On 03/20/2018 03:13 PM, andy pugh wrote:
On 20 March 2018 at 19:00, Mark <[email protected]> wrote:
Managed to get some shop time today, and was working on the lincurve stuff.
Got most of the kinks worked out, as it now loads without error, but now
getting joint following errors as soon as I try to home the machine.
The step-change in X on homing is presumably causing a step-change in Z.
Well, the Z homes first on my machine, X remains at "0" on the display
while the Z tries to home, so I'm not sure how the Z is seeing any
change at this point.
Given that your corrections are very small you can probably afford to
just make the f-error limits bigger than the correction.
No dice. Increasing the FERROR limits didn't do much of anything except
once I reached a certain point, which was not a very large change, the Z
axis licked up and the stepper motor let out a rather loud squeal.
Though if the offset component is set up correctly it should mask this
from the rest of the system.
So perhaps it is a function ordering issue.
Tried moving around the addf's with no change at all to the homing
behavior.
Here's what's happing on the machine while homing, with the original
lines in the main hal file and no lincurve or offset added:
###########################
net zpos-cmd axis.2.motor-pos-cmd => stepgen.2.position-cmd
net zpos-fb stepgen.2.position-fb => axis.2.motor-pos-fb
###########################
Machine seeks the Z home switch (up direction) at normal speed which
takes about 2 - 3 seconds, comes off the home switch (down direction)
for about 1 second, then slowly seeks the home switch again (up
direction) taking about 4 - 5 seconds, then sets itself to the offset
position (down direction) taking about 2 seconds and Display resets the
Z position to 2.4". Then the X axis homing commences. Normal normal,
no problems, been doing this a since the machine has been up and running.
Now, we add in the lincurve and offset hal coding changing some of the
original hal coding in the main hal file to this:
###########################
net z-raw axis.2.joint-pos-cmd => offset.0.in
net z-corr lincurve.0.out => offset.0.offset
net zpos-cmd stepgen.2.position-cmd <= offset.0.out
net zpos-fb stepgen.2.position-fb => offset.0.fb-in
net zpos-fb-offs offset.0.fb-out => axis.2.motor-pos-fb
###########################
and this is where the trouble begins.
Start to home the machine again. Machine seeks the Z home switch (up
direction) at normal speed which again takes about 2 - 3 seconds, and
then comes off the home switch as previous but less than the time and
distance than normal and throws a joint following error. Adjusting the
FERROR in the ini file didn't make a bit of difference.
There's something hinky in how offset is taking the offset provided by
lincurve and feeding it to the axis.2.motor-pos-fb. With the original
main hal file line: zpos-fb stepgen.2.position-fb => axis.2.motor-pos-fb
no issues. With these two lines added after lincurve and offset in
play, problems: net zpos-fb stepgen.2.position-fb => offset.0.fb-in and
net zpos-fb-offs offset.0.fb-out => axis.2.motor-pos-fb.
Perhaps I'm not netting the right pins to the right pins?
Mark
------------------------------------------------------------------------------
Check out the vibrant tech community on one of the world's most
engaging tech sites, Slashdot.org! http://sdm.link/slashdot
_______________________________________________
Emc-users mailing list
[email protected]
https://lists.sourceforge.net/lists/listinfo/emc-users
