Going crazy on profile of surface?

[KENAT]

I can’t show an image for IP reasons but I have what is effectively an aluminum ‘vacuum plate’. It’s circular with a flat on one edge. It has a series of vacuum grooves, radial ones and ones going out like spokes, looks like a spiders web! The surfaces this creates are what the item being ‘vacuumed’ sits on, held in place by vacuum in the groove.

I’ve called out the outer ‘ring’ surface as datum A and made all 339 (yes I checked and that’s how many surfaces are created by the grooves) profile of a surface .0002 to A (we’ve apparently had samples made so I assume the tolerance is held) based on what ASME Y14.5M-1994 says in 6.5.6.1. The prototype print had flatness which isn’t correct for multiple surfaces.

Question 1 does this make sense, have I applied the standard correctly?

At the moment the proto print has the grooves dimensioned with +- dims. The cross sections of all 39 grooves are R.03 +- .01. The ‘spokes’ are 15° spacing +-.5° running from the inner radial groove to the outer one (which at their outermost point is +-.047 based on outermost ring having diameter of 10.75 if my math’s right, taking into account the stack of the 23 x 15° dimensions then the last groove could be off 11.5° or 23*.047, more than one inch!). The radial grooves had diameters +-.005 and no coaxiality control. A couple of holes diameter .063 position diameter .015 (on proto print) connect the grooves to vacuum on the underside.

Question 2, given the tolerance build up from the +- dimensions I’m thinking a surface profile tolerance would actually be better, and it would address the coaxiality. I’m thinking .030 which roughly matches the current radial groove tolerance would make sense. Sound reasonable?

On the far side it has 3x ½ inch long V grooves for a kinematic mount.

Question 3 Would surface profile sound good for these as well?

Sorry for the questions but someone far more experienced/knowledgeable than I looked at an earlier draft and didn’t add any of these controls. I think most of the features were added since he looked at the earlier print but the 3 V grooves were probably there and the top was I think already 9 surfaces but flatness was called out.

Thanks all,

KENAT, probably the least qualified checker you'll ever meet...

 

[dingy2]

Is there a function and relationship where you applied GD&T? Is so, go for it. If not, I would not apply GD&T.

I learned GD&T so many years ago by reading Lowell Foster's book and the first thing that got me was the function and relationship approach which I use today.

There are some out there that have all dimensions basic and various levels of profiles, etc. while others have absolutely no application of GD&T where it would be of great value.

Dave D.

http://www.qmsi.ca

 

[KENAT]

Well,

The top side with the grooves in it needs to be very flat. A thin article is put on it and then measured down to sub micron/um level so yeah, the 'flatness' is important from a function point of view.

The grooves need to be fairly evently spread to apply even vacuum to prevent bowing of the item going on there. They also need be be positioned accurately enough that the grooves line up with the vacuum holes.

The 3 grooves on the back need to align with the 'balls' they sit on so need to have adequate control on location. I was going to use position but because they are V grooves I thought profile of a surface may be better. The sides of the groove will be what sits on the balls so seems to reflect function better to control these than just the obround where they meet the surface.

I'll admit I'm concerned about putting all the profile of a surface given how the designers here and some of our manufacturers respond to it but I think the function requries it and wanted to bounce it off some people who had some idea.

KENAT, probably the least qualified checker you'll ever meet...

 

[KENAT]

Appologies. When I said radial I meant circumfrential.

Attatched a simplified tiff showing the centerline of the grooves.

KENAT, probably the least qualified checker you'll ever meet...

 

[ringman]

Have you considered a checkerboard pattern for the grooves?. It would seem to provide a better distribution of the air passage grooves.

 

[KENAT]

Not my design and it's already been prototyped & proven.

It was a modification of an earlier design which just had 2 circumfrential rings and a 'cross' groove.

If designing from scratch I agree it's probably not the optimum layout. In fact the main chuck design is a lot better but also more expensive or something, this one is used during calibration etc, I'm not an expert but it's something like this.



KENAT, probably the least qualified checker you'll ever meet...