Controlling Radius Size Within a Profile

[jszymkowski]

What is the best way to control the size of a radius within the boundaries of a profile?

In the attached sketch, the nominal line in blue is flanked by the min and max profile boundaries. An undesirable condition is shown in pink. What controls are best utilized to ensure that the corner radius of the measured part does not exceed the min/max radius of the profile limits? Specifying a locational tolerance for the center of the radius seems logical, but presents it's own limitations when measuring a semi-circle less than 180°

I can't seem to attach the file, so here's a direct link:

http://img.photobucket.com/albums/v208/jszymkowski/web/Profile.jpg

 

[drawoh]

jszymkowski,

Why not just apply a tolerance on the radii that is tighter than the profile tolerance? The inspector would have to check both the profile tolerance and the radius value. A true position tolerance would not mean anything.

Alternately, you could apply a tighter profile tolerance to the radii. What are you trying to accomplish?

JHG

 

[jszymkowski]

Wouldn't the application of a tighter tolerance profile around the corner radii result in a sudden step from the end of one profile at the tangency to the beginning of the next? Perhaps tangency is implied?

The application of a tolerance to the radius in addition to the all-around profile should accomplish my goal. The goal being to prevent a large radius occuring at MMC or a very small radius occuring at LMC such as depicted in the picture.

I had thought of the solution you provided, but it seemed somehow unorthodox and perhaps invalid.

Thanks.

 

[TheTick]

It almost looks like you would need to double-dimension the radius: once w/ basic for the profile and once with a tolerance.

 

[fcsuper]

Have you looked Controlled Radius (CR) in the ASME Y14.5M-1994 standard (or are you taking this into account)?

Matt Lorono
CAD Engineer/ECN Analyst
Silicon Valley, CA
Lorono's SolidWorks Resources
Co-moderator of Solidworks Yahoo! Group
and Mechnical.Engineering Yahoo! Group

 

[jszymkowski]

A controlled radius requirement would not prevent the condition I wish to avoid. It would, however, cause the heads of people in our Quality department to explode.

 

[weavedreamer]

Here is a thought:

 

[weavedreamer]

Now that I look at your second post, jszymkowski, that may be what you were suggesting. Perhaps moving the D and E further out on the 'flats' coupled with a Basic Dim locating D and E would cover it.

 

[drawoh]

jszymkowski,

A smaller profile tolerance on the radii allows a step between the radii and the straight sides. Does this matter?

What geometry are you trying to achieve?

JHG

 

[powerhound]

Why do you think that CR would not prevent the condition you are trying to avoid? Your arc would start and end tangent to the profile per the tangency rule and no flats or reversals are allowed along the profile it must be one sweeping arc. See Fig. 2-19. The line in pink would violate the CR specification if CR were to have been applied to the drawing.

Powerhound, GDTP T-0419
Production Supervisor
Inventor 2008
Mastercam X2
Smartcam 11.1
SSG, U.S. Army
Taji, Iraq OIF II

 

[ringman]

I might suggest that you back up and determine the size of the radius to be functional. Drop the profile tolerance on the area of the radius and add a controlled radius. That would appear to be more appropriate.

I think your concerns are slightly misguided, in applying the profile and additionally the radius.

Hope this might be of some help.

 

[blywpr]

Not sure how you have this dimensioned, but if you have profile all around then you would get the results shown in your sketch. The only option, that I can think of, and correct me if I am wrong, but I would profile each edge individulally, use an edge to create a datum and true position the radius from that datumed edge. I think this would control the radius with respect to where the edges lie if this is what you are looking for.

 

[Xplicator]

The scenario you describe unfortunately is one of those shortcomings in the current standard regarding transitions. The draft of the future Y14.5 standard has I believe attempted to address this with the nonuniform surface of profile callout. See fig 8-9 and 8-11. It uses basics R dims and offset centers to phantom lines to depict a nonuniform tolerance zone to the transition.

Remember that even though we all do what we can to adhere to the standard, there are times when a situation calls for a very acceptable option and this is a standard symbology supplement note (see para 3.2) to achieve what we need for functionality. Also another higher level example of further clarification or supplement is the GM/Chrysler auto industries addendum.

For your particular scenario I would use a Local NOTE on the drawing and a detail view pictorial depiction of the tolerance zone using basics and phantom lines as earlier stated. I believe if you have access to, and my description isn't clear enough, the aforementioned figure from the draft standard is a catalyst for you to get the idea on how to apply this.

As long as you are clear and unambiguous, it is perfectly acceptable to supplement or amend the standard to address your particular situation.

I hope this helps.

 

[ringman]

Xplicator,

What if the radius between D and E were controlled with a controlled radius rather than Profile?

 

[Xplicator]

Ringman,
The controlled radius in my understanding, vaguely defines a refined surface finish on the radius feature itself. It does not in my opinion address the transitions at the two tangent surfaces. Now one could interpret it different then me and there in lies the problem it is not in its current state a clear enough spec and leads to ambiguity. Therefore, I suggested to the OP what I would do, to specify unambiguously the functional intent. We are always allowed this option for things that aren't clearly defined in the standard.

Which, by the way we don't really know what the actual function of this radii is. So as always with out all the info, this is purely speculation.

 

[fcsuper]

I suggested CR as well. It takes the tangent into account. The problem I see is at MMC and LMC, where the CR would allow the radius to be outside of the desired profile.

I would suggest making the radius itself basic and rely solely on profiles if this is the concern. It also elimenates the apparent double dim'ing.

Matt Lorono
CAD Engineer/ECN Analyst
Silicon Valley, CA
Lorono's SolidWorks Resources
Co-moderator of Solidworks Yahoo! Group
and Mechnical.Engineering Yahoo! Group

 

[ringman]

Jxsymkowski,

I believe that if you take a look at Par 2.15.2 of Y14.5 you will agree that CR will do task you are desiring.

 

[TheTick]

I need to recant my last post. I was unaware of CR. After reading up on CR, it seems to be what you need.

 

[ringman]

Jszymkowski,

What is there about the original profile tolerance that causes a problem?. Is it interference, your misgiving about the resultance appearance or concerns for the inspectors being able to verify?

 

[Xplicator]

Regarding the CR method. Look at the the last sentence in para 2.15.2. If this "out" doesn't confirm the ambiguity of this particular control, I don't know what does. Hence, my advice to Jszymkowski is to spell out and depict what you need if it is that critical to the function.

Again, hope this helps. Have nice memorial weekend.