One Lifepo4 cell that won't hold surface charge

[morph999]

I got a weird problem. One of my Thundersky cells charges up to 3.65v then recedes back to 3.35v. AFter my last run, I checked all my cells and they were all at 3.27v . I've charged all of them to 3.65v and all my other cells are around 3.59v or 3.60v but this one will go up to 3.65v but then quickly recede back down to 3.35v. Strange. Is it damaged ?

I'm going to get some celllog 8s monitors from hobbyking. I might not even run my batteries again until I get those monitors. The most I've used on this pack has been 13 AH and they are 20AH cells so this is odd. I thought damaged cells were supposed to be 2.3v or something like that, not 3.35v.

I'm planning on buying the $15 celllog 8 monitors from hobbyking. Do I still need to buy wires for that or does it come with wires? Thanks. I admit that some kind of cell monitoring is important.

I guess another way of seeing if it's damaged is to count how many AH used and how many AH can go into the pack. Right? I mean if I use 10 AH and I count 5 hours of 2 amp charging then the cell is probably okay ?

 

[liveforphysics]

Seems like we predicted this...

Ask the battery expert Jack Ricard about it, and what to do about it. Maybe some 5C charging on TS cells would help.

 

[Russell]

With LiFePO4 the charger current is pretty much constant up to about 95% of full so if you know the charger current you could get a rough estimate of the cell's capacity by draining it and timing how long it takes to recharge but using a wattmeter during charging would give you the best results. Unfortunately a WU type self-powered wattmeter won't work for a single cell unless you power it externally.

Holding the surface charge may not be terribly important since it represents a miniscule amount of capacity. What is important is the cell voltage when it is drained. What I would do is go for a short ride using a couple Amp-hours then check the suspect cell volts and compare it to the others. If it is close, say within 0.02 of the average, then go for a longer ride and once again check the cell voltage after the ride and compare it to the others. If the cell stays close to the others you shouldn't have much to worry about.

The nice thing about your battery is if you do find the cell is bad it's easy to replace :wink:

-R

 

[morph999]

Seems like we predicted this...

Ask the battery expert Jack Ricard about it, and what to do about it. Maybe some 5C charging on TS cells would help.

What's with all the hate?

 

[olaf-lampe]

We are currently building a 100s TS pack and there are several cells that behave strange in one or the other way. I wouldn't be concerned unless it stays that way after a few cycles. Make sure, you'll drive the pack straight from the charger and don't let them sit for weeks now. That could lead into severe unbalanced cells.

Don't worry
Olaf
<edit> I thought, we were talking about a new pack...

 

[morph999]

With LiFePO4 the charger current is pretty much constant up to about 95% of full so if you know the charger current you could get a rough estimate of the cell's capacity by draining it and timing how long it takes to recharge but using a wattmeter during charging would give you the best results. Unfortunately a WU type self-powered wattmeter won't work for a single cell unless you power it externally.

Holding the surface charge may not be terribly important since it represents a miniscule amount of capacity. What is important is the cell voltage when it is drained. What I would do is go for a short ride using a couple Amp-hours then check the suspect cell volts and compare it to the others. If it is close, say within 0.02 of the average, then go for a longer ride and once again check the cell voltage after the ride and compare it to the others. If the cell stays close to the others you shouldn't have much to worry about.

The nice thing about your battery is if you do find the cell is bad it's easy to replace :wink:

-R

okay thanks. So it would actually show a different voltage after going for a ride? I checked it on my last ride and all the cells were 3.27v . Today, I'll see if I can go for a ride and I'll do what you said. I think I'm also going to tape my multi-meter to the cell so I can see what it says when I discharge. I got a road with no traffic so I'll ride down that one.

 

[morph999]

I was recently sick for 2 weeks so I didn't ride. Then I topped all my cells off before riding just to make sure I charged them all. Then I went for a ride and the headwind was really eating up a lot of battery. I used 6.3 AH to the destination. Then I used about 6.7 AH on the way back. Usually, the ride back only uses 3.5 AH but I must have really hit the throttle hard or something. So 6.3 AH + 6.7 AH = 13 AH. Which is only about 65 % of the pack. I mean come on. If the cells are going to die with only 65 % usage, What good are they. It was an accident. I meant to only use 10 AH but I was only going 10 miles total so I was surprised to look down and see so much AH used.

