BMS for 2 6p16s hi current A123 packs in parallel????

[GreatScott!]

Hello, from Canada, my name is Jon.

Been glancing around this site for a while now with speratic posts here and there. Learning alot, pretty good site. Cheers for some great replys.

I have a home made electric motorcycle that I would eventually like to equip with LIPO. (will post pictures once graphics are done) I've asked around and hinted on the forums here before if a large pack, hi current type BMS is possible, have yet to get a straight answer. Hope this hasn't been answered already and is worthy of a new post.

Anyway, I have consulted with a few design engineers of LIPO EV manufacturers and concluded that it would just cost too much for my purposes. 5-12G's. Looking for a DIY solution as my bike and my current SLA packs are DIY specials. The impression I get from most sources is that there really isn't a DIY solution for a pack this large. Designing and building custom hardshell compartments, assembly and wiring, and charging them will not be a problem, I have that figured out; creating a monitoring and safety system while discharging them I need help accomplishing. For instance, will a good BMS just monitor or actually affect the discharge dynamically???????Is a BMS for discharge even necessary???? This stuff is the last hurtle I need to get by before I can build my pack.

The components are as follows:
The Motor:
48v 14,400Watt E-tek
http://www.electricvehiclesusa.com/product_p/mo-et-2401.htm
The controller:
http://www.electricvehiclesusa.com/product_p/co-axe4844.htm
This controller has the plug brake option on it which can limit the current to the motor when activated. It is used for regenerative breaking but I am not using it at the moment.

The batteries are 4X20ah 12 v tysonic batteries connected in series for 48v.

The building block packs I want to build will be 6p A123 cells slid into a tight fitting hard shell casing with six gauge wires popping out the top for quick connection along with smaller gauge for balance taps. This will be considered one cell of 3.3v 13.8Ah, the building block of the system. They will be about 3” wide/deep, 1.25”long and 7”high. I will then butt 16 packs like this tightly together, connected in series for a total pack voltage of 52v and 13.8ah. Charging will be accomplished by one 10s charger for the first 10 packs and a 6s charger for the rest, all with balance taps connected. Eventually I will add another identical pack to this one, connected in parallel for a total of 52v and 28ah.
The controller can put pull out 400amps from the batteries, I’ve ran it at 300amps this summer and will probably bring it down to 260 for next season; not too sure what it pulls under heavy load but I doubt it pulls the full amperage setting for long, Thats more than an average house service panel!!yikes. Either way it can pull a lot of amperage, I assume more than the average 40A E-pedal bike controllers featured on this site. That means big wires for discharge and heavy duty BMS circuitry/wiring to consider.

I imagine the best people to respond to this post will be the BMS guys, any thoughts? Perhaps there is an A123 rep lurking here somewhere who can bestow some inside information. If hi current is a problem in design, the same principles should apply as with lesser amped e pedal assist BMS systems, no? Just need to figure this out.

 

[GreatScott!]

Here is a screen shot of the Alltrax controller software, how cool is this controller, ya can actually hook it up to a pc and adjust different parameters.
Anyway, worth mentioning that an undervoltage parameter can be set which I would believe would cut the controller if the pack voltage dropped any lower than the setting. Same goes for overvoltage. This enough BM for a Lipo setup? Someone tell me it's that easy.

cheers.

 

[Jozzer]

Its not that easy, you need to monitor each cell, as when a cell runs out it does so very quickly.

 

[fechter]

The individual cell monitoring circuit that ggoodrum makes could be interfaced to an Alltrax controller. The idea is when any individual cell gets down to the minimum allowable voltage, it will trigger a signal to the controller to cut the throttle. Under normal conditions, you would hope that it never gets to the trip point, but you need that as a safety or else you could destroy a very expensive battery (not to mention you could get stuck with no power at in a bad location).

The low voltage cutout on the Alltrax could also be used, but it would be redundant if you have the individual cell monitoring cutout.

Another function of a BMS is to limit the maximum current, which you can do with the Alltrax, so you don't need to worry about that function.

 

[GreatScott!]

hi there,

You say the LVC setting can be used on the Alltrax controller but it will be redundant......, that's exactly the point i'm trying to figure out, would it be enough to manage the pack without a BMS for each individual cell?? I'm assuming the LVC setting on the controller will detect only the overall pack voltage. So you are saying that individual cells can vary in voltage as the pack discharges which would require a circuit relay on each individual cell? does this happen often? If individual cells voltage differences don't change drastically relative to each other during discharge, what's the point of the BMS, just set the controller LVC to 40V, about 2.5v per cell and learn when the pack will cut out to gauge ride times. No?

 

[Doctorbass]

GreatScott,

The monitoring of cells is not for every single cells. If you use 6p 20s for exemple, you only need to monitor each of the 20 group of 6p individually.

Each cells in the 6p will have the exact same voltage and if all cells of each parallel group have the same history and capacity and internal resistor, their individual current that contribute for each 6p group should be the same.

What is also important about building a pack is the heat distribution betewwn each 6p group. the cells in the middle of the pack will have a higher temperature than the cells around... so their internal resistor will be different and their current too. Be carefull to the heat management inside the pack.

The GGoodrum and Bob LVC circuit are excellent and maybe also the silicium 15s BMS circuit too if you mod it for higher current.

Doc

 

[GreatScott!]

Hi there,

Yep. By individual cells I meant the 6p packs considered as cells. I would imagine I could get everything balanced with a balance/charger before building the larger pack cells. ok, great thanks.

Alright, so heat will be an issue? but will it effect the batteries enough to cause problems. I'm assuming most fairly big packs featured on this site have no internal cooling mechanism as they are wrapped tightly in heat shrink; so then I imagine that would be the point of the BMS circurts, to redistribute load while discharging? that would make sense.

Ok then Doc, appreciate the efforts. Thank you. I will PM Ggoodrum.