cincinnati axis spindle question
- PCW
-
- Away
- Moderator
-
Less
More
- Posts: 18537
- Thank you received: 5081
06 Mar 2013 00:01 #30930
by PCW
Replied by PCW on topic cincinnati axis spindle question
A DC motor can be a servo motor depending on the drive
(Many if not most 1970's through 1980s CNC systems used DC servo motors)
(Many if not most 1970's through 1980s CNC systems used DC servo motors)
The following user(s) said Thank You: jlviloria
Please Log in or Create an account to join the conversation.
- jlviloria
- Offline
- Elite Member
-
Less
More
- Posts: 311
- Thank you received: 12
06 Mar 2013 00:46 #30931
by jlviloria
Replied by jlviloria on topic cincinnati axis spindle question
and in my case:
PLUS MAXPAK VS RELIANCE DRIVE SPINDLE
RELIANCE RPM III DC MOTOR.
believe that functions as a servomotor?
thanks
PLUS MAXPAK VS RELIANCE DRIVE SPINDLE
RELIANCE RPM III DC MOTOR.
believe that functions as a servomotor?
thanks
Please Log in or Create an account to join the conversation.
- andypugh
-
- Away
- Moderator
-
Less
More
- Posts: 23316
- Thank you received: 4945
06 Mar 2013 00:51 #30932
by andypugh
It has 4-quadrant operation and (optionally) tacho feedback, so it ought to.
Do you have an encoder on the spindle?
Replied by andypugh on topic cincinnati axis spindle question
believe that functions as a servomotor?
It has 4-quadrant operation and (optionally) tacho feedback, so it ought to.
Do you have an encoder on the spindle?
Please Log in or Create an account to join the conversation.
- jlviloria
- Offline
- Elite Member
-
Less
More
- Posts: 311
- Thank you received: 12
06 Mar 2013 02:08 #30934
by jlviloria
Replied by jlviloria on topic cincinnati axis spindle question
Yes Andy, also has a sensor that indicates the position for the orientation. something that worries me is the backlash, as can fit if a spindle is always going to have some mismatch?
thanks
thanks
Please Log in or Create an account to join the conversation.
- andypugh
-
- Away
- Moderator
-
Less
More
- Posts: 23316
- Thank you received: 4945
06 Mar 2013 08:05 - 06 Mar 2013 21:35 #30946
by andypugh
Replied by andypugh on topic cincinnati axis spindle question
This is the HAL file I have been experimenting with on my lathe. It allows the spindle to be used both on the normal mode, and as an axis.
To switch between modes you should define two custom M-codes.
M101
!#/bin/bash
halcmd sets position-mode 1
M102
!#/bin/bash
halcmd sets position-mode 0
Initially I used the hal "orient" function, but it isn't quite right for the job, and I am not sure it is in the 2.5 release yet. This version of the HAL file uses conversion to int and a sample-hold to round-down to the nearest full rev of the encoder.
Some parts of the file are very specific to my 7i73 configuration. These will obviously need to be edited.
I have set the HAL file up using names= rather than count= for all components. This means that the file can be a completely separate HAL file is TWOPASS is set in the INI file.
However, this is only possible if the other HAL files use names= too.
It isn't that hard to swap. You can just change (for example) "loadrt and2 count=2" to "loadrt and2 names=and2.0, and2.1" and all the pin names stay the same in the rest of the file.
The HAL file references parameters in the INI file under a new [SPINDLE] section. This needs to exist, and contain the patameters.
The HAL file provided closed loop control of both speed and position.
I don't think position feedback on the C-axis is right in velocity mode, you may need to set the f-error to more than 180 degrees. (The C axis is graduated in degrees, the conversion is done using the often-overlooked gain pins in the sum2 component).
This code is for a VFD with 0 to 10V control and FWD REV buttons. If the drive uses -10 to +10V then the ABS function can be skipped.
This code is very unlikely to work on any machine without editing.
To switch between modes you should define two custom M-codes.
M101
!#/bin/bash
halcmd sets position-mode 1
M102
!#/bin/bash
halcmd sets position-mode 0
Initially I used the hal "orient" function, but it isn't quite right for the job, and I am not sure it is in the 2.5 release yet. This version of the HAL file uses conversion to int and a sample-hold to round-down to the nearest full rev of the encoder.
Some parts of the file are very specific to my 7i73 configuration. These will obviously need to be edited.
