View Full Version : conical helix interpolation

05-18-2011, 03:43 AM
Hello forum,

Maybe there is somebody here that can help me with my problem.
The search function did not really help me out…

I want to carry out a helix interpolation on a Sinumerik 840D Powerline…. Which is supposed to be weird.

Having an initial diameter of 25 mm, the final diameter should be 29.5mm

Can anybody help me how I would need to program it?

Other data:

Starting point Z4, X0, Y12.5 (for 25mm diameter)
Ending point Z32, X0 Y14.75 (for 29.5mm diameter)
Gradient: 2mm per helix rotation

Really appreciate any help!
Best regards

05-18-2011, 01:12 PM
I believe that it can’t work with circular functions, but I am happy to accept I’m wrong.
I would try it with incremental polar G1 sets, the resolution depends on how detailed you want your part to be.
An example with the resolution of 0,2 per step

N50 G1 X0 Y0 ;desired position XY
N55 G110 X0 Y0 ; define Pole to current position
N60 G0 Z4 ;Startposition Z
N65 G1 RP=12.5 AP=90 ;Startpos XY
N70 HELIXSTART: ;Startlabel
N75 G1 RP=IC(2.25/(14*1800)) AP=IC(0.2) Z=IC(2/1800)
N85 REPEAT HELIXSTART P=((1800*14)-1) ;repetition
N90 G1 RP=0 AP=0 ; move away XY
N95 G0 Z100 ;move away Z

Try it out.

If you need it more frequent , its possible to parameterize it.

05-18-2011, 05:36 PM
Thanks for your quick help!
I will try it out tomorrow

I was thinking that circular function might not work…. In the end I discovered the evolve interpolation (INVCW, INVCCW)
According to the Sinumerik handbook, an involute can be processed in the room with that. Although then I have the problem that I don’t have a linear radius instead I have a involute… or am I thinking wrong?

But your solution sound sufficient for what I want to do.
0,2 ° resolution for the single movement should be enough.

Best regards

05-21-2011, 11:58 AM
Maybe it works with OFFN, it's worth a try!

06-14-2011, 07:18 AM
What about this:

def real vors, turn
def real dma=80
def real dmi=20
def real ta=-1
def real ti=-20
def real zust_e_max=5
def real zust_z_max=2
T="sphere 8" D1
S2000 M3
G0 X0 Y0
G0 Z2
G1 Z=ti F=vors
G1 X=dma/2 Y0 G41 offn=(dma-dmi)/2
G3 X=dma/2 Y0 Z=ta I=AC(0) J=AC(0) turn=drehungen offn=0
G3 X=-dma/2 Y0 I=AC(0) J=AC(0) offn=(dma-dmi)/2
G1 X0 Y0 Z2 G40 offn=0
G0 Z50 M5

If you program in the first G3 set a movement in Z+, then you mill a funnel from inside to outside.