Perpendicularity

[randy64]

I am seeing a conflict on how this works.

I am in a training class right now (TCI, Al Neumann). He states that Perpendicularity can be referenced to datums that are not perpendicular to the surface being held perpendicular. He has an example in his workbook showing it that way also. He refers to Perpendicularity, Parallelism and Angularity as Orientation controls (which they are), but seems to indicate that they are really flatness controls with orientation and that Perp, Par and Ang are just our perception of the direction of orientation.

In Alex Krulikowski's book "Fundamentals of GD&T" he explicitely states that the surface being called Perpendicular must be perpendicular to the datums it is being referenced to in the feature control frame.

Which is it? I've always gone with Krulikowski's interpretation. If we didn't, we could just get rid of Perp, Ang and Par and call it all Orientation.

What says you?

Thanks.

 

[powerhound]

I'm not sure if I understood what you said in your first paragraph but profile isn't limited to basic dimensions only, as shown in Fig. 6-18. You can use +/- tolerancing in conjunction with a profile of a surface callout. Regarding the issue of combined tolerances, that's up to the engineer to figure out.

Size has everything to do with it's orientation. The distance of the surface from B cannot exceed it's size tolerance at any point so this provides a measure of orientation control to B. If you want to orient it more tightly to B only but not necessarily A, then you would use parallelism, if you want it oriented more tightly to A but not necessarily to B, then you use perpendicularity. If you need them both oriented more tightly, you use profile of a surface. None of these require basic dimensions and they are all fully supported in the 1994 standard.

Powerhound, GDTP T-0419
Production Manager
Inventor 2009
Mastercam X3
Smartcam 11.1
SSG, U.S. Army
Taji, Iraq OIF II

 

[MechNorth]

Sorry guys, been off visiting & such for a day or two. OK, I have SEEN an example like Al Neuman has in his material, as referenced above. It IS counter-intuitive, but not specifically indicated as illegal in the '94 standard. And I was given the same explanation, i.e. that it is viewed as an Orientation control rather than just a Perpendicularity control. I can accept it, because perpendicularity is just a special case of orientation, but I would never suggest it to anyone as the way to do it, nor endorse it as a practice to be taught when there are far clearer ways to achieve the goal.

There is a similar but different issue when you have a composite profile control with 3 or more levels. Typically the first level controls the location, and any subsequent levels with datums referenced will refine the orientation. It takes a bit to wrap one's head around the idea that on each refinement level you are primarily focusing on the last datum in the DRF, while also including the preceding datums .. it can be a mind bender, but usually is the best way (in the applications that I've seen) of achieving the goals. When the surface is already located using a +/- tolerance, then the profile control (single segment or composite) will be refining orientation and form.

Personally, I would prefer that both parallelism and perpendicularity be eliminated and just Angularity in ALL orientation situations. It's not a huge cognitive leap to see that two lines look 90-deg or 180-deg apart, so there really isn't a need for the other two controls.

Keep in mind that just because something ISN'T specifically illustrated or included in the text of the standard, that does not mean that it's illegal. An application needs to violate aspects of the standard to be an illegal application.

The example cited in Fig. 6-35 is clearly a valid application of two datums referenced in a perpendicularity control, and it makes sense because the surface IS perpendicular to both datums, as designed. Good reference.

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

http://www.profileservices.ca

TecEase, Inc.

http://www.tec-ease.com

 

[powerhound]

I guess my issue lies in the fact that the standard says that perpendicularity is specified perpendicular to a datum. In every subparagraph of 6.6.4.1 the tolerance zone is specified perpendicular to a datum. Nowhere does it state that it can be specified parallel to anything, whether for orientation or not. Does that really mean we can add that to it and it should be okay? There is no symbol for perpenposiparalangulicity but I'm making one up, putting it on a print, and supporting it with the 94 standard cause nothing in it says I can't...

Seriously though...this has been an excellent thread. I'm glad these only come around every now and then, otherwise I'd never get any work done or sleep.

Powerhound, GDTP T-0419
Production Manager
Inventor 2009
Mastercam X3
Smartcam 11.1
SSG, U.S. Army
Taji, Iraq OIF II

 

[ProfDon]

Hi All,
This was an interesting thread. Having attended the Y14.5 meetings where this was discussed I found it interesting that you all had the identical discussion. The line about "Relation to more than one datum feature is specified to stabilize the tolerance zone in more than one direction." was added to clear this issue up in the 1994 standard. I know, you have to go by what is in the Standard since we can't all be at the meetings. Well, look at where this statement is added. It applies to all orientation tolerances--including perpendicularity. Even with the addition of this statement, some folks remained troubled as is evidenced by this thread. So, the Alternative Practice of using angularity for all orientation controls was added in 2009. Yes, a composite profile tolerance could have been used as another way to control the surface. But, using an orientation control guides the reader where profile of a surface is not well understood.
Bottom line, Al is correct since adding additional datum references to any orientation tolerance, including perpendicularity, stabilize the tolerance zone relative to the datum reference frame. MechNorth explained this very well.

