Paul & Sabrina's cheap DIY 144v motor controller

[SEBART PL]

Originally Posted by jyanof

... I see the pictures previously posted of your controller design in this post. The power section does not appear to be similar to the original Revolt design.
Good luck in your efforts!

Thank you for your information...60 C degree ambient temperature it is very impressive that controller still work well.

Yes you have absolutly right previous design is not mine, but I have come back to original design and even, improved it. I am worry for safety now during the tests driving, so why I want to have put lower temperature parametrs before I get trimed my controller. This time on road I will be driving with temperature indicator to see what are normal values, then I set up proper thermal cut off.

[badfishracing]

Anyone every experience a problem with the stock P&S controller with traces blowing off? Well, not quite stock. I've got a 400A hall effect sensor, and running 168V of lead, 181V right off the charger. It sat out overnight in a hard rain, but I don't believe anything was wet. I precharged for 10 seconds, and when the contactor pulled in, it popped, and smoke came rolling out of the only opening in the case. Now it's possible that the precharge resistor wire was loose, and it didn't actually precharge. At the end of the power board, closest to the Motor output, it looked like the + and - trace over the capacitors shorted together. Some missing trace, and some molten copper. Now it was near a screw head that holds the control board on, but the screw doesn't look arced. Just very black. Most of that section of the power board is black. Looks like I'll have a little less capacitor current capability due to some missing trace, but otherwise it looks like I can just fire it back up.

Any thoughts??

[mora]

Do you have old style copper power board? I had it fail similarly even though precharge circuit worked. Voltage at terminals wouldn't rise as there was "load" at controller. Load was a small short on power board and when I forced main contactor on it resulted in smoke and molten copper too. You might need to replace some power components. I believe your repair bill will be small; new style power board and replace all the broken components. I'd replace everything just in case but it might not be necessary. Similar short should not happen anymore with newer style power board. This quick capacitor charging reduces its life.

Sometimes my precharge contactor simply refuses to operate. I monitor pack voltage at controller terminals so I know what is going on with precharge. If voltage doesn't rise as expected I cycle ignition key and precharge sequence starts over. Too bad this 500A controller doesn't actually know when precharge is actually done. Newer control board (1000A version) can monitor voltage if I'm not mistaken. This missing functionality doesn't bother me though.

[badfishracing]

Thanks for the reply. I assembled the controller as an entire kit. About 3 years ago I guess? I suppose it was just a combination of things that led to an arc across the + and -. I can't say that any component is damaged.

This is in a pulling garden tractor. The track that we were at over the weekend was the same track that destroyed my transmission last year. Over the winter I did a full rebuild (had an excuse to, so I went for it), and went to a single speed transaxle. Still working the bugs out of the new setup. Really wanted to see what it would to at 168V and about 450 battery amp limit. Hopefully about 50hp. The big dogs in our club run about 70hp.

So with one last weekend on the season, I'm itching to get it out for a couple more hooks.

[MPaulHolmes]

Ya, the etched copper power board had B+ and B- right on opposite sides of the board, with only 1/16" of FR4 separating them where the holes were. If anything got on that FR4 to reduce the resistance a bit, a sustained arc could have happened. 168v is plenty for an arc welder, and it sounds like the copper vaporized once the arc started. Joe had something like that happen. I haven't heard of it happening with the power boards that have the solder mask. Man, I'm sorry about that!

[badfishracing]

Tried it Wednesday night at 12V with no problems. Had to help an ICE competitor with his wiring last night. I'll try 180V tonight but it looks promising.

[SEBART PL]

I have recently installed diodes STTH6002CW (total 13 diodes) with M3 screw and M4 pad with silicon pad isolating from radiator. This is what happend:
- before precharge diodes showed one direction flow
-when I precharged 88 V I have received voltage on M- bar and diodes showed two direction voltage drop
I have thougth one of diodes is bad so I started to loose screws of each one - and let say number 4 showed new future when it is screwed shows two direction flow, so I have decided to cut it from diodes series.
This situation happened again with diodes number 5, then 7 and finally 6. Now I have took off M4 pad and give M3 pad couse looks like narrow pad give short after precharge.

I have finally one direction flow after precharge but I have lost hope that this diodes are in 100% working condition and I worry that is my main key for blowing mosfet problem. Now I have hook up 9 diodes with 0.288 V drop.

Do you suggest change all diodes for new and do not screw new one so hard?

Ok, I have anserw, couse I have decided to pull a car with diodes and I got short circuit again 5 mosfet blowed, diodes need to be change for new and mounted exactly as couguar design.

[SEBART PL]

Oki doki , my controler works, during test I have destroyed 5 mosfets which was too fast with others and 17 still works looks like they open at the same time- if it is true I am lucky.

My question is to the designers of power board can I put capacitor and resistor between M- and B- to lower fall down current to protect mosfets from induction and voltage spikes?

or is any software parametr whcih could drop current softly when throttle is rapidly depress ? I am not so sure what is for this responsible ki, kp or crr .

[64jeep]

Hey, I learned a fun lesson the other day with my IGBT version of the controller that would be applicable to anyone with this controller. I never adjusted the deadband in the throttle position. I also use a limit switch on the pot box to control a contactor on the B+ side of the controller. As I figured out recently, the PWM duty cycle starts to rise a teeny tiny bit before the switch clicks. I drove my car for 3 months like this without noticing it until one morning I was trying very slowly ease onto the pedal to back down my driveway when all of a sudden the wheels spin out and leave a foot of rubber on the concrete! Lucky for me, the IGBTs were able to handle this short burst of current, so my controller is still intact. The reason that happened is I pushed the throttle to the point that the controller was trying to supply current to the motor. Since only 1/4 amp is available through the precharge resistor, the control loop kept increasing the duty cycle. When the contactor finally closed, whatever the duty cycle was at resulted in enough current to spin out the wheels before either the controller shut it down or I took my foot off the pedal.

So let this be a warning to all of you to make sure the deadband is properly adjusted on your controller, especially if you have a limit switch operating a contactor like I do!

[jyanof]

Originally Posted by SEBART PL

My question is to the designers of power board can I put capacitor and resistor between M- and B- to lower fall down current to protect mosfets from induction and voltage spikes?

or is any software parametr whcih could drop current softly when throttle is rapidly depress ? I am not so sure what is for this responsible ki, kp or crr .

There isn't a software parameter that affects inductive voltage spikes.

You could:
- increase the gate resistor value; the mosfets will run hotter though
- put the diode, mosfet, and capacitor really close together in you power board layout
- add some sort of snubber, like an RC snubber as you mention - this will reduce efficiency somewhat

speaking of capacitors, I don't see any in your picture... are they just hidden in the picture?