Quote:
Originally Posted by DJBecker
The voltage decision isn't trivial, but there is an easy to understand breakpoint.
You can get really good MOSFETs rated up to about 200V. Above that they get expensive and inefficient, and IGBTs are a better solution.
A 200V MOSFET will support a pack voltage up to about 160V, perhaps 170V if you are really careful with the design and capacitors. And they are easily paralleled for high current or better efficiency.
Once you move to IGBTs, you might as well really increase the voltage. 600V ones are not much more expensive or less efficient than 300V ones. And since they don't parallel very well -- you can only use one or two devices instead of a dozen MOSFETs -- you'll want higher voltage/lower current.
Understand the specs of the devices you are getting. Cheap used IGBTs are almost always hard to drive and slow to switch.
|
The reason to go with 400 volts on the pack is because of regulations. The competition we're entering in limits pack voltage to 400 volts. The previous motor controller we used was "rated" to 250 volts. but it was almost certainly based on mosfets and not igbts....(i say almost cause it was hard to tell from the ashes that were left over after the caps caught fire and the battery bar arced to the bottom of the box.)
I've looked at specs for 1200 volt igbt's and 600 volt igbt's. the 600 volt ones have only about half the forward voltage drop from collector to emitter under full current as the 1200 volt rated ones. this would mean only half the power dissipated in the device. What is the reason behind igbt's not being able to be run in parallel as well as mosfets? driving current to turn them on? space available in the controller box? inconsistencies between devices?
*edit*
actually i just found this:
http://www.irf.com/technical-info/appnotes/an-990.pdf
i'll read through it and see if i have any questions