INTELLIGENT WORK FORUMS
FOR ENGINEERING PROFESSIONALS

Log In

Come Join Us!

Are you an
Engineering professional?
Join Eng-Tips Forums!
  • Talk With Other Members
  • Be Notified Of Responses
    To Your Posts
  • Keyword Search
  • One-Click Access To Your
    Favorite Forums
  • Automated Signatures
    On Your Posts
  • Best Of All, It's Free!

*Eng-Tips's functionality depends on members receiving e-mail. By joining you are opting in to receive e-mail.

Posting Guidelines

Promoting, selling, recruiting, coursework and thesis posting is forbidden.

Jobs

Tolerance of Positoin - feature positioned to itself ???

Tolerance of Positoin - feature positioned to itself ???

(OP)
Please review attached drawing for comment on the tolerance of position callouts circled.

1. I do not understand how a FCF can reference the feature positioned to itself in a datum reference block. (e.g. datum A is positioned with reference to itself in A-B)

2.  I believe the location tolerance and the basic dimension  relavent to it should have the same number of decimal places. (Not shown on the drawing) (e.g. if tolerance in FCF is .XX then its basic dimensions should be XXX.XX NOT XXX) Is this in the standard?

DesignBiz

"Quality is in the details"
 

RE: Tolerance of Positoin - feature positioned to itself ???

In your example, the drawing is establishing an A-B axis through the center of the tube, using batums A and B as the datum features for establishing the axis. You're not calling the datum to itself, as much as you are making it part of the A-B axis.

 Looking at the bends of this part, I would not use this axial dimensioning method. I would use profile-of-a-surface instead, to datums A and B, using the same A and B datum features.

Regarding the number of places, since this is metric and unilateral dimensioning, it is IAW Paragragh 2.3.1(a) of  Y14.5-1994.

RE: Tolerance of Positoin - feature positioned to itself ???

1. I've never been a fan of FCF's that reference a datum feature that includes the considered feature.  This method is not described in Y14.5.  I have seen it in GD&T textbooks, but only for coaxial features and with MMC references on everything.

The presumed intent of the positional tolerances on features A and B is to control their locations and orientations "relative to each other" without making either one the datum feature.  Which is fine.  This will allow them to function well together as multiple datum feature A-B.

To me, controlling A and B relative to each other doesn't require referencing the A-B datum feature.  If the datum reference were left off, then the rule of simultaneous requirements would tie A and B to each other.

This is all I have time for today.  More tomorrow.

2. This one is in the standard.  The same number of decimal places are not required for millimeter tolerances, only for inch tolerances.  See 2.3.1 (d) and 2.3.2 (d).

Evan Janeshewski

Axymetrix Quality Engineering Inc.
www.axymetrix.ca

RE: Tolerance of Positoin - feature positioned to itself ???

An example of this situation is on page 193 of the standard in fig. 6-51 with the exception that circular and total runout are used rather than positional. We have the example but no confirming verbiage.

Unfortunately, this is one of the areas in the standard that just does not make any sense. How could one reference a datum that is not already established??

Using the example that I mentioned, one could extrapolate that positional could also be used as shown in your example. I don't like it personally but it does look somewhat legal.

 

Dave D.
www.qmsi.ca

RE: Tolerance of Positoin - feature positioned to itself ???

dingy2,

   The example on fig 6-51 makes sense to me, and I can see how I would fixture it.  In the absence of a long, round feature, two short, concentric, widely separated features make a good centre axis.  I have done this on complex shafts.

   The OP's drawing makes no sense to me.  I cannot see how I would fixture and inspect it.  If only one end had a positional tolerance, it would start to work for me.

   I observe that datums A and B locate the part completely.  I do not see a purpose for datum C, unless it controls the part's flexibility somehow.  I would call up datums A and B separately in the FCF, and define the datum B feature WRT datum A.

               JHG

RE: Tolerance of Positoin - feature positioned to itself ???

(OP)
I do appreciate the direction to the section for the decimal place question and to pg 193 by Dingy2.

I would like to make clear this is NOT my idea of dimensioning an exhaust tube using the standard.
It is the companies standard template for dimensioning tubes. hammer I dont like it at all personally.soapbox

Like Dingy2 I had the same question. How could one reference a datum that is not already established?
Seems like a paradox to me. ponder

I can appreciate the logic Ron and Evan apply to make sense of the callout, however this is not a rotating
shaft that would make sense to create a single datum using both end diameters or 2 bearing diameters.  This is
an exhaust tube and could be done more more simply uses 3 datum references in a more conventional DRF.

Thanks for all of the commentaries !
 

DesignBiz

"Quality is in the details"
 

RE: Tolerance of Positoin - feature positioned to itself ???

(OP)
ewh,

The thread you refer to is actually one I started. It is even more bizarre to me when the 2 features that are used to create a single datum are at compound angles. Our tube ends are fairly often at compound angles to each other.

We beat that horse, til about dead. deadhorse

Thanks for your input.  

DesignBiz stpatrick2

"Quality is in the details"
 

RE: Tolerance of Positoin - feature positioned to itself ???

Sorry about that, didn't check the details.  I just knew that a similar issue had been addressed.
Yes, that one was beat to death.

"Good to know you got shoes to wear when you find the floor." - Robert Hunter
 

RE: Tolerance of Positoin - feature positioned to itself ???

Quite honestly, I don't see a problem with the OP drawing.  Per the other thread, once you've got your datums established, you can locate the position of any feature of size.  Essentially, you know where in space the feature of size is supposed to be, and you just verify it as you would normally.  If the original question was really how do you verify position of Datum-A back to itself, you aren't.  You are verifying the position of datum-feature-A to Datum A-B.  Two entirely separate things.  Review the definitions of Datum and Datum Feature.

Jim Sykes, P.Eng, GDTP-S
Profile Services  www.profileservices.ca
TecEase, Inc.  www.tec-ease.com

RE: Tolerance of Positoin - feature positioned to itself ???

Well said! Jim,

I had started a response to the http://www.eng-tips.com/viewthread.cfm?qid=235809 thread but suspended it because I couldn't say with as few words as well as Evan does what I thought.

It is function!!! that must drive the selection of datum features to predict whether other features located, oriented, or formed well enough to guarantee that they perform as designed. The two ends of the tube (I suspect) are certainly the business ends of the primary datum feature... if the mating parts are more rigid than the tube itself – then theoretically they stop all six degrees-of-freedom since they are nominally not coaxial... practically rotation about a line intersecting the midpoints of each A-B and translation along that line would be a problem for inspection R&R unless some "procedure" or  "over constraint via additionally defined features"  remedies the variation.

