Touch Probe problem (wavy STL output)

[jgrove255]

Hello All,

I've been having an issue (or what I deem as an issue) while scanning with my touch probe and was wondering if anyone else has had a similar problem. Irregardless of the X,Y scanning resolution, the resultant STL file exhibits a regular wave type pattern in it. The object being scanned is a rather mundane 16" x 12.5" area of a flat, 1/8" board and the surrounding spoil board...Pretty boring, however there is a purpose to this madness that I'll share if I can get it all worked out!

Here's a hopefully clear description of what I'm witnessing, and I believe the cause of the wavy pattern, while scanning. First the probe will descend and touch the surface (the control panel shows that the 'probe is activated') and will then 'drag' the probe the predetermined distance of the step resolution. Next, the probe will then lift slightly from the surface (the control panel shows the probe in an inactive state and the noticeable, read slight, deflection of the probe while in contact with the surface is now gone) and then again moves the predetermined distance of resolution where it again lowers the probe and starts the aforementioned process over again, ad infinitum. The troughs of the wave seem to correspond to the probe's contact with the actual surface, while the crests seem to correspond to the non-contact state. I'm guessing that when the probe transitions from its activated to unactivated state there may be a 'bounce' from the mechanical nature of the probe 'springing' from its deflected to undeflected state, thus recording a point while moving up in the Z-axis. However, I would assume that the control box/software would have some sort of 'debounce' implementation (either hardware or software based) to counteract the mechanical switching nature of the probe. In all, just a wild guess to try to explain what might be going on...

The settings I used in the various trials of the scans varied in resolution (.08"-.5") and Z scan limits (generally held a 1/4" buffer above the scanned area), while all others remained the same.

Scan Limits
X: 16
Y: 12.5
Z: .375 -> .5

Step Sizes:
X: 0.08 -> 0.5
Y: 0.08 -> 0.5

Scan Velocity:
10 in/min

Here's a screen capture of the resultant STL file scanned at 0.08 X,Y step size:

@ 0.08 X,Y Resolution

Here's another screen capture of the resultant STL file scanned at 0.125 X,Y step size:

@ 0.125 X,Y Resolution

and finally, here's a side view showing the sinusoidal wave pattern:

Side View

For clarification, in the picture above, you can see the pattern exhibited on the top surface of the 1/8" board... and perhaps more clearly, towards the bottom of the model, the sinusoidal pattern exhibited on the spoil board.

Hopefully you guys get the gist of my explanation and will perhaps be able to shed some light on the issue.

Thanks in advance!
James Grove

P.S. I'm using the most recent control panel software: 1.5.0.23f with the HD version of the Shark.

[Shipfitter]

Have you used this probe before?

[ohiococonut]

I only used my touch probe once when I first got it to make sure it worked so it's been a while. I have to assume the pattern you are getting is due to the resolution you are using. The finer the detail you are trying to simulate, the finer the resolution you would use. It's supposed to be able to duplicate down to .001. On a larger project you may not require exact detailing in .001 increments. On something such as a quarter it may be necessary and you most likely won't see this type of pattern. Try setting your step over rate down to .001-.005 and scan a flat surface. If there's a pattern there you may have to zoom way in to even see it and it will be much tighter. Those are my thoughts

[Shipfitter]

I read somewhere that there was a break-in for this probe. It took me 3 times before the darn thing started doing what it was supposed to..