smartprobe - configuration questions

More
22 Jul 2017 01:40 #96255 by choppero
N24 (LOGEOPEN/home/operator/gcode/PROBE DATA/proberesults1.txt)
was added to my file and ended with the (LOGCLOSE) at the end of the file. No data entered into PROBE DATA/proberesults.txt.
I don't understand what your comment "magic comment" refers to. Is it sarcasum? What wold be the syntax be for your (LOG,...) "magic comment"

Please Log in or Create an account to join the conversation.

More
22 Jul 2017 07:14 #96258 by andypugh
(LOG, ...) uses the same syntax as (DEBUG, ...) described in the link I gave
linuxcnc.org/docs/html/gcode/overview.html#_probe_logging

Positions and probe results are stored in the numbered parameters described here:
linuxcnc.org/docs/html/gcode/overview.html#_parameters

They are known as "magic comments" because they are commands that follow the format of comments but actually have an effect. (Which comments shouldn't, really)

Please Log in or Create an account to join the conversation.

More
23 Jul 2017 02:16 - 23 Jul 2017 02:17 #96301 by choppero
Thank you. I have poked around the the G code and I begin to understand the structure. The LOG would not write to my chosen directory of PROBE DATA but when I changed the destination to logdata,viola. Now I believe I'll have to try to change smartprobe. I think to probe a shaft radius, I would need to probe from the Y direction and step Z up from a start position. Probing down causes all kinds of unknown deflection as the tip moves into the radius. I think i would also need to change the Z 0 to half the dia. of the tip. to get an approximate location. Then there is the matter of the diameter of the tip as it hits Y. Would doubling the dia in the tool table have the effect of measuring actual Y or would making it Zero be better?
Last edit: 23 Jul 2017 02:17 by choppero.

Please Log in or Create an account to join the conversation.

More
24 Jul 2017 00:08 #96327 by choppero
With you help I can use this program as it was intended and I just tried using an edge finder to find the radius of a shaft with just three points. I used a known shaft and found the edge finder got it right-on. I did get to learn more about the linuxcnc program from which Pathplot is based. So I think this is a win win.

Thanks again for the pointers.
Tom

Please Log in or Create an account to join the conversation.

More
25 Jul 2017 17:50 #96405 by andypugh

I would need to probe from the Y direction and step Z up from a start position. Probing down causes all kinds of unknown deflection as the tip moves into the radius.


If you probe down fairly quickly then probe up very slowly the tip deflection when contact is re-made should be very small.

Please Log in or Create an account to join the conversation.

More
25 Jul 2017 21:36 #96424 by choppero
I don't understand where probing up will make contact. I have tried manually probing Y at different heights using the probe Y function in pathpilot which does touch a second time very slowly. Compaired to the three points with my edge finder, the results were very dissapointing (the plot of the 6 points I struck was near .02 off in radius. I have no idea at what point on the ruby tip the actual contact is being made.

Please Log in or Create an account to join the conversation.

More
25 Jul 2017 22:49 #96430 by andypugh
G38.2 probes to the work until the probe internal contacts break.
G38.4 probes away from the work until the internal contacts make. This ought to be fractionally more accurately positioned, more so at very low speed.

Please Log in or Create an account to join the conversation.

Time to create page: 0.090 seconds
Powered by Kunena Forum