ASME Y14.5-Next may be better for nubies...

[ptruitt]

Perhaps some of us come to GD&T with too much baggage. I am hoping that ASME Y14.5-Next will take away some of the redundant symbology and hand out datums only when simultanaity is insufficient. Less is often better, in my opinion. A thread 100 posts long with many very capable minds in turmoil is not something to allow in the next Y14.5. The more nubies we can save from such a grim future, the better.

Peter Truitt
Minnesota

 

[MechNorth]

Interesting, Peter. Which symbols are redundant, and I'm not sure that I follow re the datums?
Definitely like to see some simplification, but I fear that the simpler we make things, the less they accommodate the diverse real-world situations we end up discussing here.

Jim Sykes, P.Eng, GDTP-S
Profile Services

http://www.profileservices.ca

TecEase, Inc.

http://www.tec-ease.com

 

[Belanger]

A worthy goal, but I too think that fewer symbols would only result in more debates here!

John-Paul Belanger
Certified Sr. GD&T Professional
Geometric Learning Systems

http://www.gdtseminars.com

 

[fsincox]

First, I thank the people who did help me with our discussion.

Peter,
I saw your comment in the other thread; I did not want to reply there as I didn’t want to hijack it. I must admit I have never understood why people object to the freedom of discussion of relevant issues in public forums? From what I have heard the committee meetings tend to spark the same kind of raucous discussions. Is that kind of fun only to be reserved for the elites? It must be hard putting up with us inferior minds, the Nazis had a solution for that, CH showed me they had a mobile version, too.
I would have welcomed your input since you have all the answers. Did you also bother to notice that I, myself, did not want to handle the issue in the proposed manner, but was basically handed this as a fait accompli. I was just told to vet it, “This is the way inspection wants to do it, is it legal?” typical office politics type of stuff. I have said here before how surprised I am how controversial and political things like GD&T, or going metric, really is when you are the one who is trying to implement it.
I have seen proposals for simplifying GD&T, they were going to unify in 1994 you know, what we got instead is something that looks like ISO but doesn’t pass the sniff test, I prefer to judge by the fruits!
I have long thought parallel and perpendicular were unnecessary, that concept went over here like a lead balloon. I would love to see them dump runout for just profile, now that we can officially use it with toleranced dimensions, after all we don’t believe in dictating process or inspection methods. It is not the fault of the working engineers that the committee confused the concept of concentricity (TIR) and runout. People wonder how did this issue get confused, simple, they screwed it up. First they tried to hide it by attempting to replace concentricity completely with position in 1982 at that time it became practically redundant. Then after not succeeding in erasing it from the collective memory, by trying to defining it away, telling us it is just so impractical to actually inspect no one would want to do it. Do they really think every point on a surface gets checked, too? Maybe they really aren’t as smart as they think they are?
Judging by the fruits of what we actually see in ISO, I personally thank God that foreign engineers are not OK with, “Well the center is in there somewhere” kind of thinking and want to know “how much out of round?” or “how much eccentric?”). Even when they adopted the ISO’s idea of specifying a “common zone” for a discontinuous feature they had an opportunity to actually move in the direction of harmonization what did they do, NO, they created “continuous feature”, only works with features of size!
There are too many vested special interests that get gouged when they try. Everyone has their own baggage, I have said here before: “what we in the field need is tools to get the job done”, if that means more tools so be it, if we don’t have the tools to do the job, what good is it anyway.
I hope the next one IS better, I think 2009 is better; I don’t think it is simpler.
Frank

 

[MechNorth]

As Frank seems to understand (perhaps having been involved in voluntary standardization before), there are significant personalities, egos and agendas at play when groups of users get together to reach an agreement on how to do something. It's not that people want to water down the product, just that consensus is only gained thru negotiation and trade-offs. Thus, what we as individuals may perceive as bad decisions and worse, are typically the results of trade-offs. I pushed for a few things and got one or two, but had to support others on their particular focal points.

Jim Sykes, P.Eng, GDTP-S
Profile Services

http://www.profileservices.ca

TecEase, Inc.

http://www.tec-ease.com

 

[fsincox]

Jim,
You guys did a good job with 2009, I like it. I do understand how hard it is, congress is the perfect example (no offence intended to you, Jim). Judging by the lack of ISO books out there they must just use you guys and say "but we do it a little different". I sure would like to see more on metric general tolerancing and independency fleshed out. That is for sure.

