LifePO4 GBS Amp Hour Testing 2.5v to 3.6v per cell

[createthis]

OK back to how to measure Ri of a battery. It is extremely easy to do and even easier if you have a PL8 to work with. Here is a link to a Sticky located here. However I will go through it personalized to what you have. We have to make one change to the method I linked to because it needs modified to measure lithium cell. We perform the test with Lithium at Partial State of Charge in the range of 30% to 70 SOC range. Just somewhere in the middle. Short story around 3.3 volts.

OK conditions to be met are:

Battery in PSOC range
Battery temp normal romm temp
All voltages are taken directly from the battery TERM POST

Test Equipment and Hardware required are:

PL 8 set up to discharge at 1 and 10 amps.
Good quality DMM with a minimum 3.xxx resolution or better.
Pencil and paper
Calculator
Protein Computer armed and dangerous.

We are going to use the Delta Voltage/current to find Ri testing at 1 and 10 amps. Exactly how a $3000 Digital Low Resistance Ohm meter does it with high current for accuracy and resolution.

Method

Set your volt meter to read 3 volt range directly on the battery Term Post and from no place else.
set up PL8 to discharge at 1 amp.
Connect to battery, and apply 1 amp discharge.
While on Discharge read and record battery voltage and call it V1. Example V1 = 3.297 volts
Record Current 1 as I1 = 1 Amp.

Repeat test at 10 amps.

Record voltage, call it V2 = 3.270 volts
Record I2 = 10 Amps

OK now it is just a little math

1. Find the difference in voltage by taking V1 - V2 = Delta Voltage. so Dv = 3.297 - 3.270 = .027 volts

Solve for Delta Current by taking I2 - I1 = DI = 10 amps - 1 amps = 9 amps

Now solve for resistance of cell R= Dv / Di = .027 volts / 9 amps = .003 Ohms or 3 mili-ohms.

I was all wired in parallel (yeah, I said I wouldn't, but the voltages were more divergent this time around) and ready to rewire in series, charge, and forget about it. However, I think I might make a video about this procedure instead.

[Sunking]

OK back to how to measure Ri of a battery. It is extremely easy to do and even easier if you have a PL8 to work with. Here is a link to a Sticky located here. However I will go through it personalized to what you have. We have to make one change to the method I linked to because it needs modified to measure lithium cell. We perform the test with Lithium at Partial State of Charge in the range of 30% to 70 SOC range. Just somewhere in the middle. Short story around 3.3 volts.

OK conditions to be met are:

Battery in PSOC range
Battery temp normal romm temp
All voltages are taken directly from the battery TERM POST

Test Equipment and Hardware required are:

PL 8 set up to discharge at 1 and 10 amps.
Good quality DMM with a minimum 3.xxx resolution or better.
Pencil and paper
Calculator
Protein Computer armed and dangerous.

We are going to use the Delta Voltage/current to find Ri testing at 1 and 10 amps. Exactly how a $3000 Digital Low Resistance Ohm meter does it with high current for accuracy and resolution.

Method

Set your volt meter to read 3 volt range directly on the battery Term Post and from no place else.
set up PL8 to discharge at 1 amp.
Connect to battery, and apply 1 amp discharge.
While on Discharge read and record battery voltage and call it V1. Example V1 = 3.297 volts
Record Current 1 as I1 = 1 Amp.

Repeat test at 10 amps.

Record voltage, call it V2 = 3.270 volts
Record I2 = 10 Amps

OK now it is just a little math

1. Find the difference in voltage by taking V1 - V2 = Delta Voltage. so Dv = 3.297 - 3.270 = .027 volts

Solve for Delta Current by taking I2 - I1 = DI = 10 amps - 1 amps = 9 amps

Now solve for resistance of cell R= Dv / Di = .027 volts / 9 amps = .003 Ohms or 3 mili-ohms.

[createthis]

First read up on the manual. But when you set up a Balance Charger, the charger has a Default Charge Termminate setting. Default I think is C/10 So if you select 10 amps charge rate, the Charge Terminates when the current tapers off to C/10 or 1 amp if you selected 10 amps charge current. With me so far?

