Amberwolf's first RC LiPo...experimental repair thread

amberwolf

Administrator
Staff member
Joined
Aug 17, 2009
Messages
40,859
Location
Phoenix, AZ, USA, Earth, Sol, Local Bubble, Orion
Mdd0127 donated these to me, after having some terrible luck in shipping and whatnot in buying them originally some time back.
View attachment 11

One six-cell pack has obvious physical damage
DSC05507.JPG

DSC05508.JPG
and a dead cell (which is NOT one of the damaged ones!), and the other has just a dead cell. The two 5-cell packs don't have anything instantly apparent, but may have problems later.

The end result idea is to make a small pack that can be used for 80V 5Ah or 40V 10Ah, for various smaller/lighter bike experiments.


Tonite there was only time to measure cells, take pics, and make this thread. Later will come pack surgery to remove the dead cells, and then cycling them to determine actual capacity, etc.

Using this little cell meter donated to me by Methods a long time back,
DSC05501.JPG
I did a quick check of each pack via the balance connectors, which will be numbered in the order pictured, from now on:

Note that the Maxpro meter doesn't read properly sometimes, so while it's good for a quick check I don't really trust it. For instance, it shows only three cells on one pack, but that is because the fourth cell is dead, so it ignores everything above that, and also shows total pack voltage for only those three, not the real total. On another pack the total voltage shows way too high for the cells taht are present.

1:
DSC05503.JPG

2:
DSC05504.JPG

3:
DSC05505.JPG

4:
DSC05506.JPG


So for real cell-level measurements I used this more trustworthy meter
DSC05509.JPG
that Methods sent along with some of those LVC/HVC boards
DSC05510.JPG
which will be used to alert me while using the packs in testing on the road.

Voltages are listed from most positive to most negative:

1: 19.05V
0.01
3.83
3.80
3.82
3.79
3.81

2: 18.91V
3.78
3.80
0.01
3.79
3.80
3.79

3: 19.5V
3.88
3.87
3.86
3.88
3.86

4: 19.28V
3.86
3.86
3.85
3.86
3.87

So the cells that still have voltage are all still "good", not crashed low. Further testing needed to find out how good.


Between uses and testing, the LiPo Storage Facility is used:
DSC05511.JPG
and I make sure to engage the Anti-LiPo-Ignition Safety Device when they're in there, by removing the knob that lets me turn the oven on. ;)
DSC05512.JPG
 
The wrinkles in those cells won't make a lick of difference.

Your 5minute fix would be just soldering a chunk of copper across the 0v cells, turning them into 5s packs, and ignore the 0v cell that for some reason hasn't puffed yet (maybe it never got a formation charge?)
 
I suppose that would work ok; the space and weight of the dead cells is negligible, and I can easily rewire the balance connectors to go around the dead cells.

What are the chances the dead cells could fail in a way that could damage the other cells, while shorted like that? (especially compared to the risk of damaging the other cells during the removal of the dead ones, and/or damaging the tabs during the move of each one down the line on the PCB for the one with the dead middle cell).
 
liveforphysics said:
Your 5minute fix would be just soldering a chunk of copper across the 0v cells, turning them into 5s packs, and ignore the 0v cell that for some reason hasn't puffed yet (maybe it never got a formation charge?)

That ignores the balance connector. Chargers would see a 0V cell be rather confused or indignant. You need to re-do the balance leads.

I bet the 0V cell has an fused open tab... but then that often shows up as a wacko voltage reading on an adjacent cell.
 
texaspyro said:
That ignores the balance connector. Chargers would see a 0V cell be rather confused or indignant. You need to re-do the balance leads.
amberwolf said:
and I can easily rewire the balance connectors to go around the dead cells.
Cell-to-cell voltages appear normal, so I'm guessing that the tabs are ok on these cells and just the cells themselves are dead--either self-discharged or never formation-charged in the first place for some reason at the factory. I haven't tested to see if there is any current flow if power is applied across the cell, yet. If there is none, I'd assume a blown tab, perhaps, but if there is current flow it shoudln't be the tab.

