Advanced Search

ok, I got this screen, I didn't do any more procedures here. What to do now? restart linux or is there any other procedure?

I have a sneaky suspicion that the W5500 has trouble keeping up with the small packets.
I’ve noticed that when there is an issue with the packets there is quite a long response time. It’s difficult as the packets seem to have issues at random intervals, there doesn’t seem to be anything consistent to it.
Tho the code doesn’t seem to do the required software reset, but one would think if it was an issue it would show up quite early in the piece.
Another option would be a soft core to handle the Ethernet interface. From looking at the Mesa code the softcores they use are to handle the smart serial and Ethernet comms.
But in all honesty the bridge works great.

Note that 'halui.axis.jog-speed' needs to be set for plus minus jogging:
 

I think I know what the solution is but I just cant seem to get it to work. Fusion360 has a post processor training guide and in it they discuss a section called polar interpolation, which is exactly what I need, it replaces one of the linear axis with a rotary axis and outputs strictly polar coordinates. If I go to prebuilt millturn posts such as haas 35st, doosan millturn, etc - and use that post to process my machine kinematics and toolpaths - it does output seemingly correct polar gcode. However, their post processors are significantly more complicated and have a variety of gcodes that are not relevant to linuxcnc. HOWEVER, when I try to manually integrate polar interpolation to the linuxcnc milling post, I am unable to get it to work, it still always outputs cartesian coordinates and or says the direction of my milling is incorrect.

Hm, as usual I'm concerned about unwanted path distortion:

 

I presume these splines were added by the algorithm, if so it seems like a lot of distortion for 'G64 P0.1 Q0.1'?

Hello, I just heard of this software,
Seen it in Youtube video
"Building The 5-Axis CNC Machine in a Small Room: A DIY Journey. (Part12)"

Does the software reduce the RPi's cpu load?
or,
Does the RPi still run X and OpebGL and Axis gui?
(I _think _ nhose are the alternatives )

Thx
TomP

sudo apt purge linuxcnc-uspace linuxcnc-uspace-dev

Thank you Mark!

Here is a picture of the back side of the panel, 7i73 reading switches, 2 potentiometers and mpg encoder.
It also drives 30 LEDs under switches (multiplexed). The multiplexing is currently done in hal but I will try to write a component for it.

 

And this is a custom power detection board. Basically just 12 AC optocouplers and a relay output to detect a open breaker or motor overloads.

 

Also do you insert pictures in posts as attachments or someway different. On some pictures even if I compress them, forum still complains about large file size.

i tryed RMII with the tangnano9k, bitfile generation works, but get no connection
better go with esp32:

github.com/multigcs/LinuxCNC-RIO/tree/main/UDP2SPI-Bridge

i prefer the WT32-ETH01, cheap and available

andypugh wrote:

I knew there was another HAL component that might be relevant, I don't know how I missed it:

linuxcnc.org/docs/stable/html/man/man1/sendkeys.1.html
 

 

Andy,

I need some clarity on a few points for sendkeys.

At this point I have an Arduino board that can trigger events in halshow, So when I activate an Arduino connected push button, the jog-x-pos and the halui.axis.x.plus changes from yellow to red. I expect the actual X-axis to jog when I press that button, however other than the yellow to red indication nothing happens. (the system is NOT in Estop, and the machine is enabled on the manual control tab. If I mouse click on the "+" button, the x axis does jog.

Do I need to enable/invoke SENDKEYS, to actually get the yellow to red indication of jog-x-pos to do something?

Assuming that I do need to enable SENDKEYS, I need a little help with a few points in the SENDKEYS documentation.
 Instructions say "Create a new entry in .B/etc/udev/rules.d/99-input.rules"

I'm not sure where ".B" is?

I did find       /etc/udev/rules.d/70-persistent-cd.rules
So neglecting the missing ".B"  I assumed this is the file I'm supposed to update?
However when I add "sudo echo KERNEL=="uinput", GROUP="uinput", MODE:="0660" | sudo tee /etc/udev/rules.d/70-persistent-cd.rules" to that file I get the attached error message. So lots of possible mistakes here: wrong location, wrong file, adding the uinput group did not work even though it seemed to (see last attachment) the discrepancy between 99-input.rules ,77-persistant-cd.rules , 88-input.rules?

Please advise.
 

 
 

Hi Arciera,

Ok, I will not worry about it.

I tried a free gcode file, a griffins logo. File about 14000 lines.
www.cnccookbook.com/g-code-examples-file...for_Milling_Machines

In the file no G64 P.. Q.. was given. Raw output :
 

Then i added G64 P0.1 Q0.1
The output : 


Then without the velocity display :
 

The purple are all clothoids. Cyan is spline, White is line.

File specs :
items G64 P.. Q.. before:14151 after:28292 diff:-14141
So the gcode file is increased by 14141 segments.

Maybe test some more files?

 

meister wrote:

At the moment a UDP2SPI bridge (esp32) is still the best solution.
 

This one?
WT32-ETH01 ESP32 Ethernet Development Board 

Don't worry about it, it's really not important.

@meister

Late on and I got it to work. However and as you've pointed out, the comm needs fine tuning! Can I bother you once more and ask for a detailed hardware desc for the ESP32 you've mentioned? Is there a UDP module in particular you'd like us to use and dev around? I'm also happy to get a RMII breakerboard as a second option. Both would be ideal though. All-in-all it's working well. Congrats!!

Visual of the w5500 output:

Yes, i saw some posts but they did not share any clear results.

Thank you.