Sidebar: If you think that this stuff is confusing now wait until they allow restrictions on the degrees-of-freedom" that datum features datum features may stop!

Paul
 

RE: Tolerance of Positoin - feature positioned to itself ???

(OP)

I have seen both threads which I have initiated go thru some fairly deep interpretations relative to
the average designer and engineers basic knowledge of ASME Y14.5m 1994.

In the case of the exhaust tube example (both threads) I don't see the need for adding the complexity
of the A-B multiple feature datum. To me this tube as many tubes is a relative non-critical component.
The two ends connect to other tubes to zig and zag from 'point a' to 'point b' as a conduit for exhaust flow.
These tubes can easily be a quarter of an inch off and they fit. I have been told here at the company, that the A-B callout developed because "both ends" of the tube run are equally important. That really isn't an effective argument from my point of view. In the case of a gear that may have a spline that connects to a rotating shaft and external teeth that mesh with another gear; which of those features is not equally as important? My point is that on an average component it is common to have many features that are equally important in the overall mounting and function of the part.

 From my point of view this dimensioning could be done with a primary datum using an end of the tube diameter
(Preferably the one that the xy bend coordinates originate from) and with a secondary datum as the opposite end of
the tube. A simple vertical line between the A and B vs. the "dash" between them in this FCF could help an awful lot more people involved in the design, manufacture and inspection of the part,  understand the drawing AND still do the job. After all, isn't this the main objective?
I don't argue using the proper callout regardless of it being a little more difficult to understand, if the part function warrants it. I just don't see it in this case.

What I have found here (this company) is that when I started asking questions about the tube callouts, is that absolutely NO ONE knew what the callout was supposed to mean nor how it was inspected. It gets deeper. I ask the checker what this callout means to him and he said whatever the GD&T group put on the drawing then it was okay with him. The GD&T group, consists of folks
that are not very good at even getting the basics correct. Basically straight out of college with a 40 hour GDT training class.  I asked the supplier how they inspect the part and there has been no response for a couple of months. I ask how the 1st article inspection was achieved and no one knows... back to the supplier.... they should know, and absolutely no response to my
inquiries. It's actually baffling from an  "old school' point of view to get a part designed, manufactured and inspected.

Bottom line is that no one here that knows what the callout means well enough to explain it to anyone else
and therefore pretty much it is "disregarded entirely".  Reminds me of the story about the emperor without any cloths.

The company has an on-site GDT consultant who continually had been frustrated trying to explain it to me (the thread(s) I posted) and ended up with a response, "it's in the computer".  This totally is disheartening to me. I have used computers (CAD) to do design work since 1977 and I KNOW the answer is with the designers and engineers.... not in the computer!

I was on the board before computers came around as a tool and I do like GD&T, however there is an old design rule ," keep it simple"; the ole "KIS"rule, and I won't even use the second "S".

In light of this I dont see a need to go beyond the fundamentals of GD&T at this facility, for this callout. I have seen suppliers automatically add $10k
to quote a part simply because it was dimensioned with GD&T. I dont want to give up on the standard, however as in the case of the  tube run, I think common sense and simplicity should prevail so that at least some people can understand the drawing. To me it is better to use a callout that some understand if the callout is appropriate, rather a more sophisticated callout that no one does.

I really do appreciate the comments made especially by the senior certified guys I find it extremely interesting as to the extrapolations or "extensions" that Evan, Jim, Norm, Paul and others have proposed and hope to hear more about it from this website.

DesignBiz stpatrick2

"Quality is in the details"
 

RE: Tolerance of Positoin - feature positioned to itself ???

DesignBiz's last comment brings up a question that I have been thinking of. What would be the difference in interpetation of using [A]B]C] as he suggests and using [A-B]C] as shown on the drawing?

Peter Stockhausen
Senior Design Analyst (Checker)
Infotech Aerospace Services

RE: Tolerance of Positoin - feature positioned to itself ???

I have a question:

How does one go about establishing an AXIS (LINE) between a set of skewed lines?  An axis is usually determined by 2 points.

If I am reading the drawing correctly the datum features A and B are diameters resulting in a pair of datum axes.

RE: Tolerance of Positoin - feature positioned to itself ???

A star for that, DesignBiz.
The explanations given here for this type of GD&T are good and valid; however, if a simpler scheme can be employed and does not effect the end use of the part, I'm all for it.  If you happen to have studied GD&T extensively, this type of scheme will seem the best, but a majority of people who have to interpret these drawings will draw a blank when they see it.  If they can't interpret it, how can they achieve it?  These are non-critical parts and a more conventional, uniformly interpreted approach seems to me to be a better way to go.  

"Good to know you got shoes to wear when you find the floor." - Robert Hunter
 

RE: Tolerance of Positoin - feature positioned to itself ???

This is also what bothered me about this drawing, ringster.

We generate some compound bent tube parts. On our tube drawings, all our configuration changes are set up to X,Y,Z basic dimensions in a table (X,Y,Z being 0,0,0 at datum A end)for tube bending machines. We put a datum A and B respectively on the two ends, and everything in between is to a loose profile tolerance.

RE: Tolerance of Positoin - feature positioned to itself ???

(OP)
Ringster,

This posted drawing was in ref to a feature positioned referencing itself in the FCF, however it is the same drawing callout posted on a prior thread debating the A-B; from cylindrical features of size at compound angles to each other; which would be the 2 axes that I believe you are referring too.
This was my confusion from the beginning of the other thread that a single resulting axis from these two at compound angles to each other is geometrically impossible.

I have accepted the "extensions" of the standard presented by Axym, MechNorth, and others, however it is not readily discernible with average knowledge of the standard, and still gives me rogue thoughts that bring me back to how is this a single axis from 2 at compound angles, or skewed  lines as you put it? It can not be and therefore the only sense that can be made of it is by the "extension" that really locks down all DOF's in one datum reference block and is NOT a 'single datum' itself.

Still this does not set well with me.  

DesignBiz stpatrick2

"Quality is in the details"
 

RE: Tolerance of Positoin - feature positioned to itself ???

Designbiz,

By what authority are these EXENSIONS legitimized. I don't quite understand the how this would be 'discernable' to anyone.  Average or especially above average knowlege of the STD.

