One Million Datum Targets? 4

[ikrayg]

I have a plastic part that attaches to another part by using several hoop snaps. The coworker I am working with is insisting on using a datum target at each hoop snap location to establish datum A since the snaps are functional features. The resulting drawing ends up having 12 datum A targets. Has anyone heard of doing it that way instead of using three target and profile of a surface control frames on each hoop snap?

 

[KENAT]

Well, unless I'm missing something you definitely can't have 12 datum A. Each datum identifier should only be used on one datum (though it can be shown in more than one view).

Can the snaps not be tied back to the main datums?

KENAT, probably the least qualified checker you'll ever meet...

 

[ctopher]

I agree with KENAT.
Having 12 datum A's will cause issues with the mfr/shop and inspection. It is not standard practice.

Chris
SolidWorks/PDMWorks 08 3.1
AutoCAD 08
ctopher's home (updated Aug 5, 2008)
ctopher's blog
SolidWorks Legion

 

[TheTick]

You need batter form control to ensure your latch points are in place.

 

[ikrayg]

Oh no, it's one Datum A as the main datum established by 12 datum targets; A1, A2,... A12. He says inspection will clamp down each of the 12 datum targets then measure the part. The part is ABS plasic.

 

[drawoh]

ikrayg,

If your primary datum is a plane, it is defined by three datum targets. I do not think a snap is very good for this. A snap would define a centre would it not? Twelve datum targets have no geometric meaning to me.

You should use some other feature as your primary datum. Probably, you can use snaps as your secondary and tertiary datums. Locate your snaps with positional tolerances.

JHG

 

[ctopher]

Does this help?

http://www.mccannsales.com/book/8_06/considerationsalumcastings.pdf

http://gdtseminars.com/blog/2008/03/06/dimensions-with-datum-targets/

http://www.tec-ease.com/tips/june-07.htm

http://www.tec-ease.com/tips/april-98.htm

http://www.draftingzone.com/shoppingzone/dmpgd.pdf

Chris
SolidWorks/PDMWorks 08 3.1
AutoCAD 08
ctopher's home (updated Aug 5, 2008)
ctopher's blog
SolidWorks Legion

 

[KENAT]

So mis read the OP, when using datum targets simplistically you have 3 datum targest to create your primary datum plane, 2 points to create your secondary datum plane & 1 point to create your tertiary. It's all about fixing all 6 degrees of freedom

Take a look at Ctophers links.

I tend to agree with drawoh look at using another feature if appropriate, or just pick 3 of the snaps if you must use them.

KENAT, probably the least qualified checker you'll ever meet...

 

[ewh]

As the others have noted, more than three points to define a datum is overkill.

When the people fear their government, there is tyranny; when the government fears the people, there is liberty. - [small]Thomas Jefferson [/small]

 

[Jabberwocky]

Far be it from me to get in over my head on a GD&T Analysis, but if the part is plastic and can deform then would it not be reasonable to "overdefine" your Datum A by specifying each hoop snap as part of that datum? Maybe the part will be differently warped if only 11/12 points are in contact.

 

[PaulJackson]

Well said Jabberwocky!

 

[KENAT]

Wouldn't that be constraining the snaps to datum A. You can't guarantee 12 points being coplaner unless you constrain them to a feature simulating the datum, or something like that right? You wouldn't actually have 12 datum points.

KENAT, probably the least qualified checker you'll ever meet...

 

[PaulJackson]

Exactly correct Kenat... All twelve of the datum feature surface areas A1, A2...A12 under constraint to the twelve target points or areas that make up the datum feature simulator A would serve to establish the primary.

 

[KENAT]

Would you identify them as datum targets anymore though? I would have thought this would be confusing.

Wouldn't a note on affected dimensions about the restraining be more appropriate, see section 6.8 of ASME Y14.5M-1994 and figure 6-54?

KENAT, probably the least qualified checker you'll ever meet...

 

[PaulJackson]

A note specifying the force required to restrain the part on the twelve targets would certainly be appropriate if the hoop snaps were not engaged to accomplish the restraint.

It seems although that designing the target areas to reflect the mating part's attachment features and engaging the hoop snaps for restraint might better reflect the functionally "installed" condition of the part for inspection scrutiny.

 

[MechNorth]

If I'm understanding the responses properly, I strongly disagree with them. They are datum targets, and are representational of the contacts during actual use ... therefore all 12 should be indicated as datum targets. You might not (as opposed to will not) make contact with all 12 targets in a non-restrained condition, but it is legal and appropriate. There is no limitation to the maximum number of points to be used to establish the primary, secondary or tertiary datum planes. Your ASME Y14.5M-1994 standard (soft-cover book), placed on a desk, will have far more than 3 points of contact. If the workpiece is designed to make contact at more than 3 potential locations, then they should all be datum targets. There's NOTHING in the standard that requires you to make contact at all datum targets, and in reality you likely won't mate intimately with all 12 targets, but they are there in case.

It seems a common misconception that you need to have exactly 3 points of contact to establish a primary datum plane... it is a MINIMUM of 3 points of contact to establish the plane, not a MAXIMUM! If it were a maximum, you could not legitimately use an entire planar surface to generate a planar datum, you'd be obliged to use datum target points.

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

http://www.profileservices.ca

TecEase, Inc.

http://www.tec-ease.com

 

[KENAT]

You're right MechNorth. It does say "A primary datum plan is established by at least three target points or areas not on a straight line". I'm still unclear how you can have more than 3 unless it's constrained, even if only by gravity making it sag as in the case of the paperback ASME std. If the twelve are significantly non co-planer how do you pick which ones to derive the plane from?

Anyway I'm confused so I guess I should shut up, sit back & learn.

KENAT, probably the least qualified checker you'll ever meet...

 

[ringster]

4.5.8 PATTERN OF FEATURES in ASME Y14.5 allows the use of multiple features to establish a datum. The illustration is only four holes, but is it too much of a stretch to allow 12 snaps to do the same? Curious minds might want to know. Then again........

 

[MechNorth]

Kenat, imagine a cast planar surface. How many points of contact are possible on that surface ... only 3? There are a potentially infinite number of points of contact, but you need only 3 minimum to establish the plane, and therefore the origin of measurement.

I'm glad you raised the question that moves this to the next consideration ... what is your origin of measurement? If all 12 target simulators (let's say pads for convenience sake, though also applicable for line or point contact) are created at the same level (i.e. are coplanar within gagemaker tolerances), then you could measure from any of the pads. If the pads are at different levels (i.e. not establishing a single plane, but rather a multiplicity of planes) then you need to indicate your origin of measurement on the drawing. Again for convenience, you would typically indicate the surface of one of the pads; you get to choose which one. On the drawing, you would put a datum callout on an extension line from that feature in the profile view. So, on your drawing you would then have your 12 datum target callouts (A1 thru A12), and a traditional datum callout symbol (A). The datum plane is thus identified as originating at that single datum target. Less conveniently, you can establish your origin of measurement at a basic offset distance from any of your datum targets, and label it such as "ORIGIN OF MEASUREMENT FOR DATUM A"

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

http://www.profileservices.ca

TecEase, Inc.

http://www.tec-ease.com

 

[ikrayg]

It does make sense to clamp down at all twelve datum target location to simulate the installed state since it is a semi-ridged plastic part, but the remaining issue I have is how can you verify that each hoop snap is within tolerance to each other? On this part the snaps are at different planes to each other (in two's, it a symmetrical part).

This has really been helpful. Thanks for all the responses.