Discussion:
[Emc-users] Homing To Index Following Error
Todd Zuercher
2017-06-26 16:11:51 UTC
Permalink
I gave Master a try using only 4 joints for my XYZW machine, and that did not make any difference.
I still get a following error on the W (joint3) every first try. (unless the W is parked against the limit switch at machine turn on.)

I don't understand why it always seems to work right for the Z axis, but never for the W. When I scope the homing of the Z the plots look the same, except for the obvious spike of following error on the W, and the time between when the index-enable changes and the step in the position feedback. The Z shows they happen together in halscope, and W seems to be separated by a couple of base-thread cycles.

Is that the source of the problem? And what if anything can I do about it?
--
========================================

Todd Zuercher
mailto:***@embarqmail.com

========================================
andy pugh
2017-06-26 16:34:21 UTC
Permalink
Post by Todd Zuercher
I still get a following error on the W (joint3) every first try.
Is W a second spindle axis?

I wonder if you are the only one actually homing a W-axis?
--
atp
"A motorcycle is a bicycle with a pandemonium attachment and is designed
for the especial use of mechanical geniuses, daredevils and lunatics."
— George Fitch, Atlanta Constitution Newspaper, 1916
Peter C. Wallace
2017-06-26 18:57:44 UTC
Permalink
Date: Mon, 26 Jun 2017 12:11:51 -0400 (EDT)
Reply-To: "Enhanced Machine Controller (EMC)"
Subject: [Emc-users] Homing To Index Following Error
I gave Master a try using only 4 joints for my XYZW machine, and that did not make any difference.
I still get a following error on the W (joint3) every first try. (unless the W is parked against the limit switch at machine turn on.)
I don't understand why it always seems to work right for the Z axis, but never for the W. When I scope the homing of the Z the plots look the same, except for the obvious spike of following error on the W, and the time between when the index-enable changes and the step in the position feedback. The Z shows they happen together in halscope, and W seems to be separated by a couple of base-thread cycles.
Is that the source of the problem? And what if anything can I do about it?
--
Pretty sure the mix of base thread and servo thread cycles is what causes
the problem, homimg to index requires a precise function ordering and I
dont think you can have the correct function ordering when you have the base
thread accesing hardware.

I suspect the easiset solution is to choose a fast enought host CPU that
is capable of running a 4 KHz or so servo thread
========================================
Todd Zuercher
========================================
Peter Wallace
Mesa Electronics

(\__/)
(='.'=) This is Bunny. Copy and paste bunny into your
(")_(") signature to help him gain world domination.

Loading...