My A123 cells are dying... help?
 

I have an outrunner setup on a HB Lightning buggy. I'm running direct drive to the rear wheels only (no center diff) using a Turnigy 50-55 outrunner, kv=600. Running 6S A123 packs that I build and using a Turnigy 100A ESC.

So I balance the pack and then put it on the buggy. Within about 3 minutes at low throttle with some full throttle blips, the ESC cuts out for low voltage. So I shut it down. The pack comes off usually pretty cold with maybe one or two warm cells. When I check the voltages, it always turns out that only ONE cell is dead. The rest read 3.3 V. My charger won't let me charge that cell now because it has discharged too far.

I would say it's just a bad cell, but this has happened to 4 cells individually! Which means I have only had 4 runs on this thing, because every time I run it, a cell dies. Also, everything else I've read about A123s says that they usually maintain excellent balance.

Any ideas?


Also, just want to throw this out there for others who might be thinking direct-drive with an outrunner is a good idea: On paper, it looks ok, and you can find motors that seem to fit the bill if you use enough cells. My setup, for example, *should* be very hot if I could fix the pack issue. On the track, it would probably be pretty good.

But not when bashing. Here's why: consider that these motors are sensorless and so the ESC cannot do a hard, high-torque start from stall like with a brushed motor. The ESC has to roll the motor over a bit to find out where the poles are before apply full torque. This problem is worse on the outrunner because it has no gearing. So if an inrunner setup can apply full torque at say, 2 mph, then the outrunner won't be able to apply full torque until 7 mph (using a common 46/13 spur/pinion ratio). I can see this effect when I hit full throttle with the wheels off of the ground. Even with no load, the outrunner has a bit of a dead spot before it really revs up.

So no wheelies from a stop, even with 2WD. So I think my next setup will use the RC-Monster motor mount with an inrunner setup.

The ESC is to blame for the low power startups, the turnigy just can't advance the motor fast enough to give good power.

Are they new cells? a123 brand? What is the LVC set to?

Regardless of the ESC, without sensors, the outrunner is always going to have a slower start. Whether or not that is an issue with better ESCs, I do not know.

However, I now have a better job, and so I'll be getting the MMM shortly :mdr:

Yes, they are A123 brand. I'd have to check on the LVC, but that seems irrelevant given how fast they are going out of balance. I can't even get the other cells to discharge below 3.3 V!

I was thinking about the lvc setting too- on 99% of escs there are presets for lipo voltages and pack sizes, but A123 cells have a lower nominal voltage (3.4v vs 3.7v), so the lvc would kick in much too soon.

There should be a date of manufacture stamp on the A123 cells, that'll say when they were built, so you can tell how old they are.

I know they do very well without balancing, but I have found it is actually important with some chargers, as one cell always hits the peak charge voltage and leaves the others behind, so I have to balance the pack after a charge regardless..

An outrunner may spool up slower as compared to an inrunner (more moment of inertia), but this would be no-load. An outrunner has 7 times the commutation cycles per revolution as compared to a 2 pole inrunner, so the ESC can synch it just as fast as an inrunner hooked to a pinion/ spur geardown.


Has there been any consistency with how the cells die? In a certain spot in the pack maybe? Have you tried slow charging or discharging of the packs to break them in? I have been running a123 a long time and only had once cell out of 50 die.

I thought synching had more to do with the magnitude of the back emf reaching a level that is measurable by the ESC. That would make it depend on speed and independent of the number of poles.

I have slow charged/discharged them all twice before running them.

I have not noticed that they die in a particular location in the pack. The last time this happened, the voltages were 3.3, 3.3, 3.3, 3.3, 3.3, 2.6. Everything I can think of that would cause LVC (bad solder joints, too much current, etc.) should not affect the balance.

What about overheating the cells when soldering the tabs? I use a gigantic iron... I think it's 80W, and heat the cell tabs less than 5 secs to tin and then about 5 secs again to solder the wires.

Back EMF depends on magnet strength, KV, stator construction, air gap, and other factors I am probably not aware of. To generalize, an outrunner gives more feedback EMF because the kv is lower than an inrunner. When you increase the commutation cycles by a factor of 7 (though a higher slot/pole count), generally the construction will allow a reduction of KV by 7 at the same time. Basically the net sum is zero if the inrunner has 7x more geardown as compared to the outrunner.

5 seconds is a lot of time on the cells. That is probably the culprit. You shouldn't be applying heat for more than 2 seconds tops. I get my packs built with one second applications of the iron.

Are you using a gun, or an iron? Got a big tip on there?

Wow man, 5sec is a LONG time to have anything hot near the A123 cells lol. You should be using a really hot iron, and literally just touching it on there to solder ie <2sec. They get damaged pretty easily by heat from soldering.

As for the A123s keeping in balance, they do to a certain extent. There's no real point in my experience to slow charge/discharge them. All my packs remain in balance fairly well, unless you discharge lower than like 2.2v/cell, then they go out of balance real quick.

