Has anybody else set up a 5 ax post for the new DMU-60 EVO

[pullo]

I have right now ascertained , that I have to use a shift of "circa" -150 degs on the C-axis for the machining to look "right" ,  so that is the ballpark , but  what is the  precise value, ?

Gracjan

[pullo]

So ,  in order for a post to work properly for this machine ,these  vectors have to be input :

1. The vector of the spindle 0,0,1

2. The vector of the variable rot. axis (C) 0,0,1

3. For my machine ( these values will slightly differ from machine to machine  )  0.5849.... , -.5735..., 0.5735...

   which is 44.6 degs from TOP and 44.6  degs from FRONT 

with these values I tried to machine the chamfer on the part , the axis of the tool pointing towards the center 

and this is what I got  ( a totally senseless result)

 

After manually trying out  a few values , I came to the conclusion that a Shift in  the C-axis of -150 was more or less what I am looking for  (this value

of -150 gets added to every C axis output. 

With the shiftc of 150 in my post , this is what I saw on the machine :

 

 

[pullo]

I just realized that the heading is WRONG IT SHOULD BE DMU 60  EVO

well anyways back to the subject...

and this is the result on the machine , or in laymen terms , all is fine.

 

the axis of the tool intersects the axis of the cental hole of the part , etc.....

Gracjan

 

[pullo]

Seems like gcode is responsible for the 670 evo  . It is , I say again DMU 60  evo.

 

Gracjan

[gcode]

6 minutes ago, pullo said:

Seems like gcode is responsible for the 670 evo  . It is , I say again DMU 60  evo.

 

Gracjan

Yes I have webby powers that were unknown to me

I fat fingered the title to this thread while trying to copy /paste it for a google search 

[Greg Williams]

I am not a Heidenhain guy but they must be a way to move the C axis machine zero point?

[mkd]

42 minutes ago, Greg Williams said:

I am not a Heidenhain guy but they must be a way to move the C axis machine zero point?

how you figure he's got a HH control?

 Maybe if he can post up his toolpath code we can figure what he is talking about.

Plane Spatial /CYCLE 19 or M128 for the HH 5ax stuff.

[Newbeeee™]

On ‎3‎/‎15‎/‎2019 at 1:23 PM, gcode said:

Yes I have webby powers that were unknown to me

I fat fingered the title to this thread while trying to copy /paste it for a google search 

SAUSAGE FINGERS!

[mkd]

4 minutes ago, Newbeeee™ said:

SAUSAGE FINGERS!

THAT's WHAT S.........

nevermind

[jlw™]

I'm still over here wondering what he's talking about.  He's running vector code but offsetting 150°?

I googled that machine and it looks like a plain old trunnion machine to me.

 

I'm lost and over here polishing my bowling bat hoping he can slam dunk a touchdown. 

[pullo]

So many answers I have to give here....  Actually , I was hoping to get an answer from my esteemed colleagues  

Greg Williams :  "this value of -150 gets added to every C axis output." .  This is the definition of  a shift .  We are of course talking of a rotary shift.  If you mean that I could get the shift done on the control , yes I could do that probably with a Cycle Def 7.0 Datum Shift , C-150 command . The result would be the same. However , if for some reason the

operator would negate the command we'd get a crash...

mkd : yes it is a Heidenhain  control on this machine.  I was referring to machining when RTCP is turned on (M128).  The  plane spatial/Cycle 19 works fine ,  as I can sidestep the issue of outputting  the B-angle by outputting the A and C  angles.  The machine will calculate the right B angle based on A .

JLW:  I will slam dunk this probably today.  I will measure  the position of a hole on the outer axis of the the table ( for max angular precision)  in 3-axis mode and the I will position a gauge above the hole in RTCP /M128 /5-axis mode and see if the positions aline .  I realize this is a reverse engineering approach, but it's one that will get me close

to the end result and that will give me time to ask around/figure it out , whilst the customer is chugging away on his new DMU-60  evo

Edited March 18, 2019 by pullo
spelling etc

[Greg Williams]

Seems odd that you need to add 150 to the C axis, somewhere in there you will be able to set the zero position.

What post are you using?

Could always say p_out = p_out+150?

[pullo]

So , did not get an answer here ,  I will have to continue my snooping work the next time a 5-axis (M128) case comes around . ....

Gracjan

[pullo]

So it turned out that my C-150 was just one special solution for one problem , but it would not have worked in general.  It turned out that by tweaking two parameters

in  my Camaix post I was able to output the right angles  without the C shift .  

 

Gracjan

[pullo]

jlw , wanted more info about setting up the post , so here it goes :

it's a table - table  , with  the B-axis vector always provided by DMG  as in the picture:

 

other than that , nothing special about setting up a post for it .

[crazy^millman]

5 hours ago, pullo said:

in  my Camaix post I was able to output the right angles  without the C shift .  

 

Gracjan

Might throw that out there next time also. They do it completely different than just about anyone.

[crazy^millman]

dP

Edited March 18, 2019 by 5th Axis CGI
Double Post no reason to repeat myself.

[Grievous]

Just a thought....you can set your post, to give you in M128 mode(5x) the surface-normal vectors NX, NY, NZ, instead of your machine axes. Same code will work in a machine with a different kinematic also.

[Zaffin_D]

20 minutes ago, Grievous said:

Just a thought....you can set your post, to give you in M128 mode(5x) the surface-normal vectors NX, NY, NZ, instead of your machine axes. Same code will work in a machine with a different kinematic also.

I don't think you need to use LN to support different axis combos; AC code in M128 will run on an BC machine. 

I think you mean TX, TY, and TZ for the tool vector.