To fully charge a lithium battery, yours, you charge to 3.65 volts, and terminate when charge current drops to 3% of C. For you C = 100 Amp Hours so 3% is 3 amps. So if you used say a 10 amp rate, it cuts off at 1 amp thus over charging the battery somewhat. But if you selected say 30 amps the charger would cut-off at 3 amps. Perfect.

I think this explains the flat spot on the high knee in my charge graphs. Thanks. I was thinking that was overcharge, but I wasn't sure why.

[Sunking]

I guess I didn't communicate it well, but this was a charge test only, not a discharge test. The discharges were only to get each cell down to 2.5v so I could run the charge test.

OK your Discharge Test was invalid then because you do not know if they were full or not.

Hmm. Which charge algorithm should I be using in the future?

First read up on the manual. But when you set up a Balance Charger, the charger has a Default Charge Termminate setting. Default I think is C/10 So if you select 10 amps charge rate, the Charge Terminates when the current tapers off to C/10 or 1 amp if you selected 10 amps charge current. With me so far?

To fully charge a lithium battery, yours, you charge to 3.65 volts, and terminate when charge current drops to 3% of C. For you C = 100 Amp Hours so 3% is 3 amps. So if you used say a 10 amp rate, it cuts off at 1 amp thus over charging the battery somewhat. But if you selected say 30 amps the charger would cut-off at 3 amps. Perfect.

I didn't "come up with" 200 watts. I measured it with a kill-a-watt in real time and on camera.

OK then you have serious metering errors in your kill-a-watt meter. You PS and PL8 ar over 90% eficient. That means the max power going into the charger should be no more than 125 watts if charging at 100 watts.

It takes 10 hours to do a discharge. I'm not waiting around for the phone to ring. The set point is 2.5v, but I think it's rising naturally over time until I finally come around and measure it. Like you said, it's not really a big deal.

Why wait around for anything. Go to bed, go to work, go do something as the test does not need your presence. ? If it were me, I would just turn on a 1000 watt load and let them go until the Inverter cuts off on low voltage. Them wire all of them up in Parallel, set the PL8 to 2.4 volts and walk away for a day.

But you are right, no big deal if you Top Balance. If you BB you need to get the Balance done correctly.

More later when I get time.

[createthis]

Issue 1. Did you fully charge each cell before did the Discharge AH test?

I guess I didn't communicate it well, but this was a charge test only, not a discharge test. The discharges were only to get each cell down to 2.5v so I could run the charge test.

Issue 2 How did you do a full charge? This is the kicker question because if you used PL8 Default algorithm would have left you slightly undercharged. In the PL8 firmware it is programmed to terminate charge when the current tapers to C/10 of the charge current selected. That can cause you to either undercharge or over charge the battery. No problem you just have to be aware of it and how to work around it.

Hmm. Which charge algorithm should I be using in the future?

Charge Wattages you quoited do not jive with the parameters given. If you are charging at 30 amps x 3.6 volts is 108 watts. Where were you coming up with 200 watts? To be 200 watts would require 60 amps of charge current at 1S. Your numbers do not jive. I suspect I have the details wrong.

I didn't "come up with" 200 watts. I measured it with a kill-a-watt in real time and on camera. You saw me enter the settings. I agree the math doesn't jive, but I suspect you're just seeing accumulated losses.

Lastly your Internal Resistance Measurement are complete Bogus or you have terrible batteries. It is True GBS has the highest Ri of all the Chi-Coms, thus why EV guys do not use them as it robs them blind, shorter cycle life, and heating issues. GBS 100 AH cells run 3 to 4 milli-ohms. You are saying 13 to 14 milli-ohms. OK let you in on a Secret with the PL8. Its Ri measurement is JUNK. Well known fact. They use pulses rather than Delta. Result is make believe readings.If you want to know how to measure Ri, just ask. Your PL8 is Junk when it comes to Ri.

Well, ****. Ok, how do I measure Ri? I'm marginally curious.

