serious error of position

More
10 Apr 2014 22:49 #45798 by sgelettronica
ok
nema 34 motors 9,2 Nm
ma860h 51200 khz 300 microstep driver
MESA 7i43 card
50 volt dc power supply 25 amps
or used all versions from year 2011 until the version 2.5.3
all with the same defect
tried the mechanical match3 all ok
I work with this pick and place for PCB mounting, fix all errors with positioning the camera when everything is' correct machine always executes the movement in a precise manner

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

More
10 Apr 2014 23:28 #45799 by PCW
Replied by PCW on topic serious error of position
The timings would be wrong if the step polarity was wrong so

I would try this on X and Y:
Set the ustep ratio to 1/4 of what is is now and
change your scale accordingly

Set step length to 5000
Set step space to 5000
set dir setup to 10000
set dir hold to 10000

and see if your symptoms change
if no improvement, then try inverting the step polarity

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

More
10 Apr 2014 23:57 #45800 by sgelettronica
tried in all ways
400 microstep, 4.7059 scale
Steplen, stepspace in this calibration to ensure the lowest ratio f_err
error remains the same.
and 'from 2011 that work with this error, I came to the conclusion that the software introduces this error.
no one noticed this error, because everyone is using screws

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

More
11 Apr 2014 00:05 #45801 by PCW
Replied by PCW on topic serious error of position
Please use the numbers I suggested
post your updated hal and ini files here
and position test results

Thanks

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

More
11 Apr 2014 05:05 #45822 by cncbasher
Replied by cncbasher on topic serious error of position

thanks for the reply.
what you say is not 'possible otherwise all machines with stepper motors would not be used, however, admit and accept your thoughts:
a ride in my car, is equivalent to 75 mm of displacement thus:
75/200 = 0.35 mm steps that should be the maximum error in accordance with your thinking logically in a single movement. according to your reasoning the error must be random, LinuxCNC the error repeats perfectly in the usual points to the usual extent. and 'to consider that the return to zero and' always perfect from any point.
tests: installed Mach3 does not have this defect.


i doubt , i feel more a question of understanding the error , more fully , are you also using 300 microsteps ?
please upload your hal and ini files .
also can you specify the type of belt and pulleys used , dia , teeth etc , as this also will have an effect on the problem , so the more information the better .
and the actual number of microsteps actually used .

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

Time to create page: 0.067 seconds
Powered by Kunena Forum