MSP432 and TI-RTOS: PID Library Part 1 - Intro

RE: MSP432 and TI-RTOS: PID Library Part 1 - Intro

fmilburn — Thu, 02 Mar 2017 23:02:37 GMT

Jan,

I have worked my way through your examples up to here and they are great! I did skip the one on DACs because I don't have that part but it was easy to follow.

Regarding your comment on this particular PID code:

You should (really: should) read Brett's blog before continuing here: Improving the Beginner’s PID – Introduction « Project Blog.

I fully agree. It is a great tutorial with some very good links behind it as well. Sometime back I built a sous vide controller using Brett's library and placed everything in an enclosure. It controls a slow cooker not shown in the photo. The temperature probe is inside a broken off whisk bought from the dollar store so as to isolate it from the walls of the cooker and food. There is no mains inside the enclosure - I use an external relay to drive the heating element in the cooker. Being a mechanical engineer, that seemed the safe thing to do [View:/resized-image/__size/16x16/__key/commentfiles/f7d226abd59f475c9d224a79e3f0ec07-11ba5e54-da32-447c-8b71-1617e8faf227/5722.contentimage_5F00_3.png:16:16]

[View:/resized-image/__size/620x551/__key/commentfiles/f7d226abd59f475c9d224a79e3f0ec07-11ba5e54-da32-447c-8b71-1617e8faf227/3678.contentimage_5F00_205636.jpg:620:551]

This is of course a very slow responding system. It was a great learning experience and fun playing with the tuning parameters. In the graph below I reset the temperature part way through to see how it responded. There was a bit of overshoot but it settles to less than 0.5 F error quickly.

[View:/resized-image/__size/620x222/__key/commentfiles/f7d226abd59f475c9d224a79e3f0ec07-11ba5e54-da32-447c-8b71-1617e8faf227/6811.contentimage_5F00_205637.jpg:620:222]

Anyways, on to your example. Thanks again for the series.

RE: MSP432 and TI-RTOS: PID Library Part 1 - Intro

Jan Cumps — Tue, 24 Jan 2017 21:02:07 GMT

A few sneak previews of the library in action.

- MSP432 generates PWM, and an RC filter converts that to DC.

- MSP432 ADC samples that DC

- PID changes PWM duty cycle so that the sampled value matches a preset level.

[View:/resized-image/__size/802x601/__key/commentfiles/f7d226abd59f475c9d224a79e3f0ec07-11ba5e54-da32-447c-8b71-1617e8faf227/8156.contentimage_5F00_205633.png:802:601]

two captures with different PID parameters.

proportional 1, integral 1, differential 0, steady

[View:/resized-image/__size/804x482/__key/commentfiles/f7d226abd59f475c9d224a79e3f0ec07-11ba5e54-da32-447c-8b71-1617e8faf227/8078.contentimage_5F00_205634.png:804:482]

proportional 1, integral 0, differential 0: nervous

[View:/resized-image/__size/805x482/__key/commentfiles/f7d226abd59f475c9d224a79e3f0ec07-11ba5e54-da32-447c-8b71-1617e8faf227/7573.contentimage_5F00_205635.png:805:482]

in part 2 I'll post the firmware and the RC filter.

RE: MSP432 and TI-RTOS: PID Library Part 1 - Intro

DAB — Sun, 22 Jan 2017 20:04:51 GMT

Nice post Jan.

It reminds me of some of the control algorithms I developed in the mid 1980's to control a FLIR device and slave it to the movement of a helicopter.

I had to constantly update the FLIR look angle depending upon the attitude changes made by the pilot.

Very challenging, especially since I only had an Intel 8085 process to work with.

DAB

RE: MSP432 and TI-RTOS: PID Library Part 1 - Intro

Jan Cumps — Sun, 22 Jan 2017 14:04:41 GMT

This part of the blog series is also a good excuse to try out the LED board that [mention:4f57fc9d538949ad9eb336ddb9469bb8:e9ed411860ed4f2ba0265705b8793d05] gave me (I've gone medieval on the PSU [View:/resized-image/__size/16x16/__key/commentfiles/f7d226abd59f475c9d224a79e3f0ec07-11ba5e54-da32-447c-8b71-1617e8faf227/4705.contentimage_5F00_1.png:16:16] ). I'm going to train myself on the internals of the LIB with that.

(I took a specialisation year on control theory in 1985-86, but I've forgotten more than 100% of it)

[View:/resized-image/__size/620x465/__key/commentfiles/f7d226abd59f475c9d224a79e3f0ec07-11ba5e54-da32-447c-8b71-1617e8faf227/3386.contentimage_5F00_205632.jpg:620:465]

The project that I'm going to control with the PID lib has i2c as input and output. This board has analog out and PWM at the input.

That will force me to keep the lib independent of the device and have it as a reusable entity, as intended by the original creator).

Also because it has 3 controllable circuits, I can use it to extend the library to support more than 1 control strategy, and see if I can switch between them.