5i20 & 7i33 puzzle
- whizzo944
- Offline
- New Member
Less
More
- Posts: 7
- Thank you received: 0
09 Jul 2018 19:12 #113885
by whizzo944
5i20 & 7i33 puzzle was created by whizzo944
Hello All,
I'm having a head-scratching moment here. I have a 5i20 & 7i33 connected to a chinese analog servo drive & motor.
I have used PnCconfig to write a .ini file using tkemc. If I set position display to actual and turn the motor, the display counts up
& down OK. I have set the following error values to 100mm If I enable the servo, I can jog back & forward, but, if I open hal meter
and display the following error, the error just goes up or down but doesn't return to zero when I stop jogging. If I jog for more than 100mm
then it faults out with following error. It seems to me as if the servo loop isn't closed somehow, but I can jog, so it sort of says the
servo loop is working. Anyone?
TIA,
Dave Winter.
I'm having a head-scratching moment here. I have a 5i20 & 7i33 connected to a chinese analog servo drive & motor.
I have used PnCconfig to write a .ini file using tkemc. If I set position display to actual and turn the motor, the display counts up
& down OK. I have set the following error values to 100mm If I enable the servo, I can jog back & forward, but, if I open hal meter
and display the following error, the error just goes up or down but doesn't return to zero when I stop jogging. If I jog for more than 100mm
then it faults out with following error. It seems to me as if the servo loop isn't closed somehow, but I can jog, so it sort of says the
servo loop is working. Anyone?
TIA,
Dave Winter.
Please Log in or Create an account to join the conversation.
- cmorley
- Away
- Moderator
Less
More
- Posts: 7774
- Thank you received: 2055
09 Jul 2018 21:06 #113892
by cmorley
Replied by cmorley on topic 5i20 & 7i33 puzzle
What version of linuxcnc are you using?
Can you post your INI and HAL file (or the whole config file)
You don't mention encoder feedback but does it count up and down too?
Chris M
Can you post your INI and HAL file (or the whole config file)
You don't mention encoder feedback but does it count up and down too?
Chris M
The following user(s) said Thank You: whizzo944
Please Log in or Create an account to join the conversation.
- tommylight
- Offline
- Moderator
Less
More
- Posts: 19412
- Thank you received: 6509
10 Jul 2018 05:48 #113928
by tommylight
Replied by tommylight on topic 5i20 & 7i33 puzzle
The following user(s) said Thank You: whizzo944
Please Log in or Create an account to join the conversation.
- whizzo944
- Offline
- New Member
Less
More
- Posts: 7
- Thank you received: 0
10 Jul 2018 07:58 #113932
by whizzo944
Replied by whizzo944 on topic 5i20 & 7i33 puzzle
Hello,
I thought that I was using 2.7.14 but the splash screen when it starts up says 2.7.0 The servo motor & drive are 3 phase
brushless, the encoder is built into the rear of the motor, as is usual these days. On the 5i20 board there are 2 red LEDs which are
on when I start the PC and go out when I start up LCNC. There is something stirring in the back of my mind which thinks there should
be a green LED on on the 5i20 when LCNC is up & running, but there isn't ? Attached is my config folder ( are they folders or
directories in Linux? )
Thanks again,
Dave.
I thought that I was using 2.7.14 but the splash screen when it starts up says 2.7.0 The servo motor & drive are 3 phase
brushless, the encoder is built into the rear of the motor, as is usual these days. On the 5i20 board there are 2 red LEDs which are
on when I start the PC and go out when I start up LCNC. There is something stirring in the back of my mind which thinks there should
be a green LED on on the 5i20 when LCNC is up & running, but there isn't ? Attached is my config folder ( are they folders or
directories in Linux? )
Thanks again,
Dave.
Please Log in or Create an account to join the conversation.
- PCW
- Away
- Moderator
Less
More
- Posts: 17905
- Thank you received: 4774
10 Jul 2018 12:33 #113938
by PCW
Replied by PCW on topic 5i20 & 7i33 puzzle
If you remove the:
setp pid.x.maxerror .0005
setp pid.y.maxerror .0005
setp pid.z.maxerror .0005
setp pid.s.maxerror .0005
Statements from the hal file, things should work better.
setp pid.x.maxerror .0005
setp pid.y.maxerror .0005
setp pid.z.maxerror .0005
setp pid.s.maxerror .0005
Statements from the hal file, things should work better.
The following user(s) said Thank You: whizzo944
Please Log in or Create an account to join the conversation.
- whizzo944
- Offline
- New Member
Less
More
- Posts: 7
- Thank you received: 0
10 Jul 2018 15:48 #113949
by whizzo944
Replied by whizzo944 on topic 5i20 & 7i33 puzzle
Hello Peter,
Looks like that did the trick. I also found lots of settings in my servo drives that needed changing, accel & decel ramps were set to S-shaped, need to be linear and as fast as possible, let LCNC handle the axis ramping. One thing I did find that surprised me was I have all the gains except P set to zero. The drives have an integrator in the speed & torque loops, and if you set any I gain in LCNC
it completely messes things up. I might be able to use FF0 gain to reduce the steady state following error.
Thanks for your help,
Dave.
Looks like that did the trick. I also found lots of settings in my servo drives that needed changing, accel & decel ramps were set to S-shaped, need to be linear and as fast as possible, let LCNC handle the axis ramping. One thing I did find that surprised me was I have all the gains except P set to zero. The drives have an integrator in the speed & torque loops, and if you set any I gain in LCNC
it completely messes things up. I might be able to use FF0 gain to reduce the steady state following error.
Thanks for your help,
Dave.
Please Log in or Create an account to join the conversation.
- PCW
- Away
- Moderator
Less
More
- Posts: 17905
- Thank you received: 4774
10 Jul 2018 15:57 #113950
by PCW
Replied by PCW on topic 5i20 & 7i33 puzzle
Typically for velocity mode drives LinuxCNC only needs FF1 and P and perhaps a bit of FF2
Basically FF1 should be adjusted to null error during constant speed motion
FF2 should be adjusted to minimize the error during acceleration and deceleration
and P should be as high as you can get without becoming unstable
Basically FF1 should be adjusted to null error during constant speed motion
FF2 should be adjusted to minimize the error during acceleration and deceleration
and P should be as high as you can get without becoming unstable
The following user(s) said Thank You: whizzo944
Please Log in or Create an account to join the conversation.
- whizzo944
- Offline
- New Member
Less
More
- Posts: 7
- Thank you received: 0
10 Jul 2018 16:57 #113958
by whizzo944
Replied by whizzo944 on topic 5i20 & 7i33 puzzle
Thanks again Peter.
Dave.
Dave.
Please Log in or Create an account to join the conversation.
Moderators: PCW, jmelson
Time to create page: 0.100 seconds