Optics quality surface finish

Optics quality surface finish

kripalbedi10
Observer Observer
792 Views
6 Replies
Message 1 of 7

Optics quality surface finish

kripalbedi10
Observer
Observer

https://youtu.be/MuL9iaoTcjU?si=WvQEdfJ5deFN8fWu

 

We tried setting the parameters as per this YT video suggestions but the file size became too huge to handle and had to split the programs which we didn't prefer as it could result in micro steps on the surface of the optics where the old and new programs ended and started.

 

My question to the experts is, how do handle such huge files if the point redistribute tolrance factor is 0.1, point separation distance is 0.015, mesh factor is 0.5 and max traingle length is 0.01. 

kripalbedi10_0-1729070038528.png

 

0 Likes
793 Views
6 Replies
Replies (6)
Message 2 of 7

kripalbedi10
Observer
Observer

kripalbedi10_0-1729070141469.png

 

 

This is how the points looked before we made the changes to the setting

0 Likes
Message 3 of 7

TK.421
Advisor
Advisor

On my programs where i need good surface finish, I use .0002 toolpath tolerance, .125 tolerance factor & mesh factor, and .03 point separation distance. It should be noted that all these numbers are in inches.   from your screen shots, it looks like you're working in mm. Could this be the reason your files are so huge? My file sizes grows as well, but a .0005 mm toolpath tolerance and a .1mm point separation/tolerance factor seems extremely small to me. if i tried using those numbers, i think my computer would crash trying to calculate it! 

maybe i am mis-interpreting your screen shots?


the numbers never lie
Message 4 of 7

rich10
Enthusiast
Enthusiast

On average I only use a tolerance of 0.01mm and a mesh of 0.1, other items unticked.

sometimes 0.005mm and a mesh of 0.1, this is mainly using a 0.01 programming tolerance.

 

I think your 0.005mm separation tolerance is forcing a straight line to be broken up into small segments, as I understand benefit no one unless 4/5axis, even then I would try 1mm if I needed to first.

Message 5 of 7

kripalbedi10
Observer
Observer

kripalbedi10_0-1729146282952.pngkripalbedi10_1-1729146308386.png

In order to reduce the program size - we tried using FIT ARC vs earlier POINT REDISTRIBUTE, The file size reduced to 50mb from 596 MB but I'm not sure even though the program mill time on both remains same at  14 hours. Can experts suggest what would be the difference in surface finish and profile precision between fit arc and point redistribute method? 

0 Likes
Message 6 of 7

rich10
Enthusiast
Enthusiast
Fit arcs works well in principle as long as the post is done right.
In general I use redistribute as many years ago we had a few jobs get damaged through fit arcs, this was a long time ago, 2007-2015, post processing was more likely the issue and beware vertical arcs to. Obviously redistribute does not have the hang-ups.
Message 7 of 7

M_Hennig
Collaborator
Collaborator

There is no reason to litter straight line segments with minimum point distance, in fact Heidenhain specifically states there should be one point in the start, and one at the end. I wish PM had the ability to use minimum point spacing only on curves and contours.