ThunderSky Packs giving out

After 5+ years my 2 24v 20ah ThunderSky packs act like their pooped out. On a full balanced charge they are having major voltage drop at only around 12 to 14a draw. I had a full set of 16 new cells that I was going to use to build a second set of packs. So instead of building a second set I used the cells to replace the tired one in my original packs. Something I noticed while changing the cells though makes me wonder if the old cells are really as bad as I think they are. These cells all have screw on connectors making cell replacement easy. But I noticed all the washers, screws and wire connections were corroded. Not really badly but maybe enough so they were taking in the charge but when asked to deliver high amp discharge they couldn't do it.

Before changing the cells I did a test by running the motor with the bike on a stand and applying the rear brake (I have a 750w Bafang mid-drive). Each pack has a CelLog8 on it so I watched them as I did this. At 1000w on the CA meter the all the cell bars on both CelLog8s dropped to zero. As soon as I let off the brake they popped bake to 3.2v. I haven't tried this again since I changed out the cells and cleaned all the connections though. It's a big risk for the controller blowing. But could it just be contaminated connections causing the cells to act defective?

I would like to do a full charge on each cell and individually load test it. I am thinking I need to do a high discharge load test on them. So what could I use for the load? I was looking at this at Harbor Freight. It's for 6v/12v but I think that's only if your interested in using the built in volt meter. I would think that a load is a load and it should work. I can connect my own digital meter to check the voltage drop while applying the load from the Harbor Freight tester. Right?

http://www.harborfreight.com/100-amp-612v-battery-load-tester-61747.html

Bob

Corroded connections can absolutely do what you're seeing - they'll be a high resistance connection, and as you pull more amps, the voltage drop will get worse.

Before doing anything else, it's certainly worth testing them after cleaning them up. My personal approach would probably look a lot like "Clean them up, put them back in the bike, go up the steepest hill I can find, and see what happens." The load tester you link is designed to pull 100A (probably at 12v), so it will pull proportionally more or less with different cells. It's probably significantly more amps than you want to pull out of one of your cells - if it's 100A at 12v, it'll pull ~200A at 24v, which is a 10C discharge rate. If you've got high rate cells, that might be OK, but I wouldn't try it for long (actually, I wouldn't do that at all to a cell of mine).

You're correct in that you can use a digital voltmeter to see the voltage drop through the load.

But, really, clean it up, put it back together, and load it up on the bike to see what it does.

Syonyk said:

It's probably significantly more amps than you want to pull out of one of your cells - if it's 100A at 12v, it'll pull ~200A at 24v, which is a 10C discharge rate. If you've got high rate cells, that might be OK, but I wouldn't try it for long (actually, I wouldn't do that at all to a cell of mine).

Click to expand...

FWIW, these are often called "Thundersag" cells for the reason taht they are really at best 1C cells (regardless of the manufacturer/vendor claims), because of the amount of voltage sag they have under higher loads.

I have 60Ah TS cells and have seen that, though it's been so long since I did those tests for that purpose I don't rmemeber exactly what the results were. (should be in one of my build threads, probably CB2, though I never have ended up using them on a bike as I didn't have enough for the pack voltage I wanted, only 10s).

Syonyk, When I posted the load tester I was assuming 2 things; 1, it was load adjustable from "0" to 100a but it's not and 2, the actual load applied to the cell would be proportional to the voltage of the cell. If this were true then the load on a single cell would be about 27a. High but not a killer. And for testing it would only be needed for a short time.

But I agree I need to remount them back on the bike and do a hill climb test. It's just that it's a PITA to reassemble the cells into a pack again when I don't know for sure if their all good or not. Swapping out the new cells I just installed last night for the old ones again will take 2 hours of rewiring. And if they fail it's 2 more hours to go back again. That's why I was looking for an easy cheap way to bench test them.

amberwolf, Yeah I've heard the ThunderSag stories many times. But (knock on wood) I've not had major problems. That said I wouldn't buy them again. But I already have a full set of 16 new replacements. Even if we considered them as 1c instead of the rated 2c they are suddenly pooping out at little over .5c. Granted I don't normally put them under high loads. But I have pulled 30+a (1.5c) on hill climbs. That's why I was thinking it might be a combination of poor cells, old cells and contaminated connections.

What makes things more confusing is this is a new bike with a new motor but my old cells. It's easy for me to grab at the air just to blame anything.

Bob

dumbass said:

Syonyk, When I posted the load tester I was assuming 2 things; 1, it was load adjustable from "0" to 100a but it's not and 2, the actual load applied to the cell would be proportional to the voltage of the cell. If this were true then the load on a single cell would be about 27a. High but not a killer. And for testing it would only be needed for a short time.

Click to expand...

Ah. It's Harbor Freight. If it load tests twice before blowing up, you're getting an extended lifespan out of it.