Hello Guest it is April 19, 2024, 12:16:54 AM

Author Topic: Losing lots of Steps  (Read 4376 times)

0 Members and 1 Guest are viewing this topic.

Offline nps

*
  •  5 5
    • View Profile
Losing lots of Steps
« on: April 17, 2009, 12:16:33 PM »
Hi all :-)

We changed our Milling-PC from a T40p Notebook with WinPC-NC to an Asus eee-Top with Touchscreen (cool device ;-)) and SmoothStepper and Mach3.

With WinPC-NC we lost some steps, maybe some hundred when it runs several hours without homing. To avoid that, we bought Mach3, a new PC, Smoothstepper and everything, set the values for Motortuning about to the same values as in WinPC-NC (sounds same, looks same, values somewhat also the same, with very very long ramps).

And guess what, with exactly the same G-Code we lose about 3 times more steps! (some millimeters!!!).

How did i know that steps were lost? .. because of a reference homing between two milling cycles. (and i measured the homing also with a dial gauge and a repeatability of about +/-0.015mm)

Are there any settings which can affect this?

Please help!

(btw.. we have a VK-Technik PA-6040 Mill)

Offline Hood

*
  •  25,835 25,835
  • Carnoustie, Scotland
    • View Profile
Re: Losing lots of Steps
« Reply #1 on: April 17, 2009, 04:23:00 PM »
do you have the latest plugin? there was a homing bug that affected some and not others in earlier plugins.
Hood
Re: Losing lots of Steps
« Reply #2 on: April 17, 2009, 05:13:30 PM »
Long ramps can leave the motor in a resonant point to long causing lost steps.

Offline nps

*
  •  5 5
    • View Profile
Re: Losing lots of Steps
« Reply #3 on: April 18, 2009, 02:13:00 AM »
do you have the latest plugin? there was a homing bug that affected some and not others in earlier plugins.

thanks for the fast reply :-).
yes, we do have the latest version. (SmoothStepper_Beta2_v015ogb.m3p) we got the smoothstepper about 2 weeks ago.
and i don't think it's a problem with the homing itself. the machine loses most steps while milling.

Long ramps can leave the motor in a resonant point too long causing lost steps.
oh.. interesting, i haven't thought about that yet. sounds like a trial-and-error work ;-).


besides that... where can i start to search for the lost steps? :-)