What Prevents Inadvertent Stress Concentrations or Crack Initiation on most prints? 7

[SBlackBeard]

I'm a designer - there is a recurring theme with some parts we've seen recently, where there is a sharp step right where we'd least want one. We battle with other departments - they say it meets the print, we say it doesn't.

For example, we might design a bracket with a fillet/blend specifically to avoid a sharp internal corner for stress reasons, but we control the size of the radius or the adjacent flat surfaces with a profile tolerance - say .015 (we use inches). I realize this allows for considerable surface variation (like they draw in the GD&T books!), but we also have a block tolerance surface finish of 63 or 125. Here's an image of it (blue is as-built, grey is as designed):

You can see there is a step near the root of the fillet. Likely, this was machined with two different tools or from two directions, so I understand HOW it happens. At prior jobs, quality would have rejected or reworked the part. At the current job, Quality says "because the step is within the surface profile tolerance, it meets print."

I originally thought block tolerance surface finish of 125 Ra should cover it. I then thought maybe we should say this is a controlled radius, but we see steps at the tangency point. The more I consider it, our surface finish should call out an Rmax that prevents this. I'm particularly concerned if this happens on aluminum parts prone to fatigue.

How do other people control for this type of manufacturing artifact with the print?

 

[3DDave]

Your understanding is a literal reading of the exact wording in the standard - is that correct or do you think it doesn't mean exactly what it says?

 

[BiPolarMoment]

Isn't an "all over" profile simultaneous despite possibly (probably?) having no explicit datum references?

 

[mfgenggear]

that's the problem with GD&T , people try to make it perfect and it is not. the more complex and the more it become interpretation.

 

[Burunduk]

Your understanding is a literal reading of the exact wording in the standard - is that correct or do you think it doesn't mean exactly what it says?

That's not my approach, but it may be yours if you suggest that the plural in the wording of the simultaneous requirement definition may invalidate a single datum reference.

When reading the standard, just like with everything else, you have to get the gist of what's written, and understand the words in their context. That's why the plural in the simultaneous requirement definition doesn't mean that two or more datum references are needed. The context of the plural wording is that the datum precedence order has to be the same, and it's difficult to talk or write about any precedence order without plural wording. But it is elementary that where only a primary datum reference is used, and it is the same one for several controls, it is still primary for all of them, in the sense that the part is mated to the simulated datum the same way it would mate to a primary simulated datum when lower precedence datums are referenced, so it is within the definition. The goal is that all specifications taking part in a simultaneous requirement are to be controlled in the same datum reference frame, and in one inspection set-up, in which datum feature simulators are involved. If that datum reference frame is established by a single datum reference, it is not less valid.

Isn't an "all over" profile simultaneous despite possibly (probably?) having no explicit datum references?

BiPolarMoment,
"All over" is a different type of pattern than a simultaneous requirement. All over doesn't require a datum reference and in its common datumless application which is supported by the standard all degrees of freedom are available to make a best fit evaluation. In a simultaneous requirement, all considered features are controlled in one datum reference frame with identical constraints of degrees of freedom.

 

[BiPolarMoment]

When reading the standard, just like with everything else, you have to get the gist of what's written, and understand the words in their context. That's why the plural in the simultaneous requirement definition doesn't mean that two or more datum references are needed.

There is an example of using only a single datum in this application in Fig. 7-47 (Y14.5-2018), although that is a datum axis so two degrees of freedom are controlled with a single reference.

 

[Burunduk]

BiPolarMoment, that's true, your example supports the point that it is useless to nitpick on the plural verbiage of the definition.

The correction to what you said is that the number of degrees of freedom constrained by the primary datum axis is 4. Only the axial translation and the rotation about the axis are unconstrained.

 

[BiPolarMoment]

Yes well...I was thinking in 2d.

 

[3DDave]

B, you've argued as a nitpicker before. Just when it suits your purpose. But it's clearly plural. File a change with the committee to fix it and support, as you have previously demanded be done, the exact wording until it is changed.

 

[Burunduk]

3DDave,
The closest I've done to nitpicking is insisting that tolerancing terms should be used according to their defined meaning.
You on the other hand don't hesitate before resorting to the most blatant type of nitpicking wherever you think it may benefit your position, even if the nitpicking ends up completely nonsensical, like dissecting the use of plural vs. singular words in a paragraph of the standard while ignoring the context, and being persistent with that even after the context was clearly explained to you. And for your information, one (1) is also a number that can describe the quantity of "common datum features referenced in the same order of precedence at the same boundary conditions". If you think there is anything wrong with the wording, you file the change. You probably already filed changes in the past with no good reason, go on and do it again.

 

[3DDave]

I have filed changes they were accepted. That's why there is an example of using three levels in a composite tolerance after nearly 40 years of showing only two.

I haven't felt the need to deal with this issue because I thought most users would be intelligent enough to know that zero is a number and that zero references means that all callouts with that same reference for position and profile of surface would be simultaneously applied. I overestimate users as you have clearly shown. So thanks for pointing out where users are unable to understand for themselves a simple concept.

 

[Burunduk]

Expanding the composite tolerance to include more than two segments was widely discussed prior to taking place in the 2009 revision, so don't take all the credit to yourself and your change file.

Most users are intelligent enough to know that if a geometrical control specified with zero datum references, it means that it has no datum references, and any concepts involving datum references are irrelevant to it. This is exactly why a simultaneous requirement as defined is not the only recognized grouping method to invoke a pattern, there is a list of them.
But I guess this doesn't help those users who are too lazy to learn the terms and understand the differentiation between a simultaneous requirement and the other types of patterns.

