Profile Tolerance regarding to other profile as Datum

[alrayo]

Before anything I have to say that I am beginner in GD&T, and it is really interesting, I am a mechanical designer and recently I have designed some components for certain stamping process, as you understand accuracy is critical and some surfaces must be controlled using profile tolerance, to illustrate the situation, attached you can find a PDF drawing with one of the components which I recently have designed, In this component the outer profile should be controlled as is stated, but I have an other very important profile(showed in “Detail B”) which shall be controlled by the profile represented by “DATUM D”, and it has to be controlled all around by this boundary. I am not sure if this kind of tolerance is correct, or even if will be interpreted as my intend.

 

[KENAT]

I don't see a drawing standard referenced on the drawing. What standards do you draw to?

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

 

[alrayo]

I am using ANSI as Dimensioning standard.

 

[Heckler]

Your Datums B, C, & D are redundant. In ASME Y14.5 - 1994 section 4.4.2 - Parts with Cylindrical Datum Feature. A cylindrical feature is always associated with two theorectical planes intersection at right angles on the datum axis

Heckler
Sr. Mechanical Engineer
SWx 2007 SP 4.0 & Pro/E 2001
o
_`\(,_
(_)/ (_)

This post contains no political overtones or undertones for that matter and in no way represents the poster's political agenda.

 

[KENAT]

So let me be more specific. As regards GD&T are you using ASME Y14.5M-1994 or an earlier version?

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

 

[dingy2]

Yup - Heckler is correct about your datums. Datum A is good and then you need a cylindrical feature for the secondary datum. It could be the OD or ID but something rather than a centre line.

Since the feature is asymetrical, there needs to be a tertiary for anti-rotation.

I would say fix the datums and then let's look at the profiles.

Dave D.

http://www.qmsi.ca

 

[KENAT]

If drawing to ASME Y14.5M you need to say so on the drawing or document listed by the drawing (14.5 paragraph 1.1.3). It may seem pedantic but can make a difference and reduces confusion.

I’d perhaps make your datum D into datum B and use the flat produced by.2972 as C if you need orientation. Datum selection needs to be based on function though which I don’t fully know.

Take a look in 14.5 at section 6.5. Figure 6-17 is vaguely similar to yours.

I don’t think you need the composite frames.


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

 

[ringman]

If ALRAYO is using ANSI, it would have to 1982 or earlier version, wouldn't it?

Briefly, how is this insert used, or installed in subsequent operations please?

 

[powerhound]

Yes, if it really is ANSI that he means then it's no more recent than 1982 but unless he has a very outdated corporate addendum to consider, there's no reason not to use the most recent standard. Considering that the tolerances precede the datums then he shouldn't be working to the 1973 standard. The only ANSI thing left is the 1982 standard and considering he used the suction cups as datum identifiers, I think we can safely assume he's working to the 1994 standard. The differences are so minimal anyway that we won't likely steer him in the wrong direction.

That being said...

alrayo,

Get rid of the datums on those centerlines, change datum D to datum B and make datum C the flat on the left side of the part in the top view. Redefine your datum reference frames on your profile callouts as necessary. If you use composite profile, each segment must have a datum reference, unlike composite position.
Once you have that done, repost a drawing for us and we can give you a little more help.

Powerhound, GDTP T-0419
Production Supervisor
Inventor 2008
Mastercam X2
Smartcam 11.1
SSG, U.S. Army
Taji, Iraq OIF II

 

[alrayo]

Thanks every one for your help, attached you can find the revised drawing attending your recommendations, additionally I have attached the drawings for components which will be matted with this component.

 

[ringman]

Alrayo,

Looking better. Some things to be considered: No need for the 'CL' on centerlines. Need a class designation for the thread callout. The Datum C definition is "troublesome" IMHO and does not go well the datum C in the FCF. A few other that the checkers should pick up on.

I assume that you have a checking group which will review your new results.

 

[powerhound]

alrayo,

You need to have a datum reference in the lower segment of your composite profile. What message are you trying to convey by the lower segment when the value is exactly the same as the upper segment and there is no datum reference? What are you trying to say when you put BOUNDARY below the datum B identifier? You don't need the reference to datum C in your FCF that controls all those countersinks.

On your print of the mating part in sector C8...the parallel and profile tolerances overlap each other. You can get rid of the parallel because the profile of a surface referenced to A controls the parallelism within the tolerance zone of .0002.

