dimensioning squareness

[mmolt]

Friends,

Question for you... The company I work for purchases a lot of flat Al sheet, of about .050" thick or so. We order these sheets from our supplier per a controlled print indicating their size. Having said that I think our prints probably lack some clarity in terms of sheet squareness. I believe the correct way to handle this is to use some GD&T, with two datums and two perpendicularity feature control frames. However, I'm wondering if that is a bit much? How do you handle this?

-Mike

 

[TheTick]

Sounds adequate.

Perhaps a single profile callout around the perimeter? This could cover size and perpendicularity at once.

 

[ivymike]

I don't handle it (haven't done prints in a long time), but i wonder whether you might be better served by putting a profile tolerance all the way around the sheet.

 

[mmolt]

I thought about the profile geometric symbol as well, but wondered how intuitive it might be for those looking at the print. I think it might be a bit confusing for those that aren't as "GD&T educated" and might not seem as logical as perpendicularity for rectangular shapes. I definitely see its use for curved surfaces...

Additional thoughts or methods?

-Mike

 

[KENAT]

What are your tolerance requirements/how 'square' do they need to be?

What drawing standard are you working to?

There is implied 90° for edges drawn at right angle. At least in ASME Y14.5-1994. This will invoke the 'block' angle tolerance. Maybe all you need to do is tighten that tolerance, although generally I'm a user of GD&T.

To do it properly you'll probably want 3 datums, to fully locate if for inspection.

As to the question of making it understandable for those without familiarity with GD&T, I've never found an easy answer for that. You need/want to use GD&T because you believe you can't adequatly describe your requirement using +- dimensions etc. However, the people who'll use the drawing wont understand the GD&T. Catch 22 that has been discussed at length on the GD&T forums before.

In future I'd post questions like this over in forum1103.

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

 

[berkshire]

mmolt (Mechanical)
An additional method would be to remove the angle at the corner of the sheet and place a diagonal dimension from corner to corner.
It is harder to get a sheet out of square then if the dimensions are correct.

B.E.

 

[mmolt]

KENAT,

The sheet defining dimension tolerances are plus/minus .030", and we would like the sheet to be square within .050".

Funny you mention the implied 90 degree comment, I've been round and round with our QA guy about this. If he had it his way we would call out every angle between intersecting lines, yeah, not good.

We mention the overall flatness of the sheet, is that the 3rd datum you're mentioning?

-Mike

 

[KENAT]

Before I try and give more help a bit more information please.

1. Roughly what size are these sheets (doesn't make much difference, more for me just to better understand the situation)?

2. What is your dimensional tolerance (this will help determine if profile would be the better tolerance)?

For your QA guys benefit, assuming you are working to ASME Y14.5M-1994 then look at Paragraphs 1.4 (i & j) & 2.1.1.2.

Of course for 14.5 to be effect you need to comply with para 1.1.3 and reference 14.5 either directly or indirectly from the drawing.

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

 

[mmolt]

KENAT,

The sheets are roughly 30"x60", and smaller depending on the usage, and the tolerance is plus/minus .030" for each of those.

I'll be forwarding our QA some information about this too.

Thanks,

-Mike

 

[KENAT]

I'll assume ASME Y14.5M-1994

Take a look at 2.7 regarding rule #1, especially 2.7.1.2 Envelope Principle.

This almost gets you what you want. If I understand what you mean by squarness then the +-.030 gets you something like squareness of .060. This is without any drawing change assuming you reference 14.5. You may want to get a second opinion on this but I think I'm correct.

Based on the numbers you give though squareness is slightly more important than overall size which might point away from profile. I've attached a scheme which may work however, no guarantees. Not sure the flatness on B is appropriate given the thickness, maybe straightness would be better.

Really the tolerancing needs to be based on function which I have no idea of.

I'm limited in my GD&T knowledge- just enough to cause trouble ;-) so if you don't get a second opinion on this forum then post in GD&T with a link back to this. Generally double posting is bad but I think in this case maybe OK.


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

 

[fcsuper]

Based on the limited information provided, KENAT's example looks good. I would likely add a second datum on the more significant front/back face and add an association to it within the GeoTols if you are concerned about edges in the Z (to the print) direction. (Would that be the missing "A" datum on this print, KENAT?)

Matt Lorono
CAD Engineer/ECN Analyst
Silicon Valley, CA
Lorono's SolidWorks Resources
Co-moderator of Solidworks Yahoo! Group
and Mechnical.Engineering Yahoo! Group

 

[KENAT]

Yep, I started making one of the large sides/faces datum A and then decided to limit my suggestion.

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