On 11/15/2013 02:05 AM, andy pugh wrote:
 > However, if the motor can move a significant proportion of a
 > "commutation period' in one servo thread, from rest, then software
 > commutation is not likely to be successful anyway.

You are right, but I don't agree to all 100%. I did run motors with BLDC 
component successfully. But success was much higher when testing motor 
with some load than free motor itself. And this could be explained by 
the bldc.init nuances. As I wrote, when bldc.init starts (I worked with 
qi mode), it starts blindly, so at the first moment the rotor adapts to 
magnetic field, causing big vibrations. Yes, they do decay later, but 
what happens when an index is nearby? Init ends right at the peak of 
vibrations - quite high velocity and accelerations at this moment.

Running free motors is convenient for testing and tuning, finding right 
encoder-offset.

And the servo thread of my test-setup was running much faster than 1 
kHz. I tested it successfully up to 5 kHz.

Another reason to get true measurement of encoder-offset is that even 
several counts of encoder makes significant influence in results. It may 
mean tripping over-current protection on the drive or not, which can 
result in servo following error. Or it may let the system (pid-output) 
run smoothly or not.

That's why I would like for bldc.init to work correctly or at least 
better. Thus, I am asking for you if you know what and how could be 
corrected in the code?


------------------------------------------------------------------------------
DreamFactory - Open Source REST & JSON Services for HTML5 & Native Apps
OAuth, Users, Roles, SQL, NoSQL, BLOB Storage and External API Access
Free app hosting. Or install the open source package on any LAMP server.
Sign up and see examples for AngularJS, jQuery, Sencha Touch and Native!
http://pubads.g.doubleclick.net/gampad/clk?id=63469471&iu=/4140/ostg.clktrk
_______________________________________________
Emc-developers mailing list
[email protected]
https://lists.sourceforge.net/lists/listinfo/emc-developers

Reply via email to