Location of a sphere

[pmarc]

I am curious your opinion about following issue:

Let's say I have a part like on attached picture, but I would not want to use spherical tolerance zone, because the location of the knob relative to datum C is not so important to me - it can be greater in this direction.

I am wondering how this requirement could be defined. Of course I have some ideas, but I would like to compare them with other opinions. Any hints?

 

[PaulJackson]

Two position callouts:

|Position|spherical diameter "BIG"|A|B|C|
|Position|spherical diameter "SMALL"|A|B|

 

[pmarc]

Thanks PaulJackson, I like this one, however your example still keeps spherical tolerance zone active.

My thinking was rather to somehow define cylidrical tolerance zone for a sphere's center location. Para. 5.15 of Y14.5M-1994 standard (or 7.4.6 of 2009 version) says that: "Where it is intended for the tolerance zone shape to be otherwise, a special indication is shown, similar to the example shown for a bidirectional tolerance zone of a cylindrical hole"

 

[TheTick]

You can use a directional tolerance. Attach tolerance frame to linear dimensions that describe the sphere size (one will need to be a reference dimension). I have done similar with slot positions.

 

[fsincox]

This thread is similar to the kind of ideas I was exploring in my thread on cylindrical zones, thread1103-271079. I ended up wishing I had titled it: "limits on cylindrical or spherical zones" as I was also interested at looking at limits on shperical zones, there. I believe you are not required to specify a spherical zone if the datum structure referenced for it does not support it. I would suspect the AB framework Paul referenced could be a cylindrical zone.
Frank

 

[KENAT]

Would you not just extrapolate out 'bidirectional tolerance zone"? This has been discussed a few times befor, maybe you can find it.

Posting guidelines faq731-376

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

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

 

[restedup]

pmarc,

There has not been any evidence that would necessitate a spherical tolerance zone for the sphere. Might you show the assembled condition?

I personally believe that three bidirectional tolerance zones would be the answer to your situation.

 

[pmarc]

restedup,

I do not have any practical example to present - I was only imagining a situation when a functional requirement determines positional tolerance zone shape to be cylindrical not spherical.

What you are proposing is making tolerance zone shape cubicoidal, so this is not exactly what I wanted to achieve.

 

[MechNorth]

I was rather concerned that the guidance was heading toward a cubic or rectangular tolerance zone.

Generally, if you want the center of a sphere controlled ... you are going to use a spherical tolerance zone. If you're only worried about the boundaries of the surface ... that's a call for profile of a surface.
In the first case, you could use a gross spherical tolerance zone for the first FCF, then a cylindrical zone for the second separate FCF (i.e. two FCFs stacked up as opposed to a composite positional control because the second level of a composite control is a refinement of orientation, not location, as in '94). In the second case, two stacked profile controls could be used; the first to control the size & location of the surface wrt A/B/C, the second to refine the size/location wrt A/B only.

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

http://www.profileservices.ca

TecEase, Inc.

http://www.tec-ease.com

 

[restedup]

Mechnorth,

I believe that is the direction that should be taken. If you consider that the datum features in this case are planar and the locus of points a given distance from the planes would result in a cubicoidal or rectangular zone, andthat is where we should be with the definiition. Perhaps if the shperical features were more that one, consideration could be given to another approach to accommodate the assembled relationship of multiples. However, there being a single sphere, not required.

The same would apply to a single hole or pin in a rectangular plate. Diametric pos tol not required. Only for the assurance of assembly when a pattern is used.

 

[MechNorth]

It's not that common that people actually want a rectangular tolerance zone ... that would be my first issue.

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

http://www.profileservices.ca

TecEase, Inc.

http://www.tec-ease.com