× Forum Header

Problem with A axis I am not able to understand

More
10 Nov 2020 08:42 #188849 by kmeld
it says:
Linux frez-desktop 2.6.32-122-rtai #rtai SMP Tue Jul 27 12:44:07 CDT 2010 i686 GNU/Linux

I need the most 8000 steps/sec for my applications.
So you say my base period now set to 100 000 should be in fact set to 1 000 000.?

Cris

Please Log in or Create an account to join the conversation.

More
10 Nov 2020 10:35 #188859 by kmeld
I just have done following test

I had let the machine run a loop program that basically is going back an forward XYZ linear axis within their limits and in the same rotating A axis in a constant direction.

after about 393 turns A motor had stall.

What I have observed:
1) Z and A motors have similar temperature (X,Y) are not accessible to test. I would say motor temperatures are ok. I can hold them bear hand.
2) I have tested torque required to rotate A axis in stall position (after switching off motor). ~3,5 Nm.
3) I have noticed strange behaviour of all axis when jogging after. So response for keyboard input for all axis (one a a time) was slow. I mean it took much longer than usually to stop the axis after I let go jog key. Normally when jogging reaction is mediate.
Sometimes I have the same behaviour when I load big G-CODE file.

Cris.

Please Log in or Create an account to join the conversation.

More
10 Nov 2020 14:04 #188875 by andypugh

kmeld wrote: So:
max jitter for base thread is 561920 ns


Has it always been that bad? 0.56mS would normally be considered completely unusable with software stepping.

You could try the latest LinuxCNC LiveCD (just boot from the image and run a latency test, don't install it) to see if your system is better with a different realtime system.

Please Log in or Create an account to join the conversation.

More
10 Nov 2020 14:08 #188878 by kmeld
not sure if it was always that bad. After setting up this system long time ago I was just using it.

Please Log in or Create an account to join the conversation.

More
10 Nov 2020 14:14 #188880 by kmeld
Anyway
I did some more testing.
After A motor stalling after doing ~393 turns in + direction I have stet up the same test but in - direction for A axis.

I have interrupted it after ~1059 turns with no problems.
After that jogging was laggy as before.

I then intended to run it again in + direction again.
The moment I hit R to run the program A motor stall.

Now it is running again.

What is the physical phenomena behind stepper motor stalling? Is this mechanical or electrical?

Cris.

Please Log in or Create an account to join the conversation.

More
10 Nov 2020 19:17 #188901 by Todd Zuercher
Stalls can be caused by electrical or physical problems, or a 3rd cause timing problems (which is sort of a physical problem).

If there is a physical problem causing too much resistance to the move, such as too heavy a load or a bad bearing, accelerating too fast, or too high velocity, (a stepper's torque drops off dramatically with RPM), it can induce a stall.

A poor electrical connection, or bad power supply or failing driver can cause a stall.

Bad timing,by inducing very brief accelerations that exceed the available torque starting the stall.

Once a stall starts, speed must be decreased until the motor can find enough torque to recover movement. (Often the step pluses must nearly stop to recover.)

Think of the load of your motor as resting in the valley of a wave. Taking a step shifts the wave to the side, the load then starts to "fall" down the wave moving to the valley. if the next step comes too soon, the load could fall over the top of the wave to the next valley instead of remaining in the desired one. The problem with resonance can be looked at the same way. imagine that between steps the load has time to roll through the valley then begin to roll back. If the timing of the roll back matches up with the next step right it can also lead to the load crossing the peak.
The following user(s) said Thank You: tommylight

Please Log in or Create an account to join the conversation.

More
11 Nov 2020 08:45 #188965 by kmeld
Is there any way I can monitor if problem is electrical?

Cris.

Please Log in or Create an account to join the conversation.

More
11 Nov 2020 09:04 #188970 by kmeld
As for the base period.

I tried to change it from 100 000 to 1 000 000 editing ini file but it did not quite work.
Program did not start giving error message.

How do i need to edit ini and hal files to change bate thread to 1 000 000 (as it was suggested).

Cris.

Please Log in or Create an account to join the conversation.

More
11 Nov 2020 10:25 #188976 by tommylight
What error mesages?
There should be none, only if you edit the files on a windows PC as that will inserts line endings that are not used in Linux.

Please Log in or Create an account to join the conversation.

More
11 Nov 2020 11:16 #188982 by kmeld
After changing base period in ini file (edit on linux) I am getting following error: (file attached).


Also I tried downloading and running live CD with linuxcnc-2.7.14-wheezy
It does boot.
but only in failsafe mode I ma getting UI.
I tried to run latency test. It does not start.
Also linuxcnc does not start giving error message saying (as far as I can tell) "rtapi kernel" is not loaded or can not be accessed.

Cris.
Attachments:

Please Log in or Create an account to join the conversation.

Time to create page: 0.113 seconds
Powered by Kunena Forum