Paul & Sabrina's cheap DIY 144v motor controller

[flores]

Originally Posted by jedsmd

Paul
I think you said in an earlier post that RTDexplorer will not be supported by this controller. Will there be some similar user interface for setting variables and monitoring the controller?
Jerald

I do software development for a living, I guess I could make something similar once I have my beta unit.

My idea would involve hooking up a Netduino/Arduino and writing a web interface for it.. what do you guys think of that?
Then you could plug in a ethernet cable and manage it in a webbrowser..

[MPaulHolmes]

You'll do. LOL. I could make it rtd compatible, but there would be some shoe horning required. And The readings wouldnt really show current, but a scale of 0-500, where 500 was 'Max motor amps', which is programmable.

[flores]

I will, it's gonna be extremely cool.

Paul, any e.t.a. on the finished product?

[MPaulHolmes]

Well, the last thing I've been waiting for arrived today (shielded 3 wire cable). The stupid end caps are irritating me at the moment. I'm trying to commit to something, but I'm not sure what's best. If you guys want to add your own end caps, and show off what you come up with, I won't start crying about it. haha. But I can also come up with something too. I'm planning on using pieces of aluminum sheet metal, with opposite ends bent out so the end caps can be screwed into the enclosure, but the screws would be accessible on both ends for removing the end caps. Or sheet metal screws, but I'm not sure how good those are.

The EEProm part isn't done, but won't take long. It doesn't have a bootloader for upgrading software at the moment. The quickest way to get it shipped out would be for you guys to get a pickit 3, and be able to upload new hex files. There's no way there won't be little quirks that we come across that require code changes. The whole thing is almost a completely different rewrite.

So, I could mail the first one as soon as next week probably, but that's assuming the quickest, no bootloader, etc...

[flores]

I already have a pickit 2, that will also work right?

EDIT: answer is yes, release notes

[MPaulHolmes]

haha. And we have our first destination chosen for shipping. Flores! I'll just do my best on the end caps, and if you don't like them, rip them off and put something else on! haha

[jedsmd]

Originally Posted by MPaulHolmes

haha. And we have our first destination chosen for shipping. Flores! I'll just do my best on the end caps, and if you don't like them, rip them off and put something else on! haha

I can make my own end caps. Would it be worthwhile to mount a couple of 3 inch fans with filters to move air through the controller?

[flores]

Originally Posted by MPaulHolmes

haha. And we have our first destination chosen for shipping. Flores! I'll just do my best on the end caps, and if you don't like them, rip them off and put something else on! haha

Haaaa. NICE.. Can't wait !!

*hooks up charger*

[MPaulHolmes]

I'm not sure if that would work long term in Olympia. I'd worry about 100% humidity at 40 degrees and raining all the time. But maybe if the control board has a good layer of conformal coating it would be OK. I'd just mount the base plate on a heatsink, since it's just a plate with no fins.

[MPaulHolmes]

I was writing on the EVTech list and showing some pictures, and Otmar (the maker of the Zilla, and hence, my ears were WIDE open) had this to say:

That looks good!

The ZIlla is also a high side drive. I did it to have a low cost shunt on the return line from the motor at B-.
Depending on what you plan for cooling, you may want to check out your heatsink/thermal resistance calcs before running it at power. My early module controllers had a heatsink about as thick as yours but full of fins and they were unreliable due to too little cooling.

I've seen that cap listed before, it sure looks interesting!

So, I can either leave it as is, but you guys would absolutely have to get a finned heatsink. It won't burn itself up. The temperature probe is right at the hot spot between 2 IGBTs. It might take a little while to get too hot, but I don't know). Alternatively, I can mill a wide channel right under the IGBTs, and bolt a piece of sheet metal to the base (with a sealant of sorts), and you run it water cooled. Water in one side, and out the other. Just a wide single channel. I can use the IGBT bolt holes already there to clamp the sheet metal sealer down.

I had just assumed that I couldn't fit it on my mill, but it can do that no problem..

The plate is 3/8" thick, which doesn't leave a huge amount of room for a nozzle hole. Maybe I could mill it down half way, like 3/16" deep.

A couple questions: 1. Is there a problem with doing a single wide channel like that? I'm OK with 90% good enough in exchange for ease of fabrication.

2. Do they make water input nozzle things with a square or rectangle shape part that mates to the aluminum plate? It would be easier to just mill the channel all the way to the edge (but narrowed) and epoxy in the nozzle. I suppose it would work if it was round too. Just fill in the space with JB weld or something.

So, in summary, Do you guys want to just try the controllers by mounting it to your own finned heatsink? Or do you want me to try water cooling?