Luxmill CNC retrofit project

Today I checked my Hal file and corrected it. You were right the corrupted pin names was one problem next to some others...

The G33.1 code (spindle / Z-axis synchronized motion) works nice. But if you type in for example M19 R45 nothing happens. The Spindle doesn't even start to turn. There is also no failure at LinuxCNC startup.

Do you have any idea what the problem could be?
Is it necessary to set an additional CW/CCW-signal for M19?

Best regards
Oehli

Spindle orient requires support for your specific spindle interface to be included in your hal file.

For many systems you would need the orient component:

man orient

@PCW
The orient component is already added to my hal file.

I tried a lot of things and googled for hours without success. I do not get it running right. Here some problems I found:

1) The closed loop spindle speed control doesn't work. If I set M3 S200 I get the following things:


--> The spindle encoder velocity (hm2_7i76e.0.encoder.08.velocity-rpm) is not equal to the spindle velocity (spindle.0.speed-out-abs), so I guess the pid.s (spindle pid) does't work right - but why?
In this case I also do not get an "spindle.0.at-speed" signal. So the G33.1 also doesn't work too.
When I increase the "MAX_ERROR = 0.2" to about 50 the G33.1 works. But this is not a solution....

2) I already mentioned I added a 2000ppr quadrature encoder to the high speed encoder input (10MHz) of my 7i76e in differential (RS-422) mode with A/A-, B/B-, Z/Z-. I have a stable but not equal (see 1) encoder velocity signal (hm2_7i76e.0.encoder.08.velocity-rpm) up to 5000 rpm. If I go further (max rpm is 6000) I loos the signal completely.
According to my Calculation the max count rate of my encoder at max spindle rpm is much lower than the possible 10MHz:

Is my calculation wrong or do I have to modify something?

3) When I use the M19 R120 the spindle doesn't even turn. I get the following :

But for spindle movement my spindle servo driver needs an spindle direction signal (CW / CCW) on
net spindle-cw => hm2_7i76e.0.7i76.0.0.output-01
net spindle-ccw => hm2_7i76e.0.7i76.0.0.output-00
but there is no spindle direction signal via M19. How can I integrate such an signal?


Hope for help!
Best regars
Oehli

I scanned through the other pages but did not see it. What kind of spindle drive are you using?

I use:
Spinde servo drive: Siemens MDX61B0022-5A3-4-0T

can you post your hal and ini files?

Siemens or SEW? Googling that number comes up with SEW drives.

Anyway, check if the drive can do the orientation part by giving it a digital signal. Most dedicated spindle drives can, some normal vfd's cannot. This is the easiest from a linuxcnc point of view.

If that is not an option, PCW is correct and you need to setup the Orient component. In that case I would use the -10v to +10v option for forward/reverse. It is probably better for controlling with the Orient component.

If you intend using orient with a ATC cycle you also need the drive to "hold" the spindle while in Orient mode.

I think it would be better to understand and test the spindle drive capabilities before deciding on which method to follow in hal.

Henk

@Henk: Sorry my fault it's a SEW not Siemens drive.

I added my Hal and Ini Files.

What do you mean with "you need to setup the Orient component"? Do I have to modify the orient code?

Short update Hal File Z402 (just for testing):
net orient-active => pid.s.enable => spindle-orient.enable => hm2_7i76e.0.7i76.0.0.output-01

If I set the CW Signal => hm2_7i76e.0.7i76.0.0.output-01 in combination with the spindle-orient.enable the spindle starts turning but doesn't stop anymore --> spindle-at-pos.out does't become "true".

Edit: I think the SEW spindle Servo drive can also work with +/-10V. But the 7i76e (on TB4) can only deliver 0...+10V due to the potentiometer functionality on TB4. Please correct me if I'm wrong.

Rather complex halfile....did it work before?

I think you can get -7.5 to +7.5 v from a 7i76. Maybe PCW will confirm otherwise check the docs