Is there a Halpin for the current feedrate

I'm using master branch and I'm trying to calculate cutting parameters in HAL for a glade display.
For this calculation I need the current feed rate but can't find it in the HAL pins. I found motion.current-vel but that is not ideal because its only useful when there is actual movement and does not allow for a distinction between rapid moves and moves where the feed rate applies.

Is there really no access to the feed value in HAL?

Thats right, the feed rate is thrown away once the gcode is tokenised. current-vel is for that segment only and may or may not be the feedrate.
Cmorely has recently published an experimental branch that publishes it on a pin. I have tried it. I thought it found its way into master. I'll try and look for it tomorrow when I am at my machine as I just compiled master today.

The other pin I would really like to see is the radius of arcs.

look for motion.fcode is it there?

oops, doe snot seem to be there. This is the fcode branch

github.com/LinuxCNC/linuxcnc/tree/feedcode_message

Ah, I just realized I'm actually on the switchkins-branch and this puts me out on a limb so to speak.
I guess I'll have to manage with the current-vel.

Thanks

My feedrate work was not accepted into 2.8 and statetags was merged into master so I have lost track if the work is useful/compatible.

Chris

cmorley wrote:

My feedrate work was not accepted into 2.8 and statetags was merged into master so I have lost track if the work is useful/compatible.

Chris

At least you motivated the release of state tags!

NOW we just need some documentation on how to acess the state tags data (which includes the feedrate) by resolving the known issues where the docs refer to state tags stating it will fix it.....
linuxcnc.org/docs/devel/html/gui/GStat.html#_known_issues

So I could use statetags with a python script but the feedrate might not be the current feedrate because the interpreter looks at the code ahead?

Aciera wrote:

So I could use statetags with a python script but the feedrate might not be the current feedrate because the interpreter looks at the code ahead?

It will be the real feedrate for the currently executing segment! so it will be accurate for right now! The read ahead has no effect as this is at a a deeper level.

When you work it out, please share how you did it.

Ok, I've played around with gstat a bit:

So I changed the example script in the doc to create a HAL pin called feedrate.current:


I then called this script from my postguihal as given in the doc and the pin got created:



And the value also changes as expected.

So this was successful.

However, I noticed that it also gives the feedrate of G0 moves which is what I don't want. This is what I already have by using "current-vel".

I'm actually looking for the "F" value, preferably the one as used in "Feed Override":



This does not change with G0 moves and also does not go to zero when the movement stops.