Continue to Site

Eng-Tips is the largest engineering community on the Internet

Intelligent Work Forums for Engineering Professionals

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

pattren of holes - datum

Status
Not open for further replies.
Madhu454

Madhu454

Mechanical
May 13, 2011
129
From my previous post on runout I got this doubt.

Is it mandatory to use M modifier if I want to use a pattern of holes as Datum?
If I use M modifier then I can use a fixed gauge and the axis of the pattern can be considered as datum.


Is it possible to have a pattern of holes as datum at RFS condition, If possible how to simulate the datum in this case? Here also the datum will be the axis of the hole-pattern only, not the axis of individual holes.

Can anyone help me to understand this?
 
Replies continue below

Recommended for you

Sort by date Sort by votes
Per 2009 edition of Y14.5 it is no longer mandatory to use pattern of holes at MMB when one wants to use it as a datum feature. It can be assigned at RMB basis too (see 4.12.4 in Y14.5-2009).

Like I said in the thread about runout, gage pins used to simulate the datum axis at RMB would have to expand simultaneously from MMB towards LMB till they get maximum possible contact with corresponding holes. They are of course located at true position as defined by basic dimensions.

If the pattern contains 10 holes and all of them are assinged as datum features then theoretically you have to use 10 expanding gage pins to simulate the datum - at least this is how I understand 4.12.4.
 
Upvote 0 Downvote
I've already suggested that in your previous post.

Look on Fig. 4.35 in ASME Y14.5 2009.

They show pins, but similar set-up could be created with custom chuck jaws for holes.

All of your holes will be one datum feature, and because chuck is adjustable, it will grab them at RFS.
 
Upvote 0 Downvote
I don't have a 94 standard with me today, but I don't recall it being required to have the MMC modifier on datum reference taken from a pattern of holes. It certainly is the most practical way to do things, but was there a paragraph that spelled it out as mandatory?

John-Paul Belanger
Certified Sr. GD&T Professional
Geometric Learning Systems
 
Upvote 0 Downvote
I do not believe there is a statement saying that a pattern of holes must be specified at MMC when used as a datum feature. Paragraph 4.5.7 (of the 94 standard) indicates to me that it can be at RFS. An example would be a 2 hole press fit dowel pin pattern that has mounting holes and other features tied back to it.

Drstrole
GDTP - Senior Level
 
Upvote 0 Downvote
J-P, I thought you just clarified this issue with Jim in the thread about runout.

Jim clearly mentioned that "The certification exam treats this particular item as a "must" (i.e. features must be at MMC for the simulator)." Did I misunderstand something?
 
Upvote 0 Downvote
The new standard certainly gives expanded treatment to RMB and LMB, but I'm pretty sure that the ASME standard has never required that a pattern which is designated as a datum must be referenced at MMC/MMB.

Not sure what Jim was getting at in reference to the certification exam.

John-Paul Belanger
Certified Sr. GD&T Professional
Geometric Learning Systems
 
Upvote 0 Downvote
When using a pattern of holes as datum features, they must be referenced at MMC in feature control frames that reference them. Please see paragraph 4.5.8 (not 4.5.7) at page 68 of 1994 standard.

SeasonLee
 
Upvote 0 Downvote
4.5.8 states "such as a pattern of holes at MMC" rather than MUST but it does make sense. One could certainly simulate the MMB around true position of a pattern of holes but in RMB?? I just don't know how we could simulate that situation.

Dave D.
 
Upvote 0 Downvote
Right -- paragraph 4.5.8 uses the MMC idea as an example, with the phrase "such as..." but that's not the same as making it mandatory.

John-Paul Belanger
Certified Sr. GD&T Professional
Geometric Learning Systems
 
Upvote 0 Downvote
4.5.8 doesn't directly require MMC, however there are no examples of RFS, and any certification exam questions regarding datums based on patterns of features deal with it as a requirement. Again, the wording is the issue. "Voluntary Standards" can't use "shall", "must", and other such definitive statments unfortunately. If they had wanted RFS and MMC to both be considered, they would not have indicated "MMC" in the "such as" statement, and they would have illustrated more than the one concept. The guys I've talked with who helped write the standards (and the certification exam questions) didn't envision the RFS usage, possibly because of the hard gauging bias of old. Someone recognized this shortcoming and proposed the change in the '09 edition. Perhaps it's insider knowledge, but that's how it is.

PMarc illustrates the right simulation technique for physical elements. If a planar primary datum is specified before the datum based on pattern of features, a vision system could probably establish the virtual simulator, as could a CMM.

Jim Sykes, P.Eng, GDTP-S
Profile Services TecEase, Inc.
 
Upvote 0 Downvote
Interesting tidbit about the certification test, Jim, but frankly that stinks. How can they write test questions, collect over 300 bucks a head, and then ding people all because they didn't "envision" that type of usage?

Sorry, but they gotta write test questions on what the standard says, not on what they meant to say! And the '94 standard says that multiple FOS may be used as a group to establish a datum -- and then they slip in a parenthetical clause saying "such as..." That's a flimsy case on which to suddenly say MMC is required.

I know you're just the messenger, so my apologies. But if that was really their thinking, then I wonder how many other certification questions are based on what they "envisioned"?

John-Paul Belanger
Certified Sr. GD&T Professional
Geometric Learning Systems
 
Upvote 0 Downvote
Well, J-P, it's hard to come up with questions, or solutions, for something that you haven't envisioned yet, isn't it? Have you never had a consulting situation where you had to extend and develop a unique application of GD&T controls? Why didn't you envision it before hand and include it in your training materials so that it was obvious to your clients?

