pattern positioning

[dnobie]

I am trying to determine if it is legal within ASME Y14.5M to locate a pattern of holes with conventional dimensioning (no datums) from an outside profile and then use GD&T to control the spacing of the pattern to itself. We are reviewing an old (80's) drawing package in which this was done extensively and are trying to determine if this is/was legal. The holes within the pattern are located with basic dimensions. The feature control frame does not include any datum references. Thank you in advance.

 

[fsincox]

KENAT,
I didn't realise you were so young. These practices were perfectly legal and shown by the committee in the standard before the '82 version.
The idea that this stuff is just going to disapear or doesn't still exist makes me wonder where you all work. I too wish these practices would all go away, but on the contrary, I think this is most engineers' concept of good GD&T, from what I see.

 

[KENAT]

I see plenty of bad GD&T, though our industry didn't really get started till the 90's.

I've seen much older drawings at previous employer but wasn't as hot on GD&T back then.

My comments were more to do with the suggestion it was legal to the 94 version of the standard. I'm not going to catagorically say it isn't but I have my doubts based around the two things I posted.

Posting guidelines faq731-376

http://eng-tips.com/market.cfm?

(probably not aimed specifically at you)
What is Engineering anyway: faq1088-1484​

 

[pmarc]

Even ISO standards (ISO 5458:1998) do not recommend using coordinate tolerances for locating pattern of holes. They say the meaning of +/- dimensions is in this case not standardized. But of course they do not give precise explanation why it is not recommended.

I can only imagine the reasons, and I see quite a lot of them...

The other thing is that 'not recommended' does not mean 'forbidden'.

 

[axym]

pmarc,

They say that the meaning of +/- dimensions in this case is not standardized. That's explanation enough for me as to why it is not recommended!

One could also argue that the +/- tolerances locate the edges relative to the holes, as much as they locate the holes relative to the edges. That's the thing with +/-, it isn't even clear what the considered feature is and what the reference feature is.

Evan Janeshewski

Axymetrix Quality Engineering Inc.

http://www.axymetrix.ca

 

[ringman1]

I believe that the +/- for locating a pattern was replaced by the composite tolerancing scheme, in the 1982 version. My opinion was that they totally disregarded the benefits of a rectangular locus for the PLTZ.

 

[Belanger]

It is not recommended because it is ambiguous. I would ask which edge the hole is measured from. Look at the dimension lines, you say, and follow them back to the two edges. OK, I say, now tell me which one takes precedence over the other. You reply that it doesn't matter.

Then you are saying that the angle between the two edges will always be guaranteed to be perfectly 90º, or else you are saying that it's OK for a drawing to be ambiguous. Either way, it's not good!

Ringman -- using GD&T over ± location does not eliminate a rectangular tolerance zone. It simply means that you'll need two feature control frames (one for each direction), and they should not have the diameter symbol in front of the number.

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

 

[KENAT]

Thankyou Belanger, I nearly posted the same but wasn't quite sure on how you'd implement it with composite so was going to think twice before posting.

Posting guidelines faq731-376

http://eng-tips.com/market.cfm?

(probably not aimed specifically at you)
What is Engineering anyway: faq1088-1484​

 

[ringman1]

Belanger,

I would conclude from your statement that you have used rectangular locus for PLTZ. If so, was it accepted as legitimate. I have felt that an example in the standard would be of benefit.

 

[Belanger]

I guess that's true; rectangular zones with composite might be a little tricky.

Hey -- fodder for a new thread!

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

 

[KENAT]

No, fodder for an old thread if I recall correctly.

Posting guidelines faq731-376

http://eng-tips.com/market.cfm?

(probably not aimed specifically at you)
What is Engineering anyway: faq1088-1484​

 

[fsincox]

Belanger,
The committee admits as much in the new 2009 standard basically any plus and minus "except for features of size" is not well defined. This is the logical extension of your point, too. What are people suposed to do redraw everything? after ever new standard? It is just not going to happen. This is why the divide between us few believers and the "people" is so great, Change without outside motivation, like the collapse of manufacturing in this country and the desperate need to be competitive, it will be in fact a VERY slow process.
Frank

 

[Belanger]

Yes, I certainly was not implying that every old drawing must be changed. But the OP was asking about the legality of doing so, and I pretty much agree with the others who have said that it may be legal, but not recommended. So those of us who are responsible for adding tolerances or modifying tolerances should be aware of the potential pitfalls.

I have indeed used composite position tolerancing without the diameter symbol for a pattern of holes, with no problems. However, it was in one direction only. The tricky part might be when another composite tolerance is applied in the other direction -- the way the FRTZFs intersect might be the sticky part, but I'll have to think about that some more. (KENAT, I did a search and didn't really see that issue addressed specifically. But I'd like to know more.)

RIngman,
I am curious why you are such a big fan of the rectangular zone (or locus). Are you envisioning holes/pins or are you thinking of its use on planar features of size?