Perhaps the proper rout to take would be to present for the @015 version of ASME Y14.5 and see if the committee finds it acceptable.  Too late for the 2009.

RE: Tolerance of Positoin - feature positioned to itself ???

I haven't finished beating the horse in the other thread!  There are still many points to make regarding the A-B issue.

I'm all for keeping things simple where possible, but there are times when the functional reality is more complex and must be acknowledged.  This is one of them.

If they changed the callout from A-B to A|B in an attempt to "simplify" things for everyone, it would be a disaster.  Why?  Because the inspection department would suddenly start rejecting all sorts of functionally acceptable parts (assuming that the parts are being inspected, and inspected correctly).  Why?  Because A|B doesn't represent function!  The part isn't fixtured on one end first and then the other.

To create the A|B datum reference frame correctly, the inspector would need to constrain two rotations and two translations using Datum Feature A, a cylinder which is roughly 5% of the overall length of the part.  Small deviations in the part would be magnified into very large apparent deviations.  A lot (most?) of the parts would not pass this artificially tight requirement.

Evan Janeshewski

Axymetrix Quality Engineering Inc.
www.axymetrix.ca

RE: Tolerance of Positoin - feature positioned to itself ???

Compound datums give me the heebiest of jeebies.  Yes, there are rare instances where they are appropriate.  The majority of the time, compound datums signal that the drafter did not know what he really wanted and does not fully understand how datums are defined and used.

RE: Tolerance of Positoin - feature positioned to itself ???

<start sarcasm>
OK, so I guess they just shut the line down and not ship parts until everyone involved understands this issue.
<end sarcasm>
[quote]assuming that the parts are being inspected, and inspected correctly[//quote]
Isn't that the situation that exists currently with no one able to explain the callouts?

"Good to know you got shoes to wear when you find the floor." - Robert Hunter
 

RE: Tolerance of Positoin - feature positioned to itself ???

If they fit at N/A and are interchngeable, you are home free.

If not, back to the drawing, excuse me, monitor screen.

Have you consicered Model Data Set definition for the parts..

RE: Tolerance of Positoin - feature positioned to itself ???

(OP)
Ringster,

The "extensions" I refer to were presented by the senior GDT certified contributors on this site. (Ref other thread.  As of now I do not know how to link that thread to this reply)

I agree that the authority for acceptance of an extension to the standard's fundamental principles is based on the individuals credentials presenting the extension.
What  may be an extension to one, may not make the muster for another.

Part inspection?  Who knows if they are being inspected "correctly" !!! Never have seen anything like this before.   Even the tubing in-house standard makes a comment regarding the amount of scrap and cut and weld required for tubes. I wonder if the lack of proper inspection could stop much of the scrap?

So as hard is it may be to imagine,   no one here can say for sure how the parts are inspected.  Somehow they are inspected and accepted and it is acknowledged that there will be considerable scrap.  
I have worked on many tubes for gas turbine flight and power generation engines.  We did not operate
like this at those companies.
Sorry,  I am not making this up and it is a Fortune 500 company.  Enough said?
 

DesignBiz stpatrick2

"Quality is in the details"
 

RE: Tolerance of Positoin - feature positioned to itself ???

It looks like we're getting into some philosophical issues here, like accommodating the limited GD&T knowledge of the people interpreting the drawings.  It reminds me of something I saw on one of those "funny rules of life" lists:

"An easily understood, workable falsehood is more useful than a complex, incomprehensible truth".  

I think that part of the problem lies in the reliance on the GD&T drawing itself.  I often encourage designers to prepare other documents to go along with their GD&T drawings - sketches, "means this" diagrams, written explanations, etc.  For the sole purpose of providing another means of communicating the intent for those who are not GD&T experts.  This can be particularly useful for datum referencing.  As we've seen from the "2 skewed cylinders" example, the GD&T can be complex for a relatively simple functional situation.  A quick sketch of how a real part would be held in a fixture would probably help a lot in this case.

Another part of the problem (IMHO) is the common requirement to make GD&T drawings that people with average GD&T knowledge (or less) can understand.  I see this all the time, where a designer has had to "dumb down" the GD&T because the downstream folks couldn't understand it.  This is more a management issue than anything - someone bending tubes on the shop floor shouldn't be expected to be a GD&T expert as well.  There is a real need for more "manufacturing engineering" people that can decode a GD&T drawing and convert it into manufacturing and inspection procedures that others can follow.  But it seems that many companies don't have this luxury.

Evan Janeshewski

Axymetrix Quality Engineering Inc.
www.axymetrix.ca

RE: Tolerance of Positoin - feature positioned to itself ???

I totally agree, Evan, but that is the reality of the world many of us work in.  I like the idea of generating additional documentation explaining any methodology used that might not be apparent just by following the drawing.  In many locations (here for example) the documentation system may not be set up to accomodate this, in which case you could add it to the drawing in some manner (general note, additional sheet, etc).

"Good to know you got shoes to wear when you find the floor." - Robert Hunter
 

RE: Tolerance of Positoin - feature positioned to itself ???

axym,

Quote:

"An easily understood, workable falsehood is more useful than a complex, incomprehensible truth".

   There was an article in the Skeptical Inquirer by Isaac Asimov entitled "The Relativity of Wrong".  We make assumptions about stuff.  Science comes up with a better, more accurate model.  This usually does not change the fact that the original assumption was an adequate approximation for people trying to do practical stuff.  For example, over short distances, the earth is approximately flat.  You do not account for the curvature when you build a floor, or draw a city map.

   Being confused and wrong is something different, entirely.  You did day "workable", but I think "falsehood" is a little strong.

               JHG

RE: Tolerance of Positoin - feature positioned to itself ???

I had a question I posted on this thread earlier:

What would be the difference in interpretation of using
[A]B]C] as he suggests and using [A-B]C] as shown on the drawing?

I read the spec over and I may have answered my own question.

To determine the Datum Reference Frame on most parts [A]B]C] your would pick up 3 points on A, 2 points on B and 1 point on C. For the drawing referenced in the original posting, this would be inserting the tube in a datum reference for A, clocking the tube by referencing B and make contact on C.

For the [A-B]C] datum call-out I would insert the tube in
a datum reference for A and one for B at about the same time, then make contact on C. The first five points would be distributed among datums A and B.

