Well you could try upgrading to Trixie which has linuxcnc 2.95 and simplescreen recorder as standards. 2.97 is included in testing so you can use APT pinning to get 2.9.7 

packages.debian.org/search?keywords=simp...e=trixie§ion=all

zippoffs wrote:

I think I got it! 

• Use ferrites/EMI filters on the lines coming in and out of the VFD - this solved my problem!

I'm happy to hear of your success! image missing /media/kunena/emoticons/smile.png
I want to add a clarification to that one bullet point: EMI filters are very good for the input side of VFDs, but don't try them on the output side unless they are listed for such use. The output side of a VFD produces such a high level of harmonics, a common emi filter will smoke. Ferrites are fine for input and output side.

i did notice you use machine zero as the G54 offset. try setting it off that and run the test.

I think I got it! 
I tried moving the AC lines by the BOB, but that didn't work, probably because I didn't move them far enough. Last night I added a ferrite on the AC lines, and I was able to run the program without failing. This morning I removed the earth ground because it was temporarily through my window, and ran the program again. It did fail. 
When I started building it I had installed an EMI filter on the main power coming in, before the breaker box. I now realize that anything coming in is negligible compared to what the VFD produces. So I moved the filter to the VFD 220v lines. I think that fixed it! I have now run the program three times without fail! (see pic)

Thank you everyone for your help!

For those looking at this later, here is what I learned:

• Interference is a BIG ISSUE
• Separate AC and DC lines as much as possible
• Use shielding on low voltage signal lines (maybe fine without if 24v)
• Use a shielded spindle cable and keep it short! 
• Ground all shielding, but only at one end
• Avoid ground loops - only have one ground connection for each component
• Use ferrites/EMI filters on the lines coming in and out of the VFD - this solved my problem!
• Ferrites may help stop interference - try putting them if other things don't work

Do these things from the beginning! It will save you a huge headache.

github.com/linuxcnc-ethercat/linuxcnc-et...631ab668516b649fcaaf

Posted the PR
Will be happy to get some feedback, thanks

Okay,I did a dumb haha. I got a little too hung up on pncconf not picking up the 7i73. So I started linuxcnc and ran halcmd show pin | grep 7i73 and they're all there. I'm just going to set my pins in my hal file and it should be good to go

Thanks for the quick response. I ran mesaflash --device 7c81 --addr /dev/spidev0.0 --spi --sserial and seems like it's communicating with the 7i73! I'm still not sure why pncconf isn't showing it? Is there some extra hal configuration that I need to do?

Here's the truncated output from the mesaflash command, just to show it is there
...
SSLBP port 0:
SSLBP Version: 1.43
SSLBP Channels: 2
SSLBP Baud Rate: 2500000
sserial device at channel 1: 7I73 (unit 0x130001F9)
SwRevision = 14
HwRevision = 1
NVBaudRate = 2.5Mb
NVUnitNumber = 0x130001F9
NVWatchDogTimeout = 50ms
.....

I could but that is a pretty huge overhead on a old pc. 

Ich würde das mit OR2 ,in der Maschinen hal versuchen

linuxcnc.org/docs/stable/html/man/man9/or2.9.html

Hi, 
i installed Linuxcnc 2.9.4 , and would like to use Mouse with bluetooth(logitech m350s).
kommand sudo apt update shows that adding this repository is not safe and disabled by default.
WHAT TO DO ?
 

Hi All,

My Maho 400E is still running an old version of Debian, with XFCE, that would have been the live CD back in about 2019. For my videos, I need to install simplescreenrecorder, as pointing a camera at the screen is lame.

I have tried:
sudo apt-add-repository ppa:maarten-baert/simplescreenrecorder
sudo apt update
This fails with "unable to locate simplescreenrecorder."

Please help with this. I seem to forget all linux knowledge the instant whatever I was trying to do gets done
Cheers,
Mark

 

да пожалуйста.
According to my observations, the queue error simply replaced the error  Probe tripped during non-probe move.

zippoffs wrote:

Wow, that's a lot to take in.

 

Sorry, I think I made too big a deal about some of this. That is a nice mill you've got. With a few tweaks it'll be fine.