Spinoff2: Using orientation to control position 1

[fsincox]

The whole idea of using orientation to control location has also come up in my past work as the “GD&T guy”. A very well respected senior inspector proposed the use of parallelism to control the location of shaft keyways. His idea was that you would line up the shaft diameter (the primary datum) and then roll the keyway, itself (the secondary datum) to center and proceed to measure the resulting parallelism of the opposed sides of the keyway. His rational was that the resulting parallelism of the sides of the keyway in this representation of the installed state was important to the key not working its way out under heavy loading. I have to admit I did not feel it was technically illegal, what do you think?

 

[CheckerHater]

Tolerance zone usually doesn't work alone.
It is combined with size tolerance and/or datum shift to form Virtual condition, ant it is VC that tells you where your part can and cannot be.
So, yes, the element is "contained" within tolerance zone, but zone itself is "floating around", so it does not position unless you explicitly command: "position!"

 

[fsincox]

CH,
Once at least one point on the surface is established the tolerance zone becomes fixed/located, all other points must fall within a width (contained), and oriented to, a zone the framework established. An orientation zone can no longer float outside it's stated value or there would be no "cointainment" at all. It must contain all of the surfaces points.
Frank

 

[drawoh]

drawoh,
I am sorry I do not agree, so far, You admit that you know what to do with the A (primary), B (secondary) framework for other features, It is in fact the same thing I or others would do. The establishment of a shaft diameter and key for clocking is in the standard as a common framework for these type of parts. I do not see why you know how to "clock" to the key "after". I was taught that process should have no relevance to how you establish a datum framework;...

You are not understanding my point.

On my sketch, Datum[ ]B is defined, but it is not referred to. The slot is defined entirely by Datum[ ]A. If I were to add holes as per the diagram is ASME[ ]Y14.5, I would almost certainly need the clocking feature. My drawing shows requirement, not fabrication procedure. It is not a complete drawing. All it was intended to do was illustrate a point.

A big enough parallel error will break your positional tolerance. The positional tolerance controls parallelism -- on my drawing to 0.2mm. The parallel specification on my drawing, tightens the specification to 0.05mm. There is no point in applying parallelism unless it does this.

My final comment was that you should watch out for your datums, especially if they are features of size. I scribbled out my sketch fairly quickly, as I was just trying to explain something. If I were serious about this, I would not use the 0.06mm toleranced diameter to control parallelism of 0.05mm. This is actually not relevant to the discussion. I was just noting a problem with my sketch.

JHG

 

[pmarc]

Frank, I give up.
Not because I admit you are right, but because I have no further arguments to show that your interpretation is wrong. In my opinion there is no stronger support of my, Dean's and Evan's standpoint than paragraph 6.2 of Y14.5-2009 saying: "An orientation tolerance does not control location of features".

The sad thing is that '94 edition did not clearly state that, so now you are somehow free to interpret the statement from 6.3.1 in a way which is suitable for you. Please just tell me that basing on your understanding of 6.3.1 you do not think that form tolerances also locate.

 

[CheckerHater]

Frank,
Just so I can understand your point better, could you please describe how "position" tolerance zone will look like in this case?

 

[fsincox]

pmarc,
I am sorry; I do hate to lose you, CH and yourself are my ISO guys. I understand arguing over and over can get tiring, as I mentioned I have a lot of thought vested into this so I do not look at it as simply a populist rule kind of issue.
I do think I have touched on a key part of this concept before in this forum. I made a statement in a point before that argued orientation tolerances do affect/confine location, based on the same concept of containment. It went over the same way as this is now and in the end I believe it was more an issue of semantics.
If I have a normal parallelism callout on a surface, the location (containment) of all points of the surface must fall in the band of width determined by the parallelism tolerance, that band is allowed to float inside the size tolerance, until any point on the surface is established, once a point is found all others are contained to fall within (2) offset planes, parallel to the referenced datum and determined by that value, no?
I think you guys may be thinking location more than what I mean, but to my way of speaking those points may no longer float just anywhere and not even anywhere within the size band but MUST be contained within the parallelism specified tolerance band width. From my point of view parallelism is forcing location at that point. I think that is all it is and is simply fundamental to the concept zone established
Frank

 

[fsincox]

pmarc,
I did misquote the standard? Containment is not equal to forced location? If the Nazis contain people in concentration camps are they not also forcing their location?