I can imagine an indignant charger...bellowing at me with smoke and such, most likely. :lol:

What does it take to do a formation charge? (I haven't looked it up yet.) If simple, I could try it here and see what happens to the bad cells, though I think I would do it with them off of the packs in case the results are more spectacular than desired.;)
 
Those 5s Zippys are good to go. I got them from someone here to pull cells out of.

Luke is right about the physically damaged cells working. All but one of the packs I got looked like that and I've been using 6 of them on the test bike. Still makes me nervous though.

I like the idea of pulling the "dead" cells out of the 6s packs and bypassing them to make them 5s. The zippys are fine and have a few test cycles on them for sure.
 
Sorry...I couldn't remember what you'd told me about them. :oops: I forget things so quickly if I don't write them down, make lists, etc. :(

I hope to have time to get back to the packs soon, but I am currently working on getting the powerchair motor working and tested with an ebike controller, to use as a middrive:
http://endless-sphere.com/forums/viewtopic.php?p=504935#p504935
 
I took the two known-good Zippy 5s packs and built them into a 10s to use to test the above motor with, so I don't have to keep dragging it to the Sorenson to use, or unhook CrazyBike2's pack to test this with it.
file.php


Bulk charged them to about 4.15V/cell with teh Sorenson, at 1C, monitoring cell voltages with that Maxpro thing, which seems to work well enough on these packs. Took almost 2Ah to get to that voltage, charged as a pack to 41.5V.

Works fine with little sag on there. I meant to also take CrazyBike2 for a spin on them today, but went with a friend to his house/shop to do some work for him on some LCD monitors, and to see if he could do some drilling/etc., on my motor project; got back late enough that I didn't want to go out and test stuff in the sub-40F cold. :|

So I will probably take them with me to work tomorrow, and test them on the way home when I am not in any hurry to do anything (as I will also have my regular pack on the bike, but if something goes catastrophically wrong I can still take whatever time is needed to get home with the bike afterward, as I'm doing nothing for Christmas Eve after work except be at home with the dogs, and have nothing planned for Christmas Day either).
 
Yeah, pretty quiet Christmas eve/day for us too. Have a biker mate in Freo hospital in his last days with pancreatic cancer, so it's not a great end to the not-so-great year.
 
:( That's not good.


I had a bit of time while waiting for beans to boil for chili and clothes to finish cycling in the washer, while dogs were out playing in the cold tonight, so I dug out the big soldering iron, the soldersucker, and the two packs with bad cells....
DSC05567.JPG

The first was easy; all I had to do was take off the most-positive cell that was at 0V (not puffed, though), and move the pack positive output over to cell 5's + tab. The dead cell had a giant solder-splash on it from manufacturing,
DSC05568.JPG

but it doesnt' seem to have affected anything other than the surface finish of the cell pouch; it was trapped between that cell and the next one so couldn't come loose to short things out.

The second...
View attachment 8
I had to take out cell 4 (if you count starting from negative) first, then transplant cell 6 down to where 4 used to be.
DSC05570.JPG
As soon as I undid the heatshrink and end tape wrap,
DSC05571.JPG
the original cell 4 puffed way up.
DSC05574.JPG
After that, just resolder the balance wires taht came off when I was remvoing the cells, and move the pack positive to the top of cell 5. Below is the stage after removing teh two cells.
DSC05575.JPG


Interestingly, these two Turnigy packs which are externally the same and AFAIK were in the same shipment from China to the USA orignally, are pretty differently constructed inside. The first pack was just cells with double-sided tape holding them together, and presumably originally also a white plastic "wrap" around them, and then the shrinkwrap.

The second pack's cells have a liner plastic around them longitudinally,
DSC05573.JPG
which seems slightly self-adherent to the cells (much more than saran wrap, but much less than tape),
View attachment 1
and which has the cell numbers printed on it.
DSC05577.JPG
The tape hodling the cells together is actually only holding the wrap together, where in the first pack it holds the cells together.

