Advanced Search

hey scott, ive been playing with it for the last 2 weeks. I dont have much to show, but i think the environment switching is working kinda. im testing with a nucleoh743, from the debug it seems to be working but serial is messed, it seems i dont know how to do a define or something,

github.com/cakeslob/Remora-STM32H7xx-PIO...43/Src/main.cpp#L116

im not an expert but i think thats exactly what its supposed to look like (rpm changing as diameter changes). assuming those portions that are the straight lines are in between jog moves. I didnt see (or hear) any weird spindle speed drop outs like before thought. i am assuming you figured out the issue?

Thanks Tommy

I think this is exactly what I was looking for.

I'm trying to use ethercat so via the onboard ethernet port to the drives.
Honestly, just because its small and would fit easily in my electronics enclosure.

I'm willing to setup a pc but would be process be any easier?

Thanks

langdons wrote:

Press the machine on button 10 seconds before releasing E-stop.

Does not work, nothing will work while e-stop is active.
See timedelay:
linuxcnc.org/docs/2.9/html/man/man9/timedelay.9.html

qtdragon_hd looks good and I've heard good things about it.

I personally can't get qtdragon to work at all.

But I am not that great at CNC stuff.

Just make sure to not accidentally saw jourself into pieces whilst attempting troubleshoot.

You have a nice HD monitor?

Touch screen is preferable (so I hear).

Press the machine on button 10 seconds before releasing E-stop.

How does the Pi communicate with (to?) the drivers?

Is there any specific reason you are using a Pi, not a PC?

For me, old PCs are very easily obtainable and work well enough (except when they randomy break for no reason).

If your CNC setup works, then don't worry and just move on.

If you are paranoid, just set your max jitter to some really high value in your config.

Don't worry about stuff unless you need to.

Okay, disable hypethreading and turbo.
But don't mess with too much stuff.


"and any further interaction is at this point a worthless endeavour on my part."
Whoa!

Don't just write off people because they did one dumb thing!

Sounds fun!

(not really)

Or different HDD.

Or different partition.

I reccomend never making an OS partition larger than 250GB so that way you might have extra space to make a dual-boot system.

My LinuxCNC PC has a 500GB HDD that has Ubuntu, Debian (For LinuxCNC), and some unpartitioned space.

I would:

1. Disconnect the drive from the 7I77 analog output (with all power off)

2. Power up the 7I77 with LinuxCNC not running

3. Check the analog output voltage of the affected channel
If  it is not near 0V, the 7I77 output has been damaged

If it is near 0V when LinuxCNC is not running, (this is the expected condition) 
It may simply be that the  feedback is reversed or there is some other tuning/setup
issue causing lack of control

I have the servo enabled pins setup, so when Linux is not running, the drives are disabled. But as far as I can remember, the servo was always spinning when i turned the mains on. I just recently found out, it would stop spinning when i disconnect the +-10 plug from the 7i77.

Yeah I think even anon realtime kernel running a text editor with 100 glx windows open would become unresponsive. Honestly I'm of the opinion that is an unrealistic and silly test, that doesn't prove anything. You're testing for a situation that no sensible person would subject a computer to whilst running gcode.
Obviously you are expecting unrealistic things, and any further interaction is at this point a worthless endeavour on my part.