TCP 5-axis kinematics
19 Jul 2019 21:15 #139927
by andypugh
No, I wasn't meaning _that_ general. I was just asking if the pivot positions (for example) were hard-coded.
(Though perhaps that would be OK, if they were hard-coded to zero, you could set the machine homing limits to suit)
Replied by andypugh on topic TCP 5-axis kinematics
But if you mean that it could handle basically any kinematic machine type, like:
table-table, head-table, head-head + non orthogonal varieties of these machines.
No, I wasn't meaning _that_ general. I was just asking if the pivot positions (for example) were hard-coded.
(Though perhaps that would be OK, if they were hard-coded to zero, you could set the machine homing limits to suit)
Please Log in or Create an account to join the conversation.
20 Jul 2019 01:47 - 20 Jul 2019 01:51 #139940
by dgarrett
Replied by dgarrett on topic TCP 5-axis kinematics
@plopes9000
No -- thanks for your report!
I've merged (cherry-picked) dgarr/homing_h
to the master branch and created a new branch
dgarr/home_index_no_encoder_reset that is based on the
updated master branch. Please 1) review the changes,
2) the documentation (.txt) files, and 3) test this
branch.
When this branch is successfully tested, it can be
merged to master but it is probably prudent to wait
a week or two in case problems surface with the homing_h
incorporation.
Branch Ref:
github.com/LinuxCNC/linuxcnc/commits/dga...dex_no_encoder_reset
Was there anything else you meant for me to test?
No -- thanks for your report!
I've merged (cherry-picked) dgarr/homing_h
to the master branch and created a new branch
dgarr/home_index_no_encoder_reset that is based on the
updated master branch. Please 1) review the changes,
2) the documentation (.txt) files, and 3) test this
branch.
When this branch is successfully tested, it can be
merged to master but it is probably prudent to wait
a week or two in case problems surface with the homing_h
incorporation.
Branch Ref:
github.com/LinuxCNC/linuxcnc/commits/dga...dex_no_encoder_reset
Last edit: 20 Jul 2019 01:51 by dgarrett.
Please Log in or Create an account to join the conversation.
- plopes9000
-
- Offline
- Premium Member
-
Less
More
- Posts: 87
- Thank you received: 18
20 Jul 2019 08:26 #139942
by plopes9000
Will test it tonight. But have just looked at your changes and all looks good - thank you!
Replied by plopes9000 on topic TCP 5-axis kinematics
@plopes9000
Was there anything else you meant for me to test?
No -- thanks for your report!
I've merged (cherry-picked) dgarr/homing_h
to the master branch and created a new branch
dgarr/home_index_no_encoder_reset that is based on the
updated master branch. Please 1) review the changes,
2) the documentation (.txt) files, and 3) test this
branch.
When this branch is successfully tested, it can be
merged to master but it is probably prudent to wait
a week or two in case problems surface with the homing_h
incorporation.
Branch Ref:
github.com/LinuxCNC/linuxcnc/commits/dga...dex_no_encoder_reset
Will test it tonight. But have just looked at your changes and all looks good - thank you!
Please Log in or Create an account to join the conversation.
- plopes9000
-
- Offline
- Premium Member
-
Less
More
- Posts: 87
- Thank you received: 18
20 Jul 2019 13:32 #139964
by plopes9000
I'm getting an error:
Replied by plopes9000 on topic TCP 5-axis kinematics
When this branch is successfully tested, it can be
merged to master but it is probably prudent to wait
a week or two in case problems surface with the homing_h
incorporation.
I'm getting an error:
paju@zx45:~/linuxcnc-dev-test/home_index_no_encoder_reset/src$ linuxcnc /home/paju/linuxcnc/configs/zx7045-prod2.9.0-test/zx7045-prod.ini
LINUXCNC - 2.9.0~pre0
Machine configuration directory is '/home/paju/linuxcnc/configs/zx7045-prod2.9.0-test'
Machine configuration file is 'zx7045-prod.ini.expanded'
invalid command name "parse_ini"
while executing
"parse_ini $inifile"
(file "/home/paju/linuxcnc-dev-test/home_index_no_encoder_reset/lib/hallib/check_config.tcl" line 151)
check_config validation failed
LinuxCNC terminated with an error. You can find more information in the log:
/home/paju/linuxcnc_debug.txt
and
/home/paju/linuxcnc_print.txt
as well as in the output of the shell command 'dmesg' and in the terminalPlease Log in or Create an account to join the conversation.
20 Jul 2019 14:50 #139969
by dgarrett
This error unlikely related to the branch but might
be caused by improper environment setup (sourcing
proper rip-environment) or maybe conflicts
between multiple rip builds and/or installs on
the same machine.
Please post:
1) output from the command: linuxcnc_info
2) tar file of the configuration directory
3) log files:
/home/paju/linuxcnc_debug.txt
/home/paju/linuxcnc_print.txt
Replied by dgarrett on topic TCP 5-axis kinematics
invalid command name "parse_ini"be caused by improper environment setup (sourcing
proper rip-environment) or maybe conflicts
between multiple rip builds and/or installs on
the same machine.
Please post:
1) output from the command: linuxcnc_info
2) tar file of the configuration directory
3) log files:
/home/paju/linuxcnc_debug.txt
/home/paju/linuxcnc_print.txt
Please Log in or Create an account to join the conversation.