CH,
I am working on it

Frank

 

[fsincox]

Dean,
My main reasons for mentioning the other gentleman were: one, to give credit where it was due, and second, to support the idea that this was not something engineering just dreamed up and “threw over the wall”. I believe it was an inspection driven concept by someone who, I had every reason to believe, held a solid understanding of inspection methods and capability.
Frank

 

[DeanD3W]

Frank,
No disrespect or offense intended here, but wow... You're not looking at this with an open mind and considering the comments that have very clearly stated "No" with regard to the question you're asking in your OP.

As you're reading very clear posts that tell you that orientation never controls location, including not for the case you're describing, I have this picture in my mind of you covering your ears with your hands and loudly yelling "LALALALALALA!" to yourself in order to avoid really considering what you're reading.

pmarc's illustration clearly showed that the slot and its two sides are absolutely not located when an orientation tolerance is applied. How could we not be all in agreement on this??

Orientation tolerances applied to the sides of a slot will control the form and orientation of each planar surface, no more than those two things though. Never (absolutely not ever) location in any fashion whatsoever.

Dean

http://www.d3w-engineering.com

 

[CheckerHater]

Frank,
You know, I like to play Devil's Advocate. A see idea I don't like and I am trying to make it work. It makes good mental exercise, and keeps me from getting Alzheimer's :-(
First thing I did I tried to create setup where Parallelism will somehow put keyway onto right place. (BTW, I still believe my idea was better than [A] primary secondary)
I submitted my second picture not as an insult, but rather honest attempt to understand - maybe if we will start with simpler example we can find where exactly our opinions split.
So I don't think you are losing anyone here.
Just keep on trying! Well-behaved people rarely accomplish anything.

 

[CheckerHater]

But please leave the Nazis alone!
One can be "contained" and moving at the same time:

http://en.wikipedia.org/wiki/Gas_van

 

[fsincox]

CH & Others,
I am sorry that was a bad analogy, I was trying to use the concept in a different sense to accentuate the effect I feel occurs. I did start this thread so do I feel I have the right to pursue my line of thinking as far as I need to get this into my head. Unless I can get someone else to understand I am not sure that actual communication has occurred. To say: “the standard does not say this so I believe it”, sounds to me more like: “I don’t like to think about this just believe it”, that I cannot do! I am also not an: “Everyone else thinks this way so I should too”, kind of guy. I, like CH, will tend to be a ”well why not?” kind of guy. Others are free to drop in or drop out as they feel, I will be sad to lose their input but, we will meet again.

If it makes you feel better to not use the term location for this effect, I am fine with that, the standard states “contained”. It is the actual effect that occurs that is the important key not the words we use.

Look at basic parallelism ASME Y14.5M-1994, Fig. 6-30. We all know the surface can float inside a larger size zone, but, once even one single point is established the zone becomes contained to .12(mm) wide, the parallelism zone value. Is that not the effect you see? (I know, possibly less if at the MMC boundary, of course)
Frank

 

[axym]

Frank,

I admire your willingness to think in new directions and entertain a controversial idea. Sometimes this leads to a useful new conceptual approach to something previously thought to be well understood, sometimes it leads to a blind alley. As Dean and the other members of this forum can attest, I've had a few controversial ideas myself which have turned out both ways. Once in a while I might see things that nobody else has, sometimes I'm just seeing things.

I think I understand the point you're trying to make with location of points within a zone, having pondered things like this on long airplane flights home from GD&T camp. One could argue that confining all of the points on a feature to a tolerance zone locates them relative to each other. Or that the points on the feature must all be located to some reference plane, but we can arbitrarily pick the reference plane. This concept has practical application in metrology and measurement data reporting, where we sometimes report the locations of surface points relative to an arbitrary reference plane to illustrate Flatness or Parallelism deviations.

