Close Tolerance Taper Pin - Correctness of Drawing and Measurement Results 1

[goldgunner]

Questions abound on this piece.
Please see attachment.
The red line detail is the metrology lab mark-up. Page 2 is the measurement results using a vision system, rather than the drawing note # 6 Since bar method.

Your input as to accuracy of the drawing - conflicting datum's or not? And, as to accuracy of the measurement methodology as to extrapolation of tolerances for the 'Angle' [# 4], as well as all other design and tolerance characteristics, is greatly appreciated.

This is looking like a deviation submittal to the user-customer for use 'as is'.
Some question whether the item is producible as drawing - specified, also.

Thanks very much, in advance.

 

[MikeHalloran]

The small end diameter is a reference dimension; no one should be inspecting it.

I'm curious about exactly how the video system measures the taper angle; I'd tend to trust the sine bar more.

In order to report the angle and the taper, did the metrology lab have to make calculations based on their raw data? I suspect as much, given that they apparently computed the nominal small diameter and marked it as basic. I don't trust calculated measurements much, especially if the calculations don't appear in the report.

I'd suggest that, if you're inclined (no pun intended) to give up the sine bar, you find out exactly how the metrology lab calculates what they reported, and redimension the drawing so that it still reflects the limits of what you can tolerate, and requires no calculations; they just write down what they read/measure, directly.








Mike Halloran
Pembroke Pines, FL, USA

 

[3DDave]

There's no way to use a vision system to measure the profile of surface tolerance on the entire surface simultaneously as required by the callout. The most practical measure is to make a precision tapered glass gage that accepts this part and is filled with an opaque liquid that fills the gaps. The visible density of the liquid can be calibrated and the gage turned with the part immobile inside and the visible density used to determine if the part meets the requirements.

The sine bar and v-block is also inappropriate as it does not inspect all the surface simultaneously. If the part was fixed between precision centers so that it could be turned without surface irregularities affecting the position of the part, that might also work.

I don't see what datum A is or is used for.

The taper should be basic.

 

[3DDave]

Also:

Isn't the .250 per foot taper redundant with the [1.193]degree dimension? How did they determine the plus-minus tolerance on a basic angle?

The profile tolerance is .0001 over a length of .75 inches and can add or subtract from the half-angle; the included angle might vary as +/- 2*atan(.0001/.75) or +/- ~.015278 degrees; they say the limit is +/- .0053 degrees. How odd.

There is also no valid application of the profile tolerance to the taper value, as indicated by the tolerance applied to dimension 3.

They don't really understand what the callouts mean, but then neither did the creator of the dimensioning scheme.

I'm not even sure of some of their math. .250/12 = .02083333, not .015625.

 

[ewh]

As for the taper, .250/16 = .015625 (the part is .750 lg, not 1.000), so Ø.234375 is more correct than the given Ø.2345.
I'm confused as to what establishes datum A or why it is included.

As an aside, I'd advise a little more discretion at posting so much company info. That company used to be fairly conservative about such matters back when I was in their employ.

"Know the rules well, so you can break them effectively."
-Dalai Lama XIV

 

[CheckerHater]

Who wants to bet that at the end of the day this pin is put in its place by sledgehammer?

"For every expert there is an equal and opposite expert"
Arthur C. Clarke Profiles of the future

 

[3DDave]

ewh - I see what you mean. An unnecessary calculation, left with the work incomplete - unlike the way you spotted my error.

CH - I'll see your bet and raise you that no one looked at the effect on interference with the tapered hole the acceptable variation in the part will produce. As usual, a d&t problem which is likely to be a stress analysis problem that no one bothered to do.

I wonder why it's a VICD. The supplier part numbers are identical to each other and to the drawing number; this isn't typical for a VICD. It's a fully described part; there's no reason to make a VICD.

I'm a little suspicious of their calculation on the V-block, but am working through the geometry to be certain.

 

[CheckerHater]

Oh yeah, placing conical pin into straight v creates some nice geometry problem. I wish I as 20 years younger or had more time on hands

But this brings me to another question: When we look at the whole contraption - the V, the sine plate, pins, the spacer - shouldn't the whole stack-up to be done to "gage-maker tolerance" which in this case should be 0.00001?

"For every expert there is an equal and opposite expert"
Arthur C. Clarke Profiles of the future

 

[goldgunner]

Thank you, Gents, for your interest and detailed responses.
Please keep it up - this remains an active issue.

[Yes, should have redacted the attachments to remove the company names. Would do so now, but I'm not finding a message edit capability yet]

Thanks again

 

[KENAT]

goldgunner, I believe you only have 24 hours to edit posts and then they lock. However, if you 'redflag' the post management may be able to switch the link to a redacted version of the drawing.

Posting guidelines faq731-376

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

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

 

[goldgunner]

Redacted drawing and inspection versions now attached.