Help needed with misbehaving custom battery pack

[TheDaver]

So a couple years back I bought the magicshine mj-808, and in addition to the crappy/dangerous battery they supplied, I built up one myself based on the turboferret 4x18650 holder (wired up as 2s2p) with TrustFire 2400mah individually protected cells. This worked great ... until about a month ago.

Recently the red light has been coming on way too quick - sometimes less than an hour into a ride (supposed to have about 3 hours of runtime with these cells). I took it home, plugged it in and pointed a fan at the lighthead, and was able to get another 100 minutes of runtime off of the 1/2 discharged battery. Weird.

So today, I charged it up, and ran it full while measuring the voltage of the pack every minute - and found a very strange occurrence. It did the normal dischange profile thing until about 100 minutes in - 7.2 volts or so, at which point the voltage started increasing. It increased steadily for another 26 minutes until it reached 7.34v. It was at this point that I noticed the red light was on (odd, because it's still putting out 7.3ish volts).

I assumed the driver knew something I didn't (low current or something?), so I measured each cell while it was still under load and found:
Cell 1: 3.57v
Cell 2: 3.56v
Cell 3: 3.81v
Cell 4: 3.82v

Where 1-2 and 3-4 are in series.

It really concerns me that cells 1-2 are so far off from cells 3-4. Is this a dangerous situation? (charge from 3-4 leaking into 1-2?) I did notice that shifting the cells under load did cause the pack voltage to change by 0.03v to 0.05v. Are these batteries toast now?

What would be a likely cause of such a situation? A poor connection to the cells in the battery pack? Overcharging from the cheap dealextreme charger?

I'm not really sure what to do at this stage. I was thinking about buying some Noalox to try to improve the connection to the battery tabs and eliminate oxidation (if that's an issue), and get a new set of higher quality 18650's - but I'm worried this could just happen again. Any suggestions would be grand.

 

[xul]

until about 100 minutes in - 7.2 volts or so, at which point the voltage started increasing.

Battery getting hotter?

 

[Battery Guy]

Battery getting hotter?

Could be, but the discharge rate should be too low (<1C) to cause significant heating. It could also be an intermittent connection somewhere in the pack.

With respect to the cells being out of balance, it could be a dangerous situation if your protection circuits in the individual cells don't kick in. You either have a weak cell in your first series string, or the first string is not being fully charged. I would take the cells out and charge them individually to rebalance the pack. If you see the same problem, then either cell 1 or cell 2 is dying.

Cheers,
BG

 

[TheDaver]

Battery getting hotter?

I didn't feel the battery get hot at all, which I presume is a good thing.

 

[TheDaver]

Could be, but the discharge rate should be too low (<1C) to cause significant heating. It could also be an intermittent connection somewhere in the pack.

With respect to the cells being out of balance, it could be a dangerous situation if your protection circuits in the individual cells don't kick in. You either have a weak cell in your first series string, or the first string is not being fully charged. I would take the cells out and charge them individually to rebalance the pack. If you see the same problem, then either cell 1 or cell 2 is dying.

Cheers,
BG

Ok, this is more along the lines of what I thought the issue was. I'll redo all the solders on the cell and throw some new batteries in - and hope for the best. I did balance all the cells only a couple charges ago, so whatever it is - it's still very much an issue. Perhaps a dead/dying cell after all. Charging/discharging them individually might be an interesting learning exercise (assuming the wiring isn't the issue).

Thanks for the insight.

-Dave

 

[Battery Guy]

Ok, this is more along the lines of what I thought the issue was. I'll redo all the solders on the cell and throw some new batteries in - and hope for the best. I did balance all the cells only a couple charges ago, so whatever it is - it's still very much an issue. Perhaps a dead/dying cell after all. Charging/discharging them individually might be an interesting learning exercise (assuming the wiring isn't the issue).

Thanks for the insight.

-Dave

Be very, very careful soldering to lithium-ion cells. This is tricky business and can easily damage the cell, and I have no idea how it might effect the protection chip.

Personally, I am fortunate to have access to a spot welder, so I never solder to cells. Others on this forum have more experience so I would defer to their expertise in this area.

Cheers,
BG

 

[TheDaver]

Be very, very careful soldering to lithium-ion cells. This is tricky business and can easily damage the cell, and I have no idea how it might effect the protection chip.

Personally, I am fortunate to have access to a spot welder, so I never solder to cells. Others on this forum have more experience so I would defer to their expertise in this area.

Cheers,
BG

Sorry, I was talking about the solder points on the TurboFerret 18650 holder - the cells themselves just slip in and make their connections via friction - my holder looks virtually identical to this: http://www.turboferret.co.uk/Photos/battery93.JPG except I've wired it up in a 2s2p configuration. So lucky for me, I never have to solder to the cells either.

I'd assume the resistance via those tabs is pretty low, and should be roughly the same for all the cells (or so I hope). Do you think a conductive (and antioxidant) grease would be a good idea there?