As I see it a [A-B]C] datum call-out could be thought of as being met if either [A]B]C] or [B]A]C] is satisfied.
 

Peter Stockhausen
Senior Design Analyst (Checker)
Infotech Aerospace Services

RE: Tolerance of Positoin - feature positioned to itself ???

PeterStock,

   The weird thing about this discussion is I claimed that I have used the datum [A-B] a couple of times, and since then I have used it twice.  In both cases, it was for a flat surface broken up by a slot.

   I still don't see the point of datum [A-B][C] on the OP's drawing.  This is not an appropriate application for this.  If you call up everything with respect to datums [A][B], you would have located the part completely, and accounted for some of the flexibility.  Even if the two datums are of equal importance, you need to fixture one of them first.  All datum C can do is control more flexibility.

   The sloppyness of the parts does not matter either.  Call up loose tolerances.   

               JHG

RE: Tolerance of Positoin - feature positioned to itself ???

(OP)
I really like this website!  It is very very interesting.
Thank you drawoh ! My point from the beginning. In this instance the A-B callout is unnecessary from my point of view. As you mention in a surface broken by a slot or a rotating shaft that may use 2 coaxial bearing surfaces to define a single axis, all make perfect sense to me. Multiple features used to define a single datum. What troubles me with the multiple datum concept as an extension of the DRF definition is just that, multiple datums in one datum reference block. It is has been said that we can go back to the 1982 standard to carry over these principles (I don't have a 1982 standard). It is confusing at best and in the case of symmetry if you went back to the 1973 standard it is a different meaning. Can we really rely on a previous standard to interpret the current one? I believe I understand the general concept that others have presented as to how the A-B constrains all the DOFs, however it does not meet the definitions in the 1994 standard. (e.g. each datum ref block in a FCF represents a single datum). I simply say documentation describing the new meaning in this case must accompany the callout declaring that the A-B is more than a single datum. I know some of you senior certified guys maybe frustrated with me, but that's my story and I'm stickin' to it thumbsup2 !!!



Can anyone tell me what the difference of an A-B vs. A|B callout fixture setup would be? and what advantage there would be in this case (2 skewed cylindrical axes) of one over another?
 

DesignBiz stpatrick2

"Quality is in the details"
 

RE: Tolerance of Positoin - feature positioned to itself ???

My opinion: as long as A aand B remain as shown, an AXIS of a cylinder at each end, neither A_B nor A/B will work.

RE: Tolerance of Positoin - feature positioned to itself ???

Disregard my previous post.  I clicked the "Submit Post" button by accident.  It looks like only one file can be uploaded in each post.

I've created a fictitious part that is loosely based on the one in the original post.  Please see the file named nominal.jpg.  Datum features A and B are the hatched areas at each end of the part.  The nominal centerline is also shown.

The other 3 files depict an imperfect actual part, being fixtured in 3 different ways according to 3 different datum referencing schemes.  Simulators A and B are basically located relative to each other, and the jaws are symmetric about the nominal centerlines at the datum areas.  I deliberately left out datum feature C, as it is not necessary since A and B constrain all 6 degrees of freedom.  I also deliberately left out anything to do with datums and datum reference frames - we can deal with that separately.

The file A-B.jpg shows the part in the controversial A-B scheme.  The idea is that both simulators have been tightened simultaneously and at the same rate.  Neither simulator aligns fully to its datum feature.  The part has problems in the middle, but does not have really large deviations anywhere.

The file AB.jpg shows the part location if A was referenced as primary and B secondary.  The A simulator has been tightened first, fully aligning with its datum feature.  Simulator B was tightened second, and only touches its datum feature on one side.  The position of the part is very different from that of the A-B scheme.  The requirement to align to the relatively short datum feature magnifies the deviations at the other end.

The file BA.jpg shows the situation with B as primary and A as secondary.  The position of the part is different again, with the deviations magnified the other way.

One key assumption in all of this is that the part remains rigid and is not deformed to conform to the fixture.  This is the default condition in Y14.5, and applies unless the drawing has a note describing the restraint condition.  Based on some of the recent posts, the real parts are most likely non-rigid and are being deformed to some degree when fixtured.  This might explain some of the assertions that there is little or no difference between AB, BA and A-B - the parts are being deformed the same way in each case.  However, if the parts were being inspected in the free state using a CMM, these datum referencing issues would be very significant.

Evan Janeshewski

Axymetrix Quality Engineering Inc.
www.axymetrix.ca

RE: Tolerance of Positoin - feature positioned to itself ???

AXYM, you can upload more than one file at a time.  However instead of using "click here to insert your file" or whatever it says, instead cut and paste the different links into you post.

KENAT,

Have you reminded yourself of FAQ731-376: Eng-Tips.com Forum Policies recently, or taken a look at posting policies: http://eng-tips.com/market.cfm?

RE: Tolerance of Positoin - feature positioned to itself ???

(OP)
Thank you Evan,

Your time to respond is appreciated. Most times for me,
graphics supporting the words can make things more clear.  

DesignBiz stpatrick2

"Quality is in the details"
 

RE: Tolerance of Positoin - feature positioned to itself ???

AXYM,

If Im reading the A-B jpg file correctly, it appears that it lacking in definition.  To establish the simulators A and B, there must be a relationship between the two of them.  As I view the sketch there is none.  Would you not have to have a distance somehow defining the separation of the two plus an angle?  Can you somehow clarify the sketch, to make the A-B DOABLE?

( my opinion; can't relate the 2 axes}

RE: Tolerance of Positoin - feature positioned to itself ???

Good reply, Evan.  Ringster, you would indeed need basic dimensions (linear distances & angles) to locate the two datum fetures wrt each other.  That establishes the theoretically perfect relationship between the datum simulators as well.

Jim Sykes, P.Eng, GDTP-S
Profile Services  www.profileservices.ca
TecEase, Inc.  www.tec-ease.com

RE: Tolerance of Positoin - feature positioned to itself ???

KENAT,

I thought I had seen posts with multiple links!  Thanks for the tip.

DesignBiz,

There are a lot of things in GD&T that I find hard to explain in words.  This was obviously one of them - my earlier attempts to explain the difference between A-B and AB in words didn't get there.  A diagram always seems to cut through.  In fact, I usually learn something when making the diagram.

Ringster,

You're right, there is a basic relationship between the two simulators.  They have a basic angle between them and a basic distance apart, and these were not shown on the diagram.  I don't actually know what they are - the simulator centerlines were just extracted from the nominal CAD geometry of the part.  

