GD&T true position of duplicate hole patterns relative to each other 2

[jokes60123]

Hi all,

Thanks for reading this.

I've attached a mark up to explain the general concept I'm having issue with. Effectively I've got a hole pattern, that repeats around a part. I've used a cross with a 4 hole pattern for this.

I've created basic datums to give part orientation, but am having issues getting the GD&T to line up with what I actually want (or at least communicating it).

The 4 hole pattern to itself is critical.
The clocking of the centerline of each of the individual 4 sets of hole patterns is critical relative to each other, but not to the exterior of the part.
The 16 hole pattern is critical relative to the center hole in the part.

I guess what I'm trying to figure out is: is it best to call out each of these separately for clarity? Or is there a way to communicate it all in one positional tolerance block?

Also, has anyone out there called out clocking using position? angularity seems incorrect for this use even though the meaning would be clear, and using an actual angular dimension (in addition to a basic) seems like I'd be dual dimensioning the part.

Any help would be appreciated.

Thanks!
Jokes60123

 

[pmarc]

Which of the 3 requirements you listed is more criticial than the other? Are all equally important or is there any hierarchy?

 

[fsincox]

I am with pmarc,
There is nothing easier with GD&T than tieing everything down "dead nuts", the complexity comes in the real definition of numbers, exactly how much to this and how much to that. This is what creates more compliaceted looking tolerance statements, an attempt to loosen the tolerance. It is just the opposite of what it appears.
Frank

 

[jokes60123]

pmarc -

Good point. Sorry should have indicated hierarchy.

The 4 hole pattern to itself is most critical. To attach a number to it, call it .006"

The 16 hole pattern relative to the center hole in the part is the second most critical characteristic. But is not as critical as the first. Using a relative term, call it .009"

The clocking of the centerline of each of the individual 4 sets of hole patterns is critical relative to each other, but not to the exterior of the part. This is actually a "separate" item in my mind currently. This is likely more critical as the first item, but is not relative to the individual holes, but the center of the existing pattern (the holes on either side of the centerine can shift toward / away equally from the centerline without issue). Mentally, I think of it more as an angular spec off of the basic dimension for the centerline of the hole pattern that must be within a 0.4 degrees window rotating around datum B in plane A. I however don't know if this should be called out as a symmetry of 0 relative to the centerline, or as a positional tolerance tied to the pattern as a whole.

Hopefully my description makes sense and clarifies things.

Jokes60123

 

[fsincox]

Orientation of hole patterns is currently handled by the restatement of datums in the multiple lines of a composite tolerance statement.
Read the section on composite tolerancing.
Frank

 

[MechNorth]

Recommend a broken-out composite position tolerance. First callout is 16X size with a position tolerance. Second callout takes a bit more work. Call out a detail view of a group of 4 holes; on the detail, call out "4X INDIVIDUALLY" and the refined position tolerance wrt Datum-A only. If you need a graphic, contact me directly. I'm not able to post graphics right now.

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

http://www.profileservices.ca

TecEase, Inc.

http://www.tec-ease.com

 

[jokes60123]

Thanks for all the advice.

I think I've gotten part / most of the way there, but inverted the mentality by keeping the 4 hole pattern tighter then using a 4 places call out with the composite call outs below. Attached is a revised copy.

The question I have left is should I be using a symmetry or a concentrically call out for the last control. I want the control to be the center-point of the 4 patterns relative to Datum B.

If this looks incorrect or implies something else, please feel free to comment / educate me.

 

[pmarc]

jokes60123,

Some comments:
- you should absolutely forget about using symmetry & concentricity in locating patterns of holes. The meaning of those two controls is completely different than for position tolerance. See paragraphs 5-12 and 5-14 of Y14.5M-1994 standard;
- you should also avoid using directly toleranced dimensions for locating the pattern of holes as this creates ambiguity about from which feature the location of holes should be measured. If you want to assign bigger tolerances in two orthogonal directions you can try to use an approach shown in fig. 5-41;
- true position of each of the holes should be defined by basic dimensions;
- you should not attach datum feature C symbol to basic dimension because the features from which datum C must be established do not exist in reality - these are imaginary axes of holes. The way it was done on your original sketch was much better. Datum can only be derived from actual surfaces of a part;
- if location of holes was defined with the use of basic dimensions and positional tolerance, basic 90 degree angle would not be needed as it is implied according to one of the fundamental GD&T rules (see paragraph 1.4).

 

[jokes60123]

pmarc,

Thanks for the advice. I will have to go track down our corporate copy of the standard to review your references. I can however tell that my drawing is not clearly communicating the intent which is the reason for my trying to redefine the dimensioning structure on the part.

I have difficulty deriving the datums from the surface of the part since the surface is not relevant to the critical functionality of the component. I've attached another pdf which shows 3 different orientations of the holes. These would all be functional and equivalently "good".

This is the crux is why I am having such a problem with trying to determine what the tolerances should be and how to appropriately call them out.

Each pattern, and the subsequent "pattern of patterns" is critical but only in relation to itself (pattern) / each other (pattern of pattern). The exterior of the part, and linear distance to the center hole of the part are only critical when viewed as inconsistencies from each 4 hole pattern relative to the others symmetrically around the part.

I hope the attached helps to clarify what I'm trying to do.