I have to agree that there is a LOT more involved in this than you seem to be taking into account, Byron. High power motor drive electronics is NOT easy to design. Consider this for a moment: if you want to keep your heatsinking requirements to a minimum, you will need to turn the IGBTs on and off very fast. This means that the current in some parts of the circuit will have rise/fall times of something like 100 ns. Now look at the di/dt which you have in 600 amps switching in 100 ns. That's 6 gigamps per second. If you have a 1 nanohenry stray inductance (which is nothing), you will see a 6V drop across it during this turn on/turn off transient. 6V is easily enough to cause any one of a number of bad things to happen: destruction of digital I/O, spurious triggering of IGBT gates, IC latchup, etc. What tends to happen in this arena is that you pile a whole bunch of FETs or IGBTs together and expect it to work and a whole string of tough problems spring up due to stray inductance, gate capacitance, etc. Since a sudden glitch can cause a short circuit (if both high and low side IGBTs are turned on at the same time), you can destroy your IGBTs in a fraction of a second. I once had a motor driver prototype running on a bench when suddenly there was a tiny "tink" noise. Before I knew what was happening, the fuse inline with the battery pack burst into two 4-inch long blue jets of flame. After I disconnected the battery, I found that a whole bunch of the FETs were toast. Something had turned on high and low side FETs together, they instantly failed shorted, which caused something like 1000 amps to flow, which exceeded the interrupting rating of my fuse, caused it to arc over, carbonizing and then igniting. If you want a learning experience, then build it yourself. If you want to save money, buy the stuff new or used. You will go through several sets of parts in your development and it will end up being 2x the cost of the whole thing new. Sean On Sun, May 25, 2008 at 6:17 PM, Olin Lathrop wrote: > Byron Jeff wrote: >> Paying $1400, or even $1000, >> doesn't >> make a lot of sense when the parts can be assembled for $150. > > Try the real world for a change. You've clearly never designed a commercial > product and dealt with it all the way into production. > > Parts 150 > Knowing which parts to put where 100 > Assembly, test, fallout, and repair 50 > Paying back cost of setting up produciton 200 > To pay for support when you call after > having done something stupid 300 > To cover liability costs when you do > something stupid but blame us anyway 100 > Cost of capital to get into this business 50 > To pay back regulatory compliance and > testing costs 100 > To make a living 300 > ---------------------------------------------- > Total $1,350 > > > ******************************************************************** > Embed Inc, Littleton Massachusetts, http://www.embedinc.com/products > (978) 742-9014. Gold level PIC consultants since 2000. > -- > http://www.piclist.com PIC/SX FAQ & list archive > View/change your membership options at > http://mailman.mit.edu/mailman/listinfo/piclist > -- http://www.piclist.com PIC/SX FAQ & list archive View/change your membership options at http://mailman.mit.edu/mailman/listinfo/piclist