PCB the tabs solder to is the same on both, as are teh tabs themselves.
DSC05572.JPG

I haven't tried charging either pack yet; I want to re-wrap them first, at least with a plastic sheet to keep tape sticky off them, and then tightly in duct tape (I don't have any heatshrink big enough).


For now, the dead cells are in the bottom of the oven, and the packs are on the racks in teh oven, as before.
 
you can let the gas out with a tiny pin prick, reseal with silicon or just clear tape, seems to work ok, no idea what the gas is.
 
whatever said:
you can let the gas out with a tiny pin prick, reseal with silicon or just clear tape, seems to work ok, no idea what the gas is.
Which I will probably try at some point, just to see if the cells are good enough at all to experiment with for non-bike use. Seeing as they were at 0V, I'm not sure I would trust them for the higher-current charging I would probably use for the bike packs, especially while assembled as part of a larger pack.

As separate cells at least I only lose the one cell and some wires if they fail spectacularly. :lol:



Got a couple 2s packs of the Turnigy 5Ah cells donated by RVD, which have had a parallel-series connection-type implementation failure. ;)

Balance leads are damaged, with one pack's yellow wire being burned completely past the white plastic endcap, so I'll ahve to open that up to repair it. The red has about 3/4" left, and the black about 2". A new connector with plenty of wire was already spliced onto it, but with the internal wire burned away farther down the insulation, it didn't have anything to connect it to the cells. :(


The other fared better, but even so the silicone insulation on the balance leads is lightly damaged from it, and it's likely there is wire damage inside even though it does have a connection.

I forget which pack was which, but main leads show ~7.9V on one and ~7.6V on the other.

Once I fix the balance leads they should make good packs to series with the 2 5s Zippys above and then be able to test CrazyBIke2 on RC LiPo, finally. :)

Camera battery died as I turned it on, and is now charging so no pics yet. :(
 
Jesus... MDD had terrible luck.. i've never seen anything like this.

I'm glad they've gone to a good home, and now you finally got some lipos.. :)
 
The worst part about the problems he had were that all this stuff was to show his investors for his project how good this stuff can all be, but so much of it came to him damaged, broken, or made wrong that it either couldn't be used at all or had to be fixed first. Some of it was delayed so long by vendors in even getting around to making it (much less shipping it) that the investors got bored and bailed, leaving him with little to work with. :( You know how some days everything just seems to go wrong? I think he's had at least a couple of years of that, lately. I can sympathize; it's happened to me, too. Things do get better, but not always before it gets so bad that it changes your outlook on everything, sometimes permanently.


Anyway, thanks to Mdd0127 (and an anonymous donation that started a process rolling ending up with a fun meetup and ride
http://endless-sphere.com/forums/viewtopic.php?p=513939#p513939
) I have a few more things to use to experiment with on the RC LiPo. There's a working 12s3p pack, and a meanwell with current limiter board, and a battery medic. The MW will make a good portable charger once I ensure nothing can rattle around inside and break off. :) And probably add a temperature-controlled fan off an old computer case.

Am still looking for more defective LiPo to repair or use for cells, connectors, or other parts to repair other LiPo with; it seems relatively easy to do the physical part so far, though I have not yet done any real cell testing or use on a bike. That might be a lot more work. :)
 
just for safety i moved the dud cells outside into th bbq grill, so they aren't in the oven with the good packs

also separated packs into separate metal boxes thugh not sealed.
 
charged 10s pack as 2 5s w venom charger ok, 2ah used in motor tests.
DSC05726.JPG


got venom pc usb s/w monitor fixed, was just bad cable. doesn't tell you much but could ve useful.
 
i found the venom charger can work with logview if you set lv to bantam bc-8 since it doesn't list the venom series (which seem to be a rebranded bc-8).

logview seems to ahve a few little bugs displaying things, and is only about half in english even when set to english, but it's usable enough, and way way better than the stuff that came with the venom. :)


i charged up those 2s packs. the one with nearly-intact balance leads (just red broken inside it's insulation, burned away in two places inside) took 5.3ah to recharge it, according to logview, at 1a max charge current, starting from 3.74v on each cell, ending at 4.20v on each cell. the other one with vaporized balance leads only took 3.3ah to do the same thing.