Lastly lets work on your terminology. You used the word SAG to describe Rested Open Circuit Voltage. They are two completely different things. Sag is the voltage drop incurred from OCV to fully loaded Current. It is proportional to Internal Resistance. Example if you have a 3.35 open circuit voltage fully charged up GBS battery with a R= .013 ohms and apply 20 amps the voltage Sags from 3.35 volts down to 3.35 - [20 amps x .013] = 3.09 volts which really sucks as that is a 8% loss of power of Voltage Sag. If you had a good 100 AH cell with only 2 milli-ohms you would only sag down to 3.31 volts or a 1.9% power loss. As you can see from this example if your cells are really 13 to 14 milli-ohms, you have sukass batteries.

OCV or open circuit voltage of a rested battery has nothing to do with Sag. Lithium is no different than any other battery, just different degrees. Example you charge a 12 volt FLA battery with 14.4 volts just like a 4S LFP. Let them rest until the surface charge bleeds off. When that happens a 12 volt fully charged Pb battery = 12.6 volts and LFP around 13.4 volts. Voltage meaningless in both batteries. Only useful if OPEN CIRCUIT and allowed to rest. Apply any load or charge and voltage is meaningless.

Fair enough. I'll work on that.

On the discharge side you left fuel in the tank. As you noted you only discharged to 2.8 volts. Empty if 2.4 to 2.5 OCV. There is that word again. Two ways to get there. set your set point below 2.5, or even better fast discharge at 10 amps until it complete, then repeat ay 1 amp to 2.5 volts until you rest at 2.5. No big deal as we are only talking 2 or 3 amp hours difference.

It takes 10 hours to do a discharge. I'm not waiting around for the phone to ring. The set point is 2.5v, but I think it's rising naturally over time until I finally come around and measure it. Like you said, it's not really a big deal.

So now you are ready to find your Magic Number. You know weakest cells is 103 AH. Do a proper Bottom Balance until the cells are OCV to 2.5 volts. Do this by connecting all cells in parallel and fully discharge them to 2.5 vpc fully rested. 2.4 to 2.5 is right, but every cells should be exactly the same at the bottom. That is done by wiring them all in parallel. Let them rest over night still connected in parallel at 2.4 to 2.5 volts. Don't worry th eline you do not cross is 2 volts. 2,5 is a lot of room for error.

I already did a video about bottom balancing, though I didn't wire them in parallel afterward: https://youtu.be/rZGLbAv2MCA I understand the reason for wiring in parallel but I'm not really sure it's necessary with so little energy in the cells, and I really don't think it's worth the trouble to remove the strapping. Not really planning to do another bottom balancing video.

I've already found the "Magic Number" as you call it for this pack. I have footage, but it's pretty boring so I'll probably just show the graphs and talk about them in the next video.

Now reconnect the cells in Series, use your PL8, at Fast Charge mode, no Balance Plug and pump in 90 to 95 AH and Terminate the charge. Disconnect th echargger and allow the batteries to rest over night. Next morning measure the rested voltage. That is the Voltage you use on your charge controller to float at.

Yeah, you can't "pump in 90 to 95 AH" with a PL8. It only measures AH, it doesn't have a setting for terminating charge/discharge based on AH as far as I can tell. Gotta pick a voltage set point instead. More on that in the next video.

[Sunking]

OK very good, I have the exact same equipment I use on Model Planes and EV from time to time.

To answer your first question about the Revolectric Power Supply it is 1320 watts not 1000. 24 volts x 55 amps = 1320 watts maximum. No big deal. In fact very good, as you now own a charger that can charge any battery of today and tomorrow up to 60 volts. Great match-up of power supply to charger. They were made for each other which makes a wonderful marriage.

See a couple of issues with your method, again may be I do not not understand exactly what you said in the video .

Issue 1. Did you fully charge each cell before did the Discharge AH test? I assume yes, if not then...

Issue 2 How did you do a full charge? This is the kicker question because if you used PL8 Default algorithm would have left you slightly undercharged. In the PL8 firmware it is programmed to terminate charge when the current tapers to C/10 of the charge current selected. That can cause you to either undercharge or over charge the battery. No problem you just have to be aware of it and how to work around it.

