Paul & Sabrina's cheap DIY 144v motor controller

[regmeister]

Hi y'all.

I was wondering if the power section could be cut in half (or smaller portions), and used to build for applications requiring less current? Would this be a possible. I am thinking 200A sections.

Or, would that be difficult?

Awsome work on the software... well of course the whole project!

Cheers,
Richard

[MPaulHolmes]

Absolutely you could do that! It would be SUPER easy. It would make things quite a bit cheaper too.

[regmeister]

Hi Paul,

Is there a list available for the power board parts? I didn't see this anywhere.

[regmeister]

that's ok, i found it on the wiki. my memory is darn poor somedays

[squiggles]

Originally Posted by jackbauer

Got my motor rpm sensor up and running today and began thinking of an rpm limiter circuit for the controller. What would you guys think about using an lm2917 frequency to voltage converter connected to the hardware overcurrent circuit in the revolt? It could be setup to drop out the pwm at a certain rpm with low component count. Quite important for myself with an 11" forklift motor.

Is there no counter input available on the micro?
If you speed sensor is giving pulses the easiest method would be for the micro to measure pulse time and compare to a minimum value.

many micros have a counter/timer input that will measure pulse width.

if (pulse_width < min_allowed)
pwm = pwm * 0.5;

Actually I see you are using the counter input T1 for the UVFault input.
Maybe UVFault could have gone to INT0 where it could trigger an interrupt when activated.

[regmeister]

Paul,

I was reading back through the thread and came across your post late September, "Replace the freewheel diodes with mosfets".

I have been reading a document from MIT for a modular controller design that may interest you.

They use a "reverse-parallel configuration" of the drivers to switch the mosfets.

there are some other interesting ideas as well. anyhoo, here it is:

http://web.mit.edu/first/kart/controller_rev1.pdf

[MPaulHolmes]

That was a very nice pdf. Thanks for posting it! One thing to note. I bought a couple of those TI DC-DCs they were using, and had planned on using it in the controller, but Otmar told me that he had used them, and they failed a lot on him. I think it's OK at 36v, but if you are a lot higher they have problems.

It was nice to see a conservative dead-time formula for the synchronous rectification. I was planning on using 2uS, but I might make it a bit bigger than that.

MIT always has stuff that is so elegantly stated and simple. You can tell the people that make the information have a very good understanding of what they are doing. I wish I could have gone to a school like that instead of south puget sound community college, but a 1.2 GPA sort of limits school options. haha.

[regmeister]

great, glad you liked it.

I was thinking of getting one of your power boards for experimenting, but as it uses the freewheel diode instead of the mosfets I will probably do something different for the time being.

However, I would like to donate to your rickshaw fund, Do you take a gift through paypal?

[MPaulHolmes]

Hey Regmeister! Yes, we have a paypal account. You could use either mpholmes@hotmail.com or billing@paulandsabrinasevstuff.com.

It wouldn't take very long to make a pcb that is synchronous rectification ready. I would have to write the GCode, and then just etch the dang thing! It would probably take a couple days given my 10 minutes of work, 2 year old bangs on garage door so I will play blocks, play blocks for 1 hour, 15 minutes of work, son has a HORRIBLY STINKY DIAPER, and won't let Sabrina change him. repeat. hahaha.

[regmeister]

is adams (I think it was him) newer pb design also a freewheel diode, and have you etched that version yet?

sorry for my ignorance. could we just modify the existing board to incorporate the third pin of the freewheel mosfet?

otherwise, I guess perhaps we could just work out a price for a custom etched power board of whatever design I could work out? within reason of course. Of course, I'd and de-ignorance myself as much as possible before something like that.

rickshaw funds headed your way soon.

ooh i just saw a skier crash on tv at Lake Louise... looked baaad!