GD&T Question about Datums

[MrComment]

Howdy folks,

Im no GD&T expert by any means, and Im alittle fuzzy about some of the FCFs I see.

I would like to be able to visualize how Datum D works in my attached drawing snipet. This way I can get a good feeling Im setting up my alignments correctly, etc.

Im trying to resolve in my head some sort of 3-2-1 alignment scheme that uses Datum A primary, and Datum D as the secondary. Almost all of my FCFs use this senerio.

Datum A which isnt shown in this view happens to be a plane on the bottom surface of this part (In this view you would be look down on the top of the part)

I guess Datum D is somehow related to the pattern of the tabs around the outside of the part? But Im not sure how these 4 features would relate to me being able to rotate and translate the part. Does this drawing seem to have legal callouts on it?

Thanks everyone in advance.

[Jan d.]

I agree, it is confusing. But looking at how the part is dimensions, I think the intend is to have A be primary flat, D to establish the origin left to righ and C to set the up and down origin.

Looking at detail Z, I think that they want the center line of P1 to be the D datum. Yes, yes yes, totally not obvious from the call that goes with Z, but looking at the top left view, it seems to me the only logical conclusion.

So I'd measure A, P1 and C and set up A, D, C as datums and go from there.



Jan.

[Big Red]

I want to learn too.
Why do we not use Datum -B- for the rotation in our alignment?

[Peter Warcholyk]

Jan,

I would agree, primary datum "A", but then secondary and tertiary, would then be "B" and "C" respectively.

Next report out "D", (P1 to P4), and then construct an alignment utilizing "D",
to report out the requirements in detail "X" and view "Y".

Regards,
ZydecoPete

[#2#]

I want to learn too.
Why do we not use Datum -B- for the rotation in our alignment?

Because datum B is not called out in the feature control frames.

[Mike Johnston]

Datum B is the secondary element so the coordinate system should be rotated to B (Y-axis?). It looks to me as though Datum -D- is only being used for your profile callouts in the "X" axis. But they are insinuating that you use "C" in your Y-axis but do not call it out in the FCF. I'd use -B- to rotate and set my X orgin and "C" to set my Y origin. That would be my primary alignment.

I would then use Datum -D- for my profile callouts. I'm not sure how you're going to legally use C for the lateral profiles. You may need to make an assumption or have it called out correctly in the FCF's.

Mike

[MrComment]

We dont use Datum B in this case because its not specified within the FCFs which only specify A and D.

Ok heres whats throwing me off though. I have 4 of those clips (P1-P4) around the outside of the part, and each clip has to go into the mating part. Whats so special about P1? If I construct my center line between middle of the P1 and P4 clips I guess I can see how that would set my rotation, but does the callout imply I should use all 4 clips together? Obviously I cant construct a line between all 4 features. Looking at the details of the P5 and P6 features the CL callout is just refering the CL of each individual feature?

I find this datum scheme cornfussing.. but I dont know how to make it any better since the 4 clips together are mating parts. I guess in a way it could be just like a bolt hole pattern, but instead of nice round bolts I have flat slotted clips? Could I then consider the Centroid of the 4 clips to be my center of rotation instead?

[#2#]

Datum B is the secondary element so the coordinate system should be rotated to B (Y-axis?).
Mike

Where does it say this?

[Peter Warcholyk]

MrComment,

Your best bet is to try to find out the intent / functionality of the part.
Is there a design or quality engineer that can assist you?

Perhaps the gd&t callout is not as it should be. This would not be the first time that this sort of an error has occurred.

Just a few more thoughts.

From the attached drawing it is not possible to determine if datum "D" is intended to be one or more of the "clips".

Regards,
ZydecoPete

[#2#]

MrComment,

Your best bet is to try to find out the intent / functionality of the part.
Is there a design or quality engineer that can assist you?

Perhaps the gd&t callout is not as it should be. This would not be the first time that this sort of an error has occurred.

Just a few more thoughts.

From the attached drawing it is not possible to determine if datum "D" is intended to be one or more of the "clips".

Regards,
ZydecoPete

It appears to be an automotive air filter housing cover. I would assume the clips are to hold it to the bottom half of the housing.

[justncredible]

P1 to P4 make a line as it is in the print and rotate to that. I do not see -A- in the picture, but I would guess it is a plane, your callouts come from -C- so I would use that plane as well, check the lenghts first. The intersection point of the line from P1-P4 to -C- make the zero.

As for what tabs to use, it falls on what the controlling features relate to, on this print everything comes off P1-P4 as a line, and off -C- as a plane.

Just the way I would check some of the part.

[#2#]

P1 to P4 make a line as it is in the print and rotate to that. I do not see -A- in the picture, but I would guess it is a plane, your callouts come from -C- so I would use that plane as well, check the lenghts first. The intersection point of the line from P1-P4 to -C- make the zero.

As for what tabs to use, it falls on what the controlling features relate to, on this print everything comes off P1-P4 as a line, and off -C- as a plane.

Just the way I would check some of the part.

Is C a plane?

I agree with your analysis for using P1 and P4 for a line (especially since there is a line connecting the two). However, basic dimensions can be added and subtracted as necessary to get another basic dimension. For example the horizontal basic dimension from P2 to P1 is 24.5. So, where the basic dimensions come from doesn't necessarily indicate the datum.

[MrComment]

About the same as a bolt pattern?

Goodluck, you are correct about the function of the part and the clips.

I think my DE is out of the office (not returning my emails ) and the QE cant help me with the GD&T.

I was thinking about it while I was on lunch break, and if I interpret the "P1<-->P4" callout as "4X 22.5 +/-0.5 (+FCF stuff+)" then I would clearly have a pattern right? In general is it legal to define my D Datum from a pattern of holes? I don’t think the mating interpretation of this part would change much, if at all were parts bolted rather then clipped together.

[MrComment]

Datum C is basically a tertiary edge point on the side of the part. Its being established by the part fixture.

However Datum C like Datum B isnt relavent to my FCFs though. Actually I havent found a single FCFs that reference Datums B or C at all. I guess they are just there to make the drawing busy looking, and futher add confussion.

[#2#]

Datum C is basically a tertiary edge point on the side of the part. Its being established by the part fixture.

However Datum C like Datum B isnt relavent to my FCFs though. Actually I havent found a single FCFs that reference Datums B or C at all. I guess they are just there to make the drawing busy looking, and futher add confussion.

One thing I have seen them do here. There have been instances where they have changed the datum structure when the drawing was revised. Rather than moving a datum (B) to another feature, they left it on the drawing, defined the new datum as C and re-dimensioned to C. They didn't want to move the datum and keep the same name but at the same time, they didn't want to have a datum A and a datum C without having a datum B on the drawing.