Waking up CALB SE LiFePO4 after years in storage

[nedfunnell]

Hello all,
I bought a partly finished electric motorcycle for my second project. It has 24 CALB SE cells. It was in storage a long time before I bought it. After I bought it, life got in the way and I didn't do anything with it for 2.5 years- except take a test ride around the block and find out that the pack voltage sagged badly under moderate load. I'm ready to work on the bike again and need to figure out if the pack is merely in a coma it can be woken up from or if permanent damage has occurred. I've killed a motorcycle pack before (22s Nissan Leaf cells, let to sit with small load from the charger, dead as a doornail) at 6.3v).

The good news: pack voltage is well within a normal range at ~66v, 2.75v/cell. I haven't checked individual cell voltages yet. When I test rode it as-received 2 years ago, I assumed the bad voltage sag was from low SOC that I didn't understand because the voltage was OK- I've since been told that voltage is not at all a reliable indicator of SOC for LiFePO4. I recharged the pack with the algorithm the PO had on the DeltaQ- #53, or 87.6v final voltage. Probably too high for 24s LiFePO4. The bad news: There was still substantial voltage sag after that, and that's the last time I touched it in two years.

So, can I revive this pack, do you think? Maybe it will perk up after some gentle cycles? What's the best way to try and revive it (restore original power capacity), and what do you think my chances are?

[Note: My original post said the voltage after storage was 79v- this was wrong, I must've mis-read the multimeter in the dim garage light and with a preschooler 'helping'.]

 

[e-beach]

LiFePo4 is well know for sagging under load. I run LiFePo4 and have done so for almost 8 years on the same pack. 1000+ cycles and still gives 80% of original watts. The pack sagged when it was new, however, it really sags now!

Charge the pack at a low amperage for a long time until you are sure it is fully charged and balanced. 3.65v per cell X 24 = 87.6v. Disconnect from the charger and let it sit over night then check every cell for voltage loss. Your bms might drain the whole pack a bit, but all the cells should all be within .1v or so of each other.

Put it back on the charger, make sure it is charged and balanced and go for a ride an see what it does.

Once your ride is finished check the cell voltages before charging and then again after charging and balancing.

Write all the numbers down to keep track of what you are doing.

Then repeat the process 3 or 4 or 5 times to get a solid idea of what the pack is doing.

If you have a blue tooth bms or some other way to know all the cell voltages by simply pushing a button, so much the better.

 

[nedfunnell]

LiFePo4 is well know for sagging under load. I run LiFePo4 and have done so for almost 8 years on the same pack. 1000+ cycles and still gives 80% of original watts. The pack sagged when it was new, however, it really sags now!

Charge the pack at a low amperage for a long time until you are sure it is fully charged and balanced. 3.65v per cell X 24 = 87.6v. Disconnect from the charger and let it sit over night then check every cell for voltage loss. Your bms might drain the whole pack a bit, but all the cells should all be within .1v or so of each other.

Put it back on the charger, make sure it is charged and balanced and go for a ride an see what it does.

Once your ride is finished check the cell voltages before charging and then again after charging and balancing.

Write all the numbers down to keep track of what you are doing.

Then repeat the process 3 or 4 or 5 times to get a solid idea of what the pack is doing.

If you have a blue tooth bms or some other way to know all the cell voltages by simply pushing a button, so much the better.

Thanks for your response. I have it on a 12A charge now and I'll check the cells when it's nearer to full so make sure I don't have imbalance that could put some cells over the maximum safe limit, and again after it rests. I like to keep cell voltages in a spreadsheet, conditional formatting makes it easy to pick out highs and lows.

There isn't a BMS on this pack at this time, so this will all be manual and I'm taking a risk with running it as is.

So when you talk about voltage sag, how much do you see? When I rode the bike initially, I saw it drop to 53v (2.2/cell) for a moment under high load. That seems way excessive to me, and I'm hoping that something I can do will help with that. However- big however here- I realized today that the dash instrument (a Curtis gage) is completely unreliable for voltage- it was telling me it was 81v when the pack was actually 68v. I don't know how that goes wrong, but apparently the voltage I was seeing during riding was inaccurate, so maybe this whole question is moot- I will transfer the CycleAnalyst from my dead first bike to this one to get a better reading on voltage.

 

[e-beach]

The only way I know to keep the cells from sagging under load is to apply less load, or get a battery pack that can handle the load better. The less load option is either keep off full throttle until you are up to speed, or get a controller that apply's less load and just go slower when from a full stop or going up a hill. You do have an opportunity to get a BMS with a blue tooth capability to know what your individual cells are doing. And having proper info from the CA will let you know how many amps your cells are capable of which will help you in choosing a suitable BMS if you chose to get one.