OK, I've annotated some dimensions on the nominal part that should allow you to reproduce the relationship of the simulators.  Please forgive the drawing - I'm no detailer and the software I'm using isn't really designed for drafting.  The dimensions and angles are of course supposed to be basic, despite the lack of boxes.  The datum areas have a length of 3.

Evan Janeshewski

Axymetrix Quality Engineering Inc.
www.axymetrix.ca

RE: Tolerance of Positoin - feature positioned to itself ???

Jim,

I moved south from MO years ago.  I wonder if you could illustrate how with JUST THE TWO datums A and B as shown you can establish a relationship between the two?

RE: Tolerance of Positoin - feature positioned to itself ???

Ringster,

I think that we might be running into some terminology problems here.  When you say "datums A and B as shown", what exactly do you mean?  Which of my diagrams are you referring to?

Evan Janeshewski

Axymetrix Quality Engineering Inc.
www.axymetrix.ca

RE: Tolerance of Positoin - feature positioned to itself ???

axym,

   My understanding is that the OP's tubes are flexible.  It is possible to fixture them to centre on Datums A and B.  It is likely that the final installation actually works that way, making the envelope dimensions at inspection, very relevant.

   If the tube is very flexible, it may be necessary to have a datum C to further control the flexibility.  Read ASME Y14.5M-1994, Section 6.8 on Free State Variation.

   If the tube is not flexible and it were my drawing, I would use the centre of A as my primary datum, the end face of A as my secondary datum.  I would then pick up the sides of B to use as a clocking feature, and my tertiary datum.

               JHG

RE: Tolerance of Positoin - feature positioned to itself ???

Ringster, who were you with up there / which ops?

I've done a "quick" drawing of what I'm talking about.  I didn't put the basic dimensions on the fixture views 'cause I really hate annotating to sketch intersection points with SolidWorks.

http://www.profileservices.ca/files/tidbits/thd1103_237732.pdf

Personally, I'm not a fan of the A-B compound datum for this kind of workpiece.  I'd prefer to use moving datum targets.  The A-B compound datum is legal though, if that's the design intent.
 

Jim Sykes, P.Eng, GDTP-S
Profile Services  www.profileservices.ca
TecEase, Inc.  www.tec-ease.com

RE: Tolerance of Positoin - feature positioned to itself ???

(OP)
FYI,

The original tube dwg posted is an example of a rigid metal exhaust tube and how these rigid tubes (typ .045"-.095" wall)are dimensioned at this company.

DesignBiz stpatrick2

"Quality is in the details"
 

RE: Tolerance of Positoin - feature positioned to itself ???

DesignBiz,

   If the tubes are rigid, then one of the possible errors is the angle between the two datums.  MechNorth's fixture will not work properly unless the tube is flexible enough to bend to the fixture geometry.  

   An alternative is to assume an MMC condition of maximum tube size sufficient to fit inaccurately bent tubes into the fixture.  This is silly though, since tube ODs are fairly accurate.  There is no way for a supplier to take advantage of the extra slop.  The real tube will be located with play.  

   You need to go with my datum suggesting above, or something similar.

               JHG

RE: Tolerance of Positoin - feature positioned to itself ???

(OP)
DrawOh,

I whole heartily agree with you and tried to convince some folks around here for the last several months. The template for tubes here remains the same, however I have convinced my project engineer that a callout with a DRF as you have also proposed is the way to go. He is okay with it.

Thanks again to all for your inputs !

DesignBiz stpatrick2

"Quality is in the details"
 

RE: Tolerance of Positoin - feature positioned to itself ???

Drawoh, you seem to be assuming an intimate fit on both datum features.  My graphic shows the basic geometries, not "actual" parts.  I can assure you that this type of fixture is used, and does work as shown to me by formed-tubing suppliers.  The workpieces are sort of "wedged" (for want of a better term) into the chucks for a best fit as opposed to an intimate fit.  An intimate fit on both datum features would indeed require flexibility in the workpiece.

Jim Sykes, P.Eng, GDTP-S
Profile Services  www.profileservices.ca
TecEase, Inc.  www.tec-ease.com

RE: Tolerance of Positoin - feature positioned to itself ???

MechNorth,

   That is sort of the MMC approach I described above.  The datum holes are large enough to accomodate some angle error.

   I would not want to call this up on my drawing.  I cannot visualize what would pass and what would fail inspection.  How do I account for the oversized fixture?

               JHG

RE: Tolerance of Positoin - feature positioned to itself ???

Drawoh, the problem with MMC is that it allows the workpiece to float within the "slop" of the clearance so that it is only "held" when datum feature reaches its virtual condition.  As you indicate, it would be difficult to validate the workpiece.  For RFS, the simulator (jaws in my graphic) close down to actually hold the workpiece.

Jim Sykes, P.Eng, GDTP-S
Profile Services  www.profileservices.ca
TecEase, Inc.  www.tec-ease.com

RE: Tolerance of Positoin - feature positioned to itself ???

DesignBiz,

I'm can't help but feel a bit disappointed that you've convinced your project engineer that the AB scheme is the way to go.  Based on the descriptions of the part and its function, the AB scheme specifies DOF constraints that are very different from how a real part would be held during assembly.  If the parts are constrained properly to the AB scheme during inspection, the results will disagree wildly with fitness for use.  The surface profile will appear to be way off when it's really not.  In my diagrams from yesterday, note how good the part looks (compared to the nominal centerline) in the A-B scheme.  And how much worse it looks in the AB scheme, and how horrendous it looks in the BA scheme.  It's exactly the same part geometry!  This is what you can expect to see.

I hope that things work out okay in spite of this.  Sometimes, the right design intent somehow gets through in spite of what the drawing actuallly specifies.  Usually through some form of tribal knowledge, where the vendor or inspection department knows what to do to make it come out right.  This can put the quality department in a precarious position, however, when applying the oversimplified GD&T correctly during inspection will result in massive nonconformances and the whole operation grinding to a halt.

Evan Janeshewski

Axymetrix Quality Engineering Inc.
www.axymetrix.ca

RE: Tolerance of Positoin - feature positioned to itself ???

(OP)
Evan,

Could you agree that the meaning of the A-B concept is not what you will find in the standard? I am referring to the definition of a DRF in section 4; the definition of a datum (1.3.3); and the instruction that a datum reference block in a FCF references a "single datum" (3.4.2 and 3.4.3).

