Paul & Sabrina's cheap DIY 144v motor controller

[duncan2]

Yep that's my thread!
The Device is great fun

I suppose this is number post three - only two to go!

I'm happy about mechanical stuff - I need to learn the electronic stuff

[duncan2]

Post Four

I need to decide what speed I actually need

The road speed limit here is 100Kph
The grass events I have been doing are relatively low speed and even the tarmac ones have a limitation on the allowable length of the straights

The only place where I hit my max speeds is at the drags - I hit 140kph at the 1/8th mile

[duncan2]

Do I want to increase my maximum speed so that I can dare try a 1/4 mile drag race?

Or am I best with a speed limiter?

[thingstodo]

Originally Posted by duncan2

I suppose this is number post three - only two to go!

I'm happy about mechanical stuff - I need to learn the electronic stuff

I'm the opposite, I'm afraid. If it has wires, I can get it to work.

But aligning mechanical stuff ... or getting a bearing pressed on .. purging the air from a clutch ... like magic!

[badfishracing]

Is the hardware overcurrent, set by VR1, an actual TRIP/Shutdown, or does it simply cut back the PWM signal?
I'm having an issue with a modified, overpowered, PS board. When I hit full amps, 511 as seen on the RTD explorer, it seems to cut way back on power, even through the PWM still says full voltage?
I was running about 1250 amps (seen as 511) at only 50% PWM, for about one second, then power fell off, PWM showed 100%, but it felt like about 25% power.

[thingstodo]

Originally Posted by badfishracing

Is the hardware overcurrent, set by VR1, an actual TRIP/Shutdown, or does it simply cut back the PWM signal?

As far as I know, it is a hardware detection circuit that disables the MOSFET driver, and it latches the over-current, and it displays a fault code, and you need to cycle power to get the Cougar to turn on the MOSFETs again.

Since it is hardware it is *FAST* ... the output from the current sensor is compared to a reference voltage with a comparator. When the voltage from the current sensor is above the reference voltage, the output of the comparator changes and the MOSFET driver is disabled.

I believe that there is a latch in there, but the micro may do the latching.
I'm a bit fuzzy on the details.

Paul will have to answer whether, under some weird set of circumstances, the High Current Fault would NOT latch, and if it didn't latch, how fast the circuit would enable after the current falls off (since the MOSFETs are turned off) before the MOSFETs are turned on again.

But this would not be normal operation!

I'm having an issue with a modified, overpowered, PS board. When I hit full amps, 511 as seen on the RTD explorer, it seems to cut way back on power, even through the PWM still says full voltage?

Modified and over-powered? How are you getting rid of the heat? You need a *LOT* of air moving across the heat sinks to get rid of that much heat.

Heat is proportional to the square of the current. I think that's one of the reasons that electric cars use such high voltages - the power electronics are easier to cool. And the copper wires don't need to be as big. And the contactors don't have to be as big

Is it possible that the modifications messed up the latch on the overcurrent somehow?

I was running about 1250 amps (seen as 511) at only 50% PWM, for about one second, then power fell off, PWM showed 100%, but it felt like about 25% power.

The PWM does not need to be very high at low revolutions to draw high amps. If the back EMF from the motor is low, lots of current can flow. But I guess that's why you built that DC Monster Controller!

What you describe sounds more like a heat issue to me. As the MOSFETs pump the amps, the temperature sensor gets hot, the PWM is limited to reduce the heat .. that sort of thing. I don't know where in the code (before or after the display section) the PWM is backed off.

*BUT*... there are others on this board who have hands-on experience with the Cougar and will give you a better idea on what to chase.

These are only my opinions, after all. I don't have an operational original Cougar 500A (although I did KILL one)

[duncan2]

Hi Paul
Are you selling the high voltage DC controller? - and where can it be bought from?
With the Soliton going away there are people looking for replacements

Mine is still going great but If there is any more advanced software - especially around the ramp up time I would be interested in updating mine before the spring

Thanks
Duncan

[badfishracing]

First off, I apologize for giving such a vague post, when 'thingstodo' posted such an informative and inquisitive reply.

Let me start over.

I have the P&S control board, but instead of the mosfet driver on the board, I run that into dual IGBT drivers. The IGBT drivers drive 6 400A IGBTs. I've derated this to 1200A. Surface mount hall effect sensor, with enough spacers so that the control board sees 511A when the actual motor current is 1200A, roughly. I've got another similar controller running 4x400A IGBTs.

This setup has worked extremely well. About my only issues was some cold solder joints where I was tacking on wires to Paul's beautiful control board.

I've been running this for a few years. This is in a custom pulling tractor. 135V x800A of headway cells. Used to be 170V but I keep losing them. 31x15.5-15 tires. 1900 pound class. 13" series motor chain driven to a ford ranger rearend. Was about 1:1 chain ratio to a 5.13:1 ring/pinion.

About a month ago, I was on a pretty good run, and at the end of the track the rear started clunking. Broke a few teeth off of the pinion gear. I replaced the 5.13:1 setup with a 3.55:1 gear, as thats the quickest/cheapest thing I could find. Changed the sprockets and ended up with a little bit faster final ratio.

Also somehow related, my forward/reverse contactor setup to switch the motor field welded in the forward state. Something must have surged or to weld that contactor AND break the rearend.

Anyways, slapped it back together to compete the next week, and try to stay in the points chase. Tractor took off then then falls flat. I didn't have the computer hooked up, so don't know the details. Did the very same thing the next day. So during the week I took out the forward/reverse contactors. Thought that may have been causing problems or a severe power loss.

Next event, same thing. Next day, finally hooked up the laptop to get a datalog. This time I observed full 511A motor current, but I've limited it to 300A battery current. So the PWM was only like 50%. Then after around one second, the PWM went to 100%, and current bounced around, as observed on the datalog. But in reality, the power just fell flat.
Now know that 511A motor current is actually like 1250. Previously, over the years, I had only seen about 1100A max.

Well, open opening up the controller this morning, I found the VR1 pot was set all the way CCW. So seems like hardware overcurrent protection would come into play if I'm pushing it all the way to 511A, although it may not work exactly like it was originally designed? Power absolutely fell off, and current on the datalog was funny and bouncy.
RECALL that I ended up gearing it a little bit faster than before, so it would logically see more amps than ever before.

So I think I'll turn up the hardware protection, and changing gearing to keep it more inline. That hopefully will do the trick.

[MPaulHolmes]

Hi duncan! Yes I can make changes to the controller code for better launches.
I've since much improved how the PI loop works on the AC controller, and that's directly transferable to the DC controller. I'll make the changes and you can test it out.

bad fish racing: It could be the hardware overcurrent protection. Yes I would twist the pot fully clockwise. What current sensor are you using? How are you doing the current sensing? Is it on just one of the 4 IGBTs?

[thingstodo]

Originally Posted by MPaulHolmes

Hi duncan! Yes I can make changes to the controller code for better launches.
I've since much improved how the PI loop works on the AC controller, and that's directly transferable to the DC controller. I'll make the changes and you can test it out.

bad fish racing: It could be the hardware overcurrent protection. Yes I would twist the pot fully clockwise. What current sensor are you using? How are you doing the current sensing? Is it on just one of the 4 IGBTs?

Welcome back Paul!

Duncan had asked if you were selling kits for the IGBT DC controller. Or perhaps built and tested controllers. If the Solitons are getting rare, this may be a 'sales opportunity' ... if you have the time.

If Duncan (and everyone else!) can play with the launch settings himself he can increase the acceleration until he is happy .. or he breaks something!