Advanced Search

Rodw

Thank you just what I needed.

I appreciate your time in creating this.

Cheers

Parameter können nicht mit HAL pins geändert werden.

Aber warum nicht einfach einen stepgen (stepgen.N.velocity-cmd) verwenden?

hm.. und es ist nicht möglich irgendwie die Frequenz zu regeln?
da ich so aktuell 'pwmgen.n.value' auf 0.5 setze, habe ich ein top Rechteck-Signal, so wie ich es brauche.

Habe es als alternative mit einem PDM-Signal versucht (pwmgen.01.output-type 3), jedoch ist da auch nicht immer ein Tastgrad von 0.5 gegeben und zudem ist das Signal nicht konstant (weiß aber noch nicht genau ob die geringe Abweichung etwas ausmacht)

 

I subscribed to them all!

I couldn’t find any video explaining the pipe cuter though. What kind of motor did you use and how did you set it up?

Hey, no problem! I appreciate the hard work you do for this project!!

I increased the deadband to 0.01 and switched the position back to Actual. The jumping of the numbers stopped and the motors don't make that weird noise anymore. I will test it with the oscilloscope, but I am sure that brought me one step more where I want to be. I wanna say that the motors are not stalling as early anymore.

I played around with the PID settings and thought this would help with the problem, but no change so far.

'pwmgen.pwm_frequency' ist ein Parameter, der ist nicht zum Einstellen der Drehzahl dazu ist 'pwmgen.n.value' .

Vielleicht hilt dir das hier weiter:
forum.linuxcnc.org/39-pncconf/38723-pncc...p-pwm-on-7i96#162320

tommylight wrote:

Anything else you look might seem cheaper, it is not, and they will be extremely limited in expansion compared to Mesa.

 

Trying to get a feel for this but I'm a bit overwhelmed.
Is it possible to use my own motion controller with LinuxCNC? I am only interested in closed-loop servo and I want to handle the PID myself. 

Craig

hab die *.hal und *.ini im ersten Post angehängt.
Benutze eine Mesa 7i96s und habe das bitfile angepasst, sodass auf dem step-gen_4 die PWM/Dir Ausgabe stattfindet. 

pwmgen.0.pwm-freq lieferte den gleichen Fehler.
Sowie ich gelesen habe ist die Frequenz für alle PWM-Generatoren gültig, darum wird die "0" weggelassen

Check in the ini TOOL_CHANGE_POSITION = 0 0 0. The third digit (Z) will be added to the length of the tool and will show the excess.

Thank you for your reply.

linuxcnc.org/docs/2.9/html/user/user-con...gramming-the-planner

I have read the following document, Naive CAM Detector section.
"Naive CAM Detector
Successive G1 moves that involve only the XYZ axes that deviate less than Q- from a straight line are merged into a single straight line."

As you said, CAM seems to only work on XYZ? Is there any other solution? except 'output longer segments'.

i get like before

My PC is a Dell Optiflex 3040

That is likely because LinuxCNC falls back to one-line lookahead for moves involving any other axis than XYZ. This means that velocity is no longer blended over multiple path segments. The motion planner sets the speed as to be able to stop the movement at the end of the next segment which means that the shorter the segment the slower the speed.
This is an internal limitation of the current motion planner, as a workaround for multi axis gcode generated by CAM it is usually recommended to output longer segments.

Hi,

I am trying to understand the Nema 23 closed loop CL57T driver. I have zero knowledge of CNC and stepper motors. I am learning with the objective of building my own CNC.

The manual for the driver: www.omc-stepperonline.com/download/CL57T_V4.0.pdf

The factory setting for the Microstep Switches are:

SW1 = ON, SW2 = OFF, SW3 = ON, SW4 = ON

This "Microstep 8" factory setting requires 1600 pulses per revolution. In my Python program, if I send in 1600 pulses the motor makes one revolution: I just observed that by eyes only.

When I set the switches to "Microstep 1": SW1 = ON, SW2 = ON, SW3 = ON, SW4 = ON

Then I send in 200 pulses and the motor makes one revolution: but it runs very fast compared to "Microstep 8".

I understand the relationship between switches and pulses. PLEASE HELP WITH THE QUESTIONS:

- When we use this motor and this driver with LinuxCNC, how do we set the Microstep Switches?

- Must all drivers' Microstep Switches be set the same?

Please pardon me if the questions are ignorant, all I have currently is LinuxCNC installed for Raspberry Pi 4B. I don't have any other relevant hardware, so I can't try anything yet. I am just trying to understand the basics.

I've attempted some settings, I have a video showing the speed at different pulses:



Thank you and best regards,

...behai.

Thank you for your reply. I carefully watched the video you sent.
The issue discussed in the video is one that I may encounter later, and it was very helpful.
However, I think I may be describing a different issue. I forgot to mention a detail when I was asking my question just now. It's actually this issue that I'm more eager to solve.

When using different step values, the degree of deceleration of the rotary axis varies. The smaller the step value, the lower the speed. However, this issue does not occur with the linear axis when using G64.

For the linear axis, regardless of whether the incremental step is 1, 0.1, or 0.01, the velocity can reach 3000 when using G64.

With the rotary axis, when the step value is 1, the vel is 2800.
When the step value is 0.1, the vel is 866.
When the step value is 0.01, the vel is 262.

For example:
G64 P0.01
G01 F3000
;Linear axis step 1
G01 X0 Y0 Z0
G01 X1 Y1 Z1
...
G01 X70 Y70 Z70; vel: 3000

;Linear axis step 0.1
G01 X0.1 Y0.1 Z0.1
G01 X0.1 Y0.1 Z0.1
...
G01 X69.9 Y69.9 Z69.9
G01 X70 Y70 Z70; vel: 3000

;Linear axis step 0.01
G01 X0.01 Y0.01 Z0.01
G01 X0.01 Y0.01 Z0.01
...
G01 X69.99 Y69.99 Z69.99
G01 X70 Y70 Z70; vel: 3000


;rotational axis step 1
G01 C0
G01 C1
...
G01 C69
G01 C70; vel: 2800

;rotational axis step 0.1
G01 C0
G01 C0.1
...
G01 C69.9
G01 C70; vel: 866

;rotational axis step 0.01
G01 C0
G01 C0.01
...
G01 C69.99
G01 C70; vel: 262

G01 C0.01

Thank you again for your reply!

phillc54 wrote:

You should have SheetCam set up to either use the materials that are in QtPlasmaC or to create the materials as required.
If you have something like below in the G-code file:

(o=0, nu=2, na=5mm Mild Steel 40A, ph=3.1, pd=0.1, ch=0.75, fr=3000)
then SheetCam is creating the materials.
http://linuxcnc.org/docs/devel/html/plasma/qtplasmac.html#plasma:magic-comments

The default material in QtPlasmaC is set on the SETTINGS tab.

 

After a search I just found this post.
I use free sheetcam for very small files.

I get the above mentioned code in any file it creates. 
 Normally on line 23

It prevents Qtplasmac from using the material I chose in Qtplasmac. 

I always delete that file. 
Are there some settings to prevent Sheetcam from creating this line. 

I want to handle material settings from Qtplasmac exclusively.