 

[AussieJester]

What's with all the hate?

ahhh maaaaybe because you yourself were rather vocal and rude regarding the whole need
for a BMS morph.

I didn't say Jack was right about all his conclusions. And GGoodrum is not right about everything either. There is nothing wrong with charging to 4.2v per cell. Even Thundersky says so. They make the batteries. Well....I'm going to leave this forum because I can see everyone has their nose so far up this GGoodrum's guys ass. My word is just dogshit. Have fun spending $250 on A BMS THAT YOU DON'T EVEN NEED. LMAO.

Hope you get it sorted anyway, get a cellog 8 for sure...

KiM

 

[morph999]

The cells would have to sag quite a bit to reach 2.5v. 12 x 2.5v = 30v. I've never seen anywhere near sag like that.

 

[morph999]

Well, if it turns out the cell is destroyed then I'll admit that a BMS is necessary.

 

[Russell]

okay thanks. So it would actually show a different voltage after going for a ride? I checked it on my last ride and all the cells were 3.27v . Today, I'll see if I can go for a ride and I'll do what you said. I think I'm also going to tape my multi-meter to the cell so I can see what it says when I discharge. I got a road with no traffic so I'll ride down that one.

ALL CELLS at 3.27V sounds great.

When I first removed my BMS I initially balanced the cells to the same voltage then I checked the individual cell voltages after a ride and again after charging. The first ride used 5.74AH of the 10Ah capacity and all cells were 3.29-3.30V. On a ride where I used 87% of the battery capacity the cells were a bit further apart with one at 3.20V, 4 at 3.23V, 8 at 3.24V and 2 at 3.25V, but still fine and dandy. After charging the voltages are spread out more but that's because when the cells get over about 3.40V the voltage rises very quickly but represents next to nothing in capacity.

The meter monitoring the one cell isn't a bad idea until you are confident it's hanging in there with the rest. I've gone 27 cycles now without a BMS and I don't worry about it any more. I do use the old BMS to balance the voltages now and then otherwise I simply try to keep my use to 80% DOD which is a good idea even with a BMS.

-R

 

[Toorbough ULL-Zeveigh]

Seems like we predicted this...

like car-stalled-on-track meets runaway train wreq.

 

[dogman dan]

Well, how many of us have gotten a pack with a bms that happend to have a weaker cell in it? But it does seem to me like a failure of a cell to hold surface charge could be an indicator of that cell failing before the others. But it may still have a bit of life left in it. It WILL be a lot easier to fix than unsoldering a ping pack, for sure.

But personally, I think bms or no bms has nothing to do with it. What Morph has wrong in my opinion is running a big ol x5 motor on 20 ah of 2c cells. I think that kind of motor needs 30 ah of low rate cells to last the 1000 cycles. But this is happening much sooner than I expected, so maybe he's just got one cell with a really high IR? I still expected him to get a year or two, 3-4000 miles if he kept to the low depth of discharge. But then again, high rate of discharge is known to kill a pack pretty quick. I got only about 100 cycles out of my nicads doing the same thing (5304 motor), and I only had a 22 amp controller on them.

As for being vocal and rude, Aint that what ES is for to some degree? The V is very polite, and boring.

 

[morph999]

If it turns out that this cell is near dead, I think I might just put my forsen hub motor on my new bike and use that instead of the 5303. The 5303 is really starting to anger me. Using 13 AH over 12 mile ride? Ridiculous. Most setups probably would have used about 6 or 7 AH. I can't help myself from flooring the throttle when I'm using 36v. When I have 48v, I tend to cut back a bit on it so that means I'm using about 43 amps when I'm doing that because my 35 amp controller is really a 43 amp controller.

I might just put my forsen hub and controller on instead and then just sell the 5303. It's getting to be a big headache. The forsen goes almost as fast as the 5303 and probably uses much less power. I've got 4 extra thundersky cells. I just need to prevent any further damage (if that is what happened).

 

[dogman dan]

Or just hang onto the 5303, and use it later when you can afford some headways. Or find somebody with a 9c 2807 that wants a clyte front hub. That person could even be me? Wow, 43 amp controller, no wonder the cell died.

