Open phillc54 opened 2 years ago
I wonder if anyone is depending on the current behaviour?
Knowing my luck there would be. :)
Maybe the answer is to create yet another pin, motion.feed_mupm ? (or something else, of your choice)
Pins are very cheap in LinuxCNC. I suggest creating two new pins, motion.feed_inches_per_minute
and motion.feed_mm_per_minute
, and updating the documentation to correctly describe the current behavior of motion.feed_upm
.
This provides full backwards compatibiliy and maximum flexibility for people who want to know the feed in either of the standard units.
Because pins are cheap, I think the units should be per second in line with all of the other velocities reported. With motion.feed_upm, its not possible to easily compare feed rate with current velocity in halscope which I have found very frustrating. Either add these pins or use them instead of Seb's suggestions.
motion.feed_inches_per_second
motion.feed_inches_per_second
How about four new pins:
motion.feed-inches-per-minute
motion.feed-inches-per-second
motion.feed-mm-per-minute
motion.feed-mm-per-second
We should probably keep motion.feed-upm
unchanged for now, but note in the docs the behavior that @phillc54 reported. The current manpage is self-contradictory, claiming both that motion.feed-upm
is in machine units, and that it's based on the program's F-word (which changes units with G20/G21).
The motion man page states that motion.feed_upm is in machine units per minute.
The actual reported value is G-Code file units per minute. e.g. a metric machine running a G20 file, feed-upm reports inches per minute rather than mm per minute.
I can change the documents if necessary but I am reporting this in the hope that the documentation is correct and the reported value can be changed.