One way perpendicularity GD&T 1

[Raddy13]

I'm trying to set up the tolerance for this O-ring groove, and the O-ring catalog says the side wall can be 0 to 5 degrees outwards; i.e., the top of the groove can be wider than the bottom, but never narrower. Is there a way to represent this with a perpendicularity tolerance? I'm thinking not because the tolerance zone would allow for 0 to -5 degrees taper, but I'm very green with GD&T, so I thought I would double check.

 

[Raddy13]

Oh, and the perpendicularity datum is the surface at the top of the groove (datum shown in another view).

 

[ctopher]

If it's 0 to 5 degrees outwards, it could or could not be perpendicular.
Helpful thread?
thread1103-337208

Chris, CSWA
SolidWorks 14
ctopher's home
SolidWorks Legion

 

[powerhound]

Do a search on "Ring Groove concentricity". I'm going to re-hash it here. The o-ring groove drawings that you will find in manuals like Parker are done in an antiquated style. Most times there is no direct translation to a GD&T format. You'll probably have to make an assumption or two at some point and just settle on it. I'm having trouble imagining a GD&T legal scheme to achieve what you're trying to achieve. Maybe some of the others can provide some input.

John Acosta, GDTP S-0731
Engineering Technician
Inventor 2013
Mastercam X6
Smartcam 11.1
SSG, U.S. Army
Taji, Iraq OIF II

 

[CheckerHater]

Is there a way to represent this with a perpendicularity tolerance? I'm thinking not

No, perpendicularity won’t help.
In fact, O-ring groove is one of the problems that won’t let GD&T people sleep at night.
The problem is that any geometrical control is based on the idea of “perfect surface” and tolerance zone constructed around it. The perfect surface can be flat, cylindrical, conical, irregular; but what all of them have in common is that they always maintain their perfect shape.
So far nobody came up with an idea how to treat the surface that may be flat and then suddenly become cone as it pleases.
But may be something will come out of this discussion. Who knows?

 

[3DDave]

Try a variable single-side profile tolerance to set the range of allowable surfaces with a profile of line for each vertical element to keep it smooth.

 

[drawoh]

Raddy13,

The Parker O-ring handbook shows a maximum angle for their groove. This allows for castings requiring draft angle. Can your manufacturing process take advantage of a 0 to 5[°][ ]tolerance?

Just pick an angle.

--
JHG

 

[dgallup]

If you are machining it they will want straight sides. Much easier to make & inspect.

----------------------------------------

The Help for this program was created in Windows Help format, which depends on a feature that isn't included in this version of Windows.

 

[CheckerHater]

The Parker O-ring handbook shows a maximum angle for their groove. This allows for castings requiring draft angle.

I've never seen this interpretation before. Can you cite your sources?

 

[JNieman]

3DDave: "Try a variable single-side profile tolerance to set the range of allowable surfaces with a profile of line for each vertical element to keep it smooth."

Unless I am mistaken, even a single-side profile tolerance will create two parallel boundaries which means that any material within these boundaries is good. It does not mean it has to go a certain direction. Unless my interpretation of your statement is incorrect.

However, instead of a single-sided profile, there is "NON-UNIFORM" profile specified in -2009.

Using this, you could basically specify your tolerable boundary by using profile of a surface. You could specify a perpendicular boundary on one side, and a boundary opposing it that basically defines a triangle. That still allows for /some/ instances where it's not perfect, but it will drastically reduce the approval of unacceptable geometry.

_________________________________________
Engineer, Precision Manufacturing Job Shop
Tool & Die, Aerospace, Defense, Medical, Agricultural, Firearms

NX8.0, Solidworks 2014, AutoCAD LT, Autocad Plant 3D 2013, Enovia DMUv5

 

[JNieman]

Apologies, I forgot to cite what I was referring to. Look at Figures 8-9 and 8-10 of Y14.5-2009

_________________________________________
Engineer, Precision Manufacturing Job Shop
Tool & Die, Aerospace, Defense, Medical, Agricultural, Firearms

NX8.0, Solidworks 2014, AutoCAD LT, Autocad Plant 3D 2013, Enovia DMUv5

 

[CheckerHater]

OK, what if we combine JNieman's "non-uniform" idea with 3DDave's "every element"?

 

[JNieman]

Oh shoot, combining the straightness pretty much takes care of issues of wild geometry in the triangular profile tolerance zone.

I think that cinches it, in my eyes. I believe the non-uniform zone defines the functional requirements, while straightness keeps it from abusing the zone with any strange concave/convex extremes. Well played.

_________________________________________
Engineer, Precision Manufacturing Job Shop
Tool & Die, Aerospace, Defense, Medical, Agricultural, Firearms

NX8.0, Solidworks 2014, AutoCAD LT, Autocad Plant 3D 2013, Enovia DMUv5

 

[fsincox]

Maybe even 2X?
Frank

 

[fsincox]

CH,
I actually like it!
1) You might need to use profile of a line to slow down the, "profiles must be defined with only basic dimensions", crowd if the width is to be defined +/-.

2) I might still reference the datum of the "derived axis" we are referencing the "perpendicularity" to (used for the basic angle).
Frank

 

[powerhound]

What keeps the angle from varying between 0 and 5 degrees as you revolve around the axis?

John Acosta, GDTP S-0731
Engineering Technician
Inventor 2013
Mastercam X6
Smartcam 11.1
SSG, U.S. Army
Taji, Iraq OIF II

 

[fsincox]

The definitions I have seen in the vendor catalogs or the SAE standards do not either.
Frank

 

[DanStro]

This may be a dumb idea but could you model the walls at 2.5 degrees and put an angularity tolerance on the walls with respect to some other face that is either vertical or horizontal, or center plane ? Of course you would need to convert the +/-2.5 degrees to a linear distance for the angularity spec?

Han primo incensus

 

[CheckerHater]

What keeps the angle from varying between 0 and 5 degrees as you revolve around the axis?

I guess we can throw in some runout either as a callout or general note. That will take care of cone surface not looking like cupcake.

As Frank noticed even standard definition is seriously deficient.

If you are machining it they will want straight sides. Much easier to make & inspect.

Who are "they"?

http://www.thinbit.com/products/grooving/oring_grooving.html

 

[fsincox]

Modeling the groove with an angle in is not popular in my shop they want us to model it square.
Frank