FCF on Reference Dimensions? 1

[serspec]

I recall coming across this before saying that there should not be a FCF on a dimension that is reference. I was trying to find where I saw it before and seem to not be able to find it. What is the rule with this?

I attached a sketch of what I'm talking about.

Thanks for the help.

 

[ewh]

This is allowable if the part features were already defined, either in a separate drawing or elsewhere on this drawing. The FCF is locating the feature of one part to that of another. Since these features have already been defined elsewhere, reference dimensions are appropriate.

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

 

[serspec]

Thanks a lot. I figured that was the case since I couldn't find it anywhere but for some reason it caught my attention. Thanks again.

 

[axym]

I'm not sure that this is allowable, even if the sizes were defined elsewhere. But I don't have any specific references to back that up.

In any case, I think it's a confusing practice at best. Why would a feature's size tolerance and geometric tolerance be shown in different sections of the drawing? The same thing goes for the datum feature's size tolerance and its datum feature label.

Evan Janeshewski

Axymetrix Quality Engineering Inc.

http://www.axymetrix.ca

 

[PeterStock]

It is not only permissible, but may be required in the case of serspec's attached sketch. The dimensions shown may be from the part were the tolerances were defined. The assembly should show the requirements at the assembly level, should the diameters not change during assembly then the dimensions used for the FCF should be reference. As alternative just use a FCF without a numerical dimension.

Peter Stockhausen
Senior Design Analyst (Checker)
Infotech Aerospace Services

 

[PaulJackson]

Paragraph 1.4 of ASME Y14.5M-1994 states that "Dimensions and tolerances apply only at the drawing level where they are specified." Without the tolerances for feature sizes the variable portion of the positional tolerance and allowable datum shift resulting from the MMC modifiers cannot be determined.

If this part an in-separable assembly then it is a new part and it can (should) include each necessary dimension Paragraph 1.4(c).

I know that this is not often the practice in design because assemblers claim that they cannot be held liable for sub-assembly "pass-thru" deviations... while the sub-assembly operation claims that the assembler introduced the deformation that caused the previously qualified feature specification to fail.

With in-separable assembly operations un-predictable things happen and when they do new specifications end-up on the assembly drawing despite all objections.

Paul

 

[serspec]

The part is an in-separable assembly and the two components diameters are called out in their part level drawings. Our policy here on assembly drawings like this is to always include the inner and outer diameters and lengths as reference on the final print. I just wasn't sure if since this particular part, or 3 parts, has a FCF associated with the dimensions, should those be regular dimensions instead of reference. This came up because they currently are regular diameter dimensions and they are all being changed to reference. Just wanted to make sure that the change wouldn't be violating any standards.

 

[KENAT]

I can certainly see this kind of thing being necessary for some assemblies. We have a lot of nominally coaxial items that have to be glued together at the assy level but don't have features to ensure alignment. To my mind this is a classic case where you might want to do this type of tolerancing.

KENAT,

Have you reminded yourself of faq731-376 recently, or taken a look at posting policies:

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

 

[PaulJackson]

From what you described… I would say that it violates the standards that I cited... but politics in quality assessment typically intervene and the necessary size tolerances are omitted… until the part fails to function as predicted… then the critical specifications are added as I stated.

Paul

 

[ewh]

If this part an in-separable assembly then it is a new part and it can (should) include each necessary dimension Paragraph 1.4(c).

The key here is the word "necessary" and how you interpret it. If the feature was created on another drawing and NO intent to remachine it is inherent in this drawing, then the dimension isn't really necessary for part definition.
I tend to agree with PeterStock that it may be better to exclude the dimensions altogether, unless there are similar features that may cause confusion.


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

 

[PeterStock]

For an inseparable assemply, the OD in the posted drawing would likey have changed from the as produced state. The new dimension should include that effect and not be reference. If control of the as assembled OD not be important then a referene dim or no dim should be used.

Peter Stockhausen
Senior Design Analyst (Checker)
Infotech Aerospace Services

 

[serspec]

The inner and outer parts are metal and they get bonded together with rubber, the OD would not change in this situation.

 

[TheTick]

The real problem is the MMC without a size. Without a size, the FCF could exist without the reference dimension. Attaching to the reference dimension is incidental.

The MMC qualifier requires a size tolerance. Without a size tolerance, the MMC (and entire FCF) are invalid, regardless of what they are attached to.

 

[KENAT]

Another time we put FCF on reference dimensions is on threaded inserts in inseperable assemblies. We call out the item, call out the thread size in reference and then attatch the locating FCF to it.

KENAT,

Have you reminded yourself of faq731-376 recently, or taken a look at posting policies:

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

 

[ewh]

Right you are, Tick. Thus the component drawings would have to be at hand to inspect the FCF.

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

 

[KENAT]

Or, could you repeat the tolerances from the component level on the assy level in the ref dim? Obviously ref dims usually aren't toleranced but is this a case where it may be appropriate.

Though, from a configuration control point of view that would potentially be a pain so having the part drawings to hand may be better.

KENAT,

Have you reminded yourself of faq731-376 recently, or taken a look at posting policies:

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

 

[ringster]

Kenat,

What is/was gained by using a FCF on the insert internal thread?
Was the insert a floating variety or fixed?

 

[KENAT]

"What is/was gained by using a FCF on the insert internal thread?"

Um, am I misunderstanding you? What was gained was that the required location tolerance is clearly stated. Position is used to take advantage of the increased tolerance from the circular tol zone and support its use on mating part.

Fundamentally, from an interface/function point of view what we care about is where the thread of the insert ends up, not where the OD or any of the other radial features on the insert end up. If you don't somehow state which dia the FCF applies to then there could be confusion or parts not fitting.

KENAT,

Have you reminded yourself of faq731-376 recently, or taken a look at posting policies:

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

 

[KENAT]

I may have realized the confusion, I'm not talking about helicoils or similar, more like pem nuts/studs and the like.

KENAT,

Have you reminded yourself of faq731-376 recently, or taken a look at posting policies:

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

 

[PaulJackson]

The inner and outer parts are metal and they get bonded together with rubber, the OD would not change in this situation.

Serspec, Your statement is a prediction! If true then the assembly process may not introduce variation and the reason to omit the size tolerance would be an economical choice. There are however, other factors that contribute variation in assembly that smart designers often fail to predict... thermal influences, effects of stress on the sub-assembly pieces, effects of process influences (forces) on the sub-assembly pieces, and other unknowns. I have witnessed many design assumptions for in-separable assemblies go awry. That is why I cautioned that unpredicted variation typically results in feature controls that are added afterwards.

TheTick, I agree.

Paul