LiFePO4 Batteries: Discussion Thread for Those Using Them as House Banks

[newhaul]

Originally Posted by GRIT

I should have mentioned, the maximum load was only 18 amps, on a 720 ah battery.

And I'm guessing that there's a "better" cell in group 3, not a bad cell.

Thanks.
Paul.

at this point imo the difference is to close to call .

[newhaul]

Ok need input . I just got offered a zantrex prosine 2.0 for 300 bucks untested .
It has a 100 amp charger built in I looked at the manuals online and can't find the information .
Is it programmable for use with my Lfp bank ( its way to big but can't pass up the deal . )

[GRIT]

I've no experience with this yet, so that was why I asked.

Are you saying that .04 mohms is fine? I only ask, because the other three groups are so close to equal.

What are the measurements on your system, for comparison?

Cheers.
Paul.

[newhaul]

Originally Posted by GRIT

I've no experience with this yet, so that was why I asked.

Are you saying that .04 mohms is fine? I only ask, because the other three groups are so close to equal.

What are the measurements on your system, for comparison?

Cheers.
Paul.

honestly I having felt the need to test them to that degree. Just .04 seems so miniscule
What is your measure after balancing? Or do you still need to do that ?
I'm running such a small bank it wouldn't make a difference 4p4s for 100ah total (25ah cells)

[OceanSeaSpray]

Originally Posted by GRIT

I've no experience with this yet, so that was why I asked.

Are you saying that .04 mohms is fine? I only ask, because the other three groups are so close to equal.

What are the measurements on your system, for comparison?

Cheers.
Paul.

Paul,

Maybe you can swap cells around to create more equal groups, but what is the voltage difference 0.04mOhms is going to cause? 4mV @ 100A. Leave it alone.
If anything, it could come from your connections, not even the cells.

Internal resistance increases considerably as the cells age, so difficult to compare, and the size of the bank also has an influence on it.
I had measured 0.42mOhms/cell on a new 2P4S 400Ah bank a few years ago, including cell links.

Regards,

Eric

[john61ct]

Yes internal resistance is a very variable measurement, depends a lot on the context and measuring instrument, too much to be of much value comparing between setups.

Certainly outliers are an indicator to take into account, comparing between cells or sub-bank groups, or changes over time though.

Actual Ah capacity of each cell is really the best functional indicator of health, comparing changing curves of volts vs Ah during charge / discharge cycling.

A good reason not to design the bank with too many cells.

Some will juggle better / worse cells to get "balanced groups" but for others the higher risks and increased attention required means best to replace bad cells with good, or at some point just replace the bank all at once.

Too many variables, no one approach fits all.

[CatNewBee]

internal cell resistance is really hard to mesure. You can only guess it by dividing voltage difference to current difference for 2 or more different loads.

Because of the cell characteristics near the shoulders measurement is almost meaningless above 90% SOC ( cell drift, imbalances) and below 20% SOC. So dont worry too much about it.

Monitor the charging over time for several cycles,, use WCIB? command to see what cell is balancing and what cell is receiving the charge. At some point the cell will be equalized, so the donating cell and the receiving cell will move across the bank.Than all is good.

If you experience always the same cell donating, this cell is the weakest. It accepts to few energy before it reaches the shoulder and empties quicker than the others. by re- arranging the cells you can ballance the packs a little.

You also can charge all cells together in parallel to full, equalized and then discharge them individually by measuring the discharged energy to a defined set point at the lower shoulder e.g 2.9V. This is the only way to find out what capacity asingle cell has, especially in a 4S4P configuration.

greetings
catnewbee

[senormechanico]

Originally Posted by newhaul

Ok need input . I just got offered a zantrex prosine 2.0 for 300 bucks untested .
It has a 100 amp charger built in I looked at the manuals online and can't find the information .
Is it programmable for use with my Lfp bank ( its way to big but can't pass up the deal . )

I purchased one several years ago. DOA.
Returned it for another. DOA in a different way.
Opened it up to see if I could find the problem.
It looked like it was home made with AWFUL soldering.
Total POS. I returned it for a refund.
It's even worse than most Xantrex stuff !
Whether or not it's programmable for LiFePo4 to me, is a moot point.

[GRIT]

Thanks guys, for your responses to my resistance Question.

Thank you, CatNewBee. I will try that command today. I will also start a log, so I can see how things progress. I'm sure the problem is; now that I have the bms, I have too much information to consider. I did quite a lot of research on LiFePo4, before purchasing, but nowhere did I see actual resistance numbers. Everyone says they should be nearly the same, but I don't know the definition of "nearly the same", yet.
Ignorance, is certainly less stressful than knowledge.

John61ct, I understand your comment regards the number of cells. Unfortunately, as hard as I tried, it was nearly impossible to get the cells I wanted (Winston), delivered to Canada. I settled for CALB, and the largest cells I could get were 180ah.

Without compromise, I wouldn't even be here.

Cheers, all.
Paul.

[john61ct]

I did not mean to imply you have too many.

Just pointing out that fewer can make monitoring / troubleshooting easier.

Too few is also possible, less redundancy.

Everything is a compromise, and as you say prosaic practicalities like availability are also an important factor.

[john61ct]

Originally Posted by newhaul

Ok need input . I just got offered a zantrex prosine 2.0 for 300 bucks untested

I'd pass. Xantrex is fine long as you **know** it's working.

Until it breaks, and then usually becomes unfixable scrap, company does not care to help even their authorized repair centers, thinks you'll just keep buying the new models.

[GRIT]

Originally Posted by john61ct

I did not mean to imply you have too many.

Just pointing out that fewer can make monitoring / troubleshooting easier.

Too few is also possible, less redundancy.

Everything is a compromise, and as you say prosaic practicalities like availability are also an important factor.

I took no offense, I agree with your statement. My personal "optimum" number of cells was 8; 400ah Winston cells.

I see my post said " I understand", when I meant to say, "I agree".

Cheers.
Paul.

[newhaul]

Originally Posted by senormechanico

I purchased one several years ago. DOA.
Returned it for another. DOA in a different way.
Opened it up to see if I could find the problem.
It looked like it was home made with AWFUL soldering.
Total POS. I returned it for a refund.
It's even worse than most Xantrex stuff !
Whether or not it's programmable for LiFePo4 to me, is a moot point.

thanks for that information . Exactly what I was looking for .
I am going to pass on this one .
Was to big for my system anyway but if it was a good deal I would have even considered doubling by bank size.

[nebster]

In my batch of 112 cells, all 100Ah measured at ~70% SOC, the mean Ri was 0.28mOhm. The outliers were 0.24mOhm and 0.36mOhm. The rest fell in roughly a normal distribution about the mean.

Hope that helps!

[GRIT]

Thank you Nebster. That does help.

Wow, 112 cells...

Cheers.
Paul.