- plopes9000
-
- Offline
- Premium Member
-
Less
More
- Posts: 87
- Thank you received: 18
20 Jul 2019 14:53 #139970
by plopes9000
You’re probably right. I was eager to test it but had only a little time and now that you say that I don’t recall having sourced the env.
Will check when I’m back home.
Replied by plopes9000 on topic TCP 5-axis kinematics
This error unlikely related to the branch but might
be caused by improper environment setup (sourcing
proper rip-environment) or maybe conflicts
between multiple rip builds and/or installs on
the same machine.
You’re probably right. I was eager to test it but had only a little time and now that you say that I don’t recall having sourced the env.
Will check when I’m back home.
Please Log in or Create an account to join the conversation.
- plopes9000
-
- Offline
- Premium Member
-
Less
More
- Posts: 87
- Thank you received: 18
20 Jul 2019 18:11 - 20 Jul 2019 18:16 #139991
by plopes9000
Honestly not quite sure what tainted that terminal window ... I had used it for the test from yesterday but I did source the new directory earlier, I can tell since when I tried to source it now it gave the usual message that it only needs to be done once "This script only needs to be run once per shell session.".
No matter, I started a new terminal and sourced the new dir and the error was gone.
So, without making any modifications to your code and simply adding HOME_INDEX_NO_ENCODER_RESET = YES to all axis that have HOME_USE_INDEX = YES, I was able to home successfully multiple times and on all axis using index pulse.
Just to be clear, had the changes not worked, the home sequence would, after finding the index, send the axis to a home position that is not the one configured in the ini file, eg. MIN=0, MAX= 400, HOME_OFFSET = 200 but final move would for example take the axis to the right most end of the AXIS position instead of the middle. Also, the travel limits would be messed up.
So the test is successful - Thanks again.
Could you please keep me updated on the success of this code making it in?
Replied by plopes9000 on topic TCP 5-axis kinematics
This error unlikely related to the branch but might
be caused by improper environment setup (sourcing
proper rip-environment) or maybe conflicts
between multiple rip builds and/or installs on
the same machine.
You’re probably right. I was eager to test it but had only a little time and now that you say that I don’t recall having sourced the env.
Will check when I’m back home.
Honestly not quite sure what tainted that terminal window ... I had used it for the test from yesterday but I did source the new directory earlier, I can tell since when I tried to source it now it gave the usual message that it only needs to be done once "This script only needs to be run once per shell session.".
No matter, I started a new terminal and sourced the new dir and the error was gone.
So, without making any modifications to your code and simply adding HOME_INDEX_NO_ENCODER_RESET = YES to all axis that have HOME_USE_INDEX = YES, I was able to home successfully multiple times and on all axis using index pulse.
Just to be clear, had the changes not worked, the home sequence would, after finding the index, send the axis to a home position that is not the one configured in the ini file, eg. MIN=0, MAX= 400, HOME_OFFSET = 200 but final move would for example take the axis to the right most end of the AXIS position instead of the middle. Also, the travel limits would be messed up.
So the test is successful - Thanks again.
Could you please keep me updated on the success of this code making it in?
Last edit: 20 Jul 2019 18:16 by plopes9000.
Please Log in or Create an account to join the conversation.
21 Jul 2019 12:10 #140069
by Stud54
Damn, you did it first. My HDs are still laying in the corner...
Replied by Stud54 on topic TCP 5-axis kinematics
Damn, you did it first. My HDs are still laying in the corner...
Attachments:
Please Log in or Create an account to join the conversation.
- plopes9000
-
- Offline
- Premium Member
-
Less
More
- Posts: 87
- Thank you received: 18
21 Jul 2019 12:52 #140074
by plopes9000
Oh wow very nice collection, now I know why they are hard to get on eBay
you’ve been snatching them all 
I still have another two but without integrated servo.
Despite what they say, better second than third;-) hurry up.
I had mine for about a year stored away, while the concept was brewing in my mind. If I knew where to get cheap custom casts I would have done it slightly different.
Nice thing is that the 5th axis can be swapped easily for any other setup, so it’s a modular design. While I’m exploring the current setup, many other ideas are brewing, many of which may never see the light of day.
Replied by plopes9000 on topic TCP 5-axis kinematics
Damn, you did it first. My HDs are still laying in the corner...
Oh wow very nice collection, now I know why they are hard to get on eBay
you’ve been snatching them all I still have another two but without integrated servo.
Despite what they say, better second than third;-) hurry up.
I had mine for about a year stored away, while the concept was brewing in my mind. If I knew where to get cheap custom casts I would have done it slightly different.
Nice thing is that the 5th axis can be swapped easily for any other setup, so it’s a modular design. While I’m exploring the current setup, many other ideas are brewing, many of which may never see the light of day.
Please Log in or Create an account to join the conversation.
24 Jul 2019 13:40 #140429
by 3D-Master
Replied by 3D-Master on topic TCP 5-axis kinematics
How does Linuxcnc handle softends with TCPC? does it automatically stop when the physical axis hits the softend or does it just go through? (until it hits the endstop)
Please Log in or Create an account to join the conversation.
Time to create page: 0.155 seconds