Calculate Scaling of Servo Encoders in PncConf bug

More
10 Oct 2011 13:54 - 10 Oct 2011 14:19 #13800 by NICKKINSMAN
Maybe this is known, but in case not wanted to let everyone know. This is for EMC version 2.4.6.X (x=32 in my case)

It appears to me the encoder scaling setup in PnCConf is wrongly performing the gearing calculation for setting up the encoder feeback counts per revolution.

This seems to be the case for any axis, and it is where you click the 'calculate' button to input the servo and encoder details. I think the error is just that the gear ratio between servo and leadscrew teeth is calculated inveserly. I am using 5 TPI leadscrew, 2000 count encoders, 14 teeth on motor and 28 teeth on leadscrew. By hand calc, this should be 80,000 counts. When you plug it in that way, as described, the calc button only gets you to some number (I can't recall) but nowhere near 80,000.

When you swap the motor and leadscrew teeth entries (opposite what it says on the program), then PnCConf calculates it exactly to the 80,000, and it checks out physically on the machine. So I think the backend calculation is inverted on the ratio.

Easy fix to this is just to change the labels and description in PnCConf so the numbers get in the right spot!

Anyway, hope this helps!

Nick
Last edit: 10 Oct 2011 14:19 by NICKKINSMAN.

Please Log in or Create an account to join the conversation.

More
10 Oct 2011 19:34 #13814 by cmorley
yes this is another bug that has been fixed just not released.
It is just as frustrating for me.:)
In fact there has been about 95 commits to 2.4.6 to fix EMC bugs!
The good news is I have heard interest of releasing 2.4.7.

If your interested look here:
git.linuxcnc.org/gitweb?p=emc2.git;a=sho...fs/heads/v2.4_branch
everything above the yellow 2.4.6 release tag will be fixed in 2.4.7

Please Log in or Create an account to join the conversation.

Moderators: cmorley
Time to create page: 0.062 seconds
Powered by Kunena Forum