Daniel Duesentrieb wrote:
> Like I wrote in my very first post I have absolutely no clue what I am 
> talking about. So please explain to me like to a child ;)
>
> I understand that system doesn't know in witch positions the servos are 
> because the encoders loose the values on a power off. I understand that 
> I have to jog every single axis to a specific position and adjust with 
> linuxcnc as my own homing process. This is maybe a pain in the a.. but 
> can be handled.
>
>   
LinuxCNC can home the joints.  They probably have a coarse position 
sensor, if
not, one could be added.  The encoders definitely have an index pulse, that
is accurate to one encoder count.  LinuxCNC has a mechanism to home
axes in sequence, and then move to a particular position offset from the 
index
pulse location.  So, this should allow the machine to be homed safely from
most positions.  But, there probably are some positions from which a homing
move might cause interference.  Possibly, you would have to jog the
robot to some initial position so the homing could be done without causing
a crash.  I might guess homing the outermost joint first, and working back
toward the base rotation joint as the last one, with the home offsets such
that the arm ends up folded close to the torso would be best.

> But first I need to be able to get the servos moving. Here is my question:
>
> Witch controller board do I need to use my original drivers? Or is it 
> not possible to use them and I need new ones?
>   
Your original Fanuc drives need 3 separate PWM signals per joint, with 5 
joints
that is 15 PWM drive signals.  As I think Peter said, even he may not have a
working solution at this time.
> What Hardware do I need to get this Robot moving asap?
>   
For ASAP, I think I do have a solution.  It would take :
2 Universal PWM Controllers                  $250               500
5 brushless PWM servo amps                    150                750
5 Fanuc encoder converters                       150                750
   
Total                                                                           
 
2000

I know that is a lot of money to spend on a speculative project.
These PWM servo amps are six-step (trapezoidal) drive, and
so not as smooth as sinusoidal drive.  But, they are simple,
and don't need powerful DSP processors to generate the
PWM commands.  I have some very similar motors here that
I use to test my drives and encoder converters.

Jon

------------------------------------------------------------------------------
October Webinars: Code for Performance
Free Intel webinars can help you accelerate application performance.
Explore tips for MPI, OpenMP, advanced profiling, and more. Get the most from 
the latest Intel processors and coprocessors. See abstracts and register >
http://pubads.g.doubleclick.net/gampad/clk?id=60135991&iu=/4140/ostg.clktrk
_______________________________________________
Emc-users mailing list
[email protected]
https://lists.sourceforge.net/lists/listinfo/emc-users

Reply via email to