Paul & Sabrina's cheap DIY 144v motor controller

[princeton]

Also, has anybody figured out the optimal number of Farads of capacitance that is required on the battery side of the controller to prevent Mosfet failure?

[princeton]

Can somebody give a comparison of total cost for a 1000 to 1200 amp controller, 200V, homemade part/tool cost vs ready-made controller cost?.

[princeton]

I still can't see any of the images of schematics (not enough posts), so I was wondering if anybody could tell me why the controller needs diodes? My (extremely simple) controller doesn't use any diodes. Are these simply used on the battery side of the controller to keep capacitor current from feeding back towards the battery?

[MPaulHolmes]

princeton: No, the motor has inductance becasue of the coils inside (well it would anyway,but it has more inductance becasue of that), so the current wants to keep flowing, and if you just shut off the mosfets, and don't give the current anywhere to go, it will piledrive through the mosfets from drain to source.

1000amp controller total cost in parts would be very similar to total cost of regular 500amp controller. Maybe an extra $20 or so in metal. So in modest bulk, probably $350, but I don't know how well it will work yet. haha.

Optimal number of farads: infinity. But that's not practical.

[princeton]

It sounds like the diodes should be located as close to the motor as possible. Why locate the diodes in the controller ? -- that would cause the current to flow back and forth down the wires from controller to motor, causing lots of EMF interference?

[princeton]

I was thinking of just using one of these capacitors on the battery side of the controller and another on the motor side to smooth out the voltages. Do you foresee any problems with this approach? Will the motor inductive feedback still be too much for the mosfets?

Digi-Key - LNY2V223MSEJ-ND (Manufacturer - LNY2V223MSEJ)

[chapper]

Princeton
Check out Forrest Cooks' PWM designs at
FC's Electronic Circuits
chapper

[princeton]

Thanks for the website. My design is similar to FC's light dimmer controller since it uses a simple square/triangle wave generator and a comparator to produce the PWM square wave. I just feed the output into a Darlington stage and then some small push-pull mosfet drivers that feed drive the final large mosfets. I get a nice PWM square wave at 20KHZ. I'm working on stability at this point (I burned up one final-stage mosfet during testing).

[dcb]

Yup, no frills pwm isn't terribly complicated. Lets dub that the "bad boy" controller (compliment of the "bad boy" charger), though best reserved for the adventureous types

The openrevolt has logic in it to make it a current controller, with pi loops and a couple over-current protections, and a large body of testing behind it, plus a lot I'm overlooking, so it easier to drive and should be safer for the average joe, if I understand it correctly.

edit: lots of info on the wiki you should be able to see:
http://ecomodder.com/wiki/index.php/Open_ReVolt

[chapper]

Let's make that "Bad Boy 1"
Here's "Bad Boy 2"... with "Active Braking"
http://talkingelectronics.com/projects/Xenon/Xenon.html

It's not that I recommend these circuits, but they're interesting and they work.
Many thanks to both Forrest Cook and Ken Stone