Pattern of holes as reference and function of frame with profile in a special condition

[roberto1brazil]

Hello everybody.
Please, I need your support regarding the sketch attached. It is a simplified drawing of a shroud of turbine. Only the main features are represented in the sketch. The aim of drawing is to supply information to machine a lobule located at 30° from hole “O” at opposite position ( section E-E). There is a fixture that locate the part 0,15 mm off center (in the angle indicated) in order to machine the internal lobule. The part is supplied with an initial internal diameter ranging from 237,65 to 237,70. The average wall thickness is around 1,45 mm. The origin reference center (B) is established by the three holes diameter 7 in the basic diameter of 282 mm.
In a practical way, the internal lobule is done moving the cutter to outside little by little until the thickness reaching near 1,3 mm thickness at position of section E-E. In fact it is done controlling the final thickness.
Doubts:
1- Why the reference B is written as B-B inside the frames?
2- What does it mean the frame identified with interrogation (?) in the sketch and what would be its function?
Thanks for any comments.

Roberto1brazil

 

[mkcski]

3DDave:

I have no idea how the Linux crowd controls revision status if they do. Revision control is critical to effective Standards. This forum is a clearinghouse for clarity of interpretation of a "fixed in time " Standard. I cannot imagine the chaos with interpretation if the Y14.5 was continuously revised - like the operating systems you mentioned. There is a public review and comment period for each release of Y14.5. The Committee members are professionals and experts. So, we accept this "representative" as opposed to "democratic" form of revision control to contain confusion.

Certified Sr. GD&T Professional

 

[3DDave]

All software is revision controlled. Big Revs and little revs. Major releases are governed by their governing bodies, leaving it to volunteers to supply the work, just like ASME, but without hiding the supporting arguments, unlike ASME.

There is no public feedback and no posting of interpretations that were argued in the meetings, making the 'public' portion not particularly public at all. Even though I participated in three previous Y14.x public reviews I was not notified of the latest version of Y14.5 until the comment period had passed, and only because someone in this forum mentioned it in passing; I'd think that would have been appropriate at the outset, but Committee members seemed uninterested in gathering information. In the past the comment period was so short and the time to get shipped was long enough that I would have to check the ASME website on a weekly basis for nearly a decade to catch the notice. This is unreasonable. When I last looked the organization was such that if I didn't already know there was a public-review-draft, I would not find it.

I agree to the professional, but suggest that 'expert' is not always the case. Experts would not leave as many areas unexplained in their standards and certainly would not let so many typos and incorrect results get to publication.

 

[axym]

mkcski,

Which meeting did you attend? I might have been in the 50-person "peanut gallery" behind the Y14.5 main table. If there were donuts, you were lucky - usually it's just coffee and tea. In the other committees, we have smaller rooms with just water and mints. I usually have to go down to the Y14.5 room and get coffee from there ;^).

Evan Janeshewski

Axymetrix Quality Engineering Inc.

http://www.axymetrix.ca

 

[mkcski]

3DDave:

Yes, I totally agree the communication is one-way and "catch as catch can". Do you know how ASME notifies the public of the availability of Review and Comment releases? Are the 14.5 guys following an ASME policy for this or is it determined by committee charter (or some other reason)?

I did request and get a copy of the Review and Comment for 2009. I submitted many comments, all of which were responded to - most rejected but several were held for discussion - by who I am not sure.

Certified Sr. GD&T Professional

 

[mkcski]

axym:

It was in Dayton OD in 2010. It was my first and only meeting. Were you attending meetings then?

Certified Sr. GD&T Professional

 

[3DDave]

mkcski, They didn't publish a list of all the submitted comments and disposition, did they? It's trivial to do that nowadays with the internet, but they chose to waste public time and potential interactions between commenters. Instead of cruising through the already submitted and accepted changes and seeing which errors and omissions were already dealt with, everyone ends up being a copy editor, by the dozens? Hundreds? It could take each reviewer 10-20 hours to fully read and try to understand the internally accepted colloquialisms, so that means an avoidable loss of hundreds to thousands of hours.