 

[3DDave]

I made a contribution that was accepted and used. What have you contributed? I don't know if it was discussed - you weren't at the meetings so you could not know. What I do know is I corresponded with Alex Krulikowski before it was in the standard - he did not indicate that it was being considered, though he was on the committee, but he did indicate that it should be acceptable.

You haven't gotten to the case where no datum references would be used and there is no pattern. Since you are good at parroting then I expect you won't figure that case out. Until you do that's it. Please don't attack my opinion in the future when you offer nothing of value in return. Thanks.

 

[Burunduk]

It was widely discussed in the industry and therefore brought to the attention of the committee at multiple occasions.

A simultaneous requirement based on no datum references - there is no such case. If you say there is, it is still up to you to provide the evidence, not up to me to find one that isn't there.

How many degrees of freedom are constrained by the "null reference"?

You were the one who attacked my opinion and the profile control suggestion I made to help the case of the OP. You couldn't explain why it would not work, so you went on to some useless nitpicking of the standard and personal accusations. Thank you for your valuable input.

 

[3DDave]

I don't see the example you should have been able to show, so you didn't write anything worth reading. If you don't understand the subject and don't offer workable solutions then go back to being a reciter of only what is written and leave the problem solving to engineers.

 

[Burunduk]

At least I offered more than writing an ambiguous note, or suggesting that any step on the surface is a violation of surface roughness requirements.

But your main engineering problem-solving suggestion was firing the vendor. Yeah, keep telling us about your workable solutions.

 

[Burunduk]

SBlackBeard,
As I recently found out thanks to a comment by chez311 on another thread, if a general datumless profile is specified per my suggestion from 29 Apr 22 08:28, and the drawing is based on the 2018 version of the ASME Y14.5 standard, the notation "INDIVIDUALLY" has to be placed following the profile requirement. Otherwise, it can be interpreted as profile All Over.

 

[powerhound]

Regarding whether or not position and profile without datum references should be considered a pattern per simultaneous requirements, that was argued heavily. I was at the meeting where it reared it's ugly head. I discussed this with a brilliant mind in this community and he gave me this analogy: He asked me who my favorite squash player was. I said I didn't know since I don't follow squash. So he asked me "So you don't have a favorite squash player?" I said "No". He said "Neither do I. Does that mean we have the same favorite squash player?"

John Acosta, GDTP Senior Level

 

[3DDave]

In the lowest level of a composite frame when there is no datum feature reference does the same thing apply? I know they say it's a refinement or some such - however -

Suppose there's an extruded bar with two patterns of holes that will be used as the datum features to finish the outside. They have no datum feature references as the entire outer surface will be machined relative to them. Are the two patterns simultaneous?

I read feature control frames left to right - the tolerance zone shape - the tolerance zone size - establishes all the related tolerance zones which are located/oriented per the basic dimensions between them, then aligned/oriented to the first reference using any relevant basic dimensions , further aligned/oriented to the second one until they are as desired.

They are not dandelion seeds blowing randomly in the wind suddenly organized by noting the Earth, but a diagram on a page being moved and located in unison according the the frame of reference the group belongs to. Unless they are related to different datum feature references they should all be on the same sheet of paper (or the CAD model of same.)

If they are loose like dandelions then I wonder where they are blown to when they are UOS.

Was this a guy named W.(B) T.? It's an interesting analogy, but being a favorite or having no selection is not the case in a drawing of a part as those are subjective opinions. There should be chains of basic dimensions and/or implied angle controls between them.

It still doesn't solve the continuity problem unless they are simultaneous and then it does so by making the part much more expensive than it might have to be. I can level a table with a match book or a gauge block; there's no need for parallelism in the millionths of an inch. I don't see an answer within Y14.5.

 

[Burunduk]

He asked me who my favorite squash player was. I said I didn't know since I don't follow squash. So he asked me "So you don't have a favorite squash player?" I said "No". He said "Neither do I. Does that mean we have the same favorite squash player?"

That's a spot-on analogy.
I am surprised it was argued, let alone heavily.
Any beginner starting to understand what a simultaneous requirement on a practical level and not just per the dry definition and how to specify it on a drawing semantically, realizes it has to do with common constraints of degrees of freedom for all the features in a simultaneous requirement.
This is why datum references are part of it. Of course it differs from a pattern established without datum references and implies an unconstrained best-fit for the grouped features.
Furthermore, even if a specific degree of freedom remains unconstrained such as the rotation in the example of the shaft with the two keyslots aligned by the simultaneous requirement, the degrees of freedom that ARE constrained allow the alignment anyway by making the tolerance zones related to each other in a common measurement even when both are not clocked around the datum axis.

I believe that the confusion that evolved into the "null datum reference" is rooted in the tendency of people to interpret concepts literally, i.e "what it sounds like". And since all patterns can be associated with simultaneity, they think: "The green cylinders are in a simultaneous requirement, but the standard says there need to be common datum references, then let's be sophisticated and say that a datum reference is there in a way even though it is not there, and call it the null datum reference". What they don't realize is that if there was no "3X" and the position tolerance would be applied separately to each feature it would become meaningless. But it would not be meaningless if it was profile, it would then become a pure form control, separate for every feature.

3D,
The low levels of different composite tolerances are not in a simultaneous requirement, by definition.

 

[3DDave]

The low levels of different composite tolerances are not in a simultaneous requirement, by definition.

Really? None of the features within that lower level are in a simultaneous requirement? You certainly mean that they are isolated from other simultaneous requirements, by definition and by default, not that within the pattern they aren't simultaneous.

As usual you changed to an argument I did not make.

When you said "none" is not a number, but argue that "one" is plural I think you need more time learning to design things and less time spent on this.

I'll look for the exception to the rule that position and profile are subject to simultaneous requirements, but only where it makes sense and not otherwise. Perhaps that can be found somewhere.