looking for BMS for high voltage and high power racing car

[vile]

Hello everyone!

First of all, sorry for my english, its my second language.

I am a member of the "e-Traxx"-team, a group of students from the university of applied science Düsseldorf and we are participating in the construction competition "Formula Student Electric" (FSE).
The objective is to build a fully electric racing car for one person which must confrom to several rules.

My question refers to the battery and charging system. We'd like to get our battery capacity to about 7-8,5kWh that allows a max. continuous power of about 80kW, that's the maximum allowed electric consumption in the car.
We've chosen the A123 20Ah LiFePo4 cells to build our battery pack, what means we will need about 106 to 128 cells (which will be in series) to achieve 7-8,5kWh. We haven't chosen the exact capacity yet.

Now the my and our question:
What kind of BMS would you recommend?
The BMS should be able to communicate with the charger and with a µC on board (controller for the complete car), for example via CAN bus. The EMUS BMS or GWL Power BMS could be the one.
The battery pack will consist of three battery containers (which may not exceed 120V DC) connected in series. The whole pack could be charged (on AC and also by recuperation when decelerating) by our motor (3x400V synchronous motor) controller - it will be the BAMOCAR-D3.

Do you recommend a seperate charger (for charging in stop) or is it in your opinion a good idea to use the motor controller? Just connect 3x400V to the controller instead of the motor and it charges.
What would be the best BMS? We have to monitor all voltages, temperatures and the current so it's important to have a bus system to the µC. A high balancing current would be nice to guarantee that there is no cell overcharged even when breaking tight and when the recuperation current is high. By the way: what's the maximum charging current for the A123 20Ah pouch cells? Haven't found it yet...
How would you contact the cells to allow high currents of ~200A? I read this thread endless-sphere.com/forums/viewtopic.php?f=14&t=38761 and I am very impressed by all these ideas; but which one is able to stand these currents?

Thank you for reading! I am curious about your answers and ideas!
Vile

 

[agniusm]

http://elektromotus.lt/en/products/battery-management-system-bms

 

[circuit]

http://elektromotus.lt/en/products/battery-management-system-bms

Thanks. More info and software downloads on wiki page:
http://www.elektromotus.lt/wiki/index.php?title=Main_Page&Itemid=155

 

[vile]

Thank you for your answers! I've already read in the emus bms wiki before. Obviously you recommend it. Interesting to see the Elcon PFC series chargers are supported and are available for every individual cell configuration. Would you recommend this charger series? Has anyone an idea if charging with the motor controller is a good solution?
And which cell connection is current proof? I'd be glad if you example write the number of the post (see my link) with a picture of your preferred type of connection.

Lots of questions, I know, but this is the first time we are trying to plan and build such a car, so we have to learn lots of basics.
Thank you!
Vile

 

[circuit]

ElCon is a bit bulky, but it is passively cooled and is waterproof. In couple of years, not one ElCon failed (out of hundred maybe). CAN-enabled ElCon is highly recommended, as this will give better charge control.

I am not sure what you mean by charging with motor controller. If regenerative braking, than no problem at all, I like regen and it is supported by Emus.

Can you please refine your question on cell connection?
Overal voltage (series cell count) should be chosen to meet motor controller's and charger's requirements and capacity (parallel cell count) to meet required vehicle range.

 

[vile]

We are going to use this motor drive/controller: http://www.unitek-online.de/en/produkte/batterie/bamocar_d3.html. The central computer will be a dSPACE MicroAutoBoxII.

I do not mean charging the battery by recuperation (we'll use it in any case), but I mean charging the battery when the car is in stop. There could be 3 contactors that disconnect the motor from the motor drive and connect 3x400V AC power supply to the drive/controller. Then it could convert the voltage to the correct level and charge the battery (with current limitation). In this case we do not have to buy and implement one more module.

I've asked here -> http://shop.lipopower.de/A123-Systems-AMP20M1HD-A-20Ah-Folienzelle-original-lange-Tabs for the cells, the cell connection and the emus bms. They sell AlMg connectors that are able to stand peak currents of 500A and they would prepare the cells to be compatible with the connectors (drilling or stamping three holes through the cell tab).

The battery pack will consist of four seperate modules/containers whith overall 128 cells in series (that's 32 cells for each container) and only one in parallel => 128S1P.
With about 64Wh per cell we will get a ~8.2 kWh battery which is able to provide the needed 80kW and to regen with high current when braking.
There is no required vehicle range, we try to get as far as possible. The overall weight will be about 220-240kg (without driver).

I hope these information aer helpful!
Vile

 

[circuit]

The idea to charge via motor controller is interesting. However there is one important detail: controller operates in PWM mode, so it needs inductance to regulate current. In drive mode, motor provides required inductance. There will be no inductance if you connect controller directly to three phase mains.
However, you could still use motor as a 3-phase inductor: take all 6 wires out of motor and hook each winding in series to AC mains. There should also be a reasonable amount of bypass capacitance on mains to avoid EMI problems. Also, motor's shaft should be stalled.
But be warned, I have never tried this, I only had this idea some time ago.

About splitting battery pack in several peaces: it is okay, Emus BMS supports this. Just write an email and explain what you are doing, as there are additional modules for pack isolation.

 

[vile]

Haven't thought about the inductance, it may be a problem. I think the best way would be to use a conventional charger that allows can bus communication and is made for charging the LiFePo4. Also there would be no need for high power contactors between motor and motor controller. Thank you very much for your ideas!

 

[megacycle]

This could be similar to what you are looking for and talking about
http://www.youtube.com/watch?v=zclbadsViUU.
Enjoy.

 

[jonescg]

Charging through the motor controller isn't too crazy an idea. It's easy if you have an AC motor controller (VFD) and the pack voltage is less than 590 V. Rectified 415 V three phase AC (as if you were using regen, but while stationary) should give you the right DC voltage.

DC motor controllers have inductance issues. Another option for charging is to buy a CurrentWays charger - they are good for high voltages and can run in series/parallel to get any voltage and power output you like.

 

[megacycle]

http://www.google.com/patents/US20120286740?dq=Charger+++inassignee:%22Renault+S.A.S.%22&hl=en&sa=X&ei=Yr64UMGgCM3otQaLxoGACA&ved=0CEAQ6AEwBA
Link to patent documents.