General machining question: how to coordinate feedrates when using an A axis?

I just added a rotary axis to my mill, and the hardware is working finally after a LOT of stupidity on my part, but now that I've got it working I'm trying to figure out conceptually how to do coordinated moves.
Like, moving from X1 Z1 A90 to X1 Z1 A180 at F1 is 1/3 the chip loading of moving from X1 Z3 A90 to X1 Z3 A180 at F1
If you're coding by hand (I don't have a cam program capable of four axis movement) do you have general guidelines for how to do this?  How do you handle linear-and-rotary moves where the feed changes during the move?  Interpolate it into a bunch of little steps?  Set the feedrate as a variable multiplied by whichever axis you're moving?

Thanks for any advice
 

Really, really not an expert... but if you want to change the effective feedrate over the length of the move, I think you need to break it up.  Workflow for G93 is something like this:

• Calculate total length of move at the working radius (arc segment length)
• Figure out where you want to change feedrate
• Divide up the move in to equal segments so that the time/distance F-word is in the same "units" so to speak.
• Calculate the inverse time for each segment so that the feedrate you want is appropriate.

This
X1 Z3 A90 F1 (start)
X1 Z3 A180 F1

Changes to this
X1 Z3 A90 (start)
X1 Z3 A120 F1
X1 Z3 A180 F2

I'm probably not explaining this well, but maybe you get the gist.  Doing this by hand will be, I suspect, miserable if you are changing the working radius in the middle of a move.

There are probably much smarter people than me who have spreadsheets or programs to figure this out, especially for coordinated 4-xis moves and not just 'engraving on a cylinder' stuff.  Hopefully they'll chime in.

That's a good explanation and kinda what I thought/feared. Yeah, I'll have to write a small program to interpolate my intentions.
Thanks!

Found this:
en.industryarena.com/forum/inverse-time-...4th-axis--69433.html

and theres a later post in the thread describing the radius-based feedrate calculation available in some controllers.

this looks like something one of the Lcnc Dev-ninjas could duplicate.  i.e. user enters working radius and G94 feedrate, and Lcnc calculates the approximate G93 feedrate.