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

[thingstodo]

Originally Posted by MPaulHolmes

Also, if you type "save" it will save the settings I hope. I haven't checked that it properly saves and loads everything with all the new settings available. So, hopefully you only have to set up the throttle once and then never again.

The save command does not appear to work. I get the 'PC Load Letter' error

[thingstodo]

Another update from Dec 6

https://youtu.be/tuJvQN6Kt3Y

This one is an AC motor test

[thingstodo]

The last update from Dec 6 edited so far

https://youtu.be/Lgn_WlgIei0

Attached are ACConfig.txt file, save-test-5.log, save-test-4.log in zipped file

EDIT - OK - NOW you can see the video ... after I made it public ...

[MPaulHolmes]

Wow! 200amp!!! We are really starting to get up there!!!!! I could never have hoped for a beta tester like you. Good job! Here's a graph of the last file with the ton of things being streamed:



Notice how nicely Id and Iq are tracking the throttle position (throttle is basically IqRef)!! Your Kp and Ki seem to be just right. You can see where the regen is pegged right at -200amp (in the middle of the graph). Do you know what happened at the end? Was the throttle moving all around or was something weird happening at the end of the test?

[e*clipse]

Originally Posted by P-hack

ok, *maybe* fire, brand name and $50 and supported by cordova (get out of my head ricardo montleban!) and a2dl bluetooth. But I'm not wild about having to root it to get to the google playstore.

cordova comes with rich corinthian apache support, instead of power hungry adobe support for phonegap. I'll play with it a bit more. ios might work too.
https://cordova.apache.org/docs/en/l...ort/index.html


will start here:
https://www.npmjs.com/package/cordov...uetooth-serial
(note the hardware requirements for IOS)

Sorry about this post - I've been away from the computer for a couple weeks. We took a train trip across Canada. I waved to thingstodo as we went by, but it was a bit early in the morning. LOL!

I really like this discussion of potential "dash boards." It sounds like there are a lot of potentials out there in the $50 range. I also found some refurbished industrial ones in that price range.

It seems to me it's long-term compatibility is one of the most important issues. For that reason, I would prefer something that uses a pretty standard operating system like Linux or an older version of windoze. I completely agree that Apple's attitude would make it difficult to work with.
Other things that would be nice:
1) Outdoor viewable screen
2) Some level of waterproofing
3) Some ability to withstand automotive temperature ranges
4) A standard wired connector, such as USB

Just my $0.02

- E*clipse

[dave koller]

Someone say Dashboard?

[thingstodo]

Originally Posted by MPaulHolmes

Do you know what happened at the end? Was the throttle moving all around or was something weird happening at the end of the test?

Yep - I was playing around ... running 3000 - 4000 rpm forward then going full regen to brake and spin reverse to about the same speed .. then accelerate to forward.

I should have graphed the log file data - it looks good!

I have not finished editing that part of the video yet. It's coming!

Did the rest of the data come out OK? The data was hard to read on the screen ...

[thingstodo]

Last update for Dec 6 - the playing around I mentioned

https://youtu.be/5m9Ze0L70uQ

Change for next set of tests:
- increase max AC amps to 600 on accel and regen
- wire in Emergency stop slap switches in case things get out of hand
- work on cooling for motor and controller

[MPaulHolmes]

All of the data looked good! There were no bad transmissions it appears. All the columns had all of their numbers. Here's a graph of the change in temperature over the last 20 seconds. The most the temperature changed in a 20 second period was 8 degC. When the temperature change goes negative, that just means that in that 20 second period, the temperature was dropping.


In converting it to degC, I started at 25degC as the lowest it will go, and then went up by 2's. It increments 25degC, 27degC, 29degC, ...,. It was fast to do a binary search algorithm with 63 elements in the array. I could have done 25degC, 26degC, ..., 25 + 63degC, but I wasn't sure if that would be a wide enough temperature range.

To test it to 600ampRMS, you will have to change a couple resistor values. The hardware overcurrent trip point is set to an instantaneous current of 600amp on any of the 3 phases. That may have been a little too paranoid, since they are 600amp IGBTs, for crying in the night. haha. I think 400ampRMS may not trip them. That would be 565amp peak, but it may trip since remember how the Id and Iq currents were sort of jaggedy when they were following IdRef and IqRef... The current sensor can measure up to +/-900amp, so we could just change the resistor divider trip point a bit. Maybe 750amp instantaneous peak? Then at least you could test 400ampRMS safely? I don't know what resistor to add. That would take a couple minutes of figgerin'. You coudl just add a resistor in parallel with one that's alrady there. You woudl need 2 of the same value resistors.

[thingstodo]

Originally Posted by MPaulHolmes

To test it to 600ampRMS, you will have to change a couple resistor values.

OK - I will have to add that to the list

The hardware overcurrent trip point is set to an instantaneous current of 600amp on any of the 3 phases. That may have been a little too paranoid, since they are 600amp IGBTs, for crying in the night. haha. I think 400ampRMS may not trip them. That would be 565amp peak, but it may trip since remember how the Id and Iq currents were sort of jaggedy when they were following IdRef and IqRef...

I will do my best to 'test' the trip point

The current sensor can measure up to +/-900amp, so we could just change the resistor divider trip point a bit. Maybe 750amp instantaneous peak? Then at least you could test 400ampRMS safely? I don't know what resistor to add. That would take a couple minutes of figgerin'. You coudl just add a resistor in parallel with one that's alrady there. You woudl need 2 of the same value resistors.

I'm sure we'll discuss this quite a bit more before I get around to soldering in extra resistors.