To fully charge a GBS 100 AH cell requires you to CC/CV charge at 3.65 vpc (default is 3.6) and Terminate when charge currents tapers to 3% of C. Well 3% of C is 3 amps. So the trick is to select a Charge Current where C/10 = 3 amps or 10 x 3 = 30 amps. Set charger to charge to 30 amps @ 3.65 volts. Terminate when current tapers to 3 amps. That is 100% fully charged.

Charge Wattages you quoited do not jive with the parameters given. If you are charging at 30 amps x 3.6 volts is 108 watts. Where were you coming up with 200 watts? To be 200 watts would require 60 amps of charge current at 1S. Your numbers do not jive. I suspect I have the details wrong.

Lastly your Internal Resistance Measurement are complete Bogus or you have terrible batteries. It is True GBS has the highest Ri of all the Chi-Coms, thus why EV guys do not use them as it robs them blind, shorter cycle life, and heating issues. GBS 100 AH cells run 3 to 4 milli-ohms. You are saying 13 to 14 milli-ohms. OK let you in on a Secret with the PL8. Its Ri measurement is JUNK. Well known fact. They use pulses rather than Delta. Result is make believe readings.If you want to know how to measure Ri, just ask. Your PL8 is Junk when it comes to Ri.

Lastly lets work on your terminology. You used the word SAG to describe Rested Open Circuit Voltage. They are two completely different things. Sag is the voltage drop incurred from OCV to fully loaded Current. It is proportional to Internal Resistance. Example if you have a 3.35 open circuit voltage fully charged up GBS battery with a R= .013 ohms and apply 20 amps the voltage Sags from 3.35 volts down to 3.35 - [20 amps x .013] = 3.09 volts which really sucks as that is a 8% loss of power of Voltage Sag. If you had a good 100 AH cell with only 2 milli-ohms you would only sag down to 3.31 volts or a 1.9% power loss. As you can see from this example if your cells are really 13 to 14 milli-ohms, you have sukass batteries.

OCV or open circuit voltage of a rested battery has nothing to do with Sag. Lithium is no different than any other battery, just different degrees. Example you charge a 12 volt FLA battery with 14.4 volts just like a 4S LFP. Let them rest until the surface charge bleeds off. When that happens a 12 volt fully charged Pb battery = 12.6 volts and LFP around 13.4 volts. Voltage meaningless in both batteries. Only useful if OPEN CIRCUIT and allowed to rest. Apply any load or charge and voltage is meaningless.

On the discharge side you left fuel in the tank. As you noted you only discharged to 2.8 volts. Empty if 2.4 to 2.5 OCV. There is that word again. Two ways to get there. set your set point below 2.5, or even better fast discharge at 10 amps until it complete, then repeat ay 1 amp to 2.5 volts until you rest at 2.5. No big deal as we are only talking 2 or 3 amp hours difference.

So now you are ready to find your Magic Number. You know weakest cells is 103 AH. Do a proper Bottom Balance until the cells are OCV to 2.5 volts. Do this by connecting all cells in parallel and fully discharge them to 2.5 vpc fully rested. 2.4 to 2.5 is right, but every cells should be exactly the same at the bottom. That is done by wiring them all in parallel. Let them rest over night still connected in parallel at 2.4 to 2.5 volts. Don't worry th eline you do not cross is 2 volts. 2,5 is a lot of room for error.

Now reconnect the cells in Series, use your PL8, at Fast Charge mode, no Balance Plug and pump in 90 to 95 AH and Terminate the charge. Disconnect the charger and allow the batteries to rest over night. Next morning measure the rested voltage. That is the Voltage you use on your charge controller to float at.

EDIT NOTE:

Give you a tip. Build you a 4S Balance Plug to plug into the PL8. Gives you the ability to monitor all 4 cells with alarming. Also give you the ability to Top Balance is so inclined later as the PL8 can Balance Charge at 3 amps per cell with the Balance Plug.

[createthis]

Testing amp hour capacity of each individual cell in a 100ah LifePO4 12v pack.

Graphs linked in the description of the video.