binary works - compilation not smoothly???
23 Jul 2018 18:37 #114715
by schorsch
Replied by schorsch on topic binary works - compilation not smoothly???
Hi,
thank you again.
I guess this was much easier than I thought: Seems one needs to have a referenced system in order to make UVW axis available...
I didn't manage two things:
a) Kinematics with rot4thaxis seems to work. the correct axis are moving (Z axis moves Z and Y joint, W axis moves Z joint) but in the display e.g. Z and W are appearing as Z axis (are moving along the angle of the A axis)
b) homing works in the correct speed (as it was before) but after this all axis are limited to vel 60. This is damn slow.
I tried to change MAXVEL of axis or joints but no difference, is there another place where one can limit the speed?? Acceleration is very low too.
Thanks
Georg
thank you again.
I guess this was much easier than I thought: Seems one needs to have a referenced system in order to make UVW axis available...
I didn't manage two things:
a) Kinematics with rot4thaxis seems to work. the correct axis are moving (Z axis moves Z and Y joint, W axis moves Z joint) but in the display e.g. Z and W are appearing as Z axis (are moving along the angle of the A axis)
b) homing works in the correct speed (as it was before) but after this all axis are limited to vel 60. This is damn slow.
I tried to change MAXVEL of axis or joints but no difference, is there another place where one can limit the speed?? Acceleration is very low too.
Thanks
Georg
Please Log in or Create an account to join the conversation.
27 Jul 2018 17:01 #114959
by andypugh
Replied by andypugh on topic binary works - compilation not smoothly???
There are velocity limits in the [TRAJ] section and you might also need to change the limits of the max-velocity slider.
Please Log in or Create an account to join the conversation.
07 Aug 2018 20:14 #115767
by schorsch
Replied by schorsch on topic binary works - compilation not smoothly???
...not so easy to find this, but finally solved....
axis-section must not be called [AXIS_0] or [AXIS_1] and so on, but [AXIS_X] and [AXIS_Y]
found some further small things:
the variable rot4thaxiskins.H1 needs to be negative in my case, since me machine origin is not the below but above the A Axis
finally the rotation orientation was swapped, easy to invert one pin
Thank you very much for you help, finally I need to get the different tool heights to work... will manage... thanks a lot
Schorsch
axis-section must not be called [AXIS_0] or [AXIS_1] and so on, but [AXIS_X] and [AXIS_Y]
found some further small things:
the variable rot4thaxiskins.H1 needs to be negative in my case, since me machine origin is not the below but above the A Axis
finally the rotation orientation was swapped, easy to invert one pin
Thank you very much for you help, finally I need to get the different tool heights to work... will manage... thanks a lot
Schorsch
Please Log in or Create an account to join the conversation.
28 Sep 2018 19:46 #118139
by schorsch
Replied by schorsch on topic 4th axis kinematics with tool length compensation
Hi,
I was finally able to get the 4th axis kinematics module (from the wiki) running and also the tool length probe. Just to remind you: I have an a axis installed on a xy table. spindle is fixed to z.
So far not working is 4th axis kinematics with tool length compensation. I figured out that G38.3 is not storing UVW coordinates…. should it??
I was able to set the cutter length compensation into the W axis, but it seems it has no effect. storing the cutter length Offset into z axis seems not be be correct as z is rotating.
I am wondering: When I am moving x,y,z axis the uvw coordinates are staying at 0, shouldn't be there any effect, too? Might there be a problem in the used kinematics module? Same vice versa.
Last Question:
I was able to figure out the actual height of the rotary axis is in z oder w coordinates and fixed in into the configuration as described in the wiki. For what do I need to probe the z and y (or the v and w) axis anymore, if the Pivot axis is fixed anyhow? From my point of view there should be either a button "Set y and z to rotary axis" or there should be a way to probe the rotary axis and then pass the probed coordinates into the kinematics module...hope you can help me to understand
Thanks
Georg
I was finally able to get the 4th axis kinematics module (from the wiki) running and also the tool length probe. Just to remind you: I have an a axis installed on a xy table. spindle is fixed to z.
So far not working is 4th axis kinematics with tool length compensation. I figured out that G38.3 is not storing UVW coordinates…. should it??
I was able to set the cutter length compensation into the W axis, but it seems it has no effect. storing the cutter length Offset into z axis seems not be be correct as z is rotating.
I am wondering: When I am moving x,y,z axis the uvw coordinates are staying at 0, shouldn't be there any effect, too? Might there be a problem in the used kinematics module? Same vice versa.
Last Question:
I was able to figure out the actual height of the rotary axis is in z oder w coordinates and fixed in into the configuration as described in the wiki. For what do I need to probe the z and y (or the v and w) axis anymore, if the Pivot axis is fixed anyhow? From my point of view there should be either a button "Set y and z to rotary axis" or there should be a way to probe the rotary axis and then pass the probed coordinates into the kinematics module...hope you can help me to understand
Thanks
Georg
Please Log in or Create an account to join the conversation.
28 Sep 2018 22:00 #118154
by andypugh
I think so. The good thing about Open Source is that you can look at the code, not the docs.
github.com/LinuxCNC/linuxcnc/blob/0f91c5...otion/control.c#L543
caret_pos_fb is an EMCPose
github.com/LinuxCNC/linuxcnc/blob/0f91c5...motion/motion.h#L665
And that does contain UVW
github.com/LinuxCNC/linuxcnc/blob/0f91c5...ml_intf/emcpos.h#L19
But, note that it seems to log the feedback values.
What version of LinuxCNC are you using? I wonder if this is a fun Joints/Axes bug...
Replied by andypugh on topic 4th axis kinematics with tool length compensation
So far not working is 4th axis kinematics with tool length compensation. I figured out that G38.3 is not storing UVW coordinates…. should it??
