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]

Ummm... picture?

 

[drawoh]

fsincox,

Your example sounds like the way I model stuff in SolidWorks. If a slot is located to a diameter, the only additional control I need on it is orientation.

Having said that...

The GD&T equivalent would be to call the shaft diameter as a datum, and locate (position) the slot to it. Now, you use parallelism, perpendicularity or angle to control the orientation.

This is a useful approach, if orientation is more important to you than location. If the key has clearance, a significant position error may be functional for you, while any angle error will cause stress concentration, even if you could assemble it.

JHG

 

[fsincox]

CH,
Lets try this, ASME Y14.5-2009, fig 4-32, pg 74.
In the example picture shown, you see a slot clearly drawn offset from the center of a base diameter (datum A) when the axis is drawn through the (2) holes as shown. If instead you forget the holes and find the centroid of the center of tha slot (keyway) and roll the picture horizontal to that "framework" the sides of the slot that were drawn perfectly parallel but offset will become centered but angled to that A (primary), B(secondary) alignment.
Frank

 

[fsincox]

Ideally keys are fit tight (RFS) so I was told, thus making size not as important as this tendency to crawl out of the slot under loading.
Frank

 

[fsincox]

I realize this is an orientation tolerance being used to control location only because of the way the framework is being established, it really is orientation as orientation, so I really could not say no. It also involves the concept of a datum to a framework including itself sort of thing which is legal IMHO.
Frank

 

[pmarc]

Frank,
Not sure if I imagine the situation correct, but coming back to your question from the title of the thread, notice that Y14.5-2009 clearly states in 6.2: "An orientation tolerance does not control location of features".

So if your situation is that centerplane of the keyway is designed to cross the axis of the shaft exactly in the center and a parallelism callout is applied to the keyway (to its centerplane or to both surfcaces), you can actually end up with the geometry like shown on 4-32 and the part will meet the print.

 

[fsincox]

You are right, basically the slot can't be off, it is just what happens to the sides when forced onto location. The orientation is not applied to the center, only the opposed sides. It did seem to be similar to what actually would be forced to happen in assembly.
Frank

 

[drawoh]

fsincox,

My interpretation of ASME Y14.5-2009 Figure 4-32 is that the holes are located from the slot. The slot is located exclusively from the centre, datum[ ]A. The slot defines the orientation of the part.

In your model, as I understand it, the hole pattern is vertical. The slot is centred on the OD, datum[ ]A, and at an angle.

If this were my drawing and the slot still was datum[ ]B, the slot would be horizontal, and the hole pattern would be at an angle.

JHG

 

[fsincox]

I am just using the picture for an example because it is a picture of a slot that is "offset" when the axes are aligned normal to the hole pattern. If there are no holes and the alignment of the axes goes throuh the centroid of the slot (roll the book about the center axis of datum A so the slot centroid point is horizontal). The sides will skew (the sides are still parallel to each other but not the newly established axes).
Frank

 

[CheckerHater]

Frank,

The way I see it, as soon as you remove round holes from Fig. 4-32 the round shaft will only be good to establish infinite number of planes, and being parallel to infinite number of planes will not locate you to anything.

I have to go home and have a drink.

 

[drawoh]

fsincox,

Let's use Figure[ ]4.32 as is.

Rotation of the part is defined by the slot, designated as datum[ ]B. We know, among other things, that the slot faces perpendicular to the drawing view, are not absolutely parallel to the outside diameter, datum[ ]A. The allowable error is a component of the size tolerance, and the position tolerance called up on the drawing.

I need to read up on the translation modifier. I have not seen that before. I don't think this affects my point, above.

I would add a parallelism specification to this if I needed parallelism much more accurate than what the existing tolerances impose.

JHG

 

[fsincox]

CH,
True, I agree there are infinate axes that are established from datum A, however, only one also goes through the centroid of the slot secondary datum B. Sorry, there is no offset dimension. the illustration was to represent the real poroduced part with the slot actually out of location, the drawing for this part would show the slot on center and no baxic dimension.
Frank

 

[fsincox]

I was only using that picture as an eample of a possible actual produced part and to give a picture to look at as CH had requested. On the drawing for this part there is no position tolerance on the slot and no basic dffset dimension, therefore. the drawing for this part would imply the slot on perfect center. The diameter is still datum A and the slot is datum B. On the drawing there would be a callout on each of the sides of the slot as follows: parallel within say .001 to datum A primary, datum B secondary. RFS would be implied.
Frank

 

[CheckerHater]

Providing that

1. You can derive datum from physical feature and not imaginary line/plane,
2. You only have 2 features,
3. You cannot reference feature to itself.

This is the only way I could imagine it to be done, and I am not 100% comfortable with it

 

[DeanD3W]

As pmarc stated above, orientation tolerances do not ever locate any element of any feature. Their tolerance zones cannot be constrained in location. Perpendicularity, Parallelism, and Angularity tolerance zones are orientation constrainable only, they are not location constrainable.

Dean

http://www.d3w-engineering.com

 

[DeanD3W]

Please pardon my run-on sentence "They are not location constrainable." should have been its own sentence.

 

[fsincox]

CH,
Thank you for the new picture it is almost right and will work good as the "part as shown on the drawing" with just a few modifications.
The changes to the picture are:
1) the parallelism callout is not on the centerplane of datum B, the parallelism callouts (2) are on each of the sides of the slot.
2)the referenced datum structure for the parallelism is: A (primary), B (secondary).
Frank

 

[pmarc]

Frank,
Okay, I guess we know how your part and dimensioning scheme look like. But what is your point exactly? Are you asking whether those parallelism callouts control position of the slot? If that's so, the answer is no. You need something else to control location of the keyway, otherwise geometrical characteristic of the feature is incomplete.

 

[fsincox]

If the callout is alowable, and I do not really see why it would not be, The control specified is simply orientation to a framework.
That was my only point here that some things are not as they seem and here is a case where orinetation ends up controlling location through the back door, so to speak.
Frank