Peter,
No offence intended, all my GD&T life I have been told it was simple and found it to be just the opposite, it is simple because they simplify it to teach it, but we have to make it work in the real world.
My real pet-peeve is that you have experts, on this committee, who teach people they don’t need to bother to use it. This is just what they all want to hear. I have seen a guy recommend not to place an orientation tolerance on small short stubby pilots diameters, that are very functional, but are hard to measure or are too short to be measured accurately. This is undercutting the basic concepts we are supposed to believe in basically to make it easier to sell product! He is not alone, his example is just public, I have seen the other experts talk to my managers and console them. It should be our job to define all the geometry, like the book says. It should be the shops job to decide how and whether to check.
No wonder you get people who think they don’t need a position tolerance on a shaft slot. It is because the people who write the principles don’t live up to their own words. I suspect the real problems lay a lot closer to home and they are just manifested by the general confusion of the poor people who need to make it work.

 

[ptruitt]

Well, once again, I need to apologize to the hard working, individuals who generously offer their advice here. I keep hoping for simple solutions to complex problems. All too often, I fail to see the problem as deeply as is required for a proper solution.

Peter Truitt
Minnesota

 

[fsincox]

Peter,
Do not even think of it anymore; just keep on: "fighting the good fight". I saw your work with profile on castings and the troubles you had with your vendor, I know your pain and I am with you, we need to stick together! I cannot promise blind obedience though!
Frank

 

[DeanD3W]

Frank,
You make a good point that design should specify properly and let the shop then decide what to inspect... I tend to agree, but this brings up a question similar to another current thread... If you're tolerancing a .030 thick sheet metal part with one planar surface as the primary datum feature and a cylindrical hole as the secondary datum feature, would you apply perpendicularity to that hole? Theoretically, of course we should, but if we do, isn't it likely that the shop will scoff at our silly spec?

I would like to hear some discussion on this point... I am very concerned about perceptions and acceptance of GD&T.

I see your point, but I may be one of those you're referring to, or I may be aligned with their approach anyway.

GD&T is the best means by which mechanical design, manufacturing, and inspection can communicate. Situations can exist, such as very robust designs with very large tolerances, or excellent process control that reliably produces parts that are known to work, where GD&T might be safely avoided (with some risk). When we apply GD&T doesn't it make sense to use the explicit communication provided to manage the risk of dimensional problems. Otherwise we need to put a perpendicularity tolerance on a hole punched in .030 sheet metal and run the risk of others viewing GD&T as a cost adder, rather than the profit adder that it needs to be.

Dean

http://www.d3w-engineering.com

 

[fsincox]

Dean,
I would love to discuss the implications on flexible parts, let's get that going, probably not in this thread I suppose. I have quite a few irons in the fire right now, so it will take me a while to catch up.
I will try to be good tempered, after all, I don't want to screw up my invitation to lunch!
Frank

 

[fsincox]

Dean,
Please point me to that thread, I agree sheet metal is not well covered in the standard, I myself would have ignored the perpendicularity there. I would like to get in on that and find out what others think and do.
Thanks,
Frank

 

[MechNorth]

Tks Frank, on behalf of the Y14.5 Committee. I wasn't on the committee, just sat in the back rows as a tolerated pest ... I mean guest. My first meeting, I raised issue with something (apparently that's not how they've done it, but I didn't know that ;~} ) and apparently my argument (go figure?!) was effective because the change was made. Several other changes done that way too, so I was happy. My "fine" work is part of the Y14.3 and the Y14.41 to be released this year.

And yes, I am one of the buggars who doesn't put an orientation control on a secondary datum fos in thin material wrt the primary datum. One of the critical factors for me is the legality of the document; if it's on the drawing, then it is enforceable in court. So, if you put a perpendicularity tolerance of .0005" on a hole in thin sheet, you (as the purchaser) better be able to inspect it accurately & repeatably.

Jim Sykes, P.Eng, GDTP-S
Profile Services

http://www.profileservices.ca

TecEase, Inc.

http://www.tec-ease.com

 

[fsincox]

Jim,
I understand compromise must be made and you are not in the business of making enemies and I do not want to make one of you, either.

There lies the fundamental issue, is it our job to define the required part geometry, or only what can be inspected and where? Here or anywhere? Change the rules then, don't say these are the rules and then don't follow them! Say it only applies to machined parts not sheet metal, or something! I bet part of the problem is independency makes much more sense for sheet metal parts. The ISO acknowledges different tolerances for different production processes and they have independency.
Frank

 

[DeanD3W]

Frank,
I think we should apply GD&T explicitly and specifically to the more important features in the views of a drawing, with a profile in an "unless otherwise specified" general note for the features or characteristics deemed not critical enough for their own tolerance specs...