 

[axym]

I'll start with my standard opening statement of "I agree with Paul Jackson on this one". Profile would control the location and orientation of the considered feature in the Al Neumann example. Pretty much the only case in which Profile would provide the same control as an orientation tolerance would be the "everything orthogonal" geometry of Fig 6-35.

Like Jim, I could live without specific controls for Perpendicularity and Parallelism. I would like to see an orientation control that can be applied to any feature type and related to any datum reference frame. This can be done with the lower tiers of composite FCF's but those are so confusing.

BTW, I like powerhound's comment about "perpenposiparalangulicity". Didn't Julie Andrews and Dick Van Dyke sing that?

Evan Janeshewski

Axymetrix Quality Engineering Inc.

http://www.axymetrix.ca

 

[caseynick]

IMO
The example of perpendicularity in the 94 standard which shows the surface pepprpendicular to both surfaces is an example of one possible way to use the callout. The example of perpendicularity in the 09 standard, which pretty much matches the sketch that initiated this thread, is an example of another possible way to use the callout. It would be impossible for a standard to have examples of every possible way to use a particular gd+t control.

The appendix in 09 says that there has been no significant change from 94 in the perpendicularity callout.

The reason there is no secondary datum for parallelism callouts is that clocking/rotational control is irrelevant.
p.s. Prof Don What about me? I ought to get an attaboy for saying the same thing way up the thread don't you think?

 

[ProfDon]

Attaboy, caseynick.
One comment, though. If a hole is parallel to the primary datum feature and perpendicular to the secondary feature, I could see a secondary datum feature referenced in a parallelism callout in order to "stabilize" a cylindrical tolerance zone. But, like caseynick said,"It would be impossible for a standard to have examples of every possible way to use a particular gd+t control." The question I often ask is: "Does this geometric callout have one clear meaning to those who are knowledgeable of the standard?" If the answer is "yes", it is a legitimate (legal) callout even though you cannot point to one exactly like it in the Standard. It is called an extension of principle.

 

[caseynick]

Thanks.
And in regards to your comment: Of course. I should have said "....MANY parallelism callouts......"

 

[dingy2]

It is hard to believe that someone is teaching the application of perpendicularity when the feature is not perpendicular. That, in itself, is mind boggling but it does comply with the 2009 standard. No wonder people are so confused over GD&T.

I certainly can see someone training relatively knowledgeable GD&T personnel the use of angularity in this situation but perpendicular?? mmmmmmmm.





Dave D.

http://www.qmsi.ca

 

[powerhound]

Being that some seriously credentialed people are saying that the example in Al's book, which is per the 94 standard, is legit, then I'll concede that it is valid, BUT I will still never specify perpendicularity using this method on any prints that I make. It's just not easily supported, IMHO, and there are other, more clearly interpreted and supported mehtods that produce the same results.

Powerhound, GDTP T-0419
Production Manager
Inventor 2009
Mastercam X3
Smartcam 11.1
SSG, U.S. Army
Taji, Iraq OIF II

 

[ProfDon]

Hey powerhound,
I am with you. The most important thing is to clearly communicate to the reader (customer). There are a lot of tricky things you can say on a drawing that would baffle the best of them. This was an interesting thread. Attaboys to all who participated.

 

[randy64]

ProfDon, you said, "Bottom line, Al is correct since adding additional datum references to any orientation tolerance, including perpendicularity, stabilize the tolerance zone relative to the datum reference frame."

So, is Alex Krulikowski wrong when he calls such callouts "illegal"? (Reference my link above)

I sure wish someone would have spoken up in those 1994 meetings and forced some less ambiguous language. Adding the "stabilize" line seems like a weak attempt to me. In a standard of this sort it should be more explicit, IMO. I think the standard should stand on it's own, without knowledge of the backroom dealings that birthed it. If it isn't in the standard (and I still don't see Neumann's interpretation in there - in fact I think it is excluded) then it's not part of the standard. Discussions mean nothing, only what makes it into print matters.

I'm with powerhound - I don't see myself ever using Perpendicularity that way.

 

[ProfDon]

I think Alex will admit that he might have slipped in this case.
I agree and acknowledge that it has to be in print. I pointed out that the Standard explains the purpose of additional datum references where it is discussing all 3 orientation tolerances. That is pretty clear to me. But, standards like any language evolve and if this becomes an issue or winds up in court, it will be pointed out to the committee and clarified further. That is how we got the new spotface symbol this time around.

 

[KENAT]

What kind of training is this by the way. Is it "all you need to know to be on the 14.5 committee" or perhaps "advanced GD&T so you can confuse even knowledgable collegues".

If it's a really advanced course then fair enough but if it's a more basic level case, isn't this kind of thing a bit OTT?

