Paul & Sabrina's cheap DIY 144v motor controller

[thingstodo]

Originally Posted by jbman

Well, this is what I'm getting from the controller when I power it on. I verified that were on 19200 and 8N1. Any ideas?

Gibberish from controller

Hmm. ... I verified the serial settings from one of my old videos.

If I didn't know better, I'd say that the ground pin on your serial was not making good contact. The random stuff on the screen looks like about the right number of characters, but not even one of them is correct. It should hopefully be isolated from your laptop so the only place to check is between pins on the serial cable end.

You can check the voltages on the pins. Referenced to the ground pin, the transmit and receive pins should toggle 0 - 5V, which should be OK with any modern USB to serial. Pin 5 is Gnd, as you already know.

If you have a scope handy - it's easier to see the signal level on a scope than on a meter that averages the voltages.

There is a very small chance that your converter is looking for voltages that are not 0V and 5V. The old style RS232 is actually specified as -12V and +12V for signals, and high is over 8V and low is below -8V .... but now I'm digging pretty deep looking for something that might be wrong.

I feel your pain .. went through something similar with my Cougar controller. Never really did get it working very well before I killed the Cougar.

[jbman]

Originally Posted by thingstodo

Hmm. ... I verified the serial settings from one of my old videos.

If I didn't know better, I'd say that the ground pin on your serial was not making good contact. The random stuff on the screen looks like about the right number of characters, but not even one of them is correct. It should hopefully be isolated from your laptop so the only place to check is between pins on the serial cable end.

You can check the voltages on the pins. Referenced to the ground pin, the transmit and receive pins should toggle 0 - 5V, which should be OK with any modern USB to serial. Pin 5 is Gnd, as you already know.

If you have a scope handy - it's easier to see the signal level on a scope than on a meter that averages the voltages.

There is a very small chance that your converter is looking for voltages that are not 0V and 5V. The old style RS232 is actually specified as -12V and +12V for signals, and high is over 8V and low is below -8V .... but now I'm digging pretty deep looking for something that might be wrong.

I feel your pain .. went through something similar with my Cougar controller. Never really did get it working very well before I killed the Cougar.

Unfortunately, I don't have a scope. As far as testing the ground connection, what should I measure from TX and RX to ground when connected and waiting? Thanks!

[thingstodo]

Originally Posted by jbman

Unfortunately, I don't have a scope. As far as testing the ground connection, what should I measure from TX and RX to ground when connected and waiting? Thanks!

Unfortunately, the TX and RX pins 'float' when there is no serial traffic to drive them up or down. You can send data and measure the voltage at TX .. it should be around 2.5V as measured by a digital voltmeter since bits sort of average a 50% duty cycle. Not quite, since there are fixed start and stop bits .. but it's close.

If you send the same thing - like a text file of maybe 10000 of the same character ... you should get the same voltage each time you test, if everything is connected and working. If the numbers seem sort of random, then a pin is not well connected. And Gnd would be my first guess.

[pablitosax]

Hi all :
I'm new in these forum and have a querstion : why not use a simple arduino mega to build a controller and programming these with arduino code (not C/C++) . Isn't enought ?

Regards !

Pablo

[thingstodo]

Originally Posted by pablitosax

Hi all :
I'm new in these forum and have a querstion : why not use a simple arduino mega to build a controller and programming these with arduino code (not C/C++) . Isn't enought ?

Regards !

Pablo

The new arduinos have enough speed. They can do the calculations required to figure out where the rotor is and adjust the voltage and current phases. It is not simple to do or fast to convert the program, but you can do it if you like.

The arduinos don't have a lot of circuitry built in to guard against noise and power issues in a car. You have to add that stuff yourself.

Plus the arduino doesn't really have interrupts in the traditional sense. You can make it do the control. But you should not do communications - blue-tooth, serial, ethernet - at the same time if you want good control. Timing is very critical for motor control.

If you are controlling your table saw or drill press - I'd use arduino. If you are controlling a car or an airplane or a helicopter ... I would not use arduino. Perhaps I am not up to speed with the latest arduino stuff ... but I would not trust my life to an arduino.

