Paul and Sabrina's Cheap 3 Phase Inverter (AC Controller) with Field Oriented Control

[MPaulHolmes]

Ya, you are going to have to have no resistance on the lines. And You can still tune the PI loop with as low as 12vDC bus. I wouldn't do testing with 300vDC.

[shaggythegangsta]

Originally Posted by MPaulHolmes

Ya, you are going to have to have no resistance on the lines. And You can still tune the PI loop with as low as 12vDC bus. I wouldn't do testing with 300vDC.

My next tests are going to be with 48v with no bulbs in series

I think you know that I'm running the code with the dspic30f6010a .
I have another question that my encoder is 1024 ppr but I have to configure with 2048 tick per Rev. To get correct speed...
In the code I defined generic Ac induction motor

[MPaulHolmes]

Type the following in realterm (or whatever serial streaming software you are using):
stream-poscnt 1
data-stream-period 100

Then, make one complete revolution (turning the motor slowly by hand), noting the starting value for poscnt and the ending value for poscnt.

The difference between the 2 values for poscnt will be 4 times the resolution of your encoder.

[shaggythegangsta]

Originally Posted by MPaulHolmes

Type the following in realterm (or whatever serial streaming software you are using):
stream-poscnt 1
data-stream-period 100

Then, make one complete revolution (turning the motor slowly by hand), noting the starting value for poscnt and the ending value for poscnt.

The difference between the 2 values for poscnt will be 4 times the resolution of your encoder.

ticks per rev: 1024 clockwise rotation
00000,
00000,
00000,
00001,
00002,
00000,
00001,
00003,
00001,
00002,
00003,
00000,
00002,
00004,
00000,
00001,
00002,
00000,
00002,
00003,
00000,
00001,
00002,
00000,

anticlockwise:
00000,
65535,
65534,
00000,
65535,
65533,
00000,
65534,
65532,
00000,
65533,
65531,
65535,
65533,
65535,
65534,
65532,
00000,
65534,
65533,
65535,
65534,
65534,
00000,
00000,

ticks per rev: 2048 clockwise rotation

00000,
00000,
00000,
00001,
00001,
00002,
00000,
00002,
00002,
00001,
00002,
00000,
00001,
00003,
00000,
00002,
00004,
00000,
00001,
00002,
00001,
00001,
00003,
00000,
00001,
00003,
00000,
00001,
00003,
00000,
00000,
00000,
00000,

anticlockwise:

00000,
00000,
65529,
65534,
65531,
00000,
65534,
65532,
00000,
65534,
65532,
65535,
65534,
65535,
65535,
65534,
65533,
65534,
65534,
00000,
65535,
65534,
00000,
00000,

[MPaulHolmes]

OK, the encoder isn't working. The encoder ticks per revolution setting doesn't matter for this part. This should just be recording the ticks the controller is receiving from the encoder. And it's staying at 0 +/- 3 ticks.

[shaggythegangsta]

Originally Posted by MPaulHolmes

OK, the encoder isn't working. The encoder ticks per revolution setting doesn't matter for this part. This should just be recording the ticks the controller is receiving from the encoder. And it's staying at 0 +/- 3 ticks.

I will check and let you know
But signals with oscilloscope are OK on encoder... What is causing the problem

[MPaulHolmes]

I would check the pinout. Like, is qea going to index, or whatever. and do the signals go from 0 to 5v? index is ignored, by the way.

[shaggythegangsta]

Originally Posted by MPaulHolmes

I would check the pinout. Like, is qea going to index, or whatever. and do the signals go from 0 to 5v? index is ignored, by the way.

hello paul i just tested the encoder... it is working but i noticed that output A,B of encoder gives me 4 volt and on dspic pins we have: 0 - 3.8 volt.. is it ok?

[thingstodo]

Build log for my DC controller

March 16.

We got the various test stuff moved out to the golf cart:
- battery cables
- motor cables
- contactor
- controller
- throttle
- laptop to program the controller

We got everything checked, loaded up the debug code
did a first spin
then loaded the production code
repeated the spin
did a first run with the golf cart
did a speed run with max speed at 24V, at 100A max battery amps
decided not to mess with parameters, tuning
Paul released version 1.10 - stable release
I think he`s posting to Github

Raw video - before editing
https://youtu.be/xXwJUlZ6tFM

WE HAVE A SUCCESSFUL TEST!!!

[MPaulHolmes]

Hurray! It worked! Michael is so much fun to work with! Notice that the power section is identical to the AC controller. The only difference is what board plugs into the IGBTs.

shaggy:
4v? Are you powering the control board with 4 volts?