I have set the HAL file up using names= rather than count= for all components. This means that the file can be a completely separate HAL file is TWOPASS is set in the INI file.
However, this is only possible if the other HAL files use names= too.
It isn't that hard to swap. You can just change (for example) "loadrt and2 count=2" to "loadrt and2 names=and2.0, and2.1" and all the pin names stay the same in the rest of the file.
The HAL file references parameters in the INI file under a new [SPINDLE] section. This needs to exist, and contain the patameters.
The HAL file provided closed loop control of both speed and position.
I don't think position feedback on the C-axis is right in velocity mode, you may need to set the f-error to more than 180 degrees. (The C axis is graduated in degrees, the conversion is done using the often-overlooked gain pins in the sum2 component).
This code is for a VFD with 0 to 10V control and FWD REV buttons. If the drive uses -10 to +10V then the ABS function can be skipped.
This code is very unlikely to work on any machine without editing.
#Spindle
loadrt pid names=pos_pid,vel_pid
loadrt mux2 names=mode_sw
loadrt abs names=s_abs
loadrt and2 names=s_interlock
loadrt scale names=rps-to-rpm
loadrt not names=not_pos
loadrt or2 names=s_req
loadrt sample_hold names=pos_latch
loadrt conv_float_s32 names=conv.0
loadrt conv_s32_float names=conv.1
loadrt sum2 names=latched_pos
addf mode_sw servo-thread
addf pos_pid.do-pid-calcs servo-thread
addf vel_pid.do-pid-calcs servo-thread
addf s_abs servo-thread
addf rps-to-rpm servo-thread
addf s_interlock servo-thread
addf s_req servo-thread
addf pos_latch servo-thread
addf conv.0 servo-thread
addf conv.1 servo-thread
addf latched_pos servo-thread
addf not_pos servo-thread
setp hm2_7i43.0.gpio.008.is_output 1
setp hm2_7i43.0.gpio.008.is_opendrain 1
net spindle-forward => hm2_7i43.0.gpio.008.out
setp hm2_7i43.0.gpio.010.is_output 1
setp hm2_7i43.0.gpio.010.is_opendrain 1
net spindle-reverse => hm2_7i43.0.gpio.010.out
setp lowpass.0.gain 0.02
setp hm2_7i43.0.encoder.01.scale [SPINDLE]SCALE
setp hm2_7i43.0.pwmgen.00.output-type 1
setp hm2_7i43.0.pwmgen.pwm_frequency 350
setp hm2_7i43.0.pwmgen.00.scale 1
setp vel_pid.maxoutput 1
setp pos_pid.maxoutput 1
setp vel_pid.Pgain [SPINDLE]V_PGAIN
setp vel_pid.Igain [SPINDLE]V_IGAIN
setp vel_pid.FF0 [SPINDLE]V_FF0
setp pos_pid.Pgain [SPINDLE]P_PGAIN
setp pos_pid.Igain [SPINDLE]P_IGAIN
setp pos_pid.Dgain [SPINDLE]P_DGAIN
setp pos_pid.FF1 [SPINDLE]P_FF1
net spindle-vel-cmd motion.spindle-speed-out-rps => vel_pid.command
net spindle-speed-rps hm2_7i43.0.encoder.01.velocity => scale.0.in motion.spindle-speed-in
net spindle-speed-rps vel_pid.feedback
#Latch the spindle position in whole revs
net spindle-pos hm2_7i43.0.encoder.01.position => motion.spindle-revs
net spindle-pos conv.0.in latched_pos.in0
net spindle-int conv.0.out pos_latch.in
net spindle-int-out pos_latch.out conv.1.in
#subtract the whole revs from the position
net spindle-int-conv conv.1.out latched_pos.in1
setp latched_pos.gain1 -360.0
setp latched_pos.gain0 360
net spindle-pos-cmd axis.5.motor-pos-cmd => pos_pid.command
net spindle-pos-fb latched_pos.out => axis.5.motor-pos-fb pos_pid.feedback
net vel-cmd vel_pid.output => mode_sw.in0
net pos-cmd pos_pid.output => mode_sw.in1
net sw-cmd mode_sw.out => s_abs.in
net vfd-cmd s_abs.out => hm2_7i43.0.pwmgen.00.value
net spindle-forward s_abs.is-positive
net spindle-reverse s_abs.is-negative
#This signal should be set by halcmd or button or M100 command to set pos pode
net position-mode mode_sw.sel pos_pid.enable not_pos.in pos_latch.hold s_req.in0
net vel-mode not_pos.out vel_pid.enable
net spindle-index motion.spindle-index-enable => hm2_7i43.0.encoder.01.index-enable
net spindle-on-vel motion.spindle-on => s_req.in1
net guard-up <= hm2_7i43.0.gpio.020.in_not => s_interlock.in0
net spindle-on <= s_req.out => s_interlock.in1
net spindle-enable <= s_interlock.out => hm2_7i43.0.pwmgen.00.enable
Last edit: 06 Mar 2013 21:35 by andypugh. Reason: removal of stray lines
The following user(s) said Thank You: jlviloria
Please Log in or Create an account to join the conversation.
