Representation of Datum in Alternate View

[MToomey]

Hi folks,

I have been digging around for a definitive answer on this. I haven't seen much in Y14.5-2009 (although our drawings don't state any standard, but that's another conversation). I have seen some other people bring this up here and other places, but the conversation seems to get sidetracked. What I have seen is that a datum may be shown in more than one view on the same sheet, and it need not be identified as a reference.

I have a drawing here for a machining step to be done out of house after we weld it together. This machining step will put a hole (shown top right) through six lugs (bottom left) to create a hinge that will support the structure and two holes in the bottom that will allow pins to transfer the load from below. This will straddle a beam with a load hanging from it.
Keep in mind this is a snip. I think I have included all the relevant information, but these features are fully defined in terms of location in other views.

I have done two things on the attached drawing that I think are questionable:
1. I have called a datum feature of size from a reference dimension (Datum A and B)
2. I have shown a feature of size datum as a center line in another view to dimension from it (Datum B - defined in bottom view, referenced in front view)

It seems to me that it is acceptable to use a feature of size that is not controlled on this drawing as a datum as long as I do not reference material condition (it will be RMB).
I have seen people say that there should never be a datum symbol attached to a center line, but I'm not seeing a better way to represent it in the view I am using.
My intention is to have the two 4.625 dia holes centered on the plate. Does this drawing do that? Is there a better way to do that?

Thanks,
Matt

 

[MToomey]

So, after finding this thread (thread1103-424578), it sounds like it would be improper to attach a datum feature symbol to a center line. I have replaced that symbol with a note that reads "DATUM CENTER PLANE B"

 

[TheTick]

I like that solution, as it does not encourage others to add datum symbols to centerlines.

 

[chez311]

First off, your attachment includes a comma and therefore would not open without changing the extension manually. Uploaded attachments will not support special characters.

Datum feature references should NEVER be directed at theoretical lines/points/planes - only physical features. There is no need to notate if it applies to a centerplane or not - the way it is placed implies that. A datum feature reference directly in line with a width dimension applies to the width which derives a center plane as in fig 3-4(h). A datum feature reference applied to a single planar surface or offset from the width dimension applies to that surface and derives a theoretical datum plane coincident with that surface as in fig 3-3.

In addition to this requirement, placement on a theoretical feature creates ambiguity about what exactly it is derived from - on your part alone I see at least 8 or 9 potential widths that could derive this centerline, which is not solved by double referencing B as the 118 ref width.

This leads me to another point - there is no need to call out B multiple times in multiple views, if it is specified once (ie: attached to the 118 ref width) that is sufficient, anything else creates the potential for confusion (especially if it is incorrectly applied to a theoretical centerline/plane). I assume you have referenced it again on the centerline in order to "dimension" from it - this is not necessary, any dimensions from the centerline in that or another view will be implied to be from the centerplane of B.

It seems to me that it is acceptable to use a feature of size that is not controlled on this drawing as a datum as long as I do not reference material condition (it will be RMB)

I'd say thats a fair assessment, theres no restriction in Y14.5 about application to reference dimensions. I suppose a different material condition could be specified along with a note to the drawing which controls that feature, however that would be cumbersome.

I have seen people say that there should never be a datum symbol attached to a center line, but I'm not seeing a better way to represent it in the view I am using.

See above

My intention is to have the two 4.625 dia holes centered on the plate. Does this drawing do that?

Yes. The 100 basic dimension is implied centered around the shown centerline. The 50 basic dimension is not necessary but it is not wrong.

 

[MToomey]

Chez,

My apologies, I tried changing the file name and re-uploading, but I can't seem to find a way to remove the original attachment.
Thanks for your reply. I find myself adding a lot more than I need on most of my drawings because some checkers at the customer seem to be unfamiliar. Not to say anything bad - they are skilled professionals, but there seems to be a lot of confusion in communication. It seems in the attempt to be more clear I have muddied the waters.

Would it also be more clear to put the 118 ref dimension on the front view below the 100 basic dimension?

 

[chez311]

Would it also be more clear to put the 118 ref dimension on the front view below the 100 basic dimension?

Perhaps, I don't see it making much difference either way - that is as long as the dual (and incorrect) reference to B applied to the centerline is removed.

FYI - there is no basic relationship shown between the 2X 4.625 holes and the 3.750 hole datum feature C which is necessary in order to hold position to C (basic location of true position required to applicable datum references). Also it may be recommended to hold them in position to |A|B|C| because otherwise there is no rotational constraint around C.

 

[MToomey]

There is a basic relationship between the 4.625 and 3.750 holes, you just couldn't see it!
And, yes - the incorrect duplicate reference to B has been removed.

You touched on something there that I have been working to understand recently:
What does calling true position on a hole with respect to a datum that is perpendicular to that hole do? I understand controlling x and y coordinates (if I can call them that) from two perpendicular planes, but what does a third perpendicular plane define? It seems perpendicularity would still be controlled by the tolerance zone defined by the two perpendicular references.
I understand rotation about C in this instance since it is a plane and an axis, but I have seen other instances of calling three perpendicular planes in the DRF of a true position on a hole. Any thoughts?

 

[pmarc]

MToomey,

Perhaps the attached document will help you understand the role that the primary datum nominally perpendicular to the toleranced feature plays in position callouts.

https://files.engineering.com/getfi...dbf-41f5-a84b-14d711220ad0&file=pos_tol_2.pdf

Note: This does not mean that the primary datum used in position tolerances has to always be perpendicular to the toleranced feature (unfortunately, very often people think it has to).

 

[MToomey]

Aha! Now we're cooking with gasoline!

Thanks, pmarc. I knew it if I stopped thinking about it.