Why is G38.2 done in machine coordinates?

From the original Synthetos forum (2011–2025)

Started by realthunder on in TinyG Support. 4 replies, last activity Jul 10, 2016.

#

Why is G38.2 done in machine coordinates (G53). Won't it be better in the current working coordinates, and may be in relative distance mode? Is there any good reason to insist on overriding the coordinates?

I am asking because after homed, the tool head is usually at Z max, way above the work piece. The work flow I took is to jog the head to some where close to the intended start point and set the current working coordinate using G54. Now it becomes awkward if I want to use G38.2. I'll have to recall my G54 setting and manually calculate the desired Z value in machine coordinates.

This is roughly how I used it in Chilipeppr. In its auto level widget, right before it issues the first G38.2, it move the head to a predefined location using the CURRENT working coordinate. Then it issues a hard coded G38.2 Z-10 F25 to start the actual probing. In my case, as I've set G54 at a lower point, the tool head actually moves upward.

I think Chilipeppr guys forgot to mention that in order to use its auto level, one must manually set the MACHINE ZERO to some where within 10mm above the work piece. It cost me one engraving bit to realize that! At first I was blaming Chilipeppr for not mention this (or maybe I just misread the instruction). But then after some thought, I realize that the real problem may be tinyg's implementation of G38.2, more specifically the coordinate override bit.

In conclusion, I think the problem is that, in order to use G38.2, one either need to 1) manually set machine Z zero (which will screw up the soft limit if it's enabled), or 2) manually calculate the target position in machine coordinates, or 3) do not set any working coordinate at all.

So, why not just put G38.2 in the current working coorindate, or better, simply make it in relative distance mode?

#

Has the code for G38.2 been adapted to use G10 L2 coordinate systems in the v .98 branch? I sent an email regarding build 449.01 of v.97 edge and haven't received a reply yet.

I figured it would be some kind of problem with this. Are there any ideas on how to use the new homing limit switch type from build 449.01 ($zsn=4 probe) with a manual cycle? This might be made a lot easier with canned cycles and a description of the functionality of this switch type than trying to bottle it up into a single G code command.

Also, the SpDir won't flash when the machine goes into Alarm mode from a limit switch hit, dunno if that's related.

Staff #

You are down deeper in the weeds than this Forum normally has good info for.
From where are you getting build 449.01?
I don't see it posted anywhere, perhaps you ran your own build on the Edge Branch?

TinyG developers don't stop by here frequently. Your question might get attention by posting at

https://github.com/synthetos/TinyG/issues

#

Yes, it's a file built from the edge branch, where the header file that determines the $fb is on 449.01.
I can do that. Hopefully it will bring to their attention the need of work coordinate systems by machinists talking about this on both this forum, their github, and the chilipeppr community.

This topic is from the original forum (July 2016). Your reply starts a new discussion linked to it.

Reply

Different problem? Ask a new question

Back to TinyG Support or latest activity.