Normally, if you just hammer them with high amp charge they normalize and come back into balance near the end of the charge, but only if they're not really too far out of balance to start off with. As I abuse my packs in my vehicles, I've got balance taps on all of them so I can balance charge when needed.

For your particular situation, I'd suggest charging each cell up so you're sure they're all individually charged to the same voltage (this is if you don't have balance taps obviously) and then discharge them all together, either on the car, or on a charger. Set the LVC for 2v/cell just to see what happens. If you have a balance adapter you can monitor the cell voltages, and you'll see straight away when the suspect cells begin to dump voltage, which they'll do REALLY quick at somewhere near the bottom of the discharge.

If they do that, then once the pack is discharged, when you charge each cell up individually, have a look at how much mAh is put back into each. The broken cells (if they are broken) will take far less that 2300mAh back.

Hope that helps
-Sid

I have to disaggree with some things here. Ummm..........it really depends on your setup entirely.

<object width="425" height="344"><param name="movie" value="http://www.youtube.com/v/dFDrYXLkcGg&hl=en&fs=1"></param><param name="allowFullScreen" value="true"></param><param name="allowscriptaccess" value="always"></param><embed src="http://www.youtube.com/v/dFDrYXLkcGg&hl=en&fs=1" type="application/x-shockwave-flash" allowscriptaccess="always" allowfullscreen="true" width="425" height="344"></embed></object>

http://i160.photobucket.com/albums/t...m/DSC06666.jpg

There is an AXI 4130 outrunner under there. :yes:

Hey, sikeston, I know you are the outrunner guru here, but I guess I missed some of your old threads. I just went back and read some of your threads and it seems like you had the same problem at first before changing the gearing to get the low speed to work better. I am not sure if I can change diff gearing like you did, but yeah, that would be a great way to get the torque I want.

Also, with respect to wheelies, they are of course a LOT easier in a truck. I have yet to see an outrunner setup in a buggy that can wheelie, but I have seen it done with inrunners.

So are you running 8S? I'm thinking of dropping to 4S, but I need a higher kv outrunner to do that. Right now, I'm pretty limited to the Turnigy stuff because the 8 mm motor shaft fits perfectly in a Hot Bodies drive cup. But I would like to use a higher kv motor with 4S, say 800 or so.

With the soldering, I was estimating an upper limit. I am probably in the 2 sec range, but probably not down to 1 sec because I wanted to make sure the flux is burned out. I have the Weller 80 W pencil iron with the big chisel head. Looks like 1/4-3/8" wide. Definitely the correct iron.

So have any of you guys ever ruined an A123 by soldering or is this just speculation?

The Video was shot running 6S Lipo.

On the Hot Bodies Diff cup, that is what I used on the diff pinion with this setup. The other end of the cvd was 6mm, same as the shaft of the motor.

I'm sure you could find some 6mm diff cups that would be compatible if you look around.

The right outrunner in your setup will wheelie, and even make it tough to build diffs strong enough to take the abuse.

Do alot of research before you buy. Before you settle on things. :yes:

I have not found the perfect outrunner for this application, so if you find an outrunner than can do the duty on 4S, I'd be happy to hear about it. Mine is 600 kv, and so by the numbers it *should* be great. But it's not. Plus, 6S A123 packs are quite heavy for a 1/8 buggy. So I need kv = 800 or so for 4S, and motors that size are hard to find.

Another option is to get a Turnigy 50-65 400 kv motor (fits my mount and has the 8 mm shaft) and re-wind it to 800 kv. Stall torque is independent of turns, so even without a re-wind, that motor should accelerate well with a possibly limited top-end.

I would try an new esc with your current setup...MMM? Honestly you need more than 6s1p a123, maybe 2p of those cells, or go to 6s lipo, something decent like flightpower or neuenergy.

Less voltage will only drive the amp draw up, and require a custom motor... More voltage is the way to go, within reason.

Ok, looking back at your problem... Not sure why the cells are dying, where did you get them?

Also that esc is poor, but a MMM and run 5s lipo with your current motor, or go up to 6s lipo and drop down to a 520kv motor. Either way you get to buy some stuff... Some 30c flightpower cells would be nice, like the 3200mah packs for 6s. Not sure what the weight difference would be though.

I just caught the 6s part of the build. You need at least 2p of those cells to keep voltage under load with that honking motor. This will cause bad acceleration, along with the ESC. I bet you are drawing 80a or more on those little guys.

Different batteries, different ESC, get a handle on the rollout, and then it can work. I would suggest lipo over a123, unless you use them for the short charge times.

I'm with Linc on this one. Alot of your problem is your ESC.

The startup routines very from ESC to ESC and they are not created equal.

Quark does pretty good, but I think Castle has them beat even with starting an outrunner under a load like a direct drive setup does. :yes:

I can't do 6S2P! This is a buggy, not MT. Remember that I said it is already too heavy with 6S1P? I'm looking to go down to 4S, and there are plenty of people running 4S A123 with inrunners.

The cells came from brand new Hypersonic 2S2P packs. However, I bought the packs from a hobby shop on ebay.