I think so. The good thing about Open Source is that you can look at the code, not the docs.
github.com/LinuxCNC/linuxcnc/blob/0f91c5...otion/control.c#L543
caret_pos_fb is an EMCPose
github.com/LinuxCNC/linuxcnc/blob/0f91c5...motion/motion.h#L665
And that does contain UVW
github.com/LinuxCNC/linuxcnc/blob/0f91c5...ml_intf/emcpos.h#L19
But, note that it seems to log the feedback values.
What version of LinuxCNC are you using? I wonder if this is a fun Joints/Axes bug...
Please Log in or Create an account to join the conversation.
28 Sep 2018 22:42 #118158
by andypugh
Replied by andypugh on topic 4th axis kinematics with tool length compensation
Chatting on the IRC, we think it works.
What do you see as the "actual position" of the UVW axes when the probe finishes? Possibly the forward kins that you are using is not working in UVW?
What do you see as the "actual position" of the UVW axes when the probe finishes? Possibly the forward kins that you are using is not working in UVW?
Please Log in or Create an account to join the conversation.
29 Sep 2018 19:35 #118187
by schorsch
Replied by schorsch on topic 4th axis kinematics with tool length compensation
Hi,
Concerning the version:
when I started this thread in july18 it was the master (LINUXCNC - 2.8.0~pre1). Because of your help I was able to compile by myself.
I just used G38.3 (either on z or on w axis) and checked #5061 to #5069 by using print (print, #5061 …). UVW is always 0. XYZ is ok.
Futhermore I wondered, (please check here: wiki.linuxcnc.org/cgi-bin/wiki.pl?Rot4thaxiskins) About there lines in wiki.linuxcnc.org/uploads/rot4thaxiskins.c
I have only 4 Joints (and obviously the guy who has written the code, too). I don't care about B and C, but shouldn't UVW point to joints 0,1,2???
Currently I am displaying XYZAUVW in axis. Please tell me what should axis display when I'm moving X or Z axis?? Shouldn't V and W change ist values, too?
Thanks
Georg
Concerning the version:
when I started this thread in july18 it was the master (LINUXCNC - 2.8.0~pre1). Because of your help I was able to compile by myself.
I just used G38.3 (either on z or on w axis) and checked #5061 to #5069 by using print (print, #5061 …). UVW is always 0. XYZ is ok.
Futhermore I wondered, (please check here: wiki.linuxcnc.org/cgi-bin/wiki.pl?Rot4thaxiskins) About there lines in wiki.linuxcnc.org/uploads/rot4thaxiskins.c
pos->tran.x = m_x - to_x;
pos->tran.y = m_y - to_y;
pos->tran.z = m_z - to_z;
pos->a = joint[3];
pos->b = joint[4];
pos->c = joint[5];
pos->u = joint[6];
pos->v = joint[7];
pos->w = joint[8];I have only 4 Joints (and obviously the guy who has written the code, too). I don't care about B and C, but shouldn't UVW point to joints 0,1,2???
Currently I am displaying XYZAUVW in axis. Please tell me what should axis display when I'm moving X or Z axis?? Shouldn't V and W change ist values, too?
Thanks
Georg
Please Log in or Create an account to join the conversation.
29 Sep 2018 19:58 #118192
by andypugh
Replied by andypugh on topic 4th axis kinematics with tool length compensation
What _are_ V and W?
They often refer to secondary axes (Lathe tailstock for example) and sometimes they are a second set of rotated coordinates (for example drilling along an angle defined by AB axes)
They often refer to secondary axes (Lathe tailstock for example) and sometimes they are a second set of rotated coordinates (for example drilling along an angle defined by AB axes)
Please Log in or Create an account to join the conversation.
29 Sep 2018 22:56 - 29 Sep 2018 22:58 #118201
by schorsch
Replied by schorsch on topic 4th axis kinematics with tool length compensation
Hi,
as far as I understand I would assume that U is exactly X
V is exactly the second joint (meaning of the Y axis)
W is exactly the third joint (meaning of the Z axis)
But I'm a beginner in the context of multi-axis simultanous stuff. Please correct me, if I'm wrong
Shouldn't it be like this? I Always though UVW should be in the orientation of the cutter.
Let me ask a further Question. In my XYZA Coordinate System (by using the rot4thaxis kinematics) where do I have to apply the tool length Offset?? To the Z axis??? or to the W axis?
Georg
Georg
as far as I understand I would assume that U is exactly X
V is exactly the second joint (meaning of the Y axis)
W is exactly the third joint (meaning of the Z axis)
But I'm a beginner in the context of multi-axis simultanous stuff. Please correct me, if I'm wrong
Shouldn't it be like this? I Always though UVW should be in the orientation of the cutter.
Let me ask a further Question. In my XYZA Coordinate System (by using the rot4thaxis kinematics) where do I have to apply the tool length Offset?? To the Z axis??? or to the W axis?
Georg
Georg
Last edit: 29 Sep 2018 22:58 by schorsch.
Please Log in or Create an account to join the conversation.
29 Sep 2018 23:03 #118202
by andypugh
Replied by andypugh on topic 4th axis kinematics with tool length compensation
UVW are completely different to XYZ. (otherwise, what is the point?)
Partly they are nothing more than a way to program the position of "something" from inside a G-code script.
What they actually mean is something that is defined by your configuration.
In TRT XYZAB systems W is often the tool length. But it doesn't have to be.
Partly they are nothing more than a way to program the position of "something" from inside a G-code script.
What they actually mean is something that is defined by your configuration.
In TRT XYZAB systems W is often the tool length. But it doesn't have to be.
Please Log in or Create an account to join the conversation.
Time to create page: 0.103 seconds