Joint compensation on a 6 axis robot

Hi Andy,

I followed the steps and compiled the whole linuxcnc into linuxcnc-dev what seems to be successful. Then I tried the same as in the post:
forum.linuxcnc.org/forum/24-hal-componen...s-component?start=10
The steps with the
. ./scripts/rip-environment
and then the sudo comp --install genserkins.c
does not work. I'm getting the same warning during compile as before.
When I try to do the same as the user dimech did in the post the genserkins.ko is not working when I simply copy it over from linuxcnc-dev/src into the modules folder of the real system. There is the error
could not insert module ....genserkins.ko: Invalid parameters.
The same with the pumakins.ko is working fine (as for dimech).
I need to say that I'm running a linuxcnc on Debian 2.6 while the git branch is the origin/2.7 - master git branch what is not fitting 100%. I tried to enter a different branch in:
git branch --track v2.5_branch origin/v2.5_branch
git checkout v2.5_branch
something with 2.6 but this is not working - git can't find it:
fatal: Not a valid object name: 'origin/v2.6_branch'.
I did not find any info via google what are the available branches.
Do I need to install a new 2.7 system - do you think it will run this way?
I'm very sorry but I'm new to git and the whole compile setup.
Would be very kind if you can give me a hint again.

Many many thanks,
Thomas

Hi Andy,

so far I understand and I want to work this way with a new copied version of genserkins under a new name with my modifications included. But first I want to check if the original version that I compiled inside linuxcnc-dev folder works as intended.
And here I'm struggling. The compilation worked and the genserkins.ko file was there in linuxcnc-dev/src as expected.
But when I exchange the genserkins.ko in the modules folder of the orginal system with this compiles version the AXIS is starting with the error above.
I think this is maybe because of the wrong source version (2.7) from github vs. my running system on 2.6.
Before I newly install my laptop with the 2.7 wheezy I want to cross check if the issue might be somewhere else and reinstalling will not help (with the work of reinstall and setting it up from scratch).
Not sure if you can easily assess if this is the issue. If you are just gessing it is also o.k. because I don't see anything else that might be wrong.
The alternative would be that there is a git branch that fits to my current system 2.6. This would be much easier to check.

Many thanks, Thomas

Hi Andy,

thanks so much for your help. Indeed 3.) worked fine. I was first surprised after entering git checkout -b origin/2.6 that it took only a few seconds. I thought it has to sync the whole 2.6 branch but obviously only the changed files.
After that the compilation worked and the new .ko file was working fine with my 2.6 system.
I check now the modifications inside a new named file and see if it works.

So far many thanks,
Thomas

Hi Andy,

while the 3rd option worked somehow successful on one of my Debian systems A I was using the same procedure on another Debian 2.6 system B. But this was not successful. Maybe I did something a bit different. Since I don't completely understand the system and the steps I'm a bit lost.
When I copy my new .ko file (genserkinspuma.ko) from the running system A to the system B I also want to setup for compilation the .ko file works fine. When I run the makefile on the system B and copy over from the src folder into to the running system under usr/... this genserkinspuma.ko fails to start with the known errors. What is also one indication of different files is that the genserkinspuma.ko on the good system A is 298,5kb while the other system B it creates a 291,0kb file.
I'm struggling really to find out on what version the runtime system is and what is the checkout version of git is. Is there some information on what the version is?
When I do git checkout origin/2.6 - is there no need to specify the subversion like 2.6.12?
I'm a bit lost and would be nice if you could help me again. Since your hint is working on machine A I assume it should work also on the other machine B that was installed from the same Debian image as the other one (both I guess on 2.6.12)? I guess there is just one little step different.

Many thanks again,
Thomas

Hi Andy,

System A (compiling a working genserkinspuma.ko):
uname -a:
Linux R50Debian 3.4-9-rtai-686-pae #1 SMP PREEMPT Debian 3.4.55-4linuxcnc i686 GNU/Linux
git branch:
* (no branch)
master
origin/2.6
v2.5_branch

System B (not compiling correctly):
Linux debianTerminator 3.4-9-rtai-686-pae #1 SMP PREEMPT Debian 3.4.55-4linuxcnc i686 GNU/Linux
git branch:
master
* origin/2.6
origin/2.6.12
v2.5_branch
v2.6
v2.6.12

I guess the * is telling me what the active branch is. For the correctly compiling system it says "no" and for the not compiling correctly it give the branch you told to set?? I thought that I had on both system the same commands. I was following the instructions on wiki.linuxcnc.org/cgi-bin/wiki.pl?Instal..._or_8_04_from_source till step 2.5. Then your recommendation git checkout -b origin/2.6
I"m running a Debian linuxcnc 2.6 from a live-CD. For both systems the same CD. Maybe the updated version is different. On system B I installed all updates today but hopefully this does not change my version.
Does this tell you something?

Thomas

robottom wrote:

FYI - my genserkins DH config:
setp genserkins.A-0 0
setp genserkins.A-1 0
setp genserkins.A-2 431.8
setp genserkins.A-3 0
setp genserkins.A-4 0
setp genserkins.A-5 0

setp genserkins.ALPHA-0 0
setp genserkins.ALPHA-1 1.570796326
setp genserkins.ALPHA-2 0
setp genserkins.ALPHA-3 1.570796326
setp genserkins.ALPHA-4 -1.570796326
setp genserkins.ALPHA-5 1.570796326

setp genserkins.D-0 671.83
setp genserkins.D-1 -149.09
setp genserkins.D-2 0
setp genserkins.D-3 433.07
setp genserkins.D-4 0
setp genserkins.D-5 56.25

When defining D-H parameters for Genserkins, is it expecting the values to be in mm or inch? Is there a way to toggle that setting to provide inches instead of mm?

Hi Cpk_Rbt,

to my best knowledge these parameters are directly transfered to the kinematics module. There is no conversion before it hits genserkins. As you can see below parameters are in mm and rad. You would need to use another kinematics module that is expecting inch if you really need this. But why is it a problem at all. mm to inch is just a straight conversion.

Best, Thomas