XZC Lathe Rotational feed rate - Kinematics?

More
07 Jun 2015 23:31 #59597 by andypugh

ok - so inverse time does the same thing.


It really shouldn't, and it seems to be working OK for me here.

F75 is a pretty fast feed rate (it means finish the move in 0.8 seconds). Is it possible that some moves are limited by axis speed.
F6 should take 10 seconds per move regardless of what the move is, can you check that?

Please Log in or Create an account to join the conversation.

More
08 Jun 2015 02:26 #59605 by cts1085
Andy,
You were dead on. Thank you!
The inverse time fixed my rotational timing issues.

Whoo hoo!

-Tom

Please Log in or Create an account to join the conversation.

More
10 Sep 2016 12:24 #80293 by snujcnc
i face same issue (2.5.3)
and solved with inverse time feed g93
but i also try to say my angular axis (C-Axis)= linear

but with g94 same issue i face

in my fanuc 11m hmc with b axis i never face this type of issue
now i am planning to identify what exact problem is .....e

checked in 2,6.8 issue is same

Please Log in or Create an account to join the conversation.

More
10 Sep 2016 19:55 #80313 by cmorley
In inuxcnc right now, inverse time is the only easy solution.
For linuxcnc to adjust the c axis feedrate it would need to know the distance of the tool from the center of the c axis.
In a lathe this is easy as it will always be X radius, but most other machines it can be in an arbitrary position.
There is currently no way to tell the motion controller this info.
Linuxcnc's motion controller is very generalized, it does not change based on what type of machine it controls.

Is your fanuc 11m a machine that the b axis center can always be assumed?
If yes, then I would think the manufacturer changed the motion controller to reflect this.

Chris M

Please Log in or Create an account to join the conversation.

More
11 Sep 2016 04:20 #80324 by snujcnc
Is your fanuc 11m a machine that the b axis center can always be assumed?

i think fanuc calculate roatry axis center to machine's spindle face (in case of hmc machine) and according to this position they calculate b axis feed which is related to x,y,z axis feed

otherwise when we make some rotary cam on my fanuc 11 m hmc we never face issue with feedrate of b axis with other axis

same program and same cam when we manufacturer on linuxcnc system we face this issue

but still i am looking to find out what exact solution is .

Please Log in or Create an account to join the conversation.

More
14 Sep 2016 08:27 - 14 Sep 2016 08:27 #80429 by andypugh
I just remembered this:



Which might help.

If the machine always makes things of the same diameter, or if you don't mind editing the config when the diameter changes, then you could configure the Y axis to be a rotary, where the Y-word in G-code refers to a distance along the surface of the cylinder. This isn't such a silly idea if you are doing engraving work, as it makes the CAM stage much easier. Converting the rotary axis to respond to the X, Y or Z G-code commands is the only current way to get path-blending active on a rotary axis.
Last edit: 14 Sep 2016 08:27 by andypugh.
The following user(s) said Thank You: snujcnc

Please Log in or Create an account to join the conversation.

More
19 Oct 2016 08:21 #81826 by snujcnc
Dear andypugh
i attach my machine construction
i successfully make nx 10 post with G93 feed in each line with feedrate according to my component

but when five axis operation start and on my globoidal cam i found issue on surface ...
actual surface is wavy (up down slopes are there) specially when b c x y z axis move together

i set min error,backlash,feed,everything is ok

and i am using pivote distance and tool length from my cad cam software
so machine with linuxcnc no need to calculate tool length and pivote

dimension are ok but surface is only issue when i cut with 6 mm ball mill it show around 0.5 mm steps on surface of cam

now i am confuse what to next

if you can suggest me something i can try

thanks

kalpesh patel
Attachments:

Please Log in or Create an account to join the conversation.

More
20 Oct 2016 17:25 - 20 Oct 2016 17:25 #81892 by andypugh
Can you experiment with the G61 and G64 options?

linuxcnc.org/docs/2.7/html/gcode/g-code.html#gcode:g61-g61.1

It may be that a loose tolerance on G64 combined with a slow-accerating Z axis is the problem.
Last edit: 20 Oct 2016 17:25 by andypugh.
The following user(s) said Thank You: snujcnc

Please Log in or Create an account to join the conversation.

More
20 Oct 2016 18:08 #81896 by snujcnc
Thanks for suggestions
I tried g61 and g64 p0.010
But I found same result
But I try again

And I want to know that if I write g64 p0.01 means my axis move in 0.01 tolerance correct?

You wrote slow accelerate z axis is problem

But I set all axis with same acceleration
I also try to set different acceleration on different axis as per load but results are same

I am using pivote distance of my machine in nx software and linuxcnc only need to follow as per program generated by linuxcnc tool length also I enter in nx software

Is there any possibility that if pivote calculation is calculate by linuxcnc is better then nx software calculation ?

I am using pivote calculation with nx software because linuxcnc does not support accel during five axis move if I use tool length calculation and pivote distance with linuxcnc it create very zerky movement

If you can write something more on this topic I can try your suggestion to find out solution

Thanks for reading and suggestions on my topic

Thanks
Kalpesh patel

Please Log in or Create an account to join the conversation.

Time to create page: 0.090 seconds
Powered by Kunena Forum