- mariusl
-
- Offline
- Platinum Member
-
30 Jul 2013 02:41 #37246
by mariusl
Regards
Marius
www.bluearccnc.com
Replied by mariusl on topic cincinnati axis spindle question
Andy
Do I understand this correct. You set the spindle up as a rotary axis "C" and effectively switch between a velocity and position PID for this axis?
Do I understand this correct. You set the spindle up as a rotary axis "C" and effectively switch between a velocity and position PID for this axis?
Regards
Marius
www.bluearccnc.com
Please Log in or Create an account to join the conversation.
- andypugh
-
- Away
- Moderator
-
Less
More
- Posts: 23316
- Thank you received: 4945
30 Jul 2013 05:34 #37248
by andypugh
That's the idea.
I have realised that a lot of it is much simpler if you set it as a wrapped rotary axis. You don't need the latches and such (I don't think).
Bear in mind that I don't really have suitable hardware for testing this arrangement.
Replied by andypugh on topic cincinnati axis spindle question
Andy
Do I understand this correct. You set the spindle up as a rotary axis "C" and effectively switch between a velocity and position PID for this axis?
That's the idea.
I have realised that a lot of it is much simpler if you set it as a wrapped rotary axis. You don't need the latches and such (I don't think).
Bear in mind that I don't really have suitable hardware for testing this arrangement.
Please Log in or Create an account to join the conversation.
- mariusl
-
- Offline
- Platinum Member
-
30 Jul 2013 13:36 #37254
by mariusl
Regards
Marius
www.bluearccnc.com
Replied by mariusl on topic cincinnati axis spindle question
I am just finishing of my mill conversion and then I will start with the lathe. I will most certainly try this out as it will save me a whole lot of effort and money. I was planning on putting a clutch on the spindle to enable be to engage a stepper motor to position the chuck.
Regards
Marius
www.bluearccnc.com
Please Log in or Create an account to join the conversation.
- Rick G
-
- Offline
- Junior Member
-
Less
More
- Posts: 26
- Thank you received: 155
02 Aug 2013 17:47 #37351
by Rick G
Replied by Rick G on topic cincinnati axis spindle question
Andy,
This sounds very interesting.
Are you saying you are using a VFD to drive an AC motor on the spindle for positional control (rotary axis with an encoder)?
How about heat and torque in this mode?
Rick G
This code is for a VFD with 0 to 10V control and FWD REV buttons. If the drive uses -10 to +10V then the ABS function can be skipped.
This sounds very interesting.
Are you saying you are using a VFD to drive an AC motor on the spindle for positional control (rotary axis with an encoder)?
How about heat and torque in this mode?
Rick G
Please Log in or Create an account to join the conversation.
- andypugh
-
- Away
- Moderator
-
Less
More
- Posts: 23316
- Thank you received: 4945
02 Aug 2013 19:57 #37356
by andypugh
I am saying that it can be done. I am not doing it with anything of mine at the moment, though it is likely that it would be a part of my toolchange on the Mill.
The VFD on my lathe is too crude to work well in this mode, a flux-vector drive is needed really.
Replied by andypugh on topic cincinnati axis spindle question
Are you saying you are using a VFD to drive an AC motor on the spindle for positional control (rotary axis with an encoder)?
How about heat and torque in this mode?
I am saying that it can be done. I am not doing it with anything of mine at the moment, though it is likely that it would be a part of my toolchange on the Mill.
The VFD on my lathe is too crude to work well in this mode, a flux-vector drive is needed really.
Please Log in or Create an account to join the conversation.
Time to create page: 0.451 seconds