Datums on a molded part w/ draft - acceptable practice?

[prdave00]

I have an injection molded part that has a blind hole at one end. When this component was being machined I used this hole as the primary datum since it locates/orients the part in the assembly. Now that I'm looking at molding a 1-2 degree draft angle may be required for this hole (i.e. zero draft may not be practical). Can I still use it as a datum now that it's tapered?

 

[PaulJackson]

Figure how which part surfaces are engaged in the assembly and how the part is oriented and located by those features and that will point you to your answer.

You said that "the hole (cylinder) when it was machined served as the primary datum feature" that would mean that the machined hole constrains four degrees-of-freedom (two rotations orienting the axis) and (two translations locating the axis)... that would leave two degrees-of-freedom for other features to constrain if indeed all six are so constrained (one stopping rotation about the axis) and (the other stopping translation along the axis).

If the molded cylindrical hole with draft (making it a cone) still constrains four degrees-of-freedom as it mates to a geometric counterpart (a coned protrusion) then by all means keep it as the primary datum feature and apply a form tolerance for the conical surface itself to get the integrity of the coordinate system off on the right foot.

I have no idea what your part looks like or how it actually is oriented or located in the assembly... but if a machined hole can be interchanged with a drafted hole and not suffer any critical loss of coordinate system orientation... I would suspect that other features may be doing some of the orientation that the designated primary is supposed to be doing... in which case the feature has been miss labeled as primary.

Take a good look at the assembly and see what feature or set of features control the most degrees of freedom (specifically the most orientation degrees-of-freedom) and that will lead you to the answer to your question.

Paul

 

[ptruitt]

I've been designing some castings and using complex surfaces as a datum. One datum per part that constrains all the degrees of freedom. Then I relate other features to the MMB of that datum. This allows a nest to be made, simulating the datum at MMB, for the purpose of inspection. This violates the idea of functionality, because I am dimensioning the part so as to facilitate inspection. I don't really see any 'function' in the datum any more than I see the function in any of the other features, but it gets the job done. From a functional standpoint, I see no use for any datums at all.

Peter Truitt
Minnesota

 

[fsincox]

Dave,
Since we do not have the assembly to understand, Are you saying the hole in it's unfinished state locates the "machined" (your word) part?
Frank

 

[prdave00]

Hopefully the attached file helps define the parts for assembly. The common metal part on the right is machined with a self-cutting thread. The plastic part on the right is the mating part that needs to be "coaxial" with the part on the right. The top version is the machined configuration and the bottom is the molded configuration where the "cylindrical" datum features are drafted. The plastic part will be screwed down until it perpendicular face contacts the step in the machined part. Please note that the drawing is a representation of my part (so you'll have to trust me that I need them to be coaxial to some precision to function) and is incomplete (i.e. not all tolerances are provided). I should also note that the part is more "complex" than I suggested/described in the original post.

The blind hole will be slightly undersized to the screw major diameter and the counterbore pilot will be slightly oversized to the screw major diameter. Since the blind and pilot holes are not very deep, both datum features are used to establish the primary datum.

For the machined part, I envisioned a stepped expanding pin could be fabricated (if practical) could be used in inspection to simulate the datum features. Referencing datum features A & B at their maximum material boundary in the FCF may also work. The question is what I do if max draft angle is imparted whether the datum features are simulated RFS or MMB?

Hopefully the above does not confuse matters further.

 

[PaulJackson]

Cannot see the attachment Dave, reload.

 

[prdave00]

Not sure what happened with my file. Hopefully the file comes through this time around.

 

[prdave00]

One more time...

 

[ptruitt]

I might do something like the attached drawing. I forgot to add the counterbore. I would have referenced the counterbore position to datum A at MMC.

Peter Truitt
Minnesota

 

[PaulJackson]

Is the diameter .094 [Z] a .008 interference fit with .086 [A]?

Is any of the assembly sandwiched between surface [C] and the surface perpendicular to [Z] with such pressure that it would cause the self tapped threaded piece to attempt orient itself to [C]?

Paul

 

[prdave00]

Peter: Thanks. I have considered making the A-B datum apply at maximum material condition but still trying to think through whether that reflects my function.

Paul: Yes to your first question. The blind hole is undersized to the screw major diameter [A] based on the 80-90% undersizing rule of thumb per research for a Type-23 (aka Type-T)self-cutting fasteners. No to your latter question. Nothing is sandwiched between surface |C| and the step perpendicular to |Z| - think of surface |Z| as a standoff. I'm relying on the pilot to orient the threaded piece. I'm not yet even sure how effectively surface |C| will act as a hard stop to control the depth since the step is so small. I may may need to rely on assembling under a microscope and visual gage when the two features are touching.

 

[PaulJackson]

Dave,

When this component was being machined I used this hole as the primary datum since it locates/orients the part in the assembly. Now that I'm looking at molding a 1-2 degree draft angle may be required for this hole (i.e. zero draft may not be practical). Can I still use it as a datum now that it's tapered?

With the info that you have shared so far... I would say that your use of the hole as the primary is "spot on" although perhaps it is not just [A] that is primary... but rather [A-B] as you have so detailed it. The gradually increasing interference fit orients and locates the two details to each other. I wouldn't worry much about the taper in the new molded design for orientation sake... since the mating part is going to cut threads into it. Perhaps there could be a shallow c-bore .095 that encapsulates any thread-cutting debris and gives a square corner as the assembly torque reaches its max.

Paul

 

[prdave00]

Thanks Paul. I guess my question boils down to whether the draft angle make these into irregular features of size like a conical bore? Does it now control 5 degrees of freedom since the surface would theoretically come to a point and constrain translation along the axis? Maybe I need a note that says |A-B| only simulates a common axis.

 

[PaulJackson]

You do not need a note for that... that is exactly what it does.

 

[prdave00]

Paul: As soon as I posted I realized the error in my thinking.

 

[fsincox]

Guys,
Lets clear up one issue, the OP. Can a tapered feature be used as a datum?
I say yes!

Does it fit here? It looks like it might from your description.

ptruitt,
I think the taper is on the internal feature(s) I had a different picture in mind, too. As they say: "a pivture is worth a thousand words."
Frank

 

[Belanger]

Frank -- sure, a tapered feature can be used as a datum feature. See Fig. 4-3(e) of the 2009 standard.


John-Paul Belanger
Certified Sr. GD&T Professional
Geometric Learning Systems

 

[fsincox]

JP,
Thanks, I just wanted to express a clear answer to the original basic question. I realize prdave00 is interested in a more specific question.
Frank