What Is The Correct Position Tolerance Zone Of A Sphere To A Plane 1

[fsincox]

I want to position a sphere to a plane surface. The surface for the plane is established and referenced as the datum. I am positioning a spherical feature to this plane. Is the correct tolerance zone to reference in the position callout a sphere, cylindrical, or none? The tolerance control imparted by the plane to the sphere centerpoint is only a band of width as I see it because all I have establised is a plane, what do you think?

 

[MechNorth]

Evan, what of the uni-directional position control? No diameter symbol there. In that case, you are both identifying the uni-directionality of the zone AND reinforcing for inspection what the tolerance zone shape is. By leaving the SØ or Ø symbol off of the control in the OP, you are doing the same. To me, that's the more effective communication to be made, otherwise inspection is more/most likely to over-/mis-interpret the SØ or Ø as being relevant.

So, as for legality, I'd have to say it was legal either way. For practical reasons, I would tend to drop the SØ or Ø, and/or illustrate the tolerance zone on the drawing so that there was no confusion.

I thought my comment over debating it might be provocative ;~} Sometimes you just have to push the right buttons. More specifically though, I was concerned that people may not be visualizing the correct tolerance zone rather than whether or not the symbology was legal.

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

http://www.profileservices.ca

TecEase, Inc.

http://www.tec-ease.com

 

[axym]

I would say that the unidirectional position control is different, Jim. In that case, the parallel-plane tolerance zone is defined by the drawing view and by the (lack of a) tolerance zone shape modifier. The zone description isn't affected by what the datum features are, so I'm okay with it.

I still say that leaving the spherical diameter symbol off just because there is only a single datum feature is fundamentally flawed. I see your point about providing effective communication, and the temptation to simplify things for the benefit of the user. But when this crosses over into oversimplifying or introducing an incorrect concept, I get very uncomfortable. In the OP's example, I think that describing the tolerance zone as the volume between two parallel planes introduces a deeply incorrect concept for the benefit of making a certain special case a bit easier to understand.

If we describe the tolerance zone as the volume between two parallel planes, the user could easily acquire the incorrect concept that datum shift makes the tolerance zone get bigger. In the OP's single-sphere example, this gives a similar end result as a spherical zone that translates. But the single spherical considered feature is a very special case, with unique properties. If the incorrect concept was applied in other cases with a more complex feature or more than one feature, then major problems would result.

Evan Janeshewski

Axymetrix Quality Engineering Inc.

http://www.axymetrix.ca

 

[MechNorth]

Agree with your last statement, Evan, however we are talking about the OP scenario.
As an alternate, per '09, the DRF could include a "[z]" modifier to show that it only constrains the one DOF. I'm thinking that would be the best solution (with a diameter symbol on the tolerance value).

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

http://www.profileservices.ca

TecEase, Inc.

http://www.tec-ease.com

 

[Belanger]

I still don't buy this. No diameter symbol should be shown.
A feature control frame dictates what the tolerance zone will look like, and that relies heavily on the datum references, and also on how the FCF is attached (as in the case of bidirectional tolerancing shown in Y14.5).
If a sphere is being positioned at a distance above a planar datum, and that's the only datum referenced, then the only thing the FCF can do is control that distance; the sphere could be anywhere in the other directions (within other limits, presumably) and orientation is not controlled. So in this specific example,what value does the dia symbol add, other than to confuse people?

