"Peter C. Wallace" <[EMAIL PROTECTED]> wrote:
> What is happening is this:

Thanks for the explanation, that helped a lot, I think I got most of it.
:-)


> To solve your problem there are a couple of things to try:
> 
> 1. Live with it and offset you PDM

As you say, it actually works ok with the deadband.  I tuned the PID loop
and I get positional errors of up to 0.067 revolutions (25 degrees),
this is with no load on the motor shaft.  With a 1:1 coupling from the
motor shaft to a 20 tpi leadscrew that gives me +-0.004" positioning
precision (not counting mechanical slop, backlash, etc).


> 2. Use PWM instead of PDM (If the software PWM rate is not too low). The 
> ripple will be higher, but the average current in the 0 to 50% duty cycle 
> range will be will be higher as well.

I'll give PWM a try.  My pulse-generating thread runs at 20 KHz currently,
does a PWM carrier frequency of 2 KHz sound reasonable?  That'd give me
10 levels of power.


> 3. Setup your HBridges to use 'Slow decay mode' if possible. This usually 
> means that the bottom 2 switch elements are turned on in the off part of the 
> PWM cycle, (shorting out the motor) causing the stored energy on the motor 
> inductance to be 'saved' between PWM cycles. This results in lower ripple, 
> and 
> more linear current versus duty cycle.

I thought "both bottom switches on" was active braking.  How does
this help?

I don't get it, but I'll give it a try too.  :-)


> 4. Use an HBridge with current feedback

Hm, I need to read up on motor drivers...


Thanks for all your advice, I appreciate it!


-- 
Sebastian Kuzminsky

-------------------------------------------------------------------------
This SF.net email is sponsored by: Splunk Inc.
Still grepping through log files to find problems?  Stop.
Now Search log events and configuration files using AJAX and a browser.
Download your FREE copy of Splunk now >> http://get.splunk.com/
_______________________________________________
Emc-users mailing list
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users

Reply via email to