How to specify position of tapered pin?

[pmarc]

Hello,

I am wondering how to specify position of tapered pin relative to a typical datum reference frame (primary datum feature - bottom flat surface, secondary and tertiary datums - side edges of the feature). Can I use 'standard' tolerance of position callout or should I consider something else (like profile tolerance for instance)? I couldn't find any example of such issue neither in 1994 nor 2009 Y14.5 standard.

Regards
pmarc

 

[KENAT]

Are you only trying to locate an actual 'pin', as in a separate item, or is your 'pin' a machined feature that you need to control the surfaces of as well?

Posting guidelines faq731-376

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

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

 

[pmarc]

Kenat,

Let's say the pin has to fit into a hole which is also tapered (see attached picture).

http://files.engineering.com/getfil...150f-465f-b908-8a94fa2850d8&file=example1.JPG

 

[ewh]

Positional tolerance. You are locating the axis, same as when locating a hole (axis).

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

 

[KENAT]

So you're saying you're essentially positioning the hole?

I'd lean toward position too, based on it being similar to a countersink.

Posting guidelines faq731-376

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

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

 

[axym]

Oh no, don't get us started on cones again! The Y14.5 treatment of cones leaves something to be desired and people apply all sorts of "creative" dimensioning to them. I'll try to stay on topic here, KENAT ;^).

In theory, Y14.5 does not allow Position tolerances to be applied to cones. This is because cones do not meet the definition of a feature of size, and therefore do not have a well-defined actual mating envelope, and therefore do not have a well-defined axis.

In practice, it is quite common to see Position tolerances applied to cones. If the form error of the actual cone is very small in comparison to the Position tolerance (which it very often is), a reasonably repeatable axis can be derived using the centers of two circular cross sections. Again, it's not strictly Y14.5 compliant but it's workable.

The location of a cone relative to a datum reference frame can definitely be controlled using Surface Profile.

Evan Janeshewski

Axymetrix Quality Engineering Inc.

http://www.axymetrix.ca

 

[pmarc]

Evan,

Sorry for starting a cone thread once again.
But actually I was very curious how to deal with the issue since a cone doesn't have clearly defined actual mating envelope, as you said. I think profile of a surface is much better in this situation (at least looks to be more in accordance with the standard).
By the way, it is interesting that Y14.5 doesn't specifically say how to proceed with location of this kind of features (especially that such elements are quite common). Other GD&T books I have don't explain the topic either.

 

[KENAT]

axym, I didn't mean to tell you off the other day, just sharing some thoughts.;-)

As pmarc seems interested in delving deeper...

I had a quick look and as you say 14.5M-1994 doesn't appear to directly address use of position with cones even countersinks. It mentions counterbores & position but not countersinks that I see.

However, I'll question if I may why it was determined a cone is not a feature of size. Per 1.3.17 a 'set of opposed elements ... associated with a size dimension' is a feature of size. Would a cone not fit this definition, at least if specifed as a diameter at some given point it's length and an angle? The zone is crummy because you have the size variation of diameter (or distance along the cone) and the angle but does that alone disqualify it?

That said, thinking about it, surface profile would arguably be more robust and I've used it on similar features on kinematic mounts. I suppose to some extent it's going to come down to how critical it is.

At the risk of straying off topic, if cones indeed are not features of size, does this mean countersinks on holes cannot be controlled with position?

Posting guidelines faq731-376

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

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

 

[KENAT]

thread1103-110073 sort of addresses this too.

If you do a search in GD&T forum using the second button on the ribbon below the title of this thread you'll get a number of other threads on related topic, though several of them are about using the cone as the datum not exactly your case.

Posting guidelines faq731-376

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

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

 

[fsincox]

I am sorry, I do not see how you can consider this off topic, it is the topic, "position of tapered..." Now cones and tapers are not the same thing just like diameters and radii. Position away pmarc.

 

[desertfox]

Hi pmarc

I also don't understand why tapered hole centers cannot be dimensioned using position, you are dimensioning the centre axis of the hole, even if you use a profile you still have to define your profile to a datum (I presume central axis) in which case you still have to dimension where the datum central axis is on the component.

desertfox

 

[pmarc]

Hi desertfox,

Which dimension of tapered pin would you associate feature control frame with?
As long as such kind of features don't have one unique size dimension it looks strange for me to apply tolerance of position.

 

[fsincox]

So holes like those shown in the 94 staandard, pg 27, fig 2-6 middle, must really give the taper guys a headache?

 

[desertfox]

Hi pmarc

I was refering to a tapered hole not a male pin.
I thought you were positioning a tapered hole looking at some of the threads.

desertfox

 

[desertfox]

Hi again

I would tend not to use geometrical tolerances on the actual male pin I dimension it with conventional tolerances depending what was important:- angle and diameter, 2 diameters and a length etc.

desertfox

 

[fsincox]

You did say position, too though, right. I would position a protrusion or the mating hole. Taper size is fine by me as you wish. Many times the "ends" are not used to keep separate length issues from taper issues and/or the ends are not really there.

 

[NormCrawford]

Profile is usually the best way to control tapers/cones.
One could use position of 0@mmc for the two end diameters for the "coaxial" control, which effectivly does define a virtual condition.
But I go with profile.
Sometimes the internal feature can be an issue and so using runout while stating the actual "taper/slope" is sometimes used too.
As others have said, cones can be creative. But profile is the cleanest way IMHO.

Norm Crawford
GDTP-S
Applied Geometrics, Inc.

http://www.GDandT.com

 

[fsincox]

NormCrawford,
Are you speaking of control of size, location or both?

 

[NormCrawford]

Both with profile

Norm Crawford
GDTP-S
Applied Geometrics, Inc.

http://www.GDandT.com