Paul & Sabrina's cheap DIY 144v motor controller

[electricstorm]

I am just starting to learn motor theory as far as phases, inductance, etc.
In looking over the current Revolt Controller and the latest posts it occurred to me that you could possibly connect one controller and two power sections to accomplish the 1000 amps. I could be wrong though since I am still learning about the subject of controlling motors.

Could someone point out to me why this wouldn't work? I would like to learn more about it.

Thanks!

[pamcmills]

True, since the system is modular, you would just build the power section as large as you need. I think this is more of a how would you do it exercise. Let's say all MacGyver had was two 500 amp Cougar controllers but needed 1000 amps to escape the bad guys. How would he do it? Stuff like this is good exercise for the brain and keeps the creative juices flowing.

[Greg Fordyce]

Originally Posted by pamcmills

Let's say all MacGyver had was two 500 amp Cougar controllers but needed 1000 amps to escape the bad guys. How would he do it?

Wind himself a couple of inductors?


MacGyver making his escape! hahaha.

[Georgia Tech]

Originally Posted by jpirkola

Paul and other contributors of the open revolt,

Here is a video of the Cougar running our 1300 pound ultralight modified garden tractor. This is the first real sucssful run after we worked out all of the little bugs. Although we didn't win, we were within a few feet and since they ICE powered tracrtors are tuned to their limits and we are just starting . . . watch out.

We are using all 511 amps available right from the get go. We are looking to get one of the 1000 boards for next year (2000 would be better - hint hint.

So far the controller has been flawless

Thanks to all who contributied
Jim and Darin

Hey Pual this 500 amp controller is VERY impressive!! and the beuty of it is open source!! I love it..

Where are we in regards to the 1000 amp controller? What are the issues as of now?

If you need a lil help money wise with the 1000 amp controller I can help now..getting caught up a little on finaces...

[jyanof]

Originally Posted by Greg Fordyce

...This suggests that the required inductance wouldn't need to be very great for running parallel controllers at 16khz, but I wouldn't know where to start with the calculations.

Ok, I concede that adding inductors is a possible solution. As you state though, it's still not as simple as wiring up controllers in parallel, in case anyone was going to try that (though I'm curious what would happen). Even adding a couple of inductors would involve a little thought and hopefully some math.

[ElectricZX2]

Quick update on the 1000amp beta unit I've been testing. Has been working great so far, probably around 300 miles or so. Have run it past 80 mph and terrain has been a combination of city and highway miles. throttle is buttery smooth and temps haven't been to close to limit. The highest temp I've seen was 600 units on the raw reference data (don't know what that converts to) which was a 10 mile run on the highway at over 200 motor amps 95 degrees F ambient. Oh and BTW, something was not quite right in the source code because I'm seeing a maximum current of between 550 and 600 amps.
One minor (I hope) issue...under full throttle acceleration on three ocassions, I instantly lost power (like a contractor opened). The first time I actually thought the contactor opened because it happened when I went over a bump and I'm using a 24 volt source for a 36 volt contractor and after losing power I let off the throttle for a few seconds and then I have power again. The third time it happened the road was smooth so I thought maybe it was my circuit breaker...but its a manual reset so I don't think that could be...could I have hit a limit of some sort? Maybe my throttle pot resistance went too high? Its a Curtis and I think I've measures 5.2 kohms max....

[MPaulHolmes]

Hi ElectricZ2K! Sorry I haven't updated the software. My computer had a catastrophic crash, and my harddrive/motherboard is no more, so I got a new harddrive/motherboard, but the funny thing about that is, the data doesn't magically come back just because the computer works again. hahaha. Sooooo, that's been the delay on updating your software. I'm almost certain the current sensor had a spuratic bad read that caused the cuttign out. It sounds like it's a little less robust than the LEM, so I think a little error checking is needed there. To get the stuipd thing to compile again, I need to redownload Linux and make another dual boot, but I don't remember how I got all that crap set up last time, since it took Fran like 3-4 hours of walking me through
"lwerj -2 -3 -w-r-w"
and other such intuitive linux commands.

It wasn't the best time for me to lose everything. I did have carbonite, but that only saved some stuff that was on the windows version, and not the linux version. oh happy day.

I really really really need to get things ported over to AVR Studio. Hmm... Hey! Georgia Tech! Yes, it is most definitely money. hahaha. Sabrina told me that we can take some money out soon (she's in charge of the bills) to order the boards. I think it will work well, but I do have some questions:

Does anyone know how layout affects an RC snubber? Like if the R in the RC is sitting vertically on the board, is that too much leg that the current must go through for it to be effective?

[Greg Fordyce]

Originally Posted by MPaulHolmes

I need to redownload Linux and make another dual boot, but I don't remember how I got all that crap set up last time, since it took Fran like 3-4 hours of walking me through
"lwerj -2 -3 -w-r-w"
and other such intuitive linux commands.

Hey Paul, sorry to hear about your PC problems. Get hold of the latest version of Ubuntu or Linux Mint, either CD or DVD editions. They will both boot from the CD/DVD (called "live" mode) and allow you to check that everything works on your pc. Then click a icon on the desktop and they will automatically install themselves to your hard drive. Everything is handled for you, the partitioning, installation and dual boot setup. No need to type in cryptic linux commands. You just have to answer a few questions like language, timezone etc. As always when installing a new OS, backup your data first. I run Ubuntu, but like the look of Linux Mint, maybe give that one a go.

Greg

[MPaulHolmes]

Only 29 hours left for carbonite to download everything. haha. It was 33 hours 2 days ago. We are planning on getting the boards for the "1000amp" made on or around the 13th of this month! hurray! Well, I think the theoretical max current will be 140amp*12.

And this time I'm just paying for it, so when it doesn't work and is a worthless pile of yeller monkeyturds, I won't feel as horrible.
Man oh Man, this thing really needs to be checked carefully this time. If I'm not mistaken, it might be attempt #4 for me or something like that. haha.

My only concern is the different propagation delay introduced from using multiple HCPL-4506 optocouplers. according to the datasheet, they should all be within 100ns of each other. Each group of 3 mosfets will have its own isolated 15v supply. I wonder if I could practically match them. LIke buy 100 of them, and group them by propagation delay. I taught my first statistics class this last quarter, so I bet there's some random statisticky thing I could do hahaha. Boy that class was awful.

[flores]

A litttle update on my 1000A testing:

This weekend I mounted the Warp 9, the thundersky's were aready mounted..everthing runs on the 500A now.. so the last thing to do is mount the 1000A.. but I need cableing for that.. that's not in yet.. getting really close.



In the above left corner you see the 500A.. will repleace that really soon for the 1000A.