.

Very interesting work you are doing, although i'd be 'sweating it' if i was taking the cells down to 2.3V... :ROFLMAO:
 
It takes a lot of experimenting before I start to see what is going on.
I tend to look for exceptions to the rule for some reason.
Batteries have been an enjoyable challenge so far. I think I'll be messing with these things for a long time.
 
Solarfun4jim said:
......The reason for this, is that i will still have a BMV712 connected, which uses up 0.6Ah per month.
Click to expand...
I connected my BMV712 behind the security relay. It will not draw any power after an LVD. Drawback is, It needs to be synchronised when the Battery is brought back online, which is no problem for my.
Solarfun4jim said:
Very interesting work you are doing, although i'd be 'sweating it' if i was taking the cells down to 2.3V... :ROFLMAO:
Click to expand...
I understand your concern. I hope that my conclusions from that paper about LFP storage at 0 % SOC during transportion, are correct. In any case this is the limit for my worst case calculations, and there is no intention to ever get that low.
 
Hans Kroeger said:
I connected my BMV712 behind the security relay. It will not draw any power after an LVD. Drawback is, It needs to be synchronised when the Battery is brought back online, which is no problem for my.
Click to expand...
Yes, i'm still contemplating doing that myself. In my situation, it will probably come down to whether i can adapt the BMV to synchronise at say a 3.45v/cell charge up. Originally, it was set at 280Ah and a full charge to 3.55v approx, but i'm wondering if i can amend the settings such that, for example, 270Ah and 3.45v/cell becomes the point at which it re synchornises as 100% automatically. I do not wish to be charging up to 3.55v/cell once per month, just to re synch/re calibrate. All just to be evaluated once i get my new stud for my cell.
 
Solarfun4jim said:
The main thing about all your testing and fact finding on your particular cells, is that you get to know exactly when you must check on them due to discharge rate. You set a reminder on your phone or email system to go check and you know if you dont, you are risking 100's $$$

I'm still trying to get to know my pack, (but out of commission just now due to broken stud), but my eventual aim will be to move my LVD to the highest point away from 2.5v but still be functional. With my cells being 280Ah, i'm hoping to have a reasonable buffer left in the cells after LVD. The reason for this, is that i will still have a BMV712 connected, which uses up 0.6Ah per month.

Very interesting work you are doing, although i'd be 'sweating it' if i was taking the cells down to 2.3V... :ROFLMAO:
Click to expand...
First preliminary Test Results:
discharging a 100 Ah Winston LYP cell with a current of 0,5 mA from 2.56 Volts down to 2.4 Volts took a total 0.35 Ah charge being removed from the cell.
In other words:
- if you have a 100 Ah LiFeYPO4 battery, and
- your BMS has a LVD at 2.6 Volts,
- it will take "x" days for "y" mA quiescent current to discharge your battery down to 2.4 Volts
x @ y for 100 Ah battery
30 days @ 0.5 mA
15 days @ 1 mA
3 days @ 5 mA

x @ y for 200 Ah battery
60 days @ 0.5 mA
30 days @ 1 mA
6 days @ 5 mA
3 days @ 10 mA

I plan to do a similar test with an LiFePO4 cell from CALB. But again, it will take while.
cheers Hans
 
You must log in or register to reply here.
Back
Top