Stepper Motor Control with Hercules High-End Timer - Part 7: HET Assembly Language Test

RE: Stepper Motor Control with Hercules High-End Timer - Part 7: HET Assembly Language Test

Jan Cumps — Sun, 05 Jan 2020 17:24:40 GMT

I'm checking some more advanced timer functions (not related to the stepper motor exercise, more for my own learning)

TI has an example to generate dynamic PWM signals and measure timing errors, all done within the timers.

This example is interesting because it tests generating somewhat complex signals (a triangle waveform).

The signal is made by changing the duty cycle od a PWM dynamically. A discrete low pass filter (a resistor and capacitor) turns those pulses into the analog waveform.

[View:/resized-image/__size/617x370/__key/commentfiles/f7d226abd59f475c9d224a79e3f0ec07-5fd0d1b0-d612-4e9c-8e4c-a24673f8dbf5/8712.contentimage_5F00_188533.png:617:370]

Because the PWM signal has different duty cycles during a single wave cycle, this signal is interesting to test if the timer can detect precision of a signal's attributes.

The second part of the timer logic uses this capability. It puts an output pin high (and raises a software trigger too that you can use for counting failures) whe the duty cycle is outside a given range.

You see that in the blue trace.

Everything, except the optional trigger, consumes 0 CPU cycles. Really 0 cycles on the Hercules. I can stop the controller in the debugger and both timer functions will happily continue working.

[View:/resized-image/__size/620x827/__key/commentfiles/f7d226abd59f475c9d224a79e3f0ec07-5fd0d1b0-d612-4e9c-8e4c-a24673f8dbf5/1768.contentimage_5F00_188534.png:620:827]

I'll write a post on this, because TI's example needs some changes to make it work on a LaunchPad that aren't documented ...

RE: Stepper Motor Control with Hercules High-End Timer - Part 7: HET Assembly Language Test

Jan Cumps — Mon, 12 Jun 2017 10:17:09 GMT

Capture of both the SPI and PWM part:

[View:/resized-image/__size/663x427/__key/commentfiles/f7d226abd59f475c9d224a79e3f0ec07-5fd0d1b0-d612-4e9c-8e4c-a24673f8dbf5/6712.contentimage_5F00_188532.png:663:427]

Signals on all lines, but the motor doesn't move yet.

Although I thought I was sending the identical SPI commands from Hercules as those from the official' MSP example, something may be missing.

When I pull the DRV out of sleep, it consumes 13 ma instead of the 600 mA it draws with the MSP430 app...

Reset and nSLeep aren't shown here. I've measured those with a multimeter.

But this is good news by itself: there's no GAP in capabilities - the Hercules can provide the required signals.

RE: Stepper Motor Control with Hercules High-End Timer - Part 7: HET Assembly Language Test

michaelkellett — Sun, 11 Jun 2017 14:50:03 GMT

Useful blog Jan, I see that the HET has been around since 2010 (at least) and this is the first time I've noticed it !!

Motorola (in the days before On and Freescale) offered a complex programmable timer but they never supported it with decent tools so it wasn't used much.

It looks as if TI support this a lot better - I wonder if it might appear on non Hercules ARMs at some time.

RE: Stepper Motor Control with Hercules High-End Timer - Part 7: HET Assembly Language Test

Jan Cumps — Sat, 10 Jun 2017 19:37:20 GMT

Here's a capture of a TI NHET example with dynamic period.

I've compiled it from source and it's running on my controller now, an excellent opportunity to start from working code.

The scope shows how the frequency of the signal dynamically and gradually slows down and speeds up.

[View:/resized-image/__size/620x372/__key/commentfiles/f7d226abd59f475c9d224a79e3f0ec07-5fd0d1b0-d612-4e9c-8e4c-a24673f8dbf5/5633.contentimage_5F00_188529.png:620:372]

RE: Stepper Motor Control with Hercules High-End Timer - Part 7: HET Assembly Language Test

DAB — Sat, 10 Jun 2017 17:49:48 GMT

Another excellent post Jan.

Good detail and screen captures so that the reader can easily follow along.

Well done.

DAB