The review notice is the minimum per the ASME process, which seems dedicated to making sure as few people know about the process as possible. They obviously don't see me as a customer or contributor, even though they incorporated several of my changes in the last version. My email hasn't changed. There is a review page:

https://cstools.asme.org/csconnect/PublicReviewpage.cfm

I see no fillable PDF form for electronic submission purposes.

I see some are now free; it seems like it cost me $120 to get the 2009 draft. The irregular schedule is also a problem.

1966?
1973 (7 years)
1982 (9 years)
1994 (12 years)
2009 (15 years)
(2017 draft) (8 years, quite a surprise.)

 

[mkcski]

3DDave:

No. I saw no list.

Thanks for the link. Do you have to be a member to get to this page?

Your observations are enlightening. Seems ASME is way behind the digital times. Maybe they could use paragraph and figure numbers to sort/filer comments and let the public create the "thread" of comments for each.

Certified Sr. GD&T Professional

 

[3DDave]

mkcski, No membership. Wouldn't be so much a public review otherwise, but, like I noted, I would have spent nearly 15 years of checking. There used to be a good Y14.5 group on Yahoo! but that seems to be gone or dead. I probably found out about it there.

 

[axym]

mkcski,

I was at the Dayton 2010 meetings. Do you remember a presentation on zone-based definitions for runout tolerances? I did the second part, with the crazy-looking figures. At the time Y14.5 was looking at zone-based definitions for circular runout and total runout, as there were those who felt that the indicator-based definitions were too old school and needed to be modernized. I was part of a group put together to study it, and somehow got tasked with presenting the results. The reaction was basically "that's not wrong, but we hate it". It turns out that the concept of full indicator movement through 360-degree sweeps is about 50 times easier to describe and understand than the equivalent tolerance zone convolutions. So they stayed with the indicator-based definitions at that time (I see that this has since been revisited and there are zone-based definitions in the latest Y14.5 draft).

Evan Janeshewski

Axymetrix Quality Engineering Inc.

http://www.axymetrix.ca

 

[mkcski]

axym:

Unfortunately I do not remember your presentation and I do not have my notes with me at work to review. I vaguely remember Don Day giving a presentation on something about castings using a tablet and stylus (real high tech at the time), and Al Neumann and son Scott discussing something with Datums, and Jim Meadows doing something with Position. I remember being overwhelmed by the whole experience of my first Committee meeting - of being in the same room with a bunch persons who "know" GDT - and being able to follow the presentation. I keep working on my boss to send me to another meeting - fingers crossed.

Certified Sr. GD&T Professional

 

[greenimi]

........So they stayed with the indicator-based definitions at that time (I see that this has since been revisited and there are zone-based definitions in the latest Y14.5 draft).

Evan, ( or anyone for that matter)

Do you know, by any chance, why in the draft total runout will have to control cylindricity and in 2009 was not such a requirement? Just curious, why the "door" to have total runout on a cone (with basic angle) has been closed?
As someone stated before, maybe because of the "delta" ( new modifier) ?
Do you have any "inside" /backstage details?

 

[Belanger]

Greenimi -- Here's a link to an old thread that I started on that topic. Evan made some comments there on the idea. (And Jim Sykes too -- haven't seen him on the forum lately!)
[URL unfurl="true"]http://www.eng-tips.com/viewthread.cfm?qid=315985[/url]

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

 

[3DDave]

For a cone vs total runout:

A constant zone at a basic angle to an axis is equal to a profile tolerance. All that changes in the 'size' of a cone is the distance along the axis from the theoretical cone point/vertex/apex to any other features on the part; it's why cones don't have a 'size.' Any value is just a local value and can be found if the nominal cone is extended far enough from the apex.

I know that for convenience there are alternate dimensioning methods that give the appearance of size, but they really are variations of controlling where the apex of the cone is.

If the cone is a primary interface then all the other features should have locations traceable to the apex; it would be trivial to add this to the standard as the interpretation for all cones with basic tapers.

This is in contrast to a cylinder which is the same nominal size no matter how far it is extended so that a size can be determined and a perpendicular surface which has no size at all.

Since profile of a surface relative to an axis is the same as the proposed total runout, there is no need to add the interpretation to total runout; one could show that a cone surface profile could be verified the same way as proposed for conical total runout and direct seekers of conical total runout to the profile of surface.

