Advanced Search

Crumbs I was sure that the angle used in M19, was the angle after an index pulse, but can see now its after the enable edge.

The spindle position after an index pulse is the angle relative to the index pulse, not the enable.
The enable arms the hardware to detect the index and when index has  been detected by the
hardware, signal the driver that index has been detected. The counter hardware also either latches
the count at index (Mesa hardware does this) or clears  the count at index so that the spindle position
is set to 0 at the index position.

Thank you for getting back to me. That had already been changed in the original file but I forgot to change it in what I posted online. It still has the same problem. Currently it is on joint 2. The constant homing issue has left the gantry a little out of square. It seems like whichever side is closer at the start has the problem

Both joints of a gantry need negative home sequence numbers,
but your ini file has a negative and a positive:

[JOINT_1]

HOME_SEQUENCE = -3


[JOINT_2]

HOME_SEQUENCE = 3

Hello, I finnally got my cnc plasma table running been a little over a year in the making. Very thankful for all the people on this forum who helped me get it going such as Tommylight and Rodw. With this being said im not perfectly happy with the qtplasmac interface. It is light years ahead of mach 3 but still missing some. From my understanding the way it works is the qt takes over the Z axis entirely and automaticly inputs Lead ins and feeds from the material specs, Its like 70% of the way to being a cam software in itself. This kindof makes a mess if you want to use sheet cam or something similar.  you will have options double up that have to be over ridden, aswell as theres not really a great way to run sheetcam via linux as far a i know. So I am embarking on the wild journey of creating a open source cad/cam intergration into qtplasmac. I think with the modern advancements of ai I can have this project shipped in a few weekends. My qeustion for everyone on this forum is what essential features would you like to see? what improvements could be made? My goal is to give a little back to the community and make my software free/opensource or if it turns into a massive project make it so cheap that it can be sold on a one time very minimal fee to cover dev costs with no encryption so pirating is easy for anyone who cant pay the cost. 

So in the meanwhile, without working UI I tried to test comms with halcmd.

Does
4 bit IN TRUE remora.SPI-enable
4 bit IN TRUE remora.SPI-reset
4 bit OUT TRUE remora.SPI-status

actually mean there was some comms? UART doesnt show anything though.

If it's open loop, the question becomes what is changing the analog out command?

(bottom trace second plot)

Hi - I've gotten Linuxcnc to work well on several machines but each time the machine wasn't constructed with enough precision to make it reliably work the way I wanted.  On this machine I finally built it the way I should have.  I used the files from my old machine and adapted them to this machine If something is commented out it is because I haven't got to that yet.  The machine moves great and the X and Z axis move and home really good.  The Y axis moves good but I can't get it to home properly. If things are set up so that each axis has a pnp sensor on side will respond to the sensor but the other side will turn off the led acknowledging it should home but it keeps moving towards the sensor.  If I tie both sides to a single side it works as expected.  This is fine but it prevents squaring.  I've included my files so maybe somebody could see where I've gone wrong.

Peter,
No its open loop. I've got all the gains at zero bar FF0=1.
I do have an encoder though.

Not really sure whats going on now as I run into segfaults. I havent changed a thing on linuxcnc side, i just changed the config.txt on the SD Card

line 977: 2610 Segmentation fault $EMCDISPLAY -ini "$INIFILE" $EMCDISPLAYARGS $EXTRA_ARGS

Before that I haven't seen anything in the terminal or anywhere, thats why I'm a bit puzzled. Yes I compiled it from main branch, using OS6 Version for the Firmware and likewise for the hal component.

Is there a way to just test it with the halcmd? I tried loading the remora component, but struggle beyond. Show pins shows no active session.

Both the debug file and the regular logfile are basically empty with no specifics concerning the error.

Thanks!

Ehrlich ist gut. Danke für deine Einschätzung.
Was spricht gegen Kohlebürsten?
Die Lautstärke sollte man sicherlich beachten.
Die Frage für mich wäre, wie Laut es wird, wenn der Fräser durchs Material fährt.
Wasserkühlung ist mir zu aufwändig.

Danke für den Hinweis bezüglich des Dauerbetriebes.
Da werde ich bei AMB noch mal nachfragen.

I done some investigations, and it seems my driver is simplified as possible and supports only homing and CSP modes.
There are no registers for velocity and acceleration, the driver uses internal default values.
In LinuxCNC xml configuration you can create hal pins for non existent registers and you can write to them, but it not affect the driver.
My bash script is far of dreams, but works.

If this is closed loop, I would check for issues with the encoder setup
or PID setup

Is the encoder velocity used for feedback?

Sounds like you need to set the frequency scale to .390625

You should post complete serial console output. It will make diagnosis easier.
You should start LinuxCNC from terminal. And terminal window what msg you get when you press estop button in lcnc gui. Is it spi_payload = 0 or something else?
Did you compiled and installed remora-spi lcnc component from same repository?

If the interrupt endpoint is not continuously polled, you get the disconnects. But also the xhc-whb04b-6 component keeps it from disconnecting, at least in most situations. You can run that directly from the terminal, for instance with "xhc-whb04b-6 -a" to enable all logging, and you will see that the pendant will not disconnect.

So in this "easy" situation, there is not much difference between the three programs. We know that the pendant disconnects in a long-time stress test with the xhc-whb04b-6 component and LinuxCNC running. I'm not sure either of the other programs would do any better in similar situations.