John-Paul Belanger
Certified Sr. GD&T Professional
Geometric Learning Systems

http://www.gdtseminars.com

 

[MechNorth]

None at all, J-P. It adds no value whatsoever. Is it technically incorrect though; no.
As a bit of a mental exercise for a comparable situation, consider the use of a position control in place of a perpendicularity control; is it legal, yes, but is it practical, no. Is it easily understood to be just perpendicularity, no. Should it be done that way; again, no. But again, is it legal; technically yes based on the hierarchy. BTW, I was once asked specifically about the use of position to control the perpendicularity of something (not location); I answered that it was technically legal but ill advised and shouldn't be done because it takes a significantly higher level of GD&T understanding to figure it out. Some time later, I get calls from my boss & from management at the company I was teaching at indicating that I had told some of their people that it was perfectly ok for them to use. When it comes to these kinds of "technical vs practical" debates, sometimes people will use "technicality" to support a poor argument. Per my previous post, by "best", I meant as a compromise to the polarized perspectives. Personally, I would tend toward no diameter symbol, but adding the [z] modifier to further guide the user to what I expect. I don't necessarily consider that to be accommodating the lowest link down the chain, but rather to (hopefully) minimizing my downtime answering questions.

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

http://www.profileservices.ca

TecEase, Inc.

http://www.tec-ease.com

 

[Belanger]

Sorry, Jim, but I would say the standard doesn't allow position to be used to control only perpendicularity. It's not a big deal, but the check out the defintion of position and true position: Y14.5 states that position's job is to control location ... and by the way it may also do perpendicularity, etc.

So if the core purpose of a symbol is missing (i.e., location), then I think it's clearly not being done correctly per Y14.5. Fodder for another post! (If we haven't done it tangentially several times...)

As for this sphere example, Y14.5 doesn't have any black-and-white text to prohibit the dia symbol, so we agree that it's not valuable but not illegal.

John-Paul Belanger
Certified Sr. GD&T Professional
Geometric Learning Systems

http://www.gdtseminars.com

 

[fsincox]

JP,
I would agree with your final assesment, The discussion here is very similar to the discussion we have had here and I was considering printing the results out and giving them to manufacturing and inspection. We have discussed a similar issue here, before, with cylindrical zones without the complete datum framework to support them. I supported that by a reference in Y14.5.1M-1994. I wanted to see if the general opinion was still the same.
Thanks,
Frank

 

[MechNorth]

J-P, the statement is inclusive, not exclusive. Does it say that it doesn't provide the subordinate controls if the higher level control is not applicable? No. As we've said, and supported, many times on this forum; just because it's not provided directly in the standard does not mean that it's not permitted. Extension of principles does still get you there, and it is not precluded. Thus, technically it is legal. Again, though, that doesn't mean it's not a stupid thing to do. On a technical level, we need Position, Surface & Line Profile, and maybe the Runouts (though arguably Profiles can substitute)...the rest are just subsets / subordinates of these controls, aren't they? As I've been reminded many times, the high-level understanding of the full robustness of these controls came from a comprehensive understanding of the lower-level controls. Everyone (theoretically) understands the lower level controls, so you wouldn't just eliminate them for the sake of reducing the number of symbols. I agree, BUT ... if people understand the hierarchy and how the highest-level controls provide the lower level controls, then they have a far more comprehensive knowledge of GD&T.

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

http://www.profileservices.ca

TecEase, Inc.

http://www.tec-ease.com

 

[fsincox]

I stated in another thread, I think it was one on runout used on partial sufaces, that it seemed a more generic way of specify the control we all use runout for, if we wanted to be true to the "not specifying process" would be to use profile as an option. the poor abused profile of a line with tolerance dimensions would make sense also (i geuess it might have been the "use of datums with porfile of a line" thread. Dis I miss some earlier discussion on this type of subject, here?
Frank

 

[axym]

Here's another case to consider, that I'm hoping will shed some light on our discussion.

What if instead of one sphere we have a pattern of two spheres, that we want to position to the same planar surface? The two spheres are nominally equidistant from the datum plane, and the spacing of the spheres has a basic dimension. What tolerance zone shape modifier should be used in this case?

Evan Janeshewski

Axymetrix Quality Engineering Inc.

http://www.axymetrix.ca

 

[axym]

Here's another case to consider, that I'm hoping will shed some light on our discussion.

What if instead of one sphere we have a pattern of two spheres, that we want to position to the same planar surface? The two spheres are nominally equidistant from the datum plane, and the spacing of the spheres has a basic dimension. What tolerance zone shape modifier should be used for the Position tolerance in this case?

