Advanced Search

Looking good on Pi4 with SPI connection to my Remora-style PRU.

hi
i have upgrdded my kernel to 6.1.0-23-rt-amd64 from 6.1.0-13-rt-amd64
since then, i have to remove ethercat-dkms libethercat libethercat-dev ethercat-master and reinstall them and linuxcnc-ethercat on every boot
i have followed post 1 of this thread completely .
this is error code when i want to start linuxcnc
"Failed to open /dev/EtherCAT0: No such file or directory"

What exactly do you mean where you don't know how mine works?

On the picture you reposted there is no indication which remora.input is which. That's why I've used the one I've attached. Over there it's clear that for example estop input is remora.input.04, probe is input.05 etc. So the INP3 (on your picture) is really input.06.
It is correct as I've check it.

I've given up on the core machines, personally not worth the time or effort mucking around with.

One thing that may help is taking the PC apart and cleaning it.

Tested with LMDE 6 with lastest updates and still reboot issue when starting linuxcnc with 6i25 + 7i77 & lathe hw attached.
No error messages, kernel log just show normal messages and then booting up information. What to do next?
 

Kapperi wrote:

Sorry if wrong topic but would this be possible with pokeys57cnc?

It is great controller card but I would prefer not to use mach3/4

Short answer is no,
long answer, you can modify the SKR1.4 firmware to work with the RPi/SPI, but there is no ethernet version

In your configuration folder, you should see a 'qtvcp' folder.
In that folder add 'images' and 'widgets_ui' folders.

for versa:
in the images folder copy 'probe_icons' from share/qtvcp/images/probe_icons
In the folder widgets_ui copy anything with versa in the name from share/qtvcp/widgets_ui

for basic_probe same idea
in the images folder copy widgets/basic_probe_help and widgets/basic_probe_icons
In the folder widgets_ui copy anything with basic in the name from share/qtvcp/widgets_ui

if you run qtdragon with the - i or -d debugging on you should see text like this in terminal:

[QTvcp.QTVCP.QT_PSTAT][INFO] Using LOCAL widget file from: /home/chris/emc/configs/sim/qtdragon/qtdragon_tool_probe/qtvcp/widgets_ui/ (qt_pstat.py:329)
[QTvcp.QTVCP.QT_PSTAT][INFO] Using LOCAL images path from: /home/chris/emc/configs/sim/qtdragon/qtdragon_tool_probe/qtvcp/images/ (qt_pstat.py:346)

Now you can modify those files in your configuration folder and qtdragon will use them instead.

Chris

Just FYI if you just want to wrap G-code to a cylinder
www.scorchworks.com/Gcoderipper/gcoderipper.html

A 7I77 watchdog fault occurs when communication with LinuxCNC is lost.
This is not something you typically recover from without restarting
LinuxCNC, that is, it usually means LinuxCNC or the host PC has crashed.
or host communication to the FPGA card has been lost or serial communications
have been lost

 

Question 1:
Is there an output pin for an external Estop/reset safety circuit to read the status of 7i77 WDT? If not, would it be feasible to use one of the existing pin outputs by Keeping this output pin ON while the watchdog timer has NOT bit. Here where the catch 22 comes into play, when the WDT bites, the hostmot2 driver disables the digital outputs therefore my external hardwired circuit will not be able to determine the WDT output pin status (see diagram Highlighted in RED). One way I can think of to solve this puzzle is to restrict the WDT from internally disabling the digital outputs, and have WDT just set a flag and let the hardwired safety circuit take care of disabling the power to the digital outputs.       Or I am out to lunch?Question 2:
7i77 field terminal block has several terminals I am hoping to be able to keep the field power OFF to the digital outputs until the safety lockout circuit is reset and the field digital inputs and VIN powered while the ESTOP circuit is active. If you look at the wiring diagram, I need the numbers of the terminal blocks marked with question mark (highlited in green). Nomenclature:
CR# – CONTROL RELAY
MCR - MASTER CONTROL RELAY
F# - FUSE
DI – DIGITAL INPUT
DO – DIGITAL OUTPUT
PB – PUSH BUTTON
WDT – WATCH DOG TIMER
L1 - MAIN SUPPLY LINE 120 VAC
N - MAIN SUPPLY NEUTRAL
PL- Pilot Light A brief explanation to what this hardwired circuit does follows this diagram.
  The wiring circuit diagram below  is a hardwired circuit that will allow power to be removed from digital outputs, servos and spindle. The circuit achieves the following:

• If power is lost during operation the machine will not restart when power is re-established.
• If an Estop is activated or a fault occurs during operation, the circuit will inhibit (prevent) power to digital outputs, servos and spindle. The rest of the machine control circuit will stay in the ON state; this will enable you to troubleshoot and resume operation without major reinitialization of the machine.
• Machine will not restart until all faults are corrected and operator presses reset button.

This hardwired circuit is the equivalent of safety relay without guided contacts.

I would expect you use a scale of 180
and get 90 degrees for a value of 90 (2 ms width)
and -90 degrees for a value of -90  (1 ms width)

You could also set the offset to 1 and then use a 0..180 value
to get 1 to 2 ms

After some time away I got this working with the 7i96_5abobrcd firmware.  The physical pins DB25 vs. HAL pins had me twisted around.

Just a general question, on this function.  To get the servo to physically rotate 180 degrees (+-90) I need a scale of 95.  180 gave ~90degrees, and 90 gave ~190-200 degrees. Is this just the variability of the servo potentiometer?

 

Whatever size you can put the cable through and do at least 1 or 2 windings.

I would venture a guess at interference and check that first with a scope or a DVM that has a min/max memory feature.

I made a header board for 5A-75E8.2 and started testing the litexcnc firmware and encountered the following problems.
1- When I close the linuxcnc software or press the estop key, the motors start moving.
2- For homing using encoder z pulse, after receiving the z pulse instead of stopping, it makes a strange movement at full speed and then stops.
I posted this problem before but couldn't find a solution for it.
Does anyone have any suggestions for solving these problems?

I cant say anything to problem 2, but as for problem #1:

Are you trying to use software for an e-stop? This is a terrible idea! IMO anyways. An E-stop Should cut power to your drives AND send a stop signal to your software. (one exception is: sending an e-stop signal to your spindle. if you cut power it will spin down for a long time. if you send it an e-stop signal it should stop almost instantly)