Advanced Search

Thanks, Andy also made that suggestion. You are right, I am kind of used to thinking in G94. Need to reprogram my brain.

Hi,

So i finally have the tool probe and workpoint figured out (i think*).
Running the code i got from Inventor CAM, with the default LinuxCNC post I run into issues.
I attached the Z-probe and workpiece subroutines as well as the Inventor cam generated code.
It stops on the T1 M6 line (N25). But i can't seem to get it to go beyond that. Pressing the start/pause on the HB04 doesn't do a thing.
Commenting out this line moves the program up to line N60 and throws up error messages about the Z-axis running out (it wants to go to high).
To me it looks like the code has a different idea of where the tool tip is (in Z) than what it gets set at with the Z-probe routine.
Work order:
{The z-probe / toolsetter is fixed to the bed}
Mount the probe. (that's the XYZ probe)
Run the Z-probe to set it's tip.
manually move the spindle/probe to within the bounds of the workpiece probe routine. (The vice for now)
Run the workpiece probe routine (finally got it to set it's coords correctly and it moves perfectly above the corner.)
Then i manually move the spinde to a comfortable place to change out the probe for the tool (round paper lollypop stick for now).
I placed a piece of plastic on the tool probe to create extra Z offset for initial safety so the lollypop stick doesn't plunge into the workpiece (the vice).
Then i load the 19.091 cgode file and "press play".
It moves from the probe XY to a place that is NOT the coords of line N55
Then it runs line N60 and error's out on a Z-axis out of bounds. 13mm above the workpiece is well withing that actual range of my machine with the tool probed as is with the extra spacer. I checked this with the spacer on the vice and the tool touching it (simulating the tool touching Z0 on the workpiece /G54. And then move up 13mm and then i still have plenty of headroom before I run into the Z home/limit switch.

So the 2 issues:
Getting past the M6 on line N25
Correctly using the tool length (?).

* probing the workpiece often fails when i comment out the message and M0 commands, claiming the probe is in contact during rapid movement. Adding the M0 'pause' solves that. But that is not how things should be.
Also, i would like to be able to position the probe during the routine
lines:
(MSG, Plaats probe boven werkstuk\n Hou de probe ong. 5mm binnen de hoek\n en druk daarna op start)
M0                        ; Wacht op bevestiging
The M0 hold doesn't allow moving the spindle about.

Hi,
Linuxcnc works for xenomai 3 only as it is posix based (basically same code but linked to xenomai libraries instead of standard libc libraries)
Xenomai EVL, on the other hand, is not posix based and a wrapper for threads, mutex, etc should be coded

Regards

Hi Spumco,

I can't find any official documentation about caxis.comp, just many forum posts about it from you. Is this something new not in the documentation yet?

My goal is to make my CNC router use the rotary axis and simulate a lathe. I've actually already done this before moving to my Mesa 7i76e but using an external device with a dial ( amazon link ) to control the speed. I'd like to have that programatic control over the speed for obvious reasons. I actually have a YouTube video on what I did then. 

I don't need dynamic change between velocity and position controls. I'm more than happy to restart LinuxCNC with new config when changing operations.

Used keyboard matrix controller for bottoms on the controller cabinet. 
I disassemble a old keyboard and salvage the controller. the connection points has to be cleaned before soldering wires on it. "the black stuff" on the contact points are graphite with normal traces under.  

Keyboard bracket finished with room for extra bottoms and trackpad. 

         

I just discovered that my formula and the one in the QTPlasmaC documentation ids not correct for the THAD2. Those formulas work for the previous release of THAD.
PncConf has the correct formula by the looks of it. Does any have the correct formula for me?

Pictures would be nice.

Maybe just get rid of the geartrain.

A 2-speed geartrain is not particularly helpful.

Is the spindle motor variable speed?

Why do you need an encoder?

Do you really need ultra-precise spindle speeds?

I do not, i set automation off, play with voltage till i get roughly good cuts and THC properly tracking, sometimes a bit of editing the scale and offset to match the measured voltage (ranges from 97V to 135V depending on the source and nozzles), then:
- fine tune the cut height by changing the voltage in 2V increments up or down
- fine tune the feed rate
- fine tune the air pressure to get clean cuts on the bottom side (less dross)
All this should be done whenever changing material thickness/type, nozzle diameter/type, and/or plasma source eventually. Also will be influenced by the amount of moisture in the compressed air, sometimes enough to cause issues.
And i never use autovolts, mainly due to many external factors that i have no control over, like, huge power fluctuations, huge differences in material being cut, above mentioned moisture, old air compressors spewing oil through the torch, using old plasma sources with no current control, etc.
-
Another thing that will absolutely mess with cutting is the ground clamp, it has to be properly bolted to the table, no clamp.

Why 1600 steps/rev (1/8th step)?

Enter this in the stepconf? wizard:

Driver microstepping: 8(maybe 4)
Motor steps/rev: 200/pulleyRatio

Measure the pulley ratio with the actual belt for maximum accuracy, not using some measuring tape or something.

Please use rational numbers (fractions) to express the ratios. Decimals are harder to understand,

 

Unless your belt is magic, it will have a little bit of backlash and likely some slip.

Is this a timing belt, a V-belt, or something else?

You mean hybrid servo?

(stepper+encoder)

How do others determine the volt/mm if you don't use Autovolt?

You are missing this line in your kinematic component: