Spindle synchronisation: is it working the way it should?
- IIOII
- Offline
- New Member
Less
More
- Posts: 8
- Thank you received: 0
19 Dec 2017 01:03 #103325
by IIOII
Spindle synchronisation: is it working the way it should? was created by IIOII
I've recently added a quadrature encoder (specs below) to my lathe and updated the HAL to support spindle synchronisation and spindle-at-speed functionality. I have also updated AXIS to display the PYVCP spindle speed panel. The PYVCP panel is displaying spindle speed and when at speed, triggers the LED. The system seems to be working correctly however I have two questions:
1. The filtered spindle output speed jumps around a bit (+/-20 rpm) when the spindle is run at full speed (2030 rpm). Is this considered 'acceptable' variance (noise)? If not, is there a variable to tune the filtered rate?
2. The spindle output does not exactly match the requested speed. For example, M3 S1000 will result in a spindle speed of around 1040rpm. I assume that spindle syncronisation should change the output RPM to match commanded. Is this thinking correct? If so, any suggestions as to why my configuration doesn't achieve this?
Installed version: 2.7.11
Mesa 7i76e
0-10V driving VFD (Teco L510)
Encoder: au.element14.com/webapp/wcs/stores/servl...467475&storeId=10184
1. The filtered spindle output speed jumps around a bit (+/-20 rpm) when the spindle is run at full speed (2030 rpm). Is this considered 'acceptable' variance (noise)? If not, is there a variable to tune the filtered rate?
2. The spindle output does not exactly match the requested speed. For example, M3 S1000 will result in a spindle speed of around 1040rpm. I assume that spindle syncronisation should change the output RPM to match commanded. Is this thinking correct? If so, any suggestions as to why my configuration doesn't achieve this?
Installed version: 2.7.11
Mesa 7i76e
0-10V driving VFD (Teco L510)
Encoder: au.element14.com/webapp/wcs/stores/servl...467475&storeId=10184
Please Log in or Create an account to join the conversation.
- PCW
- Away
- Moderator
Less
More
- Posts: 17904
- Thank you received: 4774
19 Dec 2017 08:05 - 19 Dec 2017 08:12 #103333
by PCW
Replied by PCW on topic Spindle synchronisation: is it working the way it should?
1. Small spindle speed reading variations are normal, largely due to the fast sampling time of the velocity signal. The usual fix is to use linuxCNC's lowpass filter component to filter the speed data for RPM display. Your hal file already has this "wiring"
but the filter is disabled:
setp lowpass.spindle.gain 1.000000
( a lowpass gain of 1.00 disables the filter ) I would suggest perhaps a 1 second time constant
setp lowpass.spindle.gain 0.0063
man lowpass for more information
2. Its expected that you will have small variations in commanded and actual spindle speed unless you have a closed
loop spindle control. Again it looks like your hal file has this "wiring" but the PID parameters have only feedforward
(FF0=1.0) set and no feedback terms (all 0), so you are running open loop. Note that for spindle synchronized moves, the
actual spindle speed is not very important because the synchronization does not control the spindle speed, but rather synchronizes the axis position with the spindle position so the axis motion will track quite large spindle speed variations.
but the filter is disabled:
setp lowpass.spindle.gain 1.000000
( a lowpass gain of 1.00 disables the filter ) I would suggest perhaps a 1 second time constant
setp lowpass.spindle.gain 0.0063
man lowpass for more information
2. Its expected that you will have small variations in commanded and actual spindle speed unless you have a closed
loop spindle control. Again it looks like your hal file has this "wiring" but the PID parameters have only feedforward
(FF0=1.0) set and no feedback terms (all 0), so you are running open loop. Note that for spindle synchronized moves, the
actual spindle speed is not very important because the synchronization does not control the spindle speed, but rather synchronizes the axis position with the spindle position so the axis motion will track quite large spindle speed variations.
Last edit: 19 Dec 2017 08:12 by PCW.
Please Log in or Create an account to join the conversation.
- IIOII
- Offline
- New Member
Less
More
- Posts: 8
- Thank you received: 0
22 Dec 2017 00:31 #103447
by IIOII
Replied by IIOII on topic Spindle synchronisation: is it working the way it should?
Thanks PCW. Adjusting the lowpass gain to your recommended value worked perfectly.
Please Log in or Create an account to join the conversation.
Time to create page: 0.066 seconds