Paul & Sabrina's cheap DIY 144v motor controller

[MPaulHolmes]

OK, so it isn't a complete failure. I have been worrying for the last couple hours that it was just going to be a big hunk of junk once tried on a larger motor. By the way, the hardware current limit comes on at 500 amps, and keeps the mosfets off until about 478amps, so the actual current limit probably averages out to around 485 amps or something.

If you feel like it's starting to lose power, it's because the thermal cutback is kicking in. The micro-controller is continually measuring the temperature, and starts to limit current when the copper heat spreader reaches about 70 degC.

At 70 degC, the software current limit is 7/8*478 amps, at around 71 or 72 degC, the software current limit is 6/8*478 amps, ... all the way until 80 degC, at which point the software current limit is 0/8*478 amps (which, by the way, is 0 amps! hehe).

[MPaulHolmes]

Hey, Ben! We can still fix certain things (software related) once some problems are identified. You'll have to unbolt the top of the box, and remove the ATMega8 micro-controller, and I can ship another one with some hopefully improved performance, and you can plug in the new one (not backwards! ).

[bennelson]

I still need to get a new ammeter.

Anyone have a 0-500 analog ammeter kicking around?

My 0-100 voltmeter isn't exactly useful any more either. I yanked it out and just plugged in the multimeter (set to DC 0-200V) in its place.

So the thermal control is based on the inside of the controller? I can overclock it as long as I keep it cool?

That's it, I'm getting some liquid nitrogen and running it at 1000 amps.....

[MPaulHolmes]

I bet it could be overclocked, but the diodes would be the weak link I think. Also, you would have to change out a single resistor (dang, should have made it a variable resistor!) to increase the hardware current shutdown point. Also, the software would have to change. Also, you would need liquid cooling! Hmm... That would be awesome! With synchronous rectification, you could use mosfets instead of the diodes, and they are rated at 120 amps continuous. If the legs could hold out, which are like twice as thick as the diode legs, that would be really cool. I just don't really know how to do it yet. Also, you would need really really good batteries to be able to pull over 1000 amps.

[Christ]

Oooh! I have a single CPU likwid cooling system by AlienWare! (It's not actually by them, it's just licensed... obviously.)

It's got an upgraded cooler though, so it's got 2 90mm fan places on it. Even though it's just being cooled by 1 65mm turbo fan with a shroud... (well, it hasn't been run in like 2 years, so it's really not being cooled by anything right now.)

I love likwid cooling.

[bennelson]

The first VIDEO test of the Open Source Controller at 144V is coming your way as soon as it finishes uploading to YouTube!

OK - Here it is!

http://www.youtube.com/watch?v=_8wTaRkAGec

[TheSGC]

Originally Posted by bennelson

The first VIDEO test of the Open Source Controller at 144V is coming your way as soon as it finishes uploading to YouTube!

I want to seeee ittt!!!!!!

[celblazer]

Sweet video and no Ben going up in flames.

[bennelson]

Video is up, see post #806.


I wouldn't go up in flames. I would go up in an explosion of acid and lightning!


Apparently, somebody in our local ecology group has started referring to me "Spark".

[Christ]

Ben - in that pic, where you tied down the disconnect, and screwed down the other side... what do you call that knot? LOL.

Nice shot of the spinning tires - oddly, the Geo probably has just as much power now (relative) as it did with the ICE.