smartprobe - configuration questions

Jim,
I'm not sure about the half the part height thing but if it works for you, use it. The part I was probing was 6" at the highest and 0.25" at the narrowest using a rotary axis. I only had to add 0.05" to the height to not get the error. Essentially, just enough that the probe wasn't touching when it started to move over one step. If the probe contacted the part prior to arriving at the stepover location, it would simple rise until it lost contact again plus the 0.05". I did have to slightly adjust the micro adjustment screw on my probe and also slow down the stepper motor acceleration because the quick movements would falsely trigger my probe. My machine is also homemade and I'm using an $80 probe I got from ebay. As far as dynamically changing the variables, I'm pretty sure Linuxcnc only reads the gcode file when it is loaded. I'm glad the added lines helped but I just borrowed the changes from someone a lot smarter than me.

Joe,
Thankyou for the update. I will do more experimenting later in the week. I did find that the program did not put an end of file M02 so you have to do that manually for linuxcnc to be able to read the gcode. I only ran this once to maybe I did something wrong for the end of file. I added the M02 at the end with an editor and it showed my part in axis. I will try to keep updating here as I go. It was fun to watch it work. Next will be to actually run the gcode and see what happens,

I had to add an end of program code too only I chose M30.(not sure if it matters which one you use) I also added the following codes at the beginning
G20 G90 G64 G40
( T3 : 0.125 )
T3 M6
G0 Z0.25
F30

I will have to re-run the probing routine with a smaller stepover because I got tool marks every tool width. Something else I didn't consider...., the tool plunged straight down to target depth. Luckily, I started it very slowly and was able to stop it before it broke anything. To compensate for this, I set Z0 well above the stock, then ran the code several times dropping the Z0 slightly each time. I'm sure there's probably a better way but it worked for now.

Thanks
Joe

I am using smartprobe in Tormach pathpilot. Tormach has limited access to the file structure so I can not retrieve the probe results. Does anyone have a way to have the results writen to a directory I've made called PROBE DATA in pathpilot?

I now have the probe writing data to the PROBE DATA file. It does not make much sense though. I could use some help making a point cloud file. I wanted to probe a radius on a shaft so made only one x move. the probe came down and touched then moved in X one step Then moved Y and touched and moved back in X one step and so on. The data file is very strange. here is 4 lines:
-0.000000 -0.500000 -0.197214 9.098637 0.000000 0.000000 0.000000 0.000000 0.000000

-0.000000 -0.500000 -0.187478 9.098637 0.000000 0.000000 0.000000 0.000000 0.000000
0
.010093 -0.500000 -0.179336 9.098637 0.000000 0.000000 0.000000 0.000000 0.000000
0
.010000 -0.500000 -0.197221 9.098637 0.000000 0.000000 0.000000 0.000000 0.000000
0


N12 #1=-0 (X START)
N13 #2=.01 (X INCREMENT)
N14 #3=2 (X COUNT)

N16 #4=-.5 (Y START)
N17 #5=.05 (Y INCREMENT)
N18 #6=20 (Y COUNT)
N20 #7=0.5 (Z SAFETY)
N21 #8=-.01 (Z PROBE)
N22 (END CONFIGURATION SECTION)

N24 (PROBEOPEN/home/operator/gcode/PROBE DATA/proberesults1.txt)

The file should be a set of XYZABCUVW coordinates for each probed position.
You might want to consider using (LOG, ....) instead of probing to format the data in way more useful to you.

linuxcnc.org/docs/html/gcode/overview.html#_probe_logging