High Power Shunt regulator ?

rsisson

100 W
Joined
Oct 18, 2007
Messages
124
I am still working two projects, one Small, my "SideKick" which has managed to eat two controllers, two battery packs and lots of other bits... but my long term project is a Battery Powered Real Car...something small like a Geo Metro, Rabbit or Civic Hatchback... to take the place of the Sidekick when the weather is not so nice...

With that in Mind I have been watching LifePo4 Battery and Charger threads carefully... Assuming that LifePo4 Batterys will come down in price (a bit) between now and when I need to order batteries, the issues I have are around managing a BIG pack of 40-50 Cells... not the 12-16 needed for low voltage.

Building on what I have seen on MetricMind, EV-Austrialia, and some others, the best way to manage a BIG pack is with Shunt regulators on each cell... The issue becomes, how to get a Big Enough shunt... if you are trying to charge a 150Ahr pack, even .1C is 15Amps..ouch !

Now yes, when the pack starts to approach the shunt voltage, the current should start to throttle back, but the FIRST shunt to trip is going to take a big hit.... So... Looking at the TL431 Precision regulator (google High-Current Shunt Regulator TL431) and a couple of components (2 resistoirs, 1 adjustable resisitor, the regulator and a High power Transistor) you have a high power shunt.... This looks a LOT like the Metric Mind shunt without the LED outputs...

If we are going to rely on the Controller of Low voltage cutout, High Voltage cut out and current limiting and our charger has ANY smarts to start throttling back as soon as it gets close to the optimal pack voltage, how smart a BMS do we need.
Now I like the LED outputs, as I can see a pack that is badly imbalanced having one cell go into shunt with the pack voltage still low…worste case… and then the shunt takes the full current OUCH…but what else do we need.

Is anyone looking at building something Like the MetricMind Shunt...I have seen a couple of commercial attempts, but they want ~$45 EACH for $7 worth of parts... and times 40-50 cells that hurts....
 
If you take my approach;

http://www.endless-sphere.com/forums/viewtopic.php?f=14&t=6419

I use 2 current sources. One high current, one low current. During bulk charge, both current sources are on. As soon as the first shunt turns on, the high current source turns off, leaving only the low current source. So the first shunt does not take a big hit.

My current problem is the heat dissipated in the linear current sources. I have designed a switchmode current source and run it in simulation, but not built it yet. Will publish full details in due course when built and tested.

Amanda
 
Back
Top