 

[john61ct]

Did not read the thread.

Storing is no problem even a decade with LFP cells as long as in cool / cold ambient temps, at low SoC and isolated from vampire loads.

2.75Vpc at rest is way lower than I'd ever let a pack of mine get to, but you might be OK - critical to check isolated per-cell voltage before doing anything else. And yes 2.65V might be OK for say capacity benchmarking, or if a trip requires every Ah of range, but for longevity, fewer imbalance issues, 3.45Vpc is plenty Full, especially if holding that for a bit of Absorb / CV time.

Getting them re-balanced if far off would be next, whether at the top of charge (~3.5V), bottom (~3.0V) or at the midpoint (~3.2V) is a whole 'nother topic, but whichever way, end up paralleling all at 1S for a few days, then isolating again for a few more, see if any drop off more than the rest.

Note LFP does drop voltage after charging for up to 24 hours. 3.33 - 3.35Vpc **at rest** is 100% SoC Full.

Loading at high C-rates (over 1C) will of course drop V a lot, that in itself is not so much an indication of wear, poor SoH.

The real test is a CC load capacity test of each cell, but that's a bit more involved.

Some gentle (under 0.3C) cycling between 3.0V and 3.45V may "wake up" the pack a bit, maybe try going to that 3.65 point a couple times (immediately discharging, never let them sit Full for long!) but there is no "magic reviving" beyond that.

 

[nedfunnell]

Did not read the thread.

Storing is no problem even a decade with LFP cells as long as in cool / cold ambient temps, at low SoC and isolated from vampire loads.

2.75Vpc at rest is way lower than I'd ever let a pack of mine get to, but you might be OK - critical to check isolated per-cell voltage before doing anything else. And yes 2.65V might be OK for say capacity benchmarking, or if a trip requires every Ah of range, but for longevity, fewer imbalance issues, 3.45Vpc is plenty Full, especially if holding that for a bit of Absorb / CV time.

Getting them re-balanced if far off would be next, whether at the top of charge (~3.5V), bottom (~3.0V) or at the midpoint (~3.2V) is a whole 'nother topic, but whichever way, end up paralleling all at 1S for a few days, then isolating again for a few more, see if any drop off more than the rest.

Note LFP does drop voltage after charging for up to 24 hours. 3.33 - 3.35Vpc **at rest** is 100% SoC Full.

Loading at high C-rates (over 1C) will of course drop V a lot, that in itself is not so much an indication of wear, poor SoH.

The real test is a CC load capacity test of each cell, but that's a bit more involved.

Some gentle (under 0.3C) cycling between 3.0V and 3.45V may "wake up" the pack a bit, maybe try going to that 3.65 point a couple times (immediately discharging, never let them sit Full for long!) but there is no "magic reviving" beyond that.

Thanks for your input. I will check the cells as I have time. I don't know if the all-in-1S method is terribly practical with my cells all in a unique geometrical arrangement to fit the bike, but I suppose I can if I need to.

I am thinking of adding the spare 25th cell that came with the bike, if I can find room. It would bring the top charge from 3.65vpc down to 3.5, not to mention adding capacity. It would probably be easier than finding someone to add a new algorithm to the DeltaQ these days. Do these need compression or is that outdated advice?

 

[john61ct]

I don't know of any prismatic LFP that do not require compression.

No need to reprogram the charger if you get the termination V down to 3.5V

If you don't take the time to verify capacity for SoH, matching, balance etc

Then you either have a very unusually superior and adjustable BMS (please link), or you're rolling the dice on murdering your pack and replacing it sooner rather than later

 

[amberwolf]

in general the calb (and similar) cells are likely to have significant voltage sag even when new.

if this doesn't affect the performance of the bike, then don't worry about it.

but if it does, then you'll need to either replace the pack with something that has a better actual c-rate (current-delivery capability), less voltage sag under the specific load your bike puts on it, or add more parallel cells of the same type (chemistry) you already have, sufficient to handle the load your bike puts on it.


when reading c-rates (current capability, amps) supplied by a vendor, you may want to cut them in half, at least. it's pretty common for battery and cell suppliers to...er..."exaggerate" just a bit.



all the prismatic or pouch cells i know of require compression to operate correctly. each manufacturer has their own specification, so use whatever calb says for yours.

 

[999zip999]

Make it ez and write it down

1. 2.85v
2. ,3.5v
3. 3.2v
And so on
Charger voltage
Battery Voltage
This is your State of charge.