Maybe I have a part in an assembly and later a different model assembly with a newer and slightly physically different cover plate than the original assembly is developed. This new cover plate is superior in quality, cost, appearance, durability than the original, however with the physical differences it cannot be retrofitted to the original model without modifications. Maybe I need to drill a couple of new holes in the mating part, open up an opening that the new cover plate has a tab that fits into it and deepen a well the plate lies in. No reasonable person in engineering would say it is simply good enough to replace the new cover plate in the old assembly, UNLESS there is proper documentation to modify the old assembly to receive it.

This may not be comparable to you but it is to me. My resistance to the A-B concept is primarily based on the unwillingness at this facility to document an amendment as to what the callout means. I would disagree with you (in absence of amending documentation) that my position is to use the simplest and not the best callout for these tubes. My position is that without the amendment this A-B callout defies the standard as written. (Wonder if the 2009 standard addresses this callout?)

I do like to read what you have to write. It seems as if you are passionate about your positions relative to the standard. I respect that and will say that you do have me interested in investigating the differences for the 2 fixturing setups. For a hard tube that is usually formed by starting at one end and proceeding thru the bends until the end is arrived at, I don't believe that if you held the tube at both ends vs. holding it at one end and clocking with another feature that the tube would pass inspection based on how it is fixtured. If it was formed within dimensional requirements it will fit and function as designed regardless of how it is fixtured.

Don't feel too disappointed Evan, I am interested in reading what you write and give consideration to it.


MechNorh,

This tube is rigid and nowhere on the drawing or in it's function are we supposed to constrain the part. It is obviously connected at both ends. This isn't like a spring where one expects to use some force to position it.
No one is suggesting that it is plausible to force the part into a fixture are they? I didn't believe the standard allows for restraining a part for inspection unless noted on the drawing. Correct?
 

DesignBiz stpatrick2

"Quality is in the details"
 

RE: Tolerance of Positoin - feature positioned to itself ???

I may be inserting 'foot into mouth', but I DO believe that 2 skewed lines (axes) cannot by themselves be related to one another.  To establish a relationship there must be another feature/s added that will provide a point on one or both axes.

 

RE: Tolerance of Positoin - feature positioned to itself ???

DesignBiz,
Wording is again important here.  To constrain something is to remove its degrees of freedom, which you do with the datums.  To restrain something is to hold it in place, typically by applying forces at specified locations other than at the datum features (see 6.8).  If you are specifying datum features of size at RFS, then you most certainly are getting both constraint & restraint.  A set of jaws, for example, closes down on each end of the tube as best it can given the basic relationship between the datum features.  The workpiece cannot move (i.e. it is restrained) and if the pipe ends are skewed wrt each other, then you get the part fully constrained as well.  Try it physically if you are still dubious; this is not merely an intellectual exercise because real people are doing it with real parts in real production facilities.  The real challenge should be reconciling your belief system to the physical reality, not whether or not it is legal.  If you are going to apply GD&T based on the manufacturing method alone, then you miss the raison d'etre for GD&T which is to communicate design intent.

Jim Sykes, P.Eng, GDTP-S
Profile Services  www.profileservices.ca
TecEase, Inc.  www.tec-ease.com

RE: Tolerance of Positoin - feature positioned to itself ???

DesignBiz,

The cover plate analogy is a bit of a stretch, but I think I see your point.  If the meaning of the A-B callout may not be documented and will be misunderstood, then it isn't of much value in your organization.

Again I applaud your attempts to evaluate things in terms of the "rule of law" stated in the Y14.5 standard.  

To paraphrase what Jim said, one should base the datum referencing method on the physical reality.  In most cases, the physical reality can be closely approximated by the applying the tools provided in Y14.5.  I some cases, it cannot.  In the case of the bent tube, the physical reality is that the tube is fully constrained by two skewed cylinders without one taking DOF precedence over the other.  I believe we've established that as a fact.  But Y14.5 does not directly address this particular situation.

So what do we do?  One option is to make simplifying assumptions about the situation in order to idealize it into something that Y14.5 does deal with.  In this case we could make the assumption that the two ends of the tube are oriented and located very accurately relative to each other.  This would mean that the tube would end up in approximately the same location whether it was fixtured on both ends simultaneously or on one end first.  This would allow us to use the conventional AB scheme.  Is this a close (or even decent) approximation of the physical reality?  In the case of the bent tubes I would say no, because of the variation you described and the projection effects that I have tried to illustrate.  You say yes, and you have direct knowledge of the parts and the process.  So we can agree to disagree over this.

Another option is to take a Y14.5 tool and apply it in a way that is not described.  In this case, applying the multiple datum feature concept is one possiblility.  Others might be the pattern concept and the mathematically defined surface concept.

Evan Janeshewski

Axymetrix Quality Engineering Inc.
www.axymetrix.ca

RE: Tolerance of Positoin - feature positioned to itself ???

Possible solution to the OP?

If the end surfaces were designated as datum features, D on the end of dia B, and C on the end of dia A, as it is now.
 
We could use a DRF B[B-D[A-C[.  This would provide an axis, B, a plane normal to B, and a point at the intersection of a and D.
 
In the world of Geometrics, B might be related to the axis of the earth, C the equator and D the Observatory at Greenwich England.  All dimensions could then be related to this DRF.

??
 
 
 

RE: Tolerance of Positoin - feature positioned to itself ???

DesignBiz,

You had some questions on the A-B concept relating to datums, datum reference frames, and their definitions in the standard.  I've been putting off answering them because what I'm going to say might be taken the wrong way if I'm not careful.  Or it might portray me as someone worthy of ignoring in the future.  Either of these is probably still likely even if I am careful!

Relating the "two skewed cylinders A-B concept" to datums is where the problems start for me.  Here's the definition from 1.3.3:

Datum.  A theoretically exact point, axis, or plane derived from the true geometric counterpart of a specified datum feature.  A datum is the origin from which the location or geometric characteristics of features of a part are established.

