Advanced Search

Hello Folks;

System:
Converted knee mill with  old SEM brushed servos running on amc servo drives in openloop mode.
Mesa 6i25 + 7i77
Loop closed with old prototrak  axis scale encoders, not from the motor encoder.

No matter how I try to tune PID & FF, many times the axis moves stop off about .004 sometimes much better.
I have deadband set to 0.
My max feed is 60.0 in/min.
Have tried various accels  now around 30 in/minsq
Most PID seem to have P around 50. No I or D
FF1 from .5 to 2.0 seem to get a reasonable error halscope.
 Just being more than  about a thou off isn't good enough for real milling.
Any info to help or graphs needed just let me know. I have played around with this ever so often for a long time.
But this time, I seriously want to finish this mill
regards,
Jon P.

PCW wrote:

Might be noise coupled into the limit switch wiring


 

If the torch is firing on that video, for sure interference.

Set the networking as described in the readme, connect stuff, open the included config folder, right click on the ini file and choose "open with" then linuxcnc.

its not a random config. it is the exact machine i have, i based the purchase of mesa hardware because it has been done and the config exists. if it didn't i would of chosen a masso g3, mach3 or acorn based upgrade.
currently it is just a bare card on the desktop next to the computer, i want to get that end done before i start making the controls box.

I just connected up a MPG type encoder to inputs 0 & 1 on TB-3 of a 7i96s. I can see hm2_7i96s.0.inm.00.enc0-count moving as expected when I rotate the MPG.

So now how would I go about getting that count to actually move the Z axis?

Worth noting that I also have a functioning pair of jog buttons, I envision using either method (buttons or MPG) to move the axis manually)

Thanks

No idea, i built a NAS last year, so i can not recal if i moved the files there from the phone.
If i do find them, i will let you know.

Been 20 years since i was in Czech Republic, but that seems more like Polish! Or Slovak, maybe?

rodw wrote:

tommylight wrote:

(no Rod, your PC does not have 20 cores, Intel lied to you ... and the rest of the world, Atom cores are not worth a dime).

 
Hey all I care about is I see that count when compiling kernels and things ... rather quickly,.....

Oh hell, i just got a 14400!
It shows 16 cores! yeah riteeeeee.
No time to mess with it till tomorrow, i might do some tests and see if i can disable the economy cores and how much impact it has on performance.
And yes i know, but i am not calling them efficient cores, they use to much power to warrant that name, and they do not much work.

Soviet influence, as it seems, till 2000.
Still, i like more Silicium, so i will stubbornly keep using it.
Silicon and silicone are to messy!

Cant wrote:

Pinium it

Done.
Thank you.

langdons wrote:

But opening up stepconf every time gets really old really fast.

(ask how I know)

Once you have a basic working config, most of us leave stepconf and pncconf and manually edit the config files

the only reason why Cornholio used 6.12 kernel was to get the latest kernel in the hope that support or the Pi5 was better but the earlier one was pretty good anyway. The default Debian uses kernel 6.1 so it is bleeding edge. Ethercat includes custom kernel components so it might not work on the bleeding edge.

Its unlikely you can take a random config and make it work on your hardware.
Use pncconf and reference the example for key parameters
QTdragon is a fairly advanced GUI.
I recommend getting your machine up and running using axis before attempting to install qtdragon

Might be noise coupled into the limit switch wiring

Is there any debouncing on the limit switch signals?

What hardware do you have?

 

My guess is you have a large amount of HF noise so may need a line
filter on the VFD input, common mode choke on the VFD output + better
(lower impedance ) grounding.

The (typically 10M Ohm) load from the voltmeter should have no significant
effect on the output voltage, but in a high EMI environment, the added
capacitance of the test leads may make a noticeable change.