Andrew - I prepare GCODE in my postprocessor - it  takes 3d points from cad
3d program , and then traverses points of edges and outputs gcode.

OK, I made correction as Andrew proposes, but this transition from straight
line to edge cutting results in velocity ripple (seen in halscope on video
below) - is it possible to have more smooth transition?

Any idea how to correct this if it is possible at all?

https://youtu.be/71HJZ2YjnJU

regards


On Mon, Aug 22, 2016 at 12:35 PM, Klemen Živkovič <[email protected]
> wrote:

> Because machine has two rotating disk, that are used to clamp pipe.
> If I want to have both disk to have same angle (so they can grip
> rectangular pipe) I need the ability to separately rotate both of them.
> I also use two separated stepper drivers for this reason.
> For ilustration you can maybe check pictures of my project at:
> https://goo.gl/photos/4A623DBE1DQRwYfcA
>
> regards,
>
> On Mon, Aug 22, 2016 at 10:40 AM, Andrew <[email protected]> wrote:
>
>> 2016-08-22 2:43 GMT+03:00 Klemen Živkovič <[email protected]>:
>>
>> > Oh yes I see how you calculate it for this example on simulation.
>> > Actually I would need to make and calculate it for general case where
>> > fillet can be only quarter of length or any other arbitrary length. I
>> get
>> > line segments from 3d cad program.
>> >
>> > Need to think if calculation of g93 feed is possible from having start
>> and
>> > end point of tiny line segments (center of rotation always at x=0 and
>> z=0).
>> >
>> > If you split fillet arcs into short segments, you still have to
>> consider X
>> moves.
>> How do you prepare G-code?
>>
>> Also, why A and B if you only have one rotary axis?
>> ------------------------------------------------------------
>> ------------------
>> _______________________________________________
>> Emc-users mailing list
>> [email protected]
>> https://lists.sourceforge.net/lists/listinfo/emc-users
>>
>
>
------------------------------------------------------------------------------
_______________________________________________
Emc-users mailing list
[email protected]
https://lists.sourceforge.net/lists/listinfo/emc-users

Reply via email to