Evan Janeshewski

Axymetrix Quality Engineering Inc.

http://www.axymetrix.ca

 

[MechNorth]

Two days with a bad headache ... now I'm seeing double???

Evan, spherical tolerance zone then.

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

http://www.profileservices.ca

TecEase, Inc.

http://www.tec-ease.com

 

[axym]

Jim,

I agree that the tolerance zone should be spherical. We have a pattern of two spherical zones that are at a certain distance from the datum plane and can freely translate parallel to it.

In the original single-sphere example, we had one spherical zone that is at a certain distance from the datum plane and can freely translate parallel to it.

To me, the shape of the tolerance zone for a given considered feature should not be affected by the presence (or absence) of other considered features or by the DOF constraints that the datum features impose.

Evan Janeshewski

Axymetrix Quality Engineering Inc.

http://www.axymetrix.ca

 

[fsincox]

Evan,
If I understand you are saying the shape matchs the feature shape?
Parallel for width, cylinder for cylinder, sphere for sphere or am I oversimplifying?
Frank

 

[Belanger]

Evan, surely the presence of a second sphere changes things. That's because when two or more features are positioned with the same FCF, each inherently becomes a datum for the other. Thus, the necessity for SØ.

John-Paul Belanger
Certified Sr. GD&T Professional
Geometric Learning Systems

http://www.gdtseminars.com

 

[ptruitt]

So do we have an opportunity for spherical bonus tolerance with the position of two spheres??? (Please say "No".)

Peter Truitt
Minnesota

 

[axym]

Frank,

Yes, I would say that for Position tolerances, the default is that the tolerance zone shape matches the feature shape. Unless something else is specified, such as the unidirectional tolerance zone technique or bi-directional Position tolerancing.

J-P,

No, in my mind the presence of a second sphere doesn't change a thing. As far as each feature in the pattern being a datum for the other, I wouldn't put it in exactly those terms. I would say that neither is the datum, but they exist in the same 3D space. The relative location is controlled.

Peter,

Sorry, it's an emphatic yes. There could be bonus tolerance with the position of the single sphere too. If the Position tolerance was referenced at MMC (or LMC), then there would be bonus tolerance. The AME size of the as-produced sphere would have to be measured and compared to the MMC size.

Evan Janeshewski

Axymetrix Quality Engineering Inc.

http://www.axymetrix.ca

 

[Belanger]

Right, the relative location is controlled. (How do you find relative location? By temporarily calling one of them a datum!)

But you say you don't see the difference between having a SØ with those 2 spheres vs. not having a SØ? Without any modifier in front of the tolerance value, your basic dimension between them would be useless because the lack of a modifier means 2 parallel planes.

John-Paul Belanger
Certified Sr. GD&T Professional
Geometric Learning Systems

http://www.gdtseminars.com

 

[MechNorth]

I think that the use of the term "datum" in J-P's post didn't mean literally a "datum", but rather that the spatial relationship to each other feature is controlled, as if each were a datum to the others. The general concept is right, if the terminology may have clouded it.

I, too (mentally at least), typically go for the same tolerance zone shape as the feature ... as an initial consideration (you have to start somewhere, and that way a symbol is there for cylindrical & spherical zones).

Yes, Peter, bonus tolerances are available on spherical zones as for cylindrical or rectangular zones.

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

http://www.profileservices.ca

TecEase, Inc.

http://www.tec-ease.com

 

[axym]

J-P,

Sorry, I misunderstood what you meant by "changes things".

I agree that the spherical diameter modifier should be used when there are 2 spheres. The modifier definitely does make a difference.

But I think that the spherical diameter modifier should also be used when there is only one sphere. The presence of the second sphere doesn't change things, as far as the tolerance zone shape modifier is concerned. To me, it's spherical diameter in either case.

Evan Janeshewski

Axymetrix Quality Engineering Inc.

http://www.axymetrix.ca