Continue to Site

Eng-Tips is the largest engineering community on the Internet

Intelligent Work Forums for Engineering Professionals

  • Congratulations The Obturator on being selected by the Eng-Tips community for having the most helpful posts in the forums last week. Way to Go!

Squareness Tolerance

Status
Not open for further replies.

ModulusCT

Mechanical
Nov 13, 2006
212
So nevermind what this says about our QA department that they've asked for a squareness tolerance on this assembly... I would however like to get some feedback regarding how best to tolerance this assembly. Please see the attached image. Below is the suggestion a senior engineer gave my drafter regarding how to control perpendicularity of the assembled frame. I suggested that this is over-toleranced, and redundant as controlling one centerplane with perpendicularity is appropriate, not the two centerplanes twice.

asdfg_siapt7.png


Below is my suggestion for making the frame square. What do you all think? Personally, I'm thinking that the peice parts should be toleranced correctly to control the assembly so this whole assembly tolerance thing is new to me, however, I'm new to this position and am trying to learn how best to process everyone's problematic drawings without throwing everyone into a state of disarray - you know, by following the Y14.5 standard.

asdf_lqki46.png


Thanks in advance.
 
Replies continue below

Recommended for you

I would use all-around profile of surface, at least based on the possibilities afforded above.

Doing more requires knowing what the part mates with, otherwise it's like asking if "20" is the correct answer to an unstated math problem.
 
I agree with 3DDave - profile - unless more function and fitup info is provided

Certified Sr. GD&T Professional
 
Thanks, I had that thought too... I figured I might blow some minds if I suggested profile. In any case, this part sits in an envelope, approximately the same size and shape as this thing. I could be said that it requires a go/no-go that you could just set this thing in and measure the VC of all dimensions.

The four parts that make up the frame are cut from a sort of rubber that comes on a roll. The four pieces are cut to equal length and the assembly needs to be square. They've had some assemblies that have been coming in a quarter inch or more out of square.

I agree, profile of a surface, all around with a value less than the size tolerance. Agree?
 
No. The size tolerance MUST be BASIC. Note too, you can locate the tolerance zone relative to the BASIC profile using the U in-a-balloon symbol. See Sectgion 8.3 on page 158 of the 2009 version

Certified Sr. GD&T Professional
 
Oh yeah, I know that the dims would need to be basic dims if we went with profile. I was just talking to myself I guess about the tolerance value that we would need. Circled U - Isn't that an unequally disposed tolerance zone modifier? I'll look it up.
 
I was not sure of your knowledge level. Yes it is. Lets say you want to part to never be smaller than a certain dimension (size) you can use the U symbol to place the entire tol zone "outside" so the part can only be bigger by some amount. Also, be aware; you can allow the tolerance zone to vary as you go "around" the part using the "between" symbol.

Certified Sr. GD&T Professional
 
ModulusCT,

A squareness tolerance would make sense to me if you were willing to have sloppy dimensions otherwise. I.e. your part would the 19.44[±]0.5, with a squareness tolerance of 0.1mm. Squareness would consist of your sides being perpendicular to each other and equal in length. I do not why you would do that.

As noted above, profile controls size and squareness.

If your part goes inside something, your squareness is required only at MMB. In ASME Y14.5-2009, look in sections 8.3 Tolerance Zone Boundries. There are now several ways to make your profile zone go one way only.

--
JHG
 
Thanks everyone for the feedback. I'm having trouble wrapping my head around the unequally disposed profile tolerance (why is U better than equal bilateral?), but this is good info and I think profile is the way to go to control size and form in this case.
 
I'm just going to hazard a guess here and suggest that legacy/traditional dimensions might be the way to go with a part like this. What is it? It looks like welded angles with expanded metal mesh.

Who's going to be making these things? Is this going to be sub'd out to a podunk fab shop that's going to only have tape measures and squares? You might want to rethink the idea of applying a profile tolerance.

You could dimension the two sides, then also dimension the diagonals with a more controlling tolerance. I guarantee that's how someone will check 'squareness' if they're making a welded grate, if they had no print.
 
JNieman said:
Who's going to be making these things? Is this going to be sub'd out to a podunk fab shop that's going to only have tape measures and squares? You might want to rethink the idea of applying a profile tolerance.

On occasion, I have provided rectangular dimensions with some sort of GD&T and I have provided the diagonal dimension across the corners as a reference. This gives a guy with a tape measure a chance to check for perpendicularity. The primary against this is based on people using drafting boards and doing the calculations by hand, possibly with a slide rule.

--
JHG
 
Providing proper GD&T controls while placing extra reference dimensions to allow alternate fab methods sounds like a good way, in my personal opinion which is a reflection of the work environment I am in. I can see some quality personnel or managers throwing a fit over the idea of allowing fabrication to commence when knowing they're using reference dimensions. If drawn/dimensioned properly, though, I imagine you could successfully get your parts how you wanted them, whichever way the supplier chose to execute the task.
 
The frame is created from a roll of aluminum bar that's run through a sort of die and a small cut made in three places, like a 45 degree notch so that the part can be folded to form the square, I believe. Though, the four pieces may be seperated, I'm not entirely sure.

Adding the reference diagonal dims sounds like a pretty good idea. So, I'm setting up the bottom surface as datum A, the bottom leg, datum B, left side, datum C, and then using an equal bilateral profile of a surface tolerance (all around - .06 tol.) to control the form and size (with basic dims). They should be able to make a simple fixture with pins or a frame that checks the VC of this assembly and just drop them in to ensure that they conform to the profile tolerance and then they can take some length measurements to assure that it's not too small.
 
Status
Not open for further replies.

Part and Inventory Search

Sponsor