[buscht]

I've got my 4th axis built and have started to use it. So far, I've been just using it as an indexer and doing traditional X,Y, and Z milling.

But if I want to do full 4 axis contouring I am not sure how to get the 4th axis speed to be in sync with the other 3 axis.

For example, let's say that I am cutting at 60" per minute. My part is 3.819" in diameter. That would make the outer circumference 12".

That means that to cut in the 4th axis at 60" per minute, I'd have to spin the axis at 5 revolutions per minute. 60/12 = 5.

But if I use this code for example
G01 X12 A360 F60.
Won't it take 6 minutes to spin the 4th axis 360 degrees?

Is my logic incorrect? Or do I need to map the 4th axis as linear and use this code?
G01 X12 A12 F60

4th axis work is new to me, so any help is appreciated.
Thanks
Trent

[ger21]

I think it depends on the control. But if you issue a command like

G01 X12 A360 F60, both the X and A should get to there destinations at the same time. Whether it takes 12 seconds or 6 minutes, you'll have to check. Using the axis as linear is something I've seen on the TurboCNC group to solve this type of problem.

[mikeschn]

Trent,

Can I see a picture of your 4th axis in position? I'd like to allow sufficient room on my machine so I can add the same!

Thanks,

Mike...

[wisp]

Most controllers auto switch to degrees per minute when a 4th axis move is made. A simple formula to calculate IPM into DPM is,
DPM = 360 x IPM / circumference
this works pretty well in most cases. just be carefull if you are making small index moves and large X or Y moves. Imagine making a 1 degree index with a 10" X move, at a feed of F360, thats one RPM. the index will take about a second, but at the same time the X axis will try to move the 10" in that one second. youch.

[Geof]

Ger is correct, if you put in a G01 move that includes linear and rotary axes they get to the end at the same time. You do have to be careful if you are dealing in degrees on the rotary axis because of course the cutter is seeing the peripheral speed; the rotary might be able to make the move at the feed you program but this might translate into a heavy cut.

[buscht]

Mike, this isn't mine but I copied it so it looks real similar.
http://www.cnczone.com/gallery/showp...puser/223/sl/m

Thanks for all the replies.
Trent

[broncosis]

I should look at our coding her eat work all of our machines run
xy and a no z still three axis but A is rotory
with a centroid control
i have been putting them together for so long
never really to look at the cnc code they run thats our programmers
job really his software does a fine job with it

but I know we are running it linear where 1" is one revolutiuon
granted they are aon a 30-1 gear reduction drive and some sick 29inlb serveros
with crazy high encoder counts but I'll dig in co some of the code for a simple cylinder
here in a minute

[broncosis]

X-1 A21.1361 Y6.1475 F50.1597
X-1 A21.1388 Y6.1472 F50.1595
X-1 A21.1416 Y6.1470 F50.1597

just snipit of the code we run
this would carve a 2 inch cylinder in our machines
the code is close to 5,000 lines though
as you can see i tis only moving a few thou
for each point along y which is layed out as a
vertical axis x is horizontal and A is rotary

it doesn't make pretty squares though almost we have tried
it isn't what it was ment for with a 5/8 x 11" carbide cutter it makes
short work of foam though

[buscht]

Thanks for checking broncosis,
Trent

[ger21]

Trent, let us know what you find out. What controller are you using?

[broncosis]

not sure how much it
helps but figured it
may provide some insight

[buscht]

I am using the 4 axis Xylotex controller. If I find out anything useful, I'll report back here.

[buscht]

I ran this command;
G01X6A60F60
The machine responded exactly as Gerry predicted, the A axis turned 60 degrees in one minute and moved 6" in one minute. Again, assuming a 4" dia part, 12.5664" in circumference that would give me 2.09" per minute speed going around in the A axis.

I tried
G01X6A60F600, The speed went to 60" per minute in the X and 20.09" in the A. Not exactly in sync, but closer to the overall performance that I wanted.

It looks like I would have to not program in degrees, but map the A axis to Y and use linear movements to get matching speeds.

Other than that, I should stick with just using the A axis as an indexer.

Its not a big deal to me, I just want to make sure I understand what is and isn't possible and then work with those parameters.

Thanks
Trent

[Torsten]

A Cam system is best in figuring out the appropriate departures and feedrates for
moves involving rotary axeses.
The Feedrate is based on the length of the move considering all axses per geometry
involved.
Because the position of the rotary axes in relation to x,y,z may vary between
different machines the Cam system has to be configured for a specific machine
in order to do this correctly.
Good Luck