Thanks for the comments, everyone. I'll probably just get the MMM and see where it goes from there.

If you do try to run 4s a123, you will definately need 2p config. and that will be heavier than the 6s setup you currently run. A123 cells are heavy for their capacity, no way to get around that. The MMM should help the startup, and you are geared (kv-ed?) for around 40mph with your current setup. I would stick with the motor and battery combo you have and try the MMM. The addl programming it offers will help you get the startup right, and if you limit the punch control it will help the current spikes to the a123 cells.

From what I have heard the packs that some ebay seller had were very old, may be why your cells are dying... Plus the load on them is high...

I was waiting for you to tell us where you bought the packs. I'm now 100% certain that you have the same packs that I purchased from Hopmeup.
I bought 6 4S2P packs, and out of those, I had 5 bad cells. There are many others that have had that same thing occur from this vendor.
I was able to get two of the cells "recovered" by pulling the packs apart and charging the individual cells back up using a low ampreage Nimh setting on my Hyperion. However, I have had to be diligent with the pack they are in because it is overly suceptible to over discharge. I had a thread a while back venting about this vendor and the way they treated me, but when you do the math, the price per cell was pretty low. And I learned a LOT about creating packs. I also ended up with a GREAT soldering iron (Hakko), which if I were you, I'd get something similar. You wont regret the investment. The pencil irons just dont cut it. I have applied more than 5 seconds of high heat to a cell in certain instances and have not had an issue. My choice was to either keep heating it to get it to flow, or throw the cell out - so it was a no loose situation. And yes, that was with the Hakko. Sometimes, some vairiables make even a good iron a moot point.

Agree. Never used the Quark but the MMM handled my outrunner very nicely. Couldn't see a need for a smoother startup.

Also like said by JThiessen get a better iron. The hakkos in my opinion are flawless and the best out there for the money. I have had mine for 8 months would moderate use and still have the original tip in perfect shape. Having a good soldering station changes the game drastically and makes soldering a breeze.

!@#$!# Yes, that is who I bought them from. I think they were around $10/cell, which is the only reason I chose them over a Dewalt pack. Well, good to know, and thanks very much for chiming in.

So I guess I need new cells and a MMM.

At the moment, I am completely impressed by the Weller 80 W iron, and I can't see the reason to get a better one. With the Weller, I could easily tin/solder the A123 cells in the 1 sec suggested by people in this thread... it's just that when I learned to solder, I was taught to let it sizzle for a few seconds to ensure that no cold joint forms. In the future, I'll try to solder it as fast as possible. For more precise work, I have a nice regulated Weller station.

Again, thanks for the responses.

I have a weller station and a wahl gun (for nimhs and larger wire) and they both work great. I also have a non adjust able 40w snap on pencil style iron, and a few others... I am sure the cells were bad, and the esc is just not performing as desired. Give the MMM a try and I am sure that you will be satisfied with the performance... No need to get another motor.

I also bought 6 2s2p packs from Hopmeup on ebay. Had one cell vent and pop, after doing a few speed runs in a 6s2p config with a neu 1515/1y. Other than that they have been fine, and a lot better than the other non a123 lifepo4 cells. Have since combined them to be 3s2p packs, to run as 6s2p series pack.

The non-A123 brand cells tend to have half the discharge ratings of proper cells, so using them in a 2p or 3p setup is the only way to avoid cooking them. Shame about Hopmeup, but it sounds like he's flogging rather old packs which are near the end of the lives...

I did that also - those things are huge, aren't they! I've since taken them back to 2s2P packs.

I finally got a Dewalt pack and the MMM the other day. Based on what I've read here, I dropped down to a 5S A123 pack (from 6S) because it doesn't seem like the 6S is necessary.

So the results are that the MMM is of course very nice, very smooth, and very fast. This thing hauls tail on 5S and even on the street it has limited traction until about 15 mph or so. I couldn't get the fan on the MMM to come on (I could only go full throttle for a few seconds where I was running).

But I still cannot wheelie, and that makes me sad. The buggy comes in a 7 lbs, but I think the outrunner is mounted too far forward to be able to wheelie. That's probably a good thing for balance and 4WD traction, but it doesn't help much with RWD traction (my setup is direct drive to the rear diff, no front wheel drive).

Definitely a grip and balance issue. If you want wheelies that badly you could try setting up the suspension for weight transfer to the rear. You've got more than enough power (maybe too much) to flip it so it's down to setup.

I think I'll probably just go with an inrunner at this point. I want to run in dirt, rocks, and snow, and I haven't been able to come up with a good way to seal the outrunner to make sure bad stuff can't get in but cooling air can get in.

I think it could be the building of your packs.

1) When you built the packs do you install balancing taps? Do you balance first several uses?
2) When you disassembled the packs, the tabs can spin.. But DON'T do that. spinning the tabs is bad and will cause the cell to die a early death.

I use 6s1P pack and have no issues in my Emaxx with a 9xl and BK controller.
going to install my MMM and Medusa for the summer soon. :)

Highflier