Paul & Sabrina's cheap DIY 144v motor controller

[MPaulHolmes]

Fran had observed similar oscillation to what I had going on with a brushless DC motor he had done a couple years ago. He think's it's too hard of gate drive. Next time around I'm going to use either 15 Ohm or 22 Ohm or something like that The source at the driver and the source at the inside of the mosfet had a difference of probably a few volts. V = dI/dt*L. The L is pretty small, but the dI/dt was friggen ginormous. If I slow down dI/dt some, V is going to get smaller and it will keep the mosfet from going on/off/on/off.

Some igbts don't have this problem because of the Kelvin emitter connection, which has a separate path to the emitter inside the igbt from the power path.

[DJBecker]

Originally Posted by MPaulHolmes

Some igbts don't have this problem because of the Kelvin emitter connection, which has a separate path to the emitter inside the igbt from the power path.

IGBTs also avoid turn-off ringing because they don't (can't) turn off quickly. You have to wait for the minority carriers to decay (recombine) before the device is fully off. This absorbs some of the inductive energy that would otherwise "slosh" around the cables, capacitors and controller structure.

There is a lot of design effort around turning off IGBTs. You can't get rid of the tail end of the minority carrier recombination, but you can immediately stop generation of new carriers by a negative gate voltage, and bypassing the gate resistor with a diode.

MOSFETs don't have same issue and are relatively easy to turn off.

We have (had) a turn-off ringing issue which we "fixed" by using 39 ohm gate resistors. That costs us a little in the turn-on speed, but it's easier to change resistors and add cooling than building a better capacitor board or physically reducing the current loop area in our controller.

[electricstorm]

Hey Paul,

Sent you a payment through PayPal, but later received a returned email
message stating that a permanent error had occurred (mailbox is full) and
the email had not been sent. Just wanted to let you know and to also see
if you actually received the funds.

Sorry to post here, but I don't have enough posts to send a private message yet.

Jim Brock

[MPaulHolmes]

New driver mostly finished being soldered. Started control board. one hand type baby...

[Greg Fordyce]

So I've been busy and haven't been keeping up with the forum since probably Thanksgiving.

Originally Posted by madmike8

Is there a setup information page for the revolt controller? Something that tells what settings can be adjusted, and how? Is there a way to set a motor voltage limit? So, if I ran 120v worth of batteries, can I set the motor to see like a 80v max limit?

Thanks

Read this post and if madmike8 is still around there is a built in function for this. You need to set the motor overspeed limit, dig through the forum and wiki for instructions, but basically motor voltage is directly related to motor rpm, so setting this up will effectively limit motor voltage to whatever you want. HTH.

A update on my controller. I have one of Pauls early rev C controllers. The car has been on the road since May and has over 3500 miles since then. I was having a problem in the cold with the controller not working until it got some heat into it. I also had a problem of the contactor control not working and the controller getting warmer than I liked. I found the P channel mosfet faulty so repalced it and at the same time ran the soldering iron over all joints on the control board. That has fixed my contactor control and also seems to have sorted out the cold weather problem. As for the controller getting to hot, previously I had fitted a large heatsink in place of the steel base plate, but now I've turned the controller over so the heatsink is on top. I recon that it just wasn't getting proper airflow, my engine bay is quite cramped.

Greg

[sawickm]

Originally Posted by MPaulHolmes

New driver mostly finished being soldered. Started control board. one hand type baby...

Hi Paul,

I am hoping your new driver and controller test out well, keep us up to date when you get a chance.

-Mark

[jackbauer]

Greg , interesting your cold weather issues turned out to be solder related. Seems mine might have been the same though I had suspected the crystal. Wednesday night was very cold here (-6c) and I got a call at 1am from the alarm dialer in work. Seems we got broken into by 3 guys that watched too many re-runs of point break and dressed up like the ex-presidents! so anyway I had to hammering down the motorway with the controller temp at -6c. Never missed a beat. Gotta hit the beach and watch out for any suspicious looking surfers

[mora]

I've had my controller down to -25C multiple times. Heatsink gets barely to zero celcius even after 20km of driving. I've had few sudden power downs during the winter but it was always quickly solved by restarting the controller. I suspect it was a bad throttle pot causing that issue though. It caused some jerky starts too. I've replaced the pot but haven't had a chance to drive using it yet. I had other issues with my controller lately and am waiting for new power PCB to arrive, hehe.

Does controller shut down if throttle goes way out of range during drive?

[MPaulHolmes]

Yes. There's a fault if the throttle goes out of range, which disables the PWM. And there's no clearing it until the power goes off and on again. For those who are curious, Mora's fault was similar to Joe's. They both had an etched board and somehow there was a short from the top to the bottom of the board. Maybe due to the heatshrink around the bolts?

[mora]

I have some pictures for everybody to see:

https://picasaweb.google.com/1036324...nRevoltFailure

I'm waiting for new power PCB from Paul which should be better than etched version. Mine likely had insulation failure (heatshrink wear problem). I had two-piece heatsink as you can see from the pictures and holes had some burr and sharp edges when I opened the assembly. Those might have caused excessive wear too.