 

[Belanger]

Dave -- read that other thread. Your idea was brought up by Jim, but it's not the same as profile because profile of a surface requires a "true profile" -- thus controlling size of the cone also. Total runout is powerless to control size.

I think that thread is still open; maybe we should be posting there for that discussion.

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

 

[3DDave]

I did read it and cones don't have a size to control, so the true profile is covered and is the same as total runout. I can reduce every method of calling out a cone to a point and a taper. No size required.

Edit - threads close after a relatively short period; that one is 5 years old.

 

[Belanger]

Cones don't have a size to control? Certainly they do. It's just that the size varies as you move axially.
You say that every cone can be reduced to a point and a taper -- that's not quite correct because that would be an infinitely long cone. You need to also designate a longitudinal start/stop distance from the vertex. And guess what: that creates a definite size control.

I'll try to create a sketch showing why profile must have a basic dimension. And yes, this would lead up to the proposed "dynamic profile" modifier in the draft version of Y14.5.

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

 

[3DDave]

Nope. The location(s) where it is cut off has no effect on the cone size. Every bit is exactly as it was in the uncut portion.

The basic dimension is on the taper. The cuts are located relative to the taper. Viola. No size measurement required to verify the cone.

 

[Belanger]

Methinks we differ on what "size" means. I'm not talking about a FOS, just size (diameter).
Picture an infinite cone, starting at a perfect vertex. If I take a finite chunk from stations 10 mm to 20 mm, you're telling me that it will have the same diameter as a chunk taken from stations 20 mm to 30 mm? I don't think so.

But let's set that aside. I want to focus on why profile of a surface would be different from total runout. I'm away from my desk for a couple of days, but I'll provide an example to illustrate my point. If you have access to Y14.5's draft, check out dynamic profile. That was created because the current meaning of profile is non-dynamic (locked around a basic dimension). This is why Evan and I were always complaining about Figures 8-17 and 8-18; they make sense in one line of thinking but conflict somewhat with the text's description of how profile is to be used.

edit: I don't have a problem with Fig. 8-17 because it's not related to a datum. Thus the profile zone can move back and forth.
But Fig. 8-18 is a problem because the diameter is toleranced -- I'll follow up in a couple of days with my explanatory sketch.


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

 

[3DDave]

Right - so it doesn't have a size, it has an infinite number of them and anything in a reasonable range will be found on any cone of reasonable size and extent.

I don't have a copy because it doesn't seem important enough after I worked on 14.5, 14.41, and 14.43 and mostly got snubbed with 'already decided, not looking at it, can't give evidence for reason decision was made. Thx. Bye.'

Had it been mentioned early enough then I might have taken a shot, but finding out after the close; nope.

 

[Belanger]

To show why profile of a surface and total runout are not necessarily the same thing when applied to a cone, I had stated that profile controls size (whereas runout does not).

Profile doesn't control size in many situations where it's not applied all around (such as in Figure 8-27). But with a cone, the profile naturally wraps around the part, thus controlling size. This brings up the question of whether the diameter of the cone needs to be a basic dimension.

See the attached graphic, which consists of several profile examples. Notice the first one is applied to a flat surface: we should all agree that the "true profile" of that surface is simply zero (no curvature; that's simply a basic dim of zero). Then the second example shows a slight radius, and we should all agree that the radius must be basic.

More examples show the radius gradually wrapping around further -- and again we say that the radius is a basic dimension. When we get to the last picture we have a full circle (or cone) and doesn't it follow that the diameter should be basic?

This is why profile is different than total runout. To say that profile and runout on a cone are equivalent is to say that the diameter in my last graphic can be converted to a ± tolerance. And if that were true, then each of my previous sketches should also be allowed a ± tolerance, but the logic breaks down somewhere in that line of thinking.

The only exception to the idea of having profile on a basic diameter is Figure 8-17. There, it's perfectly fine that the diameter is given a ± tolerance because the profile zone is not related to a longitudinal datum, and can slide left/right to accommodate the size wherever it happens to be.

We'll have to see what happens with the new idea of "dynamic profile," but I suspect it will help overcome this conflict by allowing a profile zone to expand/contract.

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