Back to the first print; if the sole purpose of those countersinks on either side of the tapped hole is just for lead-in for the bolt, I don't think I would call it out so tightly. As it is, all untoleranced dimensions are basic per your note so you would need to further define the countersinks with a geometric tolerance. You should really pore over your drawing to make sure that all the dimensions that need to be toleranced are done so. I don't think you intend for that counterbore to have a profile of .0003 applied to it. Double check your reference to surface finish in note 2 versus the statement in the upper left of your title block. You have a .125 hole called out at .360 deep. Elsewhere in the print, the same hole is shown .358 - .362. You can't do it like that. Someone will make that hole .360 deep +/-.005 and you may have to buy that part anyway if it's .364 deep.


Powerhound, GDTP T-0419
Production Supervisor
Inventor 2008
Mastercam X2
Smartcam 11.1
SSG, U.S. Army
Taji, Iraq OIF II

 

[ringman]

Powerhound,

A question with regards to Datum B. What is the true geometric counterpart of datum feature B as currently indicated? If it is as I think, what is the benefit or need for C?

Kinda complicated aint it..

 

[powerhound]

A true geometric counterpart would be the axis created by the radius. In this case datum C is probably not necessary but being that it is an asymmetrical part I automatically suggested the datum reference to C. I guess I should have investigated more to see if it really was necessary.

Powerhound, GDTP T-0419
Production Supervisor
Inventor 2008
Mastercam X2
Smartcam 11.1
SSG, U.S. Army
Taji, Iraq OIF II

 

[ringman]

Alrayo,

I for one would like to see the final results of your drawing after review by your checking group. I imagine that others might like to see the same. Is that a possibility?

There are a lot of little issues that should be addressed,
but not yet mentioned my anyone. Self included.

 

[alrayo]

Attending your recommendations and after checking tolerances in our group I had made some improvements in both drawings, (I have invoked the figures 6-17, 6-18 and 6-19 to define tolerances for profile and position), any how, I am having doubts if the feature control frames stated in the “DETAIL B” are they applicable to my tolerancing intend?, in fact I want to have control of the profile shown in “DETAIL B” and this has to lie sweeping along the external profile boundary in closed loop, keeping the cross section to the corresponding axes datum as C, D, E. that’s why I am including four profile feature control frames with profile segments to be applied.
Other issue will be the matting process between the “Inner insert” Vs “Pre-Curl Die”, I have invoked the section 6.5.5.1 (Boundary Control for a Noncylindrical Feature) to define the tolerances to apply on this two components.

 

[Xplicator]

alrayo,
I applaud your persistence to get this right. As mentioned by ringman, there are other standard related issues that indeed need to be addressed on both component specs. But first to best help you, I/we need to further understand the function and the assembly sequence/process of the insert to the die. More specifically the two threaded (5/16-18-2B) holes, datum C on each component. Could you please elaborate?

 

[ewh]

I agree with Xplicator. While it is apparent that you are determined to get this right, there are several minor, yet cluttering items (unrelated to the GD&T) which are not per the standard. Some of them are:

"X.XXX REF." is not preferred, having been replaced with "(X.XXX)". The period is also not needed, unless the abbreviation can be confused with another word, such as "no" and "no.".
"THRU" is not necessary if it is shown in a view as being through the part.
The "S" modifier is now obsolete; if "M" or "L" is absent, "S" is inferred.
"...TAP..." is specifying a method - Only the thread callout is needed.
You don't need to specify "POLISH" with a 4 finish; however it is achieved, it will be "polished".
A countersink is inferred to be nearside unless specified otherwise.

These are not usually considered major issues, but can go a long way towards de-cluttering a drawing, which in turn makes it easier to interpret, thus less expensive to manyfacture in the long run.
Are you limited to one sheet B size dwgs? Breaking these views onto multiple sheets would also go a long way towards simplification.

Keep up the good fight!

 

[powerhound]

The S modifier is not obsolete. It might be in the next version of the standard but as of now it is considered an alternative practice (Rule #2a). The important point is that it can only be used on tolerances of position. See para. 2.8(b) in the standard. Why anyone would want to actually invoke that paragraph is anyone's guess but it is legal.

Powerhound, GDTP T-0419
Production Supervisor
Inventor 2008
Mastercam X2
Smartcam 11.1
SSG, U.S. Army
Taji, Iraq OIF II

 

[ewh]

Got me, powerhound. I'm starting to rely on my memory too much, and we quit using that modifier with the '94 version.