My application is a direct drive PNP part rotator 
<https://hackaday.io/project/165115/log/162673-rotator-runout>.  AKA... 
basically zero load beyond rotor inertia, and comparatively slow speeds.

The BLDC motor will be used more like a stepper... but a really smooth and 
extremely light one.

I would say there is an 85% chance I'll just be running this open-loop, and 
just following a sine wave for the amplitude (via PWM) of the 3 phases.
Kinda like how microstepping works for a regular stepper for 2 phases 
(ignoring current chopping).. that will be this for 3 phases.
It will work <https://youtu.be/667XSSxieXM>, I just need to generate the 3 
channels of PWM tied to a sine calculation or lookup table, space them 120 
degrees apart, and then tie that to a rotational axis.

This is the minimal implementation and assumes your motor can handle full 
current stalled all day and not overheat.  
A proper implementation would be also PWM one or all of the enable lines.  
Then add a reading on the coil current/back EMF of all 3 phases would give 
you what you need to do field oriented control 
<https://en.wikipedia.org/wiki/Vector_control_(motor)>.

Those could happen, just like a Hall effect rotation sensor could happen if 
I'm desperate, but I think I'll be fine with just open-loop.

I could just build it up with the other components, but I'm inherently lazy 
(usually an asset for a programmer actually) and looking for a canned 
solution before I do so.  
Or I could just add it to the existing bldc component but I'll probably 
screw that up for a while..

Thank you.

On Saturday, April 27, 2019 at 5:06:36 PM UTC-4, justin White wrote:
>
> I run a brushless spindle motor with the BLDC component and a Mesa 8i20. 
> The waveform depends on the personality selected and usually only starts in 
> trapezoidal mode if using hall sensors then switches to sinusoidal. I can't 
> much of the gritty details of setting it up and waveform wasn't a huge 
> thing when I configured my mill. I had more of a fun time getting the PID 
> setup than anything, but that's mostly due to the fact that the 8i20 is a 
> torque mode drive rather than a velocity mode drive. I haven't touched the 
> config in a while as my mill just runs now. 
>
> I'm not sure what "resembling microstepping" means for a brushless DC 
> motor. Alot of the BLDC component relies on the feedback. It's best for 
> running servo's where the drive itself can be dumb and the Linuxcnc 
> hardware/software can handle commutation based on feedback. There's a few 
> modes which can run without feedback but the usefullness depends on what 
> you're doing. 
>
> On Saturday, April 27, 2019 at 4:03:59 PM UTC-4, [email protected] 
> wrote:
>>
>> Got a project where I will be doing pwm of the three phases of a bldc 
>> motor implementing sinesoudial drive for something resembling micro 
>> stepping.  
>> I saw the bldc component but it looks to be a trapezoidal implementation. 
>> Anyone whipped this up before I do?  Basically 3 pwm lines following a 
>> sine wave 120 degrees apart.
>>
>> Thank you.
>>
>

-- 
website: http://www.machinekit.io blog: http://blog.machinekit.io github: 
https://github.com/machinekit
--- 
You received this message because you are subscribed to the Google Groups 
"Machinekit" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to [email protected].
Visit this group at https://groups.google.com/group/machinekit.
For more options, visit https://groups.google.com/d/optout.

Reply via email to