Charging/balancing options for A123

[CGameProgrammer]

I'm wrestling with the question of what BMS/charger system to use for an A123 pack I'll be building. I'd like information on the pros and cons of various options. The ones I'm aware of:

DeWalt BMS and charger
Pros: The BMS is essentially free and the charger is roughly $50 on ebay, sometimes less. No assembly required, simple to use, BMS is fairly compact (the wires going to it need some space though).

Cons:
The BMS is impractical for discharging but can refuse to charge the batteries if they get too out of sync. Requires series depth to be a multiple of 10. The BMS is destroyed if you plug it into a live (plugged-in) charger before plugging it into a battery pack, or the reverse - unplugging it from the battery pack before unplugging it from the charger. So you basically keep the BMS connected to the battery pack at all times just to be safe, even though you're only using it for charging.

Notes:
I would assume this method is safe if a multimeter and single-cell charger are also used in the event the pack is too out-of-balance to charge. Use the multimeter to find the problem cells and use the single-cell charger very briefly to bring them up to the same level as the rest. Then continue charging as normal. Still kind of a pain though, but I assume this situation only occurs if the cells are depleted beyond 80% or so. Thoughts?

DIY BMS (GGoodrum/Fechter)
Pros: Compact (...?), full-featured. Works with any series depth. Best tech support available.

Cons: DIY, several hundred components to solder in, actually not that cheap ($130 for 24s not counting shipping, and over $200 for 30s since you need to buy two PCBs), charger sold separately, you will probably need gary/fechter's help. At least I sure would.

Swarms of single-cell chargers
Pros: No BMS needed whatsoever (for charging) resulting in the smallest battery pack possible. Nothing can go wrong really. Works with any series depth, obviously.

Cons: Somewhat high initial expense. ($13 per charger so $312 for 24s.) Will need daisy-chained power strips and will be very bulky. Reportedly the VoltPhreaks charger in the above link cannot be placed adjacent to each other when charging cells in parallel because they'll overheat, making dozens of them even more bulky. Totally nonportable except individually (hope you're patient). Obviously no discharge balancing. The VoltPhreaks charger only charges at 2 amps so worst-case it takes one hour to charge for every cell you have in parallel (six hours for 6p) assuming you have one single-cell charger for every cell in series and charge them all at once.

Notes:
No idea what the power requirements are here. Is a single typical 110/120V outlet good enough or are multiple outlets probably needed for 24 7-amp single-cell chargers running at once? Are there any other disadvantages I didn't think of?

BatterySpace "PCM"
Pros: Seems to do everything needed for both charge and discharge balancing; ready-to-run.

Cons: Expensive and bulky, especially if combining two of them for a non-standard cell count (such as 28s; the 60-amp 12s and 16s combined are $350 and twice as big as just one). Charger sold separately. Possibility of breaking?

iCharger 1010B charger/balancer
Pros: Ready-to-run, fully balances up to 10s, capable of quick-charging, extremely configurable, fairly compact, charger and balancer are a single unit.

Cons: Really expensive ($190 at the above vendor for each 10s subpack), no discharge protection, seemingly limited information on it on this site (though tons on RC forums).

 

[MitchJi]

Hi,

Swarms of single-cell chargers
Pros: No BMS needed whatsoever (for charging) resulting in the smallest battery pack possible. Nothing can go wrong really. Works with any series depth, obviously.

Cons: Somewhat high initial expense. ($13 per charger so $312 for 24s.) Will need daisy-chained power strips and will be bulky. Probably nonportable except individually (hope you're patient). Obviously no discharge balancing.

Notes:
No idea what the power requirements are here. Is a single typical 110/120V outlet good enough or are multiple outlets probably needed for 24 7-amp single-cell chargers running at once? Are there any other disadvantages I didn't think of? I'm leaning in this direction.

The voltsphreaks chargers are 2-amp chargers.

Single cell chargers for LiFePO4 batteries. These are constant current constant voltage (CCCV) chargers for 3.2V LiFe batteries. Max charging current of 2 amps.

I think 25 will be fine:

AC input: 100-240V/50~60Hz 0.35A MAX

Some of the RC Balancing chargers will charge 6s to 10s (ypedal has a 10s that cost $100). They can be combined to charge larger packs. I don't know what that entails. Also Gary posted that they balance by draining the high cells, not by reducing charge to high cells.

The price, quality and charge rates sound good.

 

[CGameProgrammer]

I meant 7-watt not 7-amp. 3.65 * 2 is around 7.

Those RC packs are another option I forgot; what's involved with that? I mean, are there any complications, or is it as simple as 10s balancer + charger = correctly balanced and charged 10s pack?

 

[MitchJi]

Hi,

I meant 7-watt not 7-amp. 3.65 * 2 is around 7.

Those RC packs are another option I forgot; what's involved with that? I mean, are there any complications, or is it as simple as 10s balancer + charger = correctly balanced and charged 10s pack?