was going to do a discharge test but i am too sleepy/tired to be sure i'd stay awake for it, and don't want to fall asleep in case it over discharges since i've never tested taht part of the venom.


i also tested the 12s 3p pack on crazybike2, but accidentally cleared the ca before writing the ride data down.
 
was installing the mini limiter board on the meanwell but i apparently killed the meanwell.
http://www.endless-sphere.com/forums/viewtopic.php?f=2&t=36100
 
That sucks wolf.. i never got around to installing my meanwell board, so i just shaved the shunt and called it a day, ghetto style.. I hope you can revive it. S-350's are rare birds now..
 
this one is an sp-350-48; cant' remember what the differnece is right now. i'm sure i can fix it, once i figure out what is wrong with it. it doesn't make sense to me that it just died like that, but it could've been esd damage to a part, or it could still be a bad solder joint, or a number of other things.

i don't have to have it to charge with becuase i have the sorensons, but it would have (hopefully still will) made a great portable charger to keep on the bike, after securing down all the heavy components so they don't break off with vibration.


there's an s-350-48 for $35 shipped in the used sale section right now if you need one. if i didn't already have this one and my sorensons, i'd be sorely tempted.
 
I think I have a couple more of them so I can bring you another one next time I come down and we can compare readings, trouble shoot that one. I won't have no crazy ladies waiting on me this time either so that will be nice.

I found a guitar amp and bass amp I want to pick up down there but really need to sell my 5kw inverter or something big like that to make it happen. I thought I had my downhill bike sold but ended up giving it to a friend instead because he basically lives on his bike, broke his frame and can't afford a new one right now. I'm going to get the inverter posted up and promoted on the Phoenix CL tomorrow so hopefully it will sell soon and I can come visit again.
 
amberwolf said:
there's an s-350-48 for $35 shipped in the used sale section right now if you need one. if i didn't already have this one and my sorensons, i'd be sorely tempted.

it's a clone with 1 shunt.. :|
( but thanks anyway )
 
oh. i guess i should look at mine to see if it has only one shunt; would that mean it is a clone? i couldn't find definitive stuff about that in the various mw threads earlier, even after more than three hours of looking around.

regarding the model designations, i found:
http://www.meanwell.com/webnet/search/seriessearch.html#open
seems to show that sp series has pfc while s series is not listed, but probably is same without it. i found a spec sheet for the s-340 series that doesn't mention pfc
http://www.soaring-cn.com/UploadImages/Specification-350W%20Series.pdf
sp series doesn't show a 350, though, just a 320 then a 480, so i dunno.
http://www.meanwell.com/search/sp-320/default.htm
well, mine says it's from 2008 on the pcb, so they might not make that line anymore. assuming it's not a clone from somewhere else.


mdd0127 said:
I think I have a couple more of them so I can bring you another one next time I come down and we can compare readings, trouble shoot that one.
comparing to a working one might help, if i haven't figured it out before then. i gotta get teh scope out and see if i have oscilation anywhere or not, and if not, why, cuz it's got basic power coming in. i'm not really enthusiastic about switching supplies just cuz i tend to get lost in the circuitry maze. in principle they're easy but often i get frustrated enough with troublehsooting that i don't get htem fixed. :( this one i'm determined, though.


I found a guitar amp and bass amp I want to pick up down there but really need to sell my 5kw inverter or something big like that to make it happen. I thought I had my downhill bike sold but ended up giving it to a friend instead because he basically lives on his bike, broke his frame and can't afford a new one right now.
i have a lot of bike frames depending on the kind he needs, including a duplicate of the one that's on your ebike. most of the frames i have i will never use as frames, just as pieces for other bikes.

or maybe we could fix his original frame?
 
There are a dozen meanwell clones out there, so the water gets muddy fast.

I purchased a genuine 48v/350w one from sureleectronics.. had 3 shunts.
Most clones have one shunt.

To further add to the confusion, the new dreaded NES model contains only one shunt, and filing it down to change it's resistance seems to have not been experimented with yet.
 
Back
Top