That profile in a general note is the key to enabling only the sensible tolerance specs to go on a drawing. The requirement that "dimensioning and tolerancing shall be complete" is still met, but the drawing much more efficiently enables managing risk. A benefit of this approach is that inspection will only address the explicit and specifically applied tolerances, so no one should scoff at any impractical or unmeasurable (insufficiently repeatable) specs. The profile in a general note, with a relatively large tolerance value, serves ensure that "complete" specs are provided and also that no one needs to pay for parts with large, unexpected issues.

If there's a better approach, then I'm all ears.

Dean

http://www.d3w-engineering.com

 

[MechNorth]

Frank,
I don't take offense about contrary opinions; they may confuse me, but I don't take offense at them. I rather like hearing how other people interpret things and process things so that I have something else to contrast my thoughts against. I even, occasionally, change my mind on something!

There apparently was a sheet metal standard years ago, but there was not enough interest/participation to keep updating it, so it was removed (the ASME rules require that every standard be reviewed and revised, re-affirmed, or removed every 4 years). I had several clients a couple years ago that wanted ASME to do something again, but none were willing to participate. I offered to lead such a group and only 2 or 3 people stepped forward, and they couldn't commit at the time either. If people want it, they need to commit to it, that means manpower, time and travel for a good 4 years in many cases. As an FYI for anyone interested, note that such endeavors MUST be made under the auspices of a recognized organization such as ASME or you risk anti-trust issues (something else that scares people off).

Jim Sykes, P.Eng, GDTP-S
Profile Services

http://www.profileservices.ca

TecEase, Inc.

http://www.tec-ease.com

 

[ptruitt]

Why would you not use profile of a surface for holes in sheet metal? It would eliminate the theoretical possibility of oval holes without adding the hard-to-measure perpendicularity control.

Peter Truitt
Minnesota

 

[fsincox]

Dean,
I love it, I wrote a company book espousing such a thing 1993, I was a little more naïve then and thought it was revolutionary, my company’s response was to proceeded to ban profile tolerancing completely!
It is basically what I am advancing in this thread:
thread1103-315749
I then started toying with the idea of combining a general profile note, to an ISO “IT” tolerance grade, defined to a "MASTER FRAMEWORK". That started me on my current journey of study and interest in the ISO system of general tolerancing as I see it to be very similar in concept.
Frank

 

[Belanger]

Jim, if you choose to not use an orientation tolerance on a secondary datum feature, how shall we simulate it (assume it's referenced at MMB) to create the DRF?

John-Paul Belanger
Certified Sr. GD&T Professional
Geometric Learning Systems

http://www.gdtseminars.com

 

[MechNorth]

Well, J-P, I'm apparently missing something in your question. If you reference the secondary datum feature at MMB, then you put in a simulator pin at the MMB condition with zero perpendicularity error ... because that's effectively what you have on a thin stock situation; zero perpendicularity error. I used to believe that you had to follow the "control the datum feature" edict blindly, but then the reality of thin material situations smacked me around a bit. In MOST metrology environments, there's no way they are going to be able to have enough truly cylindrical wall to establish a tilt on the axis (there is tapering at the one end and often bulging at the other). Where then is the practical application and function of a perpendicularity control in that situation?
Now that is NOT an absolute; one client was using a pressed-in optical element in a sheet metal component; the material still wasn't substantial, but because of the physics, the perpendicularity of the axis (well, the cylindrical wall really) was critical. They had a means of inspecting that shallow hole for orientation.

Jim Sykes, P.Eng, GDTP-S
Profile Services

http://www.profileservices.ca

TecEase, Inc.

http://www.tec-ease.com

 

[Belanger]

But that's just it, Jim. I don't propose a perp tol on a hole through the thin part because it is something that's checkable. I would propose it because it defines the gage.

A zero perpendicularity tolerance is assumed in your condition. Why build a gage based on an assumption? Make the tolerance zero at MMC; that's effectively what you are saying. It would not be checked as you say (i.e., axis), because as that datum gets simulated, it will automatically be checked by that envelope.

Rule #1 relates size to form, but not size to orientation. So the practical application and function of a perpendicularity control of zero in that situation is that it merely "crosses the T and dots the i," so to speak.

Now, after all that: I don't mean to sound too pushy on this point -- I am not really a cheerleader for this. I just don't want people to think that there is no logic behind having perpendicularity on a paper-thin part.

John-Paul Belanger
Certified Sr. GD&T Professional
Geometric Learning Systems

http://www.gdtseminars.com