Imaginary Feature Datum

[iv2jm]

We just designed a part with a patterned hole pattern. Basically, a small hole pattern of 4 holes in a square is patterned 3 times around a circle for 3 linear bearings.

the center point of the pattern of 3 has a vertical centerline called out as Datum B and the horizontal centerline called out as Datum C.

If i want to measure this with a CMM arm, how can i set up my alignment with the datum that arent on features of size?

Perhaps there should be a small hole in the center specifically for this, but then the tolerances on those holes come into play.

Thanks for any help. Ill try and send out a picture when i get to something other than MSPaint.

 

[ewh]

What feature controls the vertical and horizontal datums?

 

[KENAT]

I'm not sure the GD&T is valid, you talk about the centerlines being datums. If the datum feature symbol is actually placed on the centerline this is wrong. A picture would probably help.

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

 

[MikeHalloran]

Maybe I missed a meeting. I thought a datum had to be somethng you could touch, e.g. with your cmm.



Mike Halloran
Pembroke Pines, FL, USA

 

[iv2jm]

Im somewhat new to GD&T implementation, and im not sure the GD&T on this drawing is valid either... but im working with what it says. Maybe its because i know what the part is for, that it makes more sense, but here it is. Ill put up a link- most of the other dimensions are removed for clarity, but here is the portion im confused about...

Looking at fig 4-22 in ASME Y14.5M, you can set a datum to the center of a pattern- but there isnt necessarily a physical feature to touch. So how would you inspect that?

Like i mentioned, and like you guys seem to agree with- a feature needs to be put in to actually touch off. Or the datum needs to be located at a feature- even if its not part of this pattern.

Thanks in advance

 

[MikeHalloran]

If you wanted to buld a gage that actually had B and C datum surfaces, you would put gage pins in it that engaged ... what?

Right now, the only real datum on that drawing fragment is the bottom edge.







Mike Halloran
Pembroke Pines, FL, USA

 

[KENAT]

The B & C datums are not valid per ASME Y14.5M-1994. Take look at 3.3.2(b) for how a center plane can be set as datum (fig4-13 & 4-14 may help too although they're talking about material condition)

I start to get confused by FRTZF & PLTZF in patterns of patterns so am limited in what help I can give.

Maybe take a look in the GD&T forum. forum1103

There are some very knowledgable people there (not to imply that Mike or ewh aren't, just a few more look at that forum).

If it's someone elses drawing can you go back to them and say you're having trouble interpreting their GD&T based on the standard and have them tell you what they want?

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

 

[KENAT]

iv2jm, you've got me interested now so I put a link in GD&T thread1103-204318

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

 

[ringman]

IOt appears that a course in Dinensioning and Tolerancing would be of great benefit to whoever prepared the subject drawing. Too many problems to begin to address all of them. Centerlines as datums is one, FOR SURE.

 

[PaulJackson]

I don’t know whether the 3X Ø 1.13 – THRU diameters control the location of the “linear bearings” or if the “4X fastener holes” do that job but you can declare all of the features … either 3X Ø 1.13 – THRU... or... 12 X [3X X 4X M8X1.0 (or whatever)] as your secondary datum feature location ... you can identify it as such by hanging a datum feature identifier below the position callout. That identifies the pattern of features that comprise the secondary datum feture!

The tertiary datum feature (it seams to me) would either be the Ø 5.0 diameter thru or the pattern of features that establish that rotational significance! Naturally a basic angle would describe the offset to the detailed coordinate system.

If your fixture is dependent upon the integrity of its edge entities for location to other details of the assembled fixture or component then you should use them as others have suggested.

The important thing to remember is that real things must be designated as datums (like surfaces or holes or hole patterns or registry “touch” points yada, yada, yada. Centerlines , axes, median planes and intersections of perfect geometry are not eligible for designation as datum features.

Paul

 

[fcsuper]

To clarify Paul's finally comment, unreal things (reference geometry to use 3D CAD junkies) can be used as datums as long as they are directly associated with real feaures. See paragraphs under 4.4 of the ASME. There's some good figures in there on how to develop datums.

