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

[newhaul]

Originally Posted by sparrowhawk1

Equivalent programmable replacementAttachment 194001

that reseller is really jacking the price up my last quote from the factory was $65 or so for the tracer 40amp mppt controller .
Shop around some more .

This is the controller I use I know its only 30 amp but I only use 400 watts solar
If i wanted to add solar I would split it and use 2 controllers
https://www.windynation.com/Charge-C...628?p=YzE9MTc=

[sparrowhawk1]

Originally Posted by newhaul

that reseller is really jacking the price up my last quote from the factory was $65 or so for the tracer 40amp mppt controller .
Shop around some more .

This is the controller I use I know its only 30 amp but I only use 400 watts solar
If i wanted to add solar I would split it and use 2 controllers
https://www.windynation.com/Charge-C...628?p=YzE9MTc=

Are you sure it's the same model? Do you have a link? I did look through eBay again and I can get one for 150. But it still seems to me getting a disconnect before the controller would allow me to have redundancy at a low price. And buy separate ones for low-voltage, 1 for an alarm and 1 for cut off.

[newhaul]

Originally Posted by sparrowhawk1

Are you sure it's the same model? Do you have a link? I did look through eBay again and I can get one for 150. But it still seems to me getting a disconnect before the controller would allow me to have redundancy at a low price. And buy separate ones for low-voltage, 1 for an alarm and 1 for cut off.

factory direct via b2b sales.

If you are converting to Lfp house bank the 200 bucks for the appropriate controller is a minor part of the overall cost.

Edit just got email back from supplier the new prices are out and the 40amp unit is now 140 my cost so the 150 on eBay is readionable

[sparrowhawk1]

Originally Posted by newhaul

factory direct via b2b sales.

If you are converting to Lfp house bank the 200 bucks for the appropriate controller is a minor part of the overall cost.

Thanks for your patience. I googled B2B sales tracer without results. Like I said I like the redundancy. Also I'm probably going to start out small and maybe used lithium so $200 will be a factor. I'm also thinking of making it into a portable pack for a trolling motor in the future when I want to upgrade to a larger house bank.
Cheap & light disconnects seem to be the way to go with that.
So if anyone can recommend an adjustable low voltage alarm that would be appreciated too

[newhaul]

Originally Posted by sparrowhawk1

Thanks for your patience. I googled B2B sales tracer without results. Like I said I like the redundancy. Also I'm probably going to start out small and maybe used lithium so $200 will be a factor. I'm also thinking of making it into a portable pack for a trolling motor in the future when I want to upgrade to a larger house bank.
Cheap & light disconnects seem to be the way to go with that.
So if anyone can recommend an adjustable low voltage alarm that would be appreciated too

I'm sorry I didn't mean b2b as a service o ment it as my company direct to epever
Not via some service .

[sparrowhawk1]

Originally Posted by newhaul

I'm sorry I didn't mean b2b as a service o ment it as my company direct to epever
Not via some service .

At least I know $150 with free shipping is a fair price or the controller you have if I go that route thanks again for your time and help.

[nebster]

Originally Posted by sparrowhawk1

All the charge controllers I've ever used have said not to disconnect from the battery when the solar panels are connected. So it seems to me drop-ins would be harmful to my controller. I would think the easiest way around this is to have a sensor on the battery and have a disconnect where the panels go into the charge controller. Recommendations appreciated.

You could get a Victron charge controller. It doesn't care if the battery is present or not.

[sparrowhawk1]

Originally Posted by nebster

You could get a Victron charge controller. It doesn't care if the battery is present or not.

Price? Link? Also is it adjustable to exact voltage or does it just have a lithium selection. Newhaul same question for the controller you're using. Thanks. Also in the past controllers have had small terminals, I like using heavy gauge wire so if Max gauge wire is known maybe pass it on too. If I have a programmable controller or high and low voltage disconnect I don't think I would go with a drop in.

[nebster]

Originally Posted by sparrowhawk1

Price? Link? Also is it adjustable to exact voltage or does it just have a lithium selection. Newhaul same question for the controller yourself you're using. Thanks. Also in the past controllers have had small terminals, I like using heavy gauge wire so if Max gauge wire is known maybe pass it on too. If I have a programmable controller or high and low voltage disconnect I don't think I would go with a drop in.

Fully programmable.

They have a huge range and are one of the most popular choices. Just go look them up and pull the datasheets for the size you need. Prices vary.

[sparrowhawk1]

Originally Posted by nebster

Fully programmable.

They have a huge range and are one of the most popular choices. Just go look them up and pull the datasheets for the size you need. Prices vary.

Okay thanks. I still don't understand why it wouldn't make more sense just to get a high and low voltage cutoff.

[nebster]

Originally Posted by sparrowhawk1

Okay thanks. I still don't understand why it wouldn't make more sense just to get a high and low voltage cutoff.

It might. I'm not quite sure what that means, though. Maybe draw out a system diagram and post it?

[OceanSeaSpray]

Originally Posted by JmanC

Lab cycle life is typically claimed 1600-2000 for cycling from 20%dod back up to 100%. And many 1000's for shallow 20-30%dod. But how many can be expected from cycling between 20-85 all the time? Never getting back to 100%. Is it alot more cycles than 20-100? If so roughly how much.

