B axis rotation is very slow <resolved>

[mikeman20]

I've finally got my 4 axis router running, and everything is working well, except I cant seem to figure out how to increase the feed speed of the rotary axis.

I started using an inverse time (G93) post for aspire, and that just about doubled the speed, but its still painfully slow. A toolpath that should take 40 min, takes 2 hours.


Am I missing a setting somewhere to compensate from the degrees/min based off the radius?

thanks!
-Mike

[cnckeith]

Mike, Post a fresh report.zip and follow the support request guidelines in my signature so we can help you out:-)

[lvanduyn]

Hi Mike
I am using Aspire and the inverse time post for my A axis rotary. I had the same
problem when I switched to the Acorn. My solution is to manually edit the posted gcode
to increase the feed rate. Aspire picks the feed rate from the tool params, usually
around 40 or 50 in my case, which I then edit to a value of 10,000. That is probably a bit
fast but it allows you to dial back the speed to a comfortable level using the feed override.
Anyway, as they say, works for me.
Larry

[Dave_C]

Mike,

I'm not sure if you are using MM or Imperial but either way rotary 4th axis set up a bit odd.

Instead of Inches per minute a rotary axis is in degrees per minute!

So Check you Min and Max feed rate settings and see what you have them set to.

For example if you set the Max feed rate to 150 thinking that is good, then you get 150 degrees of rotation in one minute.

You are limited on the max speed you can turn the rotary by the max pulses you can send to the stepper.

If you have a 4 to 1 ratio on the rotary then a setting of 360 will turn the table one full turn in one minute.

I don't have very much info on your setup from your post so I may be all wet and giving you unhelpful information!

Dave C.

[Gary Campbell]

Actually a rotary axis setup is normal, but not as a flat object. Its round, so rotational speeds are defined as RPM (revs per minute) would be normal. Since we wish to be precise, degrees per minute would give that. Degrees per minute are simply 360 times rpm.

If you wish to "think in flat" then to convert you need a diameter. The feedrate in inches per minute would be like paper coming off a roll, which results in a rotational speed in revs or degrees per minute.

The formula is 360 (degrees per rev) / 3.1416 (pi) X desired feedrate / Diameter of blank = degrees per minute.

Example for a 2" blank: 114.59 x 100 / 2 or 5729 deg/min. Faster for smaller blanks, slower for larger. 5729/360 = 15.9 RPM or one rev per 3.75 seconds. Not very fast if your data stream can produce the steps.

You might try 720 as a slow speed (2 rpm) or 30 seconds per full rev
7200 as a fast speed (20 rpm) or 3 seconds per rev
Depending on your step/degree rate you might go with 72,000 or as a 144,000 max rate.

Adjust as needed to your hardware

[mikeman20]

I will post a report.zip in a moment, but I think reading everyone's comments has lead me to the solution: my max deg/min rate is just limited by acorn and I need to increase it (alot).

Once I finish getting this next model together, I will get the report.zip, try to edit the acorn parameters, then report back.

[mikeman20]

report_E415F6F483F9-0312214485_2021-05-28_21-26-43.zip

(695.01 KiB) Downloaded 89 times

I edited the maximum and fast jog speed in the config menu in CNC12, and my quick aircut test seems to work!
I tried normal mode and inverse time mode, only inverse time mode increased the rotation speed. I suspect aspire cannot do the calculations in normal mode, and from what I skimmed about inverse time mode, seems to be common enough that it was invented for this exact reason

Thanks everyone for the help!