clocking question

[caseynick]

A drawing shows a cylinder with a flange on each end and each flange has a 8 hole pattern of holes in it. The cylinder o.d. is datum A. The flange face at one end is datum B. The flange face at the other end is datum C. The holes at the datum B end are controlled with a positional callout to B/A. The holes in the datum C end are controlled with a positional callout to C/A. The holes are shown in line. Does the drawing require the holes to be in line?

 

[ringman]

Don't think so. To align them you should assign a datum identifier to either one of the holes or the pattern on one flange and relate the other set of holes to it.

 

[Heckler]

ASME Y14.5M-1994

section 4.4.2 Parts with Cylindrical Datum Features - A cylindrical datum feature is always associated with two theorectical planes intersecting at right angles on the datum axis.

section 4.4.3 Rotational Orientation - To establish rotational orientation of two planes about a datum axis, a third or tertiary datum feature is referenced in the feature control frame.

So with that said, just like ringman stated above you need a tertiary datum to clock both sets of holes to or make one hole a datum and clock the other set to that datum feature.

IMO, your datum [-C-] is redundant since you have datum [-B-] in the same plane.



Best Regards,

Heckler
Sr. Mechanical Engineer
SW2005 SP 5.0 & Pro/E 2001
Dell Precision 370
P4 3.6 GHz, 1GB RAM
XP Pro SP2.0
NVIDIA Quadro FX 1400
o
_`\(,_
(_)/ (_)

Never argue with an idiot. They'll bring you down to their level and beat you with experience every time.

 

[ewh]

Datum C is not in the same plane as datum B, though it is in the same orientation. If it is critical that all holes be perpendicular to datum B, then omit datum C. If it is more critical that the holes be perpendicular to their respective flange face, then leave datum C as you have it, or control that surface relative to datum B.
As far as clocking, the holes will only be misaligned by the angular block tolerance allowed (provided it is obvious on the drawing what their orientation is)(see para 2.1.1.2). If they need to be more tightly controlled, then the addition of making one of the holes a datum and clocking the rest relative to it is a good idea.

 

[ewh]

My previous post should read
"If it is critical that all holes be perpendicular to datum B, then omit datum C, and control that surface relative to datum B. If it is more critical that the holes be perpendicular to their respective flange face, then leave datum C as you have it."

 

[powerhound]

I've had this discussion with someone before and no conclusion was ever reached. This seems like the perfect time and place to bring it up again. Does the implied 90 rule apply if you are locating features? The rule itself refers to right angles and surfaces of features but says nothing about aligning centers. In regards to the case that caseynick proposed; the part is still not fully constrained as all degrees of freedom have not been restricted. I had always understood standard practice to be to use A, B, and C datums to restrict the degrees of freedom THEN use additional datums to further define your part if necesssary. If the holes have to be aligned, I would have called datum C out as one of the holes to stop rotation of the cylinder and then position the rest of the pattern to it. A composite tolerance block could be in order here depending on the application. As ewh said, you can control the end opposite datum B with a profile of a surface or something like that if it has to be constrained with respect to the datum B surface. As you have it, the only thing controlling the orientation of those end surfaces is the implied 90 rule.

Powerhound
Production Supervisor
Inventor 11
Mastercam X
Smartcam 11.1
SSG, U.S. Army
Taji, Iraq OIF II

 

[ewh]

While I agree that it is good practice to clock the holes as suggested, I don't feel that it is an actual drawing requirement. It is more of a fool-proofing exercise.

 

[ringman]

Perhaps it would be well to know the mating relationship at assembly??

If orientation of the 2 patterns to each other is necessary, I don't believe that the title block tolerance can be applied. That would be mixing basic and plus minus.

 

[ewh]

The hole orientation on the drawing is not optional unless specifically stated. As long as one of the holes on each hub lies on the hub(s) centerline (12, 3 6 or 9 o'clock) and the spacing is noted (EQ SP for example), then the block tolerance would apply. If the holes are located with basic dimension(s), then the geometric control applies and the centerlines are assumed to be a basic 90 degrees. It really is better to be safe and specify the clocking.

 

[namdac]

Heckler,
Have ANSI memorized eh?
Namdac

 

[Heckler]

No, but like a good engineer it's one of the tools in my toolbelt that gets used a lot next to Roaks Formulas for Stress & Strain & the Machinery's HDBK.

Best Regards,

Heckler
Sr. Mechanical Engineer
SW2005 SP 5.0 & Pro/E 2001
Dell Precision 370
P4 3.6 GHz, 1GB RAM
XP Pro SP2.0
NVIDIA Quadro FX 1400
o
_`\(,_
(_)/ (_)

Never argue with an idiot. They'll bring you down to their level and beat you with experience every time.

 

[drawoh]

caseynick,

Your drawing does not require the holes to be in line because you have done nothing to control the rotation of the pattern. If it were my drawing and this mattered to me, my datum A would be the end face and datum B would be the OD. My hole pattern at the bottom would be located to datums A and B. Datum C would be one of these holes, and my top pattern would be located to datums A, B and C.

Your datums B and C are nominally parallel planes. Using both as datums is weird, but not necessarily wrong. If the planes are out of parallel, the holes will not line up. If your part is flexible, or if the location of hole to each of the planes is critical, then your drawing is right.

JHG

 

[ewh]

drawoh,
I take exception to your statement "Your drawing does not require the holes to be in line because you have done nothing to control the rotation of the pattern". If a hole is drawn at 12 o'clock on each hub, that is where are to be to be machined, as long as this is shown clearly on the drawing. The only question is what tolerance to apply. If the machinist drills one at 12 o'clock and one at one o'clock, the part does not agree with the drawing (as one hole would not be on the hub centerline). I agree that it is better to more fully control this clocking, but it is still not a drawing requirement.

 

[ringman]

esh,

If you think about it for a min, centerlines on a drawing are only 2 dimensional. For it to carry thru to the other end of the part it would have to be plane not in the plane of the flange face.

Another consideration: if you specify a feature for the orientation on the opposite end, you remove the need for the discussion that we are involved in. (time wasted for the checkers involved with the drawing.)

 

[ringman]

CLs are only one dimentional, before someone corrects. Sorry about that.

 

[ewh]

The view layout determines the planes of the drawing centerlines. Unless the views are non-orthographic, the orientation of the centerline in the left view is the same as the orientation in the right view. I definitely agree that an orientation feature should be defined, just that is is not a drawing requirement.

 

[drawoh]

ewh,

The datums for the holes on top are A - the part's OD, and C, the face the holes are located in. I do not see a rotation datum anywhere. There are lots of applications where the rotation of hole patterns is unimportant. A fabricator could take advantage of this, and use a drilling template.

More importantly, if you call the fabricator with this drawing and ask why the hell the holes don't line up, you have a weak case.

JHG

 

[ewh]

If there are no clocking dimensions, then the clocking probably isn't critical. I still stand by my statement that the part should match the drawing, though. I don't think you would have a weak case, either. Review the part and the drawing together with the fabricator. Orient the part to agree with the views. If the part does not agree with the views, the part is not per the drawing.
Why would Y14.5 have ANY statement about how to interpret centerline tolerances on a drawing if it were irrelevent?