[pablitosax]

Thank you very match thingstodo ...very clear explanation.
Yes..the arduino is very sensible/sensitive to noise and 'hangs' or 'reset' itself with small motors or inductive load,etc...
And as all run inside a loop ...all the task are enqueue to do in the same loop so it´s a bad idea to overcharge in the same loop a lot of task (bluetooth,ethernet)...

Are dead links to download the motor contoller of this thread ? DO you have another valid links ?

Regards!!
Pablo.

Originally Posted by thingstodo

The new arduinos have enough speed. They can do the calculations required to figure out where the rotor is and adjust the voltage and current phases. It is not simple to do or fast to convert the program, but you can do it if you like.

The arduinos don't have a lot of circuitry built in to guard against noise and power issues in a car. You have to add that stuff yourself.

Plus the arduino doesn't really have interrupts in the traditional sense. You can make it do the control. But you should not do communications - blue-tooth, serial, ethernet - at the same time if you want good control. Timing is very critical for motor control.

If you are controlling your table saw or drill press - I'd use arduino. If you are controlling a car or an airplane or a helicopter ... I would not use arduino. Perhaps I am not up to speed with the latest arduino stuff ... but I would not trust my life to an arduino.

[thingstodo]

Originally Posted by pablitosax

Are dead links to download the motor contoller of this thread ? DO you have another valid links ?

I would swear that I replied to this already ... must have forgot to hit Submit.

No links, but I have the PCB layout, a partial BOM, and the code for the versions of the boards that I have. I have a DC IGBT version that Paul has said is about 400V and 1400A, and an AC IGBT version that Paul has said is about 400V and 1200A. The big IGBTs will push out more power, but the reflected waves from the motor terminals may destroy them so the 1000V ( I think ) rated IGBT are running at max 400V

Paul's web site is here, and he has the bare boards for sale. You may want to send an email from his site asking about the files. Sabrina can maybe email you the files.

It should be here
https://pandspowerelectronics.ecwid....=0&sort=normal

But it seems to be down right now.

[pablitosax]

Thanks thingstodo :

400v x 1200 A it's a lot !!!! ...appears powerfull as Manzanita Zilla 2k (144v x 2000A) ,really ?
What about the software ? do you have a current limiter by soft ? .Have a video to see the use of the software ?

Regards !

Originally Posted by thingstodo

I would swear that I replied to this already ... must have forgot to hit Submit.

No links, but I have the PCB layout, a partial BOM, and the code for the versions of the boards that I have. I have a DC IGBT version that Paul has said is about 400V and 1400A, and an AC IGBT version that Paul has said is about 400V and 1200A. The big IGBTs will push out more power, but the reflected waves from the motor terminals may destroy them so the 1000V ( I think ) rated IGBT are running at max 400V

Paul's web site is here, and he has the bare boards for sale. You may want to send an email from his site asking about the files. Sabrina can maybe email you the files.

It should be here
https://pandspowerelectronics.ecwid....=0&sort=normal

But it seems to be down right now.

[thingstodo]

Originally Posted by pablitosax

400v x 1200 A it's a lot !!!! ...appears powerfull as Manzanita Zilla 2k (144v x 2000A) ,really ?

I have not gotten the AC controller anywhere near that output. These numbers are a bit of a guess. 400V seems reasonable for the max voltage of the IGBTs. 1200A per phase is more heat than is likely possible to transfer out of the IGBT to the heat sink and out to the glycol. But it should put that out for a little while.

What about the software ? do you have a current limiter by soft ? .Have a video to see the use of the software ?

There is a hardware limit, and a software limit. I have to dig out the test video to refresh my memory. I think I hit the hardware limit when it was adjusted as low as it would go - maybe - 350A or so?

https://www.youtube.com/watch?v=eImd...&index=64&t=0s

The hardware limit is not easy to calibrate. I was expecting the limit to be at least 400A.

[jackbauer]

Are the schematics and layout files available anywhere? Most of the links on the Wiki seem dead.