The cell should still have lasted longer though, chances are, the higher discharge rate that day combined with an unknown to you reduced capacity caused an overdischarge that killed it. Cell level voltage monitoring is cruicial, whether by bms or watching a meter.

 

[AussieJester]

Or just hang onto the 5303, and use it later when you can afford some headways. .

Lithium Polymer would also be a good option seeing they stay balanced
very well a permanent BMS isn't essential, might suit Morph better
than playing with high number of headway cells which do require
full-time BMS...? Cost...~50 bucks 22v 20c 5000,ah pack... could
have 44v 15ah pack and/or 66v 10ah pack for 300 bucks, 11pound all up
huge weight saving on the TS setup.

KiM

 

[dogman dan]

That's true. Shoulda just said a higher c rate battery of any kind for that x5. The cheap lifepo4 like pings etc, seem to be pretty stable at 1c, but more than that seems to find the weak cell in the pack pretty fast. His pack would do ok with a 9c and 22 amp controller. Mabye the faster wind one for him though, only 23 mph at 36v on a 9x7. But also easier to add 4 more cells with his setup than doing it to a ping. Then he'd have 27 mph on a 9x7, and 1c discharge rates.

 

[liveforphysics]

LiPo sounds like a great Morph battery option.

Charge it with a lead acid battery charger indoors. I think 1 lead acid charger per 2S group should work out nicely for that 6.5-7v HVC sweet-spot. When the puffing starts, thats how you know they're getting a good charge going on inside there.

 

[AussieJester]

^^^HAhAHA excellent advise Luke should be stickied in the
battery section IMHO

KiM

EDIT: I actually have some spare near new Lipos i can sell you Morph, 45 each
only charged to 5v once as you can see they are in excellent condition!

View attachment Undercharged_lipos.jpg

 

[morph999]

Just took the bike out. The bad cell goes to about 3.05v under full 40 amp load. That's with only about 1 or 2AH used...with 10 AH used...it's probably much worse.

In fact, I'm pretty sure that I'm killing all my cells with each time I take the bike out. It's the motor and controller, not my method of charging and using the batteries. Under full load, it's sagging to about 35.5v . That's 12 cells. That's bad. Right? I remember when I first got the cells, they'd sag to about 36.9v. That's one more volt now that it's sagging. Actually, I have no idea if I'm killing the cells or not because some sag is normal.

It's the 5303. You guys were right. It's just too big and too much of a power hog.

I'm going to sell the 5303 hub motor. I think Nomad85 might want it so I'll offer to him first. It's only got 150 miles on it.

What do you guys think I should do? I don't mind going 20 mph. I think I need about 15 AH of battery no matter what I do because I don't like pedaling. I hurt my leg recently and the 5303 has been a real help because I can go places on it without having to pedal. I can probably do that on the forsen too, though.

 

[morph999]

Just checked all the cells. They are all 3.36v and the bad one is 3.35v while at rest. And the bad cell is 3.05v under load (discharge). The whole pack sags to about 35.5v under full 40 amp load. Under about a 20 amp load, it only sags to about 38v.

So what should I do? All the cells are reading about the same under rest. Maybe there is nothing wrong after all.

 

[AussieJester]

So what should I do? All the cells are reading about the same under rest. Maybe there is nothing wrong after all.

Konions are another option get in touch with DocBass see if he can hook you up with a suitable pack to power the x5. Dogman has pointed out already what will happen to your TS cells (if it hasn't already) The Konions don't require any BMS either downside they are small cells and you need heaps of them for a pack which means alot of connections..

KiM

 

[liveforphysics]

KiM- Quit being a LiPo hog, and ship the man your special packs all ready.

 

[AussieJester]

KiM- Quit being a LiPo hog, and ship the man your special packs all ready.

Your right Luke, OK Morph.....how about 40 bucks each and ill even cut the old bullet
connectors off so you can solder new ones on yourself :lol: If i let them
go any cheaper it would be like me paying you to take them

KiM

 

[morph999]

"special packs" ? hahaha.

There might not be anything wrong with the batteries I have. That's what I'm trying to find out. What I need to know is if how I'm using them currently is sustainable for 500 - 1000 charges. If it's sustainable for 2 or 3 yrs, I'll just keep the setup that I have now.

I need to know if a sag to 2.98v per cell is okay and sustainable.