Results 1 to 11 of 11

Thread: Wall Thickness Variation - GD&T

  1. #1
    Associate Engineer
    Join Date
    Feb 2012
    Posts
    4

    Wall Thickness Variation - GD&T

    Hello Everyone,

    This is my first post, so I figured I'd ask something that has plagued me for years! I'm wondering if anyone knows of a way to properly tolerance wall thickness variation of a tube/ring? Essentially the variation of the section thickness between the Bore and the OD around the entire ring. We typically basterdize runnout and pull the datum to the side of the Bore or OD to represent the "surface". We're not satisfied that this is the most appropriate way to datum this tolerance. It is odd to think of a moving datum but that is what we have. Each radial point of the Bore or OD is the datum for the other. Any ideas would be greatly appreciated!

    -Matt

  2. #2
    Administrator Kelly Bramble's Avatar
    Join Date
    Feb 2011
    Location
    Bold Springs, GA
    Posts
    2,237

    Wink

    Welcome to Engineers Edge!

    Wall thickness or tolerance control on a tube or ring is dependant on functional requirements with an eye on manufacturability.

    If the tube is just carrying fluids from one place to another then a minimum and maximum wall thickness need only be specified. When measuring, manufacturing would not necessarily measure every possible wall thickness but would measure enough locations to have confidence in the as-manufactured part and manufacturing processes utilized.

    “basterdize runnout” – I’m not sure what you mean by this, again mechanical tolerance should by applied to meet functional requirements so that the end item part does the job it was meant to do.

    “It is odd to think of a moving datum but that is what we have.” – Moving Datum? From your description that is not what is going on. When a diameter is specified as your datum, the datum or “zero” is established most correctly from a datum simulator. For an external diameter one would collapse a ring gage or equivalent onto the as-built external diameter. The ring gage would collapse until the three outer-most areas of the as-built external diameter are contacted. Then, a zero or a measurement is made from the Ring gage surface itself to the concerned feature elements to establish the surface variations.

    So, you see the datum is not really “moving”.

    Can you post an example on this forum?

  3. #3
    Associate Engineer
    Join Date
    Feb 2012
    Posts
    4
    Sorry, I think I threw this conversation off by using the word "tube". The products I am talking about are ball and roller bearings, specifically the outer and inner rings. We have a requirement to hold the sizes to tolerances of .0002" or up to .001", while maintaining a section thickness variation of .000075" (around the entire ring). We want to allow the larger tolerance for size while constraining the thickness variation. Dimensioning the size is obviously not an issue, but we use a runout to control the wall thickness variation. This is not correct because using the Bore or OD as the datum for this runout means you are checking the feature's radial variation relative to the axis of the datum, not its wall.

    Hopefully this is clearer now. But let me use a slightly more detailed example:

    Bore tolerance .001"
    OD tolerance .0003"
    Wall Thickness Variation .000075" (technically this wall thickness variation is with respect to the track or "raceway" of the part... but to keep things simple let's just say it is to the bore and ignore the track)

    We can use a note to call out wall thickness variation, but we were hoping there was a known tool in GD&T that would cover this requirement.

  4. #4
    Administrator Kelly Bramble's Avatar
    Join Date
    Feb 2011
    Location
    Bold Springs, GA
    Posts
    2,237
    1) I think you need to post a drawing image…

    2) I’ll bet you’re reverse engineering/designing a commodity bearing.

    When you say “outer and inner ring” I think you’re talking about the bearing races or bearing assembly cage. Either way, for bearing applications what matters most is the surface variations in contact with the moving ball or roller bearing as well as the installation or mounting surface that locates the bearing assembly.

    Typically, one controls the contact surface variations with a form tolerance, such as cylindricity or circularity. I think you need to specify a cylindricity of (.000075/2) for the bearing roller contact bore.

    To control “wall thickness” one normally defines the bearing installation or mating surface as a Datum, then relates relevant contact surfaces using a circular or total runout tolerance. Position tolerance could also be used to control the wall thickness between the features. If using total runout I think you need to specify the datum as your lower tolerance “bore” feature and relate the OD using total runout of .001. – But hey, I havn’t seen a drawing

    Again, an engineering drawing speaks thousands of words…

    BTW, both circular and total runout tolerance include a level of form control (circularity and cylindricity) and may be a sufficient comprehensive control for the roller bearing contact surfaces.

  5. #5
    Lead Engineer RWOLFEJR's Avatar
    Join Date
    Mar 2011
    Location
    Rochester Pennsylvania
    Posts
    396
    Sounds like you just need to put a concentricity tolerance on it...?? Regardless of what size the I.D. and O.D. come out... you need them to be concentric within "XXXXX".
    Far as I know there isn't any "law" saying that you need to assign a dimension in your tolerancing of concentricity...?

  6. #6
    Associate Engineer
    Join Date
    Feb 2012
    Posts
    4
    This image may clear up what I'm asking for. Please excuse my MS Excelish equation.

    Concentricity isn't really going to cover the radial section deviations. It controls the axis.

    Wall Thickness Variation.jpg

  7. #7
    Associate Engineer
    Join Date
    Feb 2012
    Posts
    4
    So, I guess that image didn't help.

  8. #8
    Technical Fellow
    Join Date
    Feb 2011
    Posts
    1,043
    Quote Originally Posted by mmlj View Post
    So, I guess that image didn't help.
    Ummm, as far as I can see, you have been given the two possible variables to check. You can test for concentricity within a tolerance band. Or, you can test for wall thickness at several places around the tube to all be within a tolerance band.

    What is more important to your project? Concentric or wall thickness?

    Not sure what else we can help with, unless we misunderstand the question.

  9. #9
    Associate Engineer
    Join Date
    Sep 2014
    Posts
    1
    Please Find Wall thickness
    Diameter OD = 600MM
    dished to dsh end = 1375MM
    Material : 316L
    Design Pressure = 8 bar
    Wall thickness = ???

    Media : Water
    Last edited by sumahesh93; 09-11-2014 at 03:15 AM. Reason: media included

  10. #10
    Associate Engineer
    Join Date
    Oct 2014
    Posts
    1
    Matt,
    Have you found a solution to your GD&T problem? We have similar parts and are using runout to control radial wall variation. To measure runout, we put the part O.D in a v-block, position the measuring stylus on the I.D. Then the part is rotated 360 degrees and the variation is recorded. On the drawing, the datum is attached to the O.D. dimension extension line, not aligned to the dimension. The runout GD&T is pointing to the part I.D. Not to the standard, but the best way we have found to do it.

  11. #11
    Administrator Kelly Bramble's Avatar
    Join Date
    Feb 2011
    Location
    Bold Springs, GA
    Posts
    2,237
    On the drawing, the datum is attached to the O.D. dimension extension line, not aligned to the dimension. The runout GD&T is pointing to the part I.D.
    Per. ASME Y14.5-2009 and earlier.. That is an acceptable method for specifying a datum and indicating an applicable geometric tolerance.

    To be sure that the wall thickness is being controlled, the datum feature should have a limit tolerance reasonably controlled as well as the related I.D..

    The thinnest wall average would equal:

    [(Smallest OD) - (Largest ID) - runout] / 2

    All values should on diameter, as specified .

    Last edited by Kelly Bramble; 10-14-2014 at 08:26 AM.

Tags for this Thread

Posting Permissions

  • You may not post new threads
  • You may not post replies
  • You may not post attachments
  • You may not edit your posts
  •