Ok, just my opinion:
On the drawing you uploaded, Datums B and C are invalid as I interpret the drawing's design intent, though it could be they are associated with the holes along their planes (but I am sure that is not the design intent), in which case they are valid but awkward. The method used to develop B and C looks vaguely based on figure 4.9 of the standard. It's almost as those the designer added small holes along Datum B in order to make it valid, which of course in violation of the design intent (meaning that the datums are not based on anything important and are therefore abritary having no purpose within the design, making them pointless).

Given that assumed error, the control frame in the lower right is also nearly pointless.

Unless there's a whole lot of control framing going on in Detail A (which we can't see) most of the basic dimensions shouldn't be basic because no relationships exist to the locations of the holes and other dimensions or control frames. Just because a dimension happens start in line with a datum doesn't mean the datum is being used to identify its location. You will need a control frame or some further dimensioning to quality the dim as basic. In either case, it seems there is either too many basic dimensions or not enough, depending on what's in Detail A.

(Peeve pet, but not a requirement to fix the drawing: Thread callouts shouldn't normally show drill size unless there some criticallity involved (extremely rare from what I've seen). Listing drill size is redundant to the thread callout and may even be considered a process instruction.)

To do what I think is trying to be done here, new Datums B and C needs to be established based on meaningful features of size and the function of those datums. Many many many more positional tolerances are needed...many more (are they in Detail A?), and a completely new dimensioning scheme is necessary.

To answer your immediate question: the control frame in the lower right it is uninspectable.


Matt Lorono
CAD Engineer/ECN Analyst
Silicon Valley, CA
Lorono's SolidWorks Resources
Co-moderator of Solidworks Yahoo! Group
and Mechnical.Engineering Yahoo! Group

 

[PaulJackson]

Fcsuper,

School me... what paragraphs in 4.4 advise that reference geometry rather than actual features can be used to designate datums?

I am not attacking... I just have never found that.

Paul

 

[KENAT]

I think (hope) he means center planes (4-6 & 4-7 though I think the figures I previously stated are clearer) and planes through 2 holes (4-8 & 4-9).

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

 

[PaulJackson]

Kenat,

Since the paragraphs end at 4.6.7 I assume that you are referring to the figures 4.6, 4.7, 4.8, and 4.9. If so all of the datum designations in those illustrations refer to actual features... surfaces: planes, cylinders, complex surfaces, etc. I don't know how deep this discussion should go but there are an infinite number of 3 point planes that can be derived from a convex surface... hence the rocking datum dilemma.

Paul

 

[KENAT]

Paul, yes figure 4-6 etc (in my copy figures use - while paragraphs use . I've started using it to distinguish rather than saying figure but it's just laziness, sorry).3

I agree that the planes refer to real features but I can see how someone might think that since the planes aren't actual surfaces they are some how 'unreal'. Of course the kicker is that even if you attatch the datum to a surface you aren't actually using that surface but a plane derived from it.

You're also probably right that this isn't the place to go into this too deeply.

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

 

[fcsuper]

KENAT explained my comment well enough. ...and yes this is the wrong forum for this discussion. Just a final word on this since I'm not going to follow this discussion further: There appears to be bores along both B and C datums on the drawing. But, as I stated, these does not appear to be functionally important enough provide useful center planes that can serve as those datums.

Matt Lorono
CAD Engineer/ECN Analyst
Silicon Valley, CA
Lorono's SolidWorks Resources
Co-moderator of Solidworks Yahoo! Group
and Mechnical.Engineering Yahoo! Group

 

[iv2jm]

Yeah- Thats really the issue. To tolerance those holes to use as a datum is overkill. they are all just clearance.

As for detail A, the pattern is basic dimensioned to each other, with one of the holes basic dimensioned to the datum intersection.

I know its fundamentally wrong, but it does get the point across.

we just cant really inspect it directly. and inspecting it indirectly introduces possible error in whatever other tool we use/make.

to complicate things further, whoever machined it just cut everything on a waterjet. and i dont think it was because the gd&t was somewhat 'illegal'

Thanks for all you help guys, It makes the learning process a lot more interesting with all your input. I really appreciate it.