But the Y14.5 standard uses a more narrow definition of "location", effectively confining it to the location of entire features (or entities extracted from a feature) relative to other features or relative to datums. Inter-feature location and not intra-feature location, if you will. So the extension of the word "containment" to mean "location" is unfortunately a blind alley, at least as far as application to the current Y14.5 standard is concerned.

So keep the questions and ideas coming. Your posts have made me think of things that I wouldn't have thought of otherwise.

Evan Janeshewski

Axymetrix Quality Engineering Inc.

http://www.axymetrix.ca

 

[fsincox]

CH,
Well legally or otherwise?
Can I get the part made with a good possibility of working for you? Yes
Is the part actually functionally defined? Probably not
I work with drawings like yours all of the time. GD&T is not a popular subject with most of the people I know and work with, this is why I bother you guys. I believe the 1973 standard and earlier attempted to define “legally” what a drawing like this means. Under the newer versions of the standard they have stepped away from +/- style definition to the point where it is harder to be sure that they mean “legally”. Meanwhile, out in the real world we work with these drawings all of the time. This has been one of my many points of dissatisfaction with the situation we find ourselves in. The standard committees spend time creating new way of doing these thing which most, I see, in the industry try to avoid and then they proceed to cut out the legs from underneath those of use that have this existing work in the field.
Just a suggestion (I do not want to lose you), If this is not directly relevant to a point here, let’s start another thread for it.
Frank

 

[axym]

Frank,

I completely agree that the actual effect is more important than the words. Sometimes the words get in the way, as we often find in our debates on this forum. Sometimes we have to look away from the words, and look to the geometry. When pmarc took the time to make a simple diagram of the off-center keyway, he probably saved us all a lot of time trying to describe the same thing in words.

You're saying that the entire surface must located within a certain zone in order to conform to the Parallelism requirement, once the first point has established. Technically this might be true, but again it doesn't match Y14.5's definition of location. It's not controlling the location of the feature relative to other features or relative to a datum (a datum in Y14.5 is established from a labeled datum feature). You're using the first established surface point as a de facto datum for location of the others, which imposes a constraint that is not part of Parallelism. There is no requirement for the Parallelism zone to be located to anything, including the considered feature itself - it can freely translate.

Evan Janeshewski

Axymetrix Quality Engineering Inc.

http://www.axymetrix.ca

 

[CheckerHater]

Frank,
I already said in one of my posts, that one can make a living by producing good parts from bad prints.
But the purpose of this forum is to make 'good' prints, right?

 

[fsincox]

Evan, CH, Dean, drawoh & any others that may have followed along,
I am a supporter of standards so I can accept that the terminology is defined by them and we need to be careful when using terms. The problem is I still need a way to refer to things in our somewhat social context.
I had always referred to a zone of form (yes, I miss you already pmarc, wherever you are,) and orientation as: “all of the features points must be located within a zone of blah, blah…”. If I should not say that, I can accept that you would prefer that I use different terminology in the interest of common understanding. I might now use the word contain, because I had just looked it up, yesterday.
In the end I still need a way to refer to the effect the orientation tolerance has on the surfaces. I believe it is forced to be contained within a .XXX zone that is, as stated. It is established by the datum framework which contains the centroid of the feature itself, therefore, it is by definition on center. In my opinion, establishing a framework like this is very similar, conceptually, to the A-B framework. Some here do not like or understand A-B either, but, it has been in the standard since: probably MIL-STD-8? This in my mind means they lost that argument long ago, (It is particularly OK if you use it for runout, one of my other pet-peeves).
Frank

 

[fsincox]

I looked up the 2009 statement pmarc referenced, which is not in the 1994 standard and I agree, it clearly says orientation does not control location. I can't find their definitition of location can some one reference the page for me, 2009 is fine? I am working and I am doing the best I can to keep up.
Frank

 

[fsincox]

CH,
I agree with your statement.
The important difference I see in the (2) hole block sketch and my example is you were not using datums other than a single perpendicular plane to establish measurement, thus we are not tied to the concept of perfection, that is establised by using a datum framework.
Frank

 

[fsincox]

Sorry, It was 1966 standard.
There is no runout in MIL-STA-8 it is only my dear old friend concentricity defined like rounout another error the standard made that curses the ASME world to this day, IMHO.
Frank