I think its pretty simple but one of the reasons I suggested it is I would like to know. One advantage over the Voltsphreaks (aside from more compact) is faster charging. Probably 10 amps.

http://endless-sphere.com/forums/viewtopic.php?f=6&t=7348&p=110762&hilit=+Charger#p110760

My pride and Joy chargers.. these babies will charge it all.. 1 to 10 cell LiMn/Co -- Nimh -- nimh -- sla -- LiFe.. up to 10 amps !! expensive.. but worth every penny imo.

edit to add: These are Thunder Power 1010C chargers(pink ), they can interface with the TP210 ballancers with a data wire and ballance/discharge/monitor cell level charge..

The blue box is a 12v ( 25 amps max )power supply that runs both chargers.

They are Thunder Power 1010C chargers.. ( the 2nd one i got is a re-labeled unit sold for half the price of the 1st original one i got.. but it's identical in every way i could find even after opening them up )
http://www.endless-sphere.com/forums/viewtopic.php?f=14&t=6921&start=0

http://www.amainhobbies.com/product_info.php/cPath/2_389_390/products_id/18362

Super high charge rates are possible even with the largest 6S packs, with up to 10A @ 22.2V (6S)!

http://www.rcgroups.com/forums/showthread.php?t=737623

New Product - New High End Charger - MegaPower 960SR (12 cells A123 version )

Luke highly recommended this one:
http://www.hobbycity.com/hobbycity/...me=iCharger_106B-plus_250W_6s_Balance/Charger

Methods suggested a Chargery AC Adaptor for RC Chargers.

 

[GGoodrum]

What you really need more than anything is low-voltage protection at the cell level. The BMS options mentioned, including the "open" one that Richard and I have been doing, include this function, as well as some form of balanced charging control, but there are lots of tradeoffs, as you point out. You can also just do an LVC-only board, which is what I've done on my latest 16s5p a123 pack. If you start out with healthy a123 cells, and have parallel block/cell-level over-voltage protection, the cells stay pretty well balanced. What I do is use a bulk charger or a suitable current controlled supply, to charge the pack. About every 5th cycle, or so, I use a bunch of the VoltPhreaks 2A chargers to "top off" the cells, to make sure at least once in awhile, each block of cells is getting completely full. I only have one pack, right now, that I use one of the full BMS boards on, and that is a 16s6p a123 pack that is made up from a hodgepodge of healthy and "stressed" cells, so that each block ends up having different voltages and capacities. For this one, the full BMS makes good sense, as the BMS lets it each cell reach its own 100% level, at its own pace, just like individual cell chargers, but lets you charge at a higher rate, using a bulk charger/supply.

One other option you might consider is using a bunch of the 3.7V DC-DC convertors that Bill Zelman is offering in the "For Sale" section. These are each capable of up to 75W, so you could charge up to 20A per channel, but that high is not all that practical. You drive the front of these with pretty much any 48V supply, which are quite plentiful and cheap on ebay. If you had say, a 15A 48V supply, you could charge 16 channels at 9-10As, which is pretty good. Anyway, these modules are about $10 each, and you can get a suitable 48V supply for about $50.

-- Gary

 

[liveforphysics]

That one you linked is great for 6s charging, but for your application, I would go with this one for longer string charging.

http://www.hobbycity.com/hobbycity/store/uh_viewItem.asp?idProduct=6609&Product_Name=iCharger_1010B+_300W_10s_Balance/Charger

That charger absolutely kicks ass. I own 4 of them, and I love them all.

Charges at 10amps, great balance function, super simple and reliable to use.

 

[CGameProgrammer]

That one you linked is great for 6s charging, but for your application, I would go with this one for longer string charging.

http://www.hobbycity.com/hobbycity/stor ... ce/Charger

That charger absolutely kicks ass. I own 4 of them, and I love them all.

Charges at 10amps, great balance function, super simple and reliable to use.

Yeah I saw that and was wondering about it. What power supply do you use with it? And where do you get whatever connector is needed to plug into the balance board?

 

[liveforphysics]

Balancer leads are a $1.49.

Just cut them up and solder to them as you need.

http://www.hobbycity.com/hobbycity/store/uh_viewItem.asp?idProduct=6909&Product_Name=JST_Battery_-_Polyquest_charger_6S


I personally have some Telcom 12v power supplies that I use. They are kinda like computer power supplies, only for really big computers. Something like 12v 72amps each if I remember right. Absurdly expensive stuff to buy outright, but free if you pull them out of equipment in the PC recycle bin at work.

You can find slick regulated 12v 26amp and 12v 30amp power supplies everywhere on fleabay for cheap.

 

[fechter]

Like Gary says, if you use a swarm of single cell chargers, you should probably use a LVC only board to protect the cells during discharge.

Electric Motorsport has been using chargers that are essentially a swarm of single cell chargers all packaged into one big box. Looks nice and seems to work OK. Downside is a massive plug between the charger and battery.