As for the insider-info thought, it's something that a number of us have raised at the Y14.5 meetings, and pushed to have the background/thoughts somehow documented within or external to the standard. Some successes, some disappointments.

They are working on the next certification exams (for '09) now. I know the people writing the questions. They are well experienced in GD&T and every question is highly vetted. Of course, something will be missed in retrospect.

Jim Sykes, P.Eng, GDTP-S
Profile Services TecEase, Inc.
 
Upvote 0 Downvote
Of course I have had situations where I had to extend principles to guide a client's design. But if the test writers are making absolute statements in the questioning, then I would think it matches an absolute statement in Y14.5.

I guess we'd have to see the question itself. I know they don't do T/F questions, but suppose they wrote something like: A pattern of FOS which is used as a single datum reference must be referenced at MMC (T or F?).

If they are going to mark an answer of False as incorrect, don't you think they'd at least look at the paragraph the question comes from and see the disconnect? That's all I meant.

I don't doubt the skills of the folks writing the test, and I guess I should give them a pat on the back for such work. As we've established in several threads lately, no standard is perfect, and that includes certification tests!

John-Paul Belanger
Certified Sr. GD&T Professional
Geometric Learning Systems
 
Upvote 0 Downvote
It's also in Tec-Ease's training materials that it's a requirement. Considering the sources (Krulikowski, Day, Bakos) have been part of the Y14.5 committee for a long time, it is a good reflection of the intent.

Jim Sykes, P.Eng, GDTP-S
Profile Services TecEase, Inc.
 
Upvote 0 Downvote
I apologize for using 4.5.7 that was an error. I agree 4.5.8 indicates that MMC is recommended, I won’t argue whether or not it is intended to mean must be or can be.
However, as an extended principle using a two-hole pattern as the secondary datum feature referenced at MMC there is quite a bit of functional and practical usage. I think an example of two press fit dowel pin holes being used as a secondary datum feature is a good example. This establishes two planes perpendicular to the primary datum feature.
If more than three features were used as the secondary datum feature, I can understand this creating issues, but two are fine.
There are several examples for using patterns of holes as a secondary datum feature RFS, in their various text, see links and attachments. In my opinion the end result is to correctly define the product requirement, sometimes we have to extend principles in the standard that were not spelled out explicitly.

The first example in the attachment is referenced at MMC, but the text indicates that RFS is a viable option.

You have to read the text, because there is an erroneous picture (by intent) to supplement the other illustration.

About ½ way down the page for this.


Drstrole
GDTP - Senior Level
 
Upvote 0 Downvote
I agree that 4.5.8 only describes the MMC reference but doesn't forbid RFS or LMC.

Further, the configuration described in 4.5.7.2 and shown in Fig. 4-21 opens the door to the RFS reference. This is an example of multiple features of size, referenced RFS, used to establish a datum. The two cylindrical features shown in the figure are of unequal size, but if they were equal then this would be functionally identical to a pattern referenced RFS. So I would say that the committee's case that RFS wasn't envisioned, and that datum feature pattern must be referenced at MMC, is difficult to defend.

This is not to say that referencing a datum feature pattern at RFS is problem-free. Even in the two-feature example in Fig. 4-21, the meaning is not completely clear. 4.5.7.2 states that "the datum axis is simulated by simultaneously contacting the high points of both surfaces with two coaxial cylinders", and the figure describes the "smallest pair of coaxial circumscribed cylinders". But depending on the initial conditions (i.e. starting sizes) of the coaxial cylinders, it is possible to circumscribe the surfaces in different ways and arrive at different datum axes. So the smallest pair is not uniquely defined.

The description was improved in Y14.5-2009, but the exact behavior of the simulators for patterns referenced at RFS (RMB in 2009) remains a can of worms to this day.

Evan Janeshewski

Axymetrix Quality Engineering Inc.
 
Upvote 0 Downvote
Drstrole,
Not sure where you read in the Tec-Ease tip that RFS was viable. Quite the contrary, it recommends not doing it.

I suppose that RFS could be doable for two coaxial features, but what beyond two features? What of non-coaxial holes? That's where the issue comes in. Personally, I'm glad that they improved this a bit in '09 as it's a reality that many of us work with on a design basis. It is (imho) a shortcoming of the ASME standard that the analytics for setup are not better; I believe Evan and associates are working on that to some degree in Y14.5.1, and hopefully some improvement will happen in the fixturing arena as well.

Jim Sykes, P.Eng, GDTP-S
Profile Services TecEase, Inc.
 
Upvote 0 Downvote
Status
Not open for further replies.

Similar threads

sendithard
  • Question
Inclined datum and degrees of freedom
Replies
9
Views
4K
Burunduk
Burunduk
SeasonLee
  • Question
Datum callout question 2
2 3 4
Replies
75
Views
12K
mfgenggear
mfgenggear
SDETERS
  • Locked
  • Question
Z Datum Average of 2 Points callout
Replies
8
Views
3K
mfgenggear
mfgenggear
nmine
  • Question
Datum on hole pattern
Replies
6
Views
3K
Burunduk
Burunduk
B
  • Question
Perpendicularity callout with curved datum 1
2
Replies
23
Views
5K
greenimi
greenimi

Part and Inventory Search

Sponsor

Back
Top Bottom