Hello Guest it is March 29, 2024, 12:48:26 AM

Author Topic: Digitize Wizard Issues  (Read 8824 times)

0 Members and 1 Guest are viewing this topic.

Digitize Wizard Issues
« on: July 22, 2011, 05:02:22 PM »
Hello,

I am attempting to digitize a chair seat surface. I am entering the parameters and am successful at generating the g-code for the probe process. I have tested the probe and see the LED on the Diagnostics page change color. 

My probe follows the code correctly. I do get 'Ignore Probe' errors, which stop the process until I restart the cycle.

However, when I look at the .tap file code there is no change in the Z axis, that is, the process is not recording the surface.

I am attaching the first part of the tap file saved as a word document.

Does anyone have an idea of what I am doing wrong?

Thanks,

Vicrider
vicrider

Offline BR549

*
  •  6,965 6,965
    • View Profile
Re: Digitize Wizard Issues
« Reply #1 on: July 22, 2011, 08:53:12 PM »
The tap file is the Gcode file that drives the machine while it is probing the surface. The G31 is the probe command it is telling the machine to move down in Z untill it either trips on something or reaches the target Depth.
 THe M40 code in the beginning will OPEN the save point file that the operation will generate. It will contain all the position values for each probe trip, XY and Z.
That will give you the point cloud for the suraface . Then you will need to convert the point cloud into a 3 d contour file so it can be processed with a 3d cam program to generate a Gcode CUT file to replicate the surface.

(;-)TP
Re: Digitize Wizard Issues
« Reply #2 on: July 22, 2011, 11:22:34 PM »
Thanks, BR,

Ahhh, so the .tap file is not the result of the probe, it is the executable for the probe.....no wonder the Z values are all the same. Duh!

I need to find and look at the saved point file..... I think I may be saving it as the wrong file type. I haven't got to the contour file yet, but I have downloaded MachCloud and, if I understand the process, that's supposed to convert the point file to a contour file that my BobCad will import.

Apropos of my inexperience, I have a few other generic questions, tho.

The seat profile is 15" x 15". The maximum -Z dimension is -.75". Do you think a .2" step-over in the x and y axes will produce enough points to generate the surface? Those parameters generate 11K lines of code, which seems lengthy. This could be more of my ignorance showing tho.

The pressure exerted by the probe during the process seems to be inordinately high. It is actually dimpling the walnut original. Is this to be expected or is there a method to adjust the response time of the software to the probe signal to reduce the reaction time and thus the pressure exerted?

Hmmmmm, here's hoping.

Thank you,

Vic
vicrider

Offline stirling

*
  • *
  •  2,188 2,188
  • UK
    • View Profile
    • www.razordance.co.uk
Re: Digitize Wizard Issues
« Reply #3 on: July 23, 2011, 03:33:35 AM »
The seat profile is 15" x 15". The maximum -Z dimension is -.75". Do you think a .2" step-over in the x and y axes will produce enough points to generate the surface? Those parameters generate 11K lines of code, which seems lengthy. This could be more of my ignorance showing tho.
depends on the detail of the surface you want to capture. If there's no fine detail then you could probably get away with much greater stepover. Can't see the seat from here so hard to say. Try larger stepover and if it doesn't capture the detail you want then reduce it. Until you get a feel for it you could start with an inch stepover.

The pressure exerted by the probe during the process seems to be inordinately high. It is actually dimpling the walnut original. Is this to be expected
Absolutely NOT. The probe should trip "feather light" - something is very wrong. Either your probe spring is way to heavy OR see see below.

or is there a method to adjust the response time of the software to the probe signal to reduce the reaction time and thus the pressure exerted?
Mach's response time is "step instant". That is: Mach will start to decelerate the instant a step causes a probe trip. BUT - how long it takes your probe to actually stop is a function of your machine's acceleration and probing feedrate. If the distance it takes to decelerate causes the probe to hit end travel then you'll be forcing the probe into the surface.

Ian

EDIT: To test whether your deceleration DISTANCE is less than your probe travel (which it ALLWAYS should be if you don't want to break something), probe something soft enough to give, but firm enough to trip, like a piece of polystyrene and then compare your Z DRO with gcode var 2002 when the probe has stopped. The difference is your deceleration distance at your ACTUAL probing feedrate.
« Last Edit: July 23, 2011, 05:48:28 AM by stirling »

Offline BR549

*
  •  6,965 6,965
    • View Profile
Re: Digitize Wizard Issues
« Reply #4 on: July 23, 2011, 11:34:59 AM »
SOme Probes do NOT have much overtravel in the Z axis or as Stirling suggests has a WAY too heavy of a spring. The XY travel has a leverage on the spring so it may move easier the Z does not.

Just a thought, (;-)TP
Re: Digitize Wizard Issues
« Reply #5 on: July 23, 2011, 02:09:23 PM »
Thank you both for you time and attention,

I am certain something is wrong with the probe process, I just don't know what. The probe moves quite easily to finger touch, and trips the diagnostic screen LED, so I don't think the spring is too strong.

I am afraid I do not understand your instructions completely. I should run the probe manually against the styrene (not with the wizard?), read the DRO at the touch point and then when? I don't know what 'gcode var 2002' means.

Sorry for being so ignorant, but I do not know exactly how to make this comparison.

If you could elaborate on your instructions, it would be greatly appreciated.

Thank you,

Vic
vicrider
Re: Digitize Wizard Issues
« Reply #6 on: July 23, 2011, 02:16:45 PM »
Here's another question,

'how long it takes your probe to actually stop is a function of your machine's acceleration and probing feedrate'

if the deceleration distance is too great, what parameters do I change? I am using 30 in the Digitize Wizard feed rate bloc. Are there additional settings for the probe in a configuration page?

Vic
vicrider

Offline stirling

*
  • *
  •  2,188 2,188
  • UK
    • View Profile
    • www.razordance.co.uk
Re: Digitize Wizard Issues
« Reply #7 on: July 23, 2011, 04:22:58 PM »
Open menu item Operator/Gcode Var Monitor and enter 2002 in one of the Var address boxes. Then from the MDI screen use something like G31 Z-10 to probe the styrene or whatever (the actual value (-10) doesn't matter as long as it's well lower than the surface of the styrene). Then you'll see the value of var 2002 change. That is indicating the Z position where the trip actually occured and the Z DRO will show where the Z axis actually stops.

If the difference between these two readings (the deceleration distance) is greater than the probe travel then assuming you've set your acceleration in motor tuning to the optimum value for your machine, then all you can do is lower the feedrate at which you probe. That will reduce the deceleration distance, because just like in a car, the slower you go, the quicker you can stop.

Ian
Re: Digitize Wizard Issues
« Reply #8 on: July 26, 2011, 02:45:57 PM »
Hi Ian,

I ran the test you described today on Styrofoam several times. I noted the default feed rate on the MDI screen to be 6. The results in the Var Monitor (differences between Z travel and Var) were consistently in the .0002" range. Which was considerably less than total Z travel which was in the .09" to .1" range.

I slowed the feed rate down in the digitize wizard and I got a consistent scan without errors.

Thank you very much for your advice,

Vic
vicrider

Offline stirling

*
  • *
  •  2,188 2,188
  • UK
    • View Profile
    • www.razordance.co.uk
Re: Digitize Wizard Issues
« Reply #9 on: July 27, 2011, 04:01:26 AM »
Glad it worked out Vic.

Cheers

Ian