 

[liveforphysics]

I think the 1 charger per cell option is pretty neat. If they made 40amp single cell chargers that were tiny, I would buy them up for my own project.

To do 24s single cell chargers, you only need a 26 pin connector. I think I happen to have some quality compact 26pin connectors laying around on my project shelf somewhere. If you want, I will mail them to you for free if you need them.

 

[CGameProgrammer]

No thanks. I've decided to go with the C1010B charger/balancer. I won't need to keep a BMS on the pack, the charger is allegedly very good, and it has a fast-charge capability (bulk charge til the first cell reaches 3.6V). The cut-off voltage is adjustable too. I'll probably make a new thread on how it seems to perform with a 10s6p pack. Of course this means no discharge protection but I do have a CycleAnalyst and don't run the pack down to empty anyway.

I eventually realized the single-cell chargers would be ridiculously bulky for 30 of them, especially since reportedly they shouldn't be placed next to each other on a power strip or they'll overheat when charging multiple cells in parallel. Also they're 2-amp chargers which means 6p could take over six hours in the worst case, with no fast-charge capability.

 

[CGameProgrammer]

I've updated the original post with corrected information. I'm trying to make this thread a useful compendium of A123 charging options.

 

[GGoodrum]

I bought 16 of the 48V - 3.7V/20A DC-DC convertors from Bill Zelman (Powermed here...) and I'm amazed at how small they are. Each module is 1.5" x 2.25", including the attached heatsink, and is $10, shipped. These will take anything from 36-73V input and has a trimmable output, using a single resistor. They take 95W in and deliver up to 75W at 3.7V, or a max of 20A per module. The beauty of the variable input is that these can be used with an inexpensive ebay 48V supply, which seem to be pretty plentiful, last time I checked, or you can use them "in series" with an existing 48V SLA/LiFePO4 bulk charger. You could simply bulk charge when it is more convenient to do so, and when you want to ensure all the cells get a full charge, all you do is plug the charger into a board full of these, which would be wired to a multipin plug that connects into a matching plug on an LVC board.

I plan on mounting these on a 2-1/5" x 12" piece of aluminum (8 on each side...) this weekend and wiring them up. Eventually, I'm thinking of adding a small circuit to each channel with a red/green LED that lits up red initially and then turns green when the cell is full.

The AMP 4.2mm PE series 18-pin connectors I'm using have about a 9-10A limit on the pins, so what I need is about a 750W supply, factoring in the efficienies, so a 48V/15A supply should work well. I have an HP 0-60V 0-15A regulated supply and I have a Zivan NG1 that maxes out at 15-18A that I plan on using to test the modules.

One other point I wanted to make about RC chargers is that most, if not all, 10s chargers I've seen are usually rated "up to 10A", but usually can only manage that for 5s packs, or less. For a 10s pack the max current is usually down around 4-5A. Also, with RC chargers, you need a good 12V AC-DC supply, one that can supply 13-15A.

-- Gary

 

[erth64net]

My goal is quick and straightforward charging, with a minimum of parts/equipment, while away from home & an inexpensive balance charging option once home.

I've got a an 8-cell A123 pack, and have started charging it via an SLA's "24 volt" charger (4C24040) that's been adjusted to peak at 29.6V. Based on the basic spec sheet, and my own measurements, once it "senses" a full charge, the trickle current drops to ~3mA...so I'd think I'm safe in this regard (any risks of overcharging at a continuous ~3mA?).

Though, I'd love to also have the ability to integrate this charger with a balancer...inexpensively. Does anyone see why BMSBattery's 8s BMS wouldn't work?

It seems, for quick-charges, carrying around just the charger would be sufficient - until arriving home, and then instead of a charger<->battery connection, I'd connect the BMS between the pack's balance connectors, and the charger. In-effect, the BMS would only be connected while balance-charging, and while power is supplied by the "SLA" charger.

Does this path sound sensible?

 

[CGameProgrammer]

The problem with using the charger without a balancer is that it only knows the total pack voltage. So if it essentially stops at 28.8V, that doesn't mean every cell is 3.6V; you can have seven cells overcharged at 3.8V with the remaining cell at 2.2V, for example, or seven cells at 3.5V with the last one overcharged at 4.3V, etc.

 

[erth64net]

The problem with using the charger without a balancer is that it only knows the total pack voltage. So if it essentially stops at 28.8V, that doesn't mean every cell is 3.6V; you can have seven cells overcharged at 3.8V with the remaining cell at 2.2V, for example, or seven cells at 3.5V with the last one overcharged at 4.3V, etc.

All valid points...if you have an unbalanced pack, though from all the info I've gathered, once an M1 pack has been balanced, the likelihood it'll "unbalance" in a single day or two's time is pretty slim. Further, I understand the 28.8V issue with SLA chargers, which is why (as I indicated above ) it has been adjusted to consider 29.6 volts the peak, and not 28.8. As for the balancer...well, I guess at this point its pretty obvious you didn't read my posting. :?