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

 

[fsincox]

John,
I agree, too.
Frank

 

[Belanger]

Ringman,
After noodling this out a little more, I think I see what you are getting at. Even if we are considering a regular pattern of holes on a part, we might want to allow them (as a group) to drift north or south only a certain amount -- perhaps because of a concern about wall thickness between the hole and the edge of the part. And to use a diametrical zone would force our tolerance cylinder to stay inside of the square tolerance zone, thus losing some real estate on the corners.

So I agree that there may be value in keeping a rectangular zone, and that can be done with GD&T. I can't quite get my mind around composite tolerancing with a rectangular zone, and since I am an opponent of pure ± tolerancing for location, I'm guessing that there's a solution that lies somewhere in the middle.

Perhaps a feature control frame for position to the edges of the part (no dia symbol), and a separate frame for hole-to-hole position showing the dia symbol. Ugh.

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

 

[KENAT]

Belanger, I'm pretty sure Ringman or Ringster has brought up the rectangular tolerance zone before, and I thought at least one in the context of composite tolerance.

thread1103-242095 might be it but I thought more it had been discussed in more detail elsewhere, maybe I'm thinking of thread1103-238012 or thread1103-242404 but after a search I'm guessing thread1103-217781 was what was in my memory.

Posting guidelines faq731-376

http://eng-tips.com/market.cfm?

(probably not aimed specifically at you)
What is Engineering anyway: faq1088-1484​

 

[fsincox]

In my opinion, this all starts with the “wink, wink, I don’t really care about this dimension” thing I have been here railing against. I keep trying to argue that to say: “I don’t care about the perpendicularity of large diameter/small length pilots to their mounting face, because I can’t check it” leads to the same road. The people who locate patterns this way are doing the same thing: I don’t care about the location of the pattern from the edges just the pattern itself. Once you allow the out of “well it really isn’t that important to me” instead of “this describes the actual condition” it is hard to know where to stop, everyone has an opinion. All general tolerances concepts are based on this.
Frank

 

[dingy2]

Well Frank:

I don't think anyone has said "wink, wink, I don't care about the dimension" but sometimes GD&T trainers profess overuse of GD&T when it is not required for the design intent of the respective feature.

A simple linear +/- is sufficient and legal for the location of a pattern (PLTZF) of holes where the pattern location is not important to its function. Positional tolerances (FRTZF) may be vital inside the pattern if we have a mating part assembling to the features inside the pattern.

So many GD&T trainers profess that all features of size should have positional tolerances and all surfaces should be covered with a profile of a surface without regard to their function. It has been said here in this forum that this is better practice while the reasoning is somewhat weak.

Could this lead to higher costs for the product? Could we end up checking positional tolerances on patterns that have no bearing on the part function? Could we end up not confirming positional tolerances on certain features where there is a need since all other features of size have the same positional requirement?

I do care about each an every dimension but the application of GD&T should be predicated by the feature's function and not a shotgun approach.

Dave D.

http://www.qmsi.ca

 

[Belanger]

Dave,

Sorry to drag this thread out, but do you think it's crystal-clear when ± dimensions are used to locate a feature of size? I keep beating the drum that it's not, because the precedence of the two implied datums is not given.

And we can't say that it doesn't matter, because somebody is going to have to check the X and Y location for that feature, and there are two possible readouts. Thus we don't know which one is really the intended location.

Plus/minus dimensions are only really clear when showing a size, chamfer, or radius. It's best to use GD&T for location -- hey, if it ain't critical, then just give it a generous tolerance!


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

 

[dingy2]

JP:

Yes it is clear just as long as one comes from a secondary datum if it is a hole or from the secondary and tertiary datums if they are surfaces. Crystal Clear! I could have positional tolerances on some patterns while linear tolerances on other patterns. Some patterns would then require checking fixtures and included in the Control Plan while others would only be confirmed sporadically.

If one used positional for all features of size, how should it be interpreted? One could have a note that states positional tolerances above a certain value have no bearing upon it function, then fine, I would agree. This just doesn't happen though.

I will remain diametrically opposed on the concept of all features of size must have positional tolerances.


Respectfully,

Dave D.

http://www.qmsi.ca

 

[Belanger]

But recall that my beef is that we don't have any datum precedence.

You say that "it is clear just as long as one comes from a secondary datum if it is a hole or from the secondary and tertiary datums if they are surfaces." But may I ask how you know what the secondary datum is? I humbly submit that we don't have any idea what the secondary or tertiary datums are if you are using ± tolerancing for the location.

You also ask "If one used positional for all features of size, how should it be interpreted?" Well, I hope we agree that every feature of size must have a location tolerance. So I'm not sure what the question is asking. I simply meant that position should control a feature of size because the datums are clear -- I don't care if the zone is rectangular, cylindrical, or another shape, and I don't care if the tolerance value is 0.0001 mm or 10 mm.

Using GD&T is not a statement about the importance of a feature. It's about defining that feature in a clear way.

Maybe I'm still not seeing something, but I hope my basic idea makes some sense.

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