5 axis Robot arm Design - Linuxcnc configurarion

I am designing a 5 axis robot arm.

I am needing guidance regarding the efficient procedure.
I know what inverse and forward kinematics are.

Is there a tool available to make calculations for d-h parameters.

Actually getting the parameters is fairly easy once you get an understanding of the idea behind it. You can do it on a piece of paper. I don't know of any tools but I made a little tutorial about how to derive genserkins' modified D-H parameters for a 6 axis robot.

linuxcnc.org/docs/devel/html/motion/dh-parameters.html

Yeah!
That helped me a lot.
It has started to look somewhat easier.

First I will design the arm.

After that, I will try to calculate modified d-h parameters and get some hands-on with genserkins.
With help from @aciera, I think I will be succesfull in simulation.

this is a basic idea
base is 720mm tall
1st arm: 600mm
2nd arm: 600mm
link to mig tip: 250mm fixed

guide me further

Worm geared reduction would be very nice as there would be no need for brakes and they can have very high reduction ratios, but they do have plenty of backlash.
Since it looks to be very light and no forces are to be applied to the joints, belt reductions might be also usable but aim for the biggest ratio you can find. Alas, these would need brakes on at least 3 joints.
Worm gears have another good feature, that you can bolt the arm directly to their shaft.

tommylight wrote: Worm geared reduction would be very nice as there would be no need for brakes and they can have very high reduction ratios, but they do have plenty of backlash.
Since it looks to be very light and no forces are to be applied to the joints, belt reductions might be also usable but aim for the biggest ratio you can find. Alas, these would need brakes on at least 3 joints.
Worm gears have another good feature, that you can bolt the arm directly to their shaft.

I like the worm gears idea.

What if we could calculate backlash in advance
and tell linuxcnc to compensate for backlash.

Moreover, teeth will be pushed one side due to gravity. Backlash can be a little problem.

Belt drives with bicycle disk brakes is my other idea.

Motors will be closed loop steppers with silent drivers like trinamics.

pinder wrote:
Moreover, teeth will be pushed one side due to gravity. Backlash can be a little problem.

That is a valid point.

tommylight wrote:
That is a valid point.

Worm gears will also ease the mounting mechanism as the motor shaft is prependicular to output shaft, so it drecrease oveall length of motor and gears. Thats what I am considering untill something better comes up.

I'll also dig other forums for other people's designs. cnc-club.ru has never failed me.
Translation is a bit tricky but can be managed.

Need some help with forward and inverse mechanics
Somebody can provide some step by step guide.
I read genserkins.
Kinda Confused at this time

IIRC genserkins uses the Jacobi matrix for calculation. That together with it being written in C does make it challenging to reverse engineer as it is not well documented.

If you want a basic explanation of how kinematics can be calculated:

forum.linuxcnc.org/media/kunena/attachme.../5-axis-machines.pdf

The complete theoretical background is laid out in the book "Introduction to Robotics, Mechanics and Control" by John J. Craig. Here you find the explanation of the "modified" D-H parameter convention as used in genserkins. The book can be found on the web.

However, at the end of the day you don't really need to understand the math behind it all because genserkins.c takes care of that. All you need to figure out is the modified D-H-parameter set that describes your robots kinematic setup.

Here is my simulation of a RV-6SDL Melfa robot:

forum.linuxcnc.org/10-advanced-configura...th-simulation#158467

There is also a spreadsheet that does forward and inverse kinematic calculations for an AR2. Beware though that this uses the "unmodified" D-H-Parameter convention:

github.com/Chris-Annin/AR2/blob/master/A...nematic%20model.xlsm