Problem with rotation around rotational axis

[[Ti...]]

Hello everyone,

This is my first post, so please let me know if I've forgotten anything!

I'm currently running this program on our Spectrum with RDS and have noticed that Calypso does not display the measuring probe symmetrically to the X-axis when it is swiveled, i.e., it is not parallel to the machine coordinate system in X. The first two images were taken at the same moment, but there is a rotation of the probe on the screen of approx. 20°, which is not present on the actual machine. The contour is traced on the machine parallel to the X-axis.
When I took a closer look, I also noticed that the circular segment, which should theoretically follow the green contour (first image), is also rotated relative to the component axis. The orange path is followed. This is consistent across all/multiple circular probing operations of the program.
The base system for this can be seen in the third image with the plane rotation empty, which means it should be defined as the machine axis (+X). I chose this because the component is rotationally symmetrical and can thus be defined more easily (without probing the vise/chuck it is in).

Why are the circular segments not probed symmetrically to the X-axis when rotated? Where does the rotation of the probe come from? I think these things are related, but I don't understand the cause.

[[ni...]]

Have you any rotation in the "Sonder" button ?

[[Ti...]]

Please sign in to view this quote.

Thanks for the reply, no that is empty.

[[ni...]]

you could try to do the base system in manuel; and see after what the CMM do in automatic.

[[Qu...]]

This happened to me whenever the program doesn't have a planar rotation and especially when using cone and cylinder. It rotate the model out of of its original direction and misalign it with the machine coordinate. 

If you have a planner seat, try checking if the model still align with the machine.

 

My only fix is to remade the program since it is faster than fixing each feature again.

Edited February 5

[[Ma...]]

You can either make an alignment from another program - i like to have one program from granite and something to tell machine - this is X axis and this is Y axis.

Then you can load this alignment into your program. Now use theoretical 3d line with features alignment from this loaded one. It will lock your rotation and you don't have to scan anything additional to lock it.

I don't think that using theoretical 3d line with base alignment will solve this rotation, but sometimes it's helping.

[[Ti...]]

Please sign in to view this quote.

Do you mean when starting the cnc-programm? It doesnt seem to make any difference whether its the cnc-programm, program or manual alignment.

[[Ti...]]

Please sign in to view this quote.

So how can I avoid this when the part doesnt have any features that can restrict the planar rotation? I thought leaving it empty defaults to the Machine coordinate system?

I have a simulation seat, how can I use it so see if its still aligned?

I have redone the program (over an hour of work) only for it to rotate again...

 

[[ni...]]

can you post the element and strategy window, and the real trajectory on the part (not a drawing) like this :

[[Ti...]]

Hello Nicolas,

of course, here it is with measured points. It looks as if everything is fine but when probing the axis is rotated on the screen and the scanned segment is not the yellow one on the part but also rotated along the Z-Axis as shown in the orange sketch in my post...

[[Ze...]]

As a test I would measure a line on the vise and clock to that. Then you can see if it's CAD model related, machine coordinate issue, etc.

[[ni...]]

your cylinder is not on the base system, It was about the Kegel_50° I was asking for.

not the problem you are asking but I see you have only one height probing (Anzahl Schnitlle) on you cylinder this why you have a tilt on W2 Y/Z , you might have at least two or prefer to mesuring a circle.

[[Ti...]]

Please sign in to view this quote.

I tried to avoid using the vise to restrict the last degree of freedom since it should be really easy with the machine coordinate system and would help improve the cycle time...
 

[[Ti...]]

Please sign in to view this quote.

I'm sorry, here is the the screenshot for the cone!

Thank you for reminding me - I learned that the other day and will switch the element to a circle since I can't scan at multiple heights due to the slot being very narrow. I only want the diameter so it doesn't matter. Do I have to recreate the element or is it okay to just change it here?

[[ni...]]

 

try to recreat it, and see if it close the problem. Cause if I understand well, the scanning is doing well, the matter is just about visualisation ?

[[Ti...]]

Please sign in to view this quote.

No it's not just visualization, the segment should be scanned along the green line but it is scanned along the orange one in real life. The visualisation is just what made me look closer and realize that it isn't scanning the defined segment but a rotated one along the axis...

 

[[Ti...]]

Please sign in to view this quote.

I was told by a zeiss instructor that leaving it empty defaults to the machine coordinate system (the chosen axis on the right e.g. +X in my case) and that I dont have to import/load alignment features...

[[ni...]]

create circle in place of cylinder, and do a calibration of all stylus you used in this program.

and by two time someone ask to scan a line to clock rotation, you should try.

[[Ti...]]

Please sign in to view this quote.

Yes I changed it from cylinder to circle and all probes are calibrated.

What do you mean? I shoud use a element to restrict the 6th degree of freedom?

[[ni...]]

see previous message by Martin Jánský and another by Zen Cat; in normal situation this is not nessecery to lock the rotation with such part, but in your case it could either solution the problem or going further in analys