Hello Guest it is March 29, 2024, 04:41:45 AM

Author Topic: Tracking lost steps  (Read 4965 times)

0 Members and 1 Guest are viewing this topic.

Tracking lost steps
« on: October 24, 2011, 08:49:40 PM »
Is there a way that I could script the machine to rehome itself and report how much it has to adjust the dro's? I'm trying to figure out if I'm losing steps and if so how much...it's open loop steppers

Offline RICH

*
  • *
  •  7,427 7,427
    • View Profile
Re: Tracking lost steps
« Reply #1 on: October 25, 2011, 06:56:55 AM »
Quote
I'm trying to figure out if I'm losing steps and if so how much

I am thinking that you are meaning distance.
One needs to define the terms of step measurement, steps in terms of pulses or distance. They relate to each other but the measurement of them would be different.
There are a few  ways, namely electronic and mechanicaly or some combination of the two. No matter how it's done one needs to question
each step of the measureing  process, know the degree of accuracy they can measure, know what they are measuring. Else the data won't be worth much.

example:
Mechanicaly you will move a known distance and then compare the movement to a known standard of that distance.
Electronicaly one will need to compare commanded pulses from the computer  to steps recieved by the motor and then still compare that to
some sort of mechanical movement.

Confused....
Lets say you measure pulses, then one needs to ask....What part of the system are you really measuring? Are you comparing a pulse sent to a pulse recieved?
A pusle sent may not equal a pulse recieved since there may be a mechanical influence between the pulses. Or even elecronmechanical
such that the stepper dosen't react to a pulse ie; say gretaer than 10 micro steps.
Then the question is...WHY don't things add up?

There is a little more to all of this than a simple reply, but, i guess it depends on how anal one wants to be with it all.

RICH
Re: Tracking lost steps
« Reply #2 on: October 25, 2011, 09:47:21 AM »
Didn't think it had to sound that complicated. If I can get the distance mach3 has to adjust itself when homing, I can take that distance. figure out the number of steps it lost, and compare that to the distance it ran in the program. I feel like I can do the same thing if I put a sharp point bit in the machine in confirm 0,0 is still the corner of the spoilboard after running, but it's hard to measure and requires a tool change, not as easy as running a script that could yield more accurate results.

What about scripting the same sequence it does when it homes the machine but just not setting the DROs back to 0? - leaving them alone would give me the numbers I'm looking for.

Offline BR549

*
  •  6,965 6,965
    • View Profile
Re: Tracking lost steps
« Reply #3 on: October 25, 2011, 10:15:19 AM »
Just run verify at the end of the program It rehomes and telll you the error from the last refhome.

(;-) TP

Offline RICH

*
  • *
  •  7,427 7,427
    • View Profile
Re: Tracking lost steps
« Reply #4 on: October 25, 2011, 05:05:31 PM »
Quote
Didn't think it had to sound that complicated.
Guess I was a step ahead of you.  ;)  What are you going to adjust or fix if the number of missed steps are unsatisfactory?

RICH
Re: Tracking lost steps
« Reply #5 on: October 25, 2011, 10:10:22 PM »
I'll slow the machine back down until I'm not sacrificing accuracy for ipm. It's a cnc router, (plywood and sheet material mostly) and I've just started fooling with the motor tuning....as I make changes it would be nice to rehome after a long program or run and see if any were lost - currently I just check if X0 Y0 by jogging it there and looking at it (edge of spoilboard).
Re: Tracking lost steps
« Reply #6 on: October 26, 2011, 11:01:21 AM »
TP - I had overlooked your reply - what do you mean by 'run verify'?
Re: Tracking lost steps
« Reply #7 on: October 26, 2011, 11:28:26 AM »