KENAT,

Have you reminded yourself of faq731-376 recently, or taken a look at posting policies:

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

What is Engineering anyway: faq1088-1484

 

[MechNorth]

Interesting question, Kenat. Training material evolves over time, reflecting changes to the standards, and wider application and acceptance of the technologies involved. Sometimes we add something in because we see an increasing demand for that particular nugget of knowledge, and later may find that it's over the heads of many who aren't at an advanced level ... so we sometimes take things out too. When books are published for wide distribution, there is usually significant pressure to include more rather than less, and hopefully something "unique" for want of a better term. My thoughts are that this topic should be focused into Intro, Intermediat and Advanced materials, but I know from first-hand experience that such an approach is not appreciated by most, and they often evaluate themselves as being significantly more capable than is the case. In learning the Y14.5 standard, we re-read the standard numerous times by the linked-topic study method and keep re-reading it as we broaden our own exposure to new applications. Sometimes it takes a sudden lightning-bolt moment to bring a bothersome concept into focus.

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

http://www.profileservices.ca

TecEase, Inc.

http://www.tec-ease.com

 

[KENAT]

Funny MechNorth. I think we're maybe of a similar opinion. They had a week long training session here just before I started that, having read the material, was probably at the Intermediate level.

Based on how people have (not) implemented it etc. I think a far more basic, and probably a bit shorter, "Intro" course would have been better.

Obviously, others may have a different point of view, but for me an Intro level course would probably focus on what to me is the bread and butter of GD&T - basic hole patterns ensuring fit/function.

When I had a chance to train some new interns that's what I did. I only had an hour, no where near enough time to do it justice, but I attempted to give them an appreciation of fundamentally what tolerance was and then how to apply it to hole patterns.

I realize hole patterns aren't the be all and end all, and of course can get complicated in them selves, but the most common applications & errors I see are basic hole patterns so to me that's where the bang for the buck is.

KENAT,

Have you reminded yourself of faq731-376 recently, or taken a look at posting policies:

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

What is Engineering anyway: faq1088-1484

 

[MechNorth]

Yup, Kenat, and I find that the focused needs at each company (and even within different sections of the same company) vary wildly. Considering that, and without the "luxury" of being onsite long-term, often trainers and their materials have to cover the whole scope of the standard, though obviously some sections in greater & lesser detail. What makes it worse is that so many of the topics are intertwined that we have to cover some topics that the trainees don't think are applicable ... inevitable scope creep. Some instructors stick very tightly to their materials and timelines, while some of us divert to tangents to address the funkier applications of the class. That's what keeps the training interesting for both sides.

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

http://www.profileservices.ca

TecEase, Inc.

http://www.tec-ease.com

 

[axym]

GD&T can be a difficult thing. It's one thing to learn the basic ideas, and something else entirely to learn them well enough to get all of the details exactly right. And we all know the consequences of not getting all of the details exactly right. The devil is in the details, as they say.

I often get requests to not cover the whole standard, and to just show the students enough to do their job. But it's amazing how complicated things get even for garden-variety parts. To inspect a typical position tolerance properly you have to understand the details of axes, actual mating envelopes, cylindrical tolerance zones, patterns, simultaneity, material condition, bonus tolerance, basic dimensions, datum features, datum feature simulators, datums, datum reference frames, degree of freedom constraint, precedence and non-override, datum shift, and virtual condition. If any of these details are not treated correctly, the results can be wrong.

I've found that training that tries to keep things simple and glosses over details tends to be well received but causes confusion when applied later. Training that deals with all of the details and complexities tends to cause confusion right away! It's hard to find a balance that gives the students some job-ready GD&T skills without confusing the heck out of them. All compressed into a couple of days, of course!

Evan Janeshewski

Axymetrix Quality Engineering Inc.

http://www.axymetrix.ca

 

[KENAT]

(To the OP sorry if this is too far off topic, I'll start a new thread if need be)

I suppose implicitly, I was invisioning a situation where there are more experienced folk that can be used as resources when people start getting into more complex areas and that will be reviewing the work for accuracy. Obviously this isn't always the case.

I was thinking of a course that gets people up to the basic level, allows them to operate at that level for a while to get some real world experience, with support/review of their work while they learn. I was thinking that eventually they would probably take a slightly more advanced training and/or (I know this is crazy talk, but bear with me) spend some time getting to know the standard, teaching themself, asking questions when they find something they don't understand...

However, I realize now I was clearly in some kind of fantasy. I'm probably still at the stage that I don't really know enough to know what I don't know, but 90+% of the time seem to muddle through OK referring to the standard and thinking about things. The other 10% I either ask here, find a work around or give up;-).

KENAT,

Have you reminded yourself of faq731-376 recently, or taken a look at posting policies:

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

What is Engineering anyway: faq1088-1484

 

[MechNorth]

Now THAT's funny, Kenat! Progressive learning and internal resource mentoring !!! What do you think industry is, something intelligent & efficient? Oh, my side hurts now from such a good joke!

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

http://www.profileservices.ca

TecEase, Inc.

http://www.tec-ease.com