If you never recharge your cells to 100%, you are going to screw them up completely on the long run because the memory effect will make them almost un-rechargeable past a point and this point will keep moving lower and lower over time.
While you don't technically "lose" capacity, the result is the same until remediation can take place.

I have been running a very interesting fractional cycling experiment here for 4.5 years and now this bank stays in absorption for over an hour at the end and only reaches 70% SOC when I take it to the termination point of 14.0V and 0.03C.
Recharging the last 30% would take hours and hours at very low current. The next step now is erasing that memory and getting back to normal performance. Basically, it shows very high internal resistance when the SOC approaches this point and it doesn't behave normally at all. Other then that, it performs fine. The internal resistance is generally a bit higher than when it was new, but this is normal with aging.

I will publish all the data and results once I have collected more information about releasing the cell memory.

[sparrowhawk1]

Originally Posted by nebster

It might. I'm not quite sure what that means, though. Maybe draw out a system diagram and post it?

Well here's a low voltage disconnect that I'm going to have to get something like regardless. I was hoping there would just be a simple programmable disconnect for let's say 14 volts or 13:8.https://www.amazon.com/Victron-BPR00...069&psc=1&th=1
It also says that it can be a charge interrupter. But I want something that will measure the voltage at the battery and disconnect it between the panels and an mppt.

[nebster]

Originally Posted by OceanSeaSpray

If you never recharge your cells to 100%, you are going to screw them up completely on the long run because the memory effect will make them almost un-rechargeable past a point and this point will keep moving lower and lower over time.

I’ve read the literature on this effect (in part, thanks to you bringing it to my attention), and I still can’t really marry up what they observe with the practical impact on a real battery pack. Maybe it is so subtle that I can’t understand it, or maybe I am dense. But I will try again.

(I guess I should preface this response with the statement that I’m not personally that concerned about lifetime beyond about 1500 cycles, because for me that is ten years, and that’s way longer than I’ll need. If the effect you are observing only presents after thousands of cycles, I will have concede disinterest and let others worry about whether it’s real and worth mitigating in some way.)

That said, let me take a stab:

While you don't technically "lose" capacity, the result is the same until remediation can take place.

If there is a way to remediate the problem, does that still mean the battery is “totally screwed up”? Or do you mean, it will be screwed up only until remediation is undertaken?

I have been running a very interesting fractional cycling experiment here for 4.5 years and now this bank stays in absorption for over an hour at the end and only reaches 70% SOC when I take it to the termination point of 14.0V and 0.03C.

I know you said you will give us a detailed write-up when you’ve concluded your remediation experiments, but can you just tell us the basic parameters of the system as a starting point for now? Things like cell type, brand, pack topology, BMS and re-balancing strategy if any, what charge regime, temperatures, rates, cycling frequency?

In addition to being deeply curious about your experiment and this phenomenon (which you and only you continue to bring up over the years, somewhat strangely), part of why I ask about the basic parameters is because some of what you say could be interpreted other ways.

For example, if you are always charging to 14.0V, that is definitely not a partial charge. (You don’t say you are, but you don’t say you aren’t either. If you aren’t, what is your normal charging method?)

For another example, it seems to me that an LFP cell that has just experienced normal degradation might exhibit behavior identical to what you describe; namely, charge to a “full voltage” and only measure 70% of the original energy going in. How do you distinguish between normal deterioration and this special effect?

Recharging the last 30% would take hours and hours at very low current.

Would take, or does take? I’m guessing you have tried it, and it does work? And how many hours and hours does it take? Is it, say, 0.015C for 30% = ~20 hours? (And is that holding CV at 3.5Vpc?)

Another question I have is, why did you decide to run your experiment to 70%? (If that is what you decided; it’s not quite clear, but I am presuming.) 70% is an unusual number, because the data suggest that something like 90% is nearly as good... way more storage, fewer cycles per total energy used, and not much more heating. I would expect most people would pick 85%, or 90%, or 95%. I chose 92%.

The next step now is erasing that memory and getting back to normal performance. Basically, it shows very high internal resistance when the SOC approaches this point and it doesn't behave normally at all. Other then that, it performs fine. The internal resistance is generally a bit higher than when it was new, but this is normal with aging.

It sounds like here you are talking about the same question I had above... how do you distinguish between normal aging and the special effect?

I have a separate question as well: if the experiment is to run the pack for hundreds or thousands of short cycles to 70%, and it works fine at 70% after all those years, what would be the rationale in real life for someone to remediate the pack back to a higher value? If someone has decided to run it at 70% for life, would they care?

I will publish all the data and results once I have collected more information about releasing the cell memory.

I’m definitely looking forward to it. As I have written before, I can see memory “bumps” in my charge curve. But they are so small, and they don’t seem to have any effect on the actual performance of the pack.

I have not tried taking my cells above the 92% stopping point. I could imagine I might see a large “bump” there in the curve, since they so often reach that point. But... I can’t figure out why I would want to do this either, except for science. Best I can tell, I’m getting the same stored energy today that I did on cycle zero, with the same CV threshold, in about the same amount of time.

What I’m wondering is, at what point would I observe some kind of performance impact that might motivate this testing or some kind of “remediation” process? Maybe you have an opinion from your experiment, already.

[sparrowhawk1]

Nebster. What voltage are you stopping at to get 92%? thanks