Presumably, the purpose of datums is to provide geometry to orient and locate a coordinate system relative to.  The idea is that the datums are extracted from the simulators and then the DRF is constrained to the datums in the specified order of precedence.  This can work well for simple datum feature types:

Planar datum feature: datum plane
Cylindrical datum feature: datum axis
Spherical datum feature: datum point
Width datum feature: datum centerplane
Coplanar surface datum feature: datum plane
Coaxial diameter datum feature: datum axis

In certain combinations of these simple datum features, a DRF can be constructed on the datums in a fairly obvious way.  Three orthogonal planar surfaces (figs 4-2 and 4-3) is one example.  Primary planar surface - secondary cylinder - tertiary slot (figs 4-6 and 4-7) is another.  Primary planar surface - secondary cylinder - tertiary cylinder (figs 4-8 and 4-9) is another.  Y14.5 and a lot of GD&T textbooks tend to illustrate DRF construction using these simple cases.  On the plus side, a good majority of part interfaces are covered by these cases.  On the minus side, these are very special cases with unique simplifying properties that do not exist in more complex (but still very common) situations.  Many important effects and complexities are masked.

The explanations in Y14.5 are also full of nasty pitfalls.  There are several statements and depictions in Chapter 4 that are misleading, not generally true, or outright incorrect.  The use of certain terminology is also imprecise and inconsistent and, IMHO, has led to a huge amount of confusion over the years.  

One example, that I've mentioned before in another thread, is the diagram on the front cover.  The lines labeled "datum axis" are not datum axes, they're coordinate axes of the DRF.  The point labeled "datum point" isn't a datum point, it's the origin of the DRF.  The planes labeled "datum planes" are datum planes, but they're also coordinate planes of the DRF in this special case of 3 orthogonal planar datum features.  The datum planes are also labeled as origins of measurement, which is incorrect.  The coordinate planes of the DRF are the origins of measurement, and again they happen to coincide with the datum planes in this special case.  So misleading.

Another example is a statement made in section 4.4.2 regarding cylindrical datum features.  It states that "a cylindrical datum feature is always associated with two theoretical planes intersecting at right angles on the datum axis".  How much confusion has this caused for members of this forum alone?  The problem is that the statement is not generally true - it's only true in certain special cases.  One is a cylindrical primary datum feature.  Another is a cylindrical secondary datum feature that is nominally perpendicular to a primary planar datum feature.  A third is a cylindrical secondary datum feature that is nominally in line with a primary spherical datum feature.  There may be a few other cases that I haven't thought of.  But that's it.  In all other cases, the "two theoretical planes" on the datum axis conflict with DRF planes that have already been established.  For me, trying to apply the idea of these two theoretical planes in a general sense has been a conceptual wild goose chase.  I've tried to make sense of it but for the life of me I can't.  Based on posts we've seen on this forum, others have run into similar difficulties.

For datum feature configurations other than the aforementioned "special cases", constructing a unique DRF from the datums just isn't possible, for a variety of reasons.

For simple datum features that are not orthogonal to each other, the datums are well-defined but the origin and/or clocking of the DRF is not obvious or unique.  One example of this is the "two skewed cylinders" referenced A|B.  The primary datum axis establishes the direction of the DRF's Z axis and the X and Y origins.  Because the secondary datum axis is skewed relative to the primary, there is not an obvious and unique way to constrain the DRF's rotation about Z and or define its Z origin.

For non-simple datum features, the location of the datum is not well defined.  As a consequence, the origin of the DRF is not well defined.  One example is the hole pattern secondary/tertiary datum feature, where the location of the "datum axis" is arbitrary.  For circular or rectangular patterns, the center is an obvious, but still arbitrary, choice.  For other patterns, there is no obvious center or origin point.  There have been many discussions of the "center of the pattern" issue on this and other GD&T and CMM forums.

Some GD&T books, as well as the recently released Y14.5M-2009, expand the definition of a datum to include combinations of a point, a line, and a plane.  This allows datums to be identified for features like cones, extruded shapes, and complex surfaces.  For example, the datum for a pattern of parallel holes would be a "line on a plane" and the datum for a complex surface would be a "point on a line in a plane".  This is somewhat useful for visualizing what degrees of freedom are constrained by a particular datum feature when it is referenced as primary, but that's about it.  There is still the problem of non-uniqueness.  For the hole pattern, the locations of the plane and the line are still arbitrary.  For our two skewed cylinders referenced as A-B, the datum is a point on a line in a plane.  But the orientation and location of the plane, the orientation and location of the line in the plane, and the location of the point on the line are all completely arbitrary.  So we're not any further ahead as far as defining a unique DRF origin.  I just don't find it very useful.  Don't even get me started on how these datum types would constrain DOF's when referenced as secondary or tertiary.

So where does that leave us?  In most FCF applications, it doesn't really matter where the DRF origin is because everything is relative anyway.  If certain DRF axis directions and/or origin are desired by the designer for some reason, these can be annotated on the drawing or model in the form of a labeled set of coordinate axes related to the datum features by basic dimensions and angles.  The model defines the relationship between the datum features and the DRF, and thus defines the relationship between the datum feature simulators and the DRF.  When it comes time to build inspection fixtures or CMM programs, a unique DRF can then be constructed directly from the (physical or virtual) simulators.  Without the need for extracting combinations of points, lines, and planes for each datum feature and trying to hang a DRF on those.  I'm not saying that this "simulators to DRF" approach is easy in every case, but it is doable.

Comments?

Evan Janeshewski

Axymetrix Quality Engineering Inc.
www.axymetrix.ca

RE: Tolerance of Positoin - feature positioned to itself ???

Evan,

I read then re-read your comments.  I applaud your statements and you for being outspoken.  I feel that I agree with at least 90 percent of your comments and have had feelings as to the inadequacy of geometric background on some of the committee members in producing the standard.  

I have had some very strong misgivings about several items that you mentioned.  I was in somewhat of a contest with another person (anonymous) in the input for the 1994 update  corrections to the draft.  My recollection is that some of your comments were addressed and subsequently rejected at that time.  

 

 

RE: Tolerance of Positoin - feature positioned to itself ???

Ringster,

I appreciate the feedback.  What was the 10% that you don't agree with?

My comments were on the standard itself, and not intended as any kind of commentary on the committee members.  To be honest, I really don't know very much about most of them or what their backgrounds are.  I attended a Y14.5 meeting for the first time last year in Albuquerque, when they were in the comment review stage.  I got to see how the meetings work and how the committee operates.  It sort of reminded me of the old joke about sausage and law - you don't want to see either one being made.  Like anything made by committee, there are a lot of different viewpoints and interests to balance.  There is definitely a compromise between the "new and improved" and the "tried and true".

My comments are based on the content of the '94 version.  I don't have the final release of the '09 version yet, so I don't know for sure what the final content of the DRF section is.  I've only seen the public review drafts.  So some of these issues may have been addressed - we'll see.
 

Evan Janeshewski

Axymetrix Quality Engineering Inc.
www.axymetrix.ca

RE: Tolerance of Positoin - feature positioned to itself ???

Evan,

Reread the post "at least 90 percent".

RE: Tolerance of Positoin - feature positioned to itself ???

(OP)
I don't think my example was that much of a stretch. Just trying to get minds back to basic engineering fundamentals.  Engineering is in large part communication. It is a basic fundamental to document a design. If one is similar in many ways but is different in any way, then minimally the differences are defined by a modification drawing.
It is fairly easy to agree that the standard does not cover everything.  In this case, A-B has been explained in detail as to its meaning vs. the standards basic DRF concept.  I don't find it that hard to understand, however  my "bottom line" is that this deviation must be defined as an amendment to the standard, as all deviations should be.
 Evan, although you have walked me through your thought process on this callout meaning and have pointed out yourself that it doesn't fit the standard's current definitions; I don't see why, that my main point of documenting the deviations is any different than any other documentation of a design requiring definitions for modifications.  It's neither that I don't understand your presentation nor reject it, I simply say if this type of "extension" of the standard is used then it must be documented as an amendment to the standard.
This particular part has been talked about as being fixed at both ends and therefore both ends are equally important. I have commented that both ends indeed connect to other tubes, however if you look at most any exhaust tube (and I have mentioned before that this is what was posted, a rigid exhaust tube) it is truly held in place by hangers. I have mentioned before in almost any part that certain features can be found that are equally as important. I have contended in this case that the "equally as important" argument seems to ignore other important features of this component. The tubes are normally  of considerable length and each segment has minimum clearance at some point.  The segments must be manufactured within tolerance as well, or it won't matter if the ends come out in the correct position because it still wouldn't fit.
MechNorth has also mentioned a few days ago "design intent" is the way a component should be dimensioned. I have personally been a vocal advocate of this principle. However as I have thought more on this I find many times the manufacturing process must be "considered".  If for example I know I have a cast or molded part, it is a part of the design to consider minimal wall, draft, transitions from one thickness to another, etc. If my design does not consider the manufacturing process and its relationships to part function, then it could easily be a poor design.  

Thanks for all the comments.
 

DesignBiz stpatrick2

"Quality is in the details"
 

RE: Tolerance of Positoin - feature positioned to itself ???

DesignBiz,

I think we're both in agreement on the need for explanations of practices are aren't covered in the standard.  I'm still a bit confused about what the situation in your company is.  Is it that there is resistance to documenting those explanations, so that if something is not covered in the standard then you can't use it?

Regarding the "equal importance" of features, here are my thoughts on that.  I wouldn't use the term "importance" when it comes to choosing datum features and how they are referenced.  It all comes down to degrees of freedom and how they are constrained.  The tubes are fixed at both ends, therefore the two end cylinders are the features that constrain the tube's degrees of freedom and should be the datum features.  Each cylinder has an equal role in constraining the degrees of freedom, so they should be referenced in a way that treats them equally (i.e. A-B).  Other features on the part might be equally "important", and must be in tolerance in order for the part to function.  But these other features don't constrain the DOF's and are therefore not referenced as datum features.

Evan Janeshewski

Axymetrix Quality Engineering Inc.
www.axymetrix.ca

RE: Tolerance of Positoin - feature positioned to itself ???

DesignBiz, I've never said to ignore manufacturing and inspection.  The PRIMARY intent of GD&T is to convey design intent, not to convey manufacturing and inspection techniques.  A common problem is that manufacturing overwhelms design with their demands and the design suffers.  It seems that many / most companies want to maintain their US vs THEM department division mentality.  As part of my training material, I advocate Design Previews rather than Design Reviews.  The difference is that everyone (design, manufacturing & inspection) gets a chance for input on the controls and tolerances before the GD&T annotation is added.  After the drawing is done, it's little value to go back and shred the designers's work and self esteem.  More advanced companies will even set up design groups to solicit manufacturing and inspection input before designing the workpiece.  Some companies have integrated this into DFM processes.

Jim Sykes, P.Eng, GDTP-S
Profile Services  www.profileservices.ca
TecEase, Inc.  www.tec-ease.com

RE: Tolerance of Positoin - feature positioned to itself ???

(OP)
Axym,

The company does not want to take the time to document the amendment because it would need to be approved by a committee. Therefore it is the local opinion that it is better for the callout not to be understood, than to go thru the process of documenting it's meaning.


MechNorth,

Never claimed that 'you said' to ignore mfg or inspection, just wrote that you had mentioned that design intent is the way a part should be dimensioned; I agree, but this also got me thinking that mfg needs to be considered more some times than others. "I thought more on this..." (my previous comment) is in regard to my usual response.... "dimension for design intent/functionality"; meaning... maybe I should answer a little differently in the future.
 

DesignBiz stpatrick2

"Quality is in the details"
 

RE: Tolerance of Positoin - feature positioned to itself ???

One of the joys of text-based communication is the freedom to read in / read out subtleties.  Thus, misunderstandings.  No problem.

Jim Sykes, P.Eng, GDTP-S
Profile Services  www.profileservices.ca
TecEase, Inc.  www.tec-ease.com

RE: Tolerance of Positoin - feature positioned to itself ???

(OP)
FYI,
Although I enjoy the comments from all of the site's contributors, I find particularly interesting the detail and perspective from the inspection set-up aspect that yourself and Axym provide.


c u folks on another thread.....

DesignBiz stpatrick2

"Quality is in the details"
 

Red Flag This Post

Please let us know here why this post is inappropriate. Reasons such as off-topic, duplicates, flames, illegal, vulgar, or students posting their homework.

Red Flag Submitted

Thank you for helping keep Eng-Tips Forums free from inappropriate posts.
The Eng-Tips staff will check this out and take appropriate action.

Reply To This Thread

Posting in the Eng-Tips forums is a member-only feature.

Click Here to join Eng-Tips and talk with other members!


Resources


Close Box

Join Eng-Tips® Today!

Join your peers on the Internet's largest technical engineering professional community.
It's easy to join and it's free.

Here's Why Members Love Eng-Tips Forums:

Register now while it's still free!

Already a member? Close this window and log in.

Join Us             Close