Dismiss Notice
Join Physics Forums Today!
The friendliest, high quality science and math community on the planet! Everyone who loves science is here!

Uni-axial Tensile Testing of Deformed Bars

  1. Jun 25, 2012 #1
    ASTM provides standards for testing that's obvious enough if talk about the standard specimen profile; but idk that some specimen ,for instance round bars, are recommended to have a standard Dog Bone specimen but in case of deformed bar we don't have any special specimen.

    Tensile tests are conducted on the as manufactured bars. Although this might be due to the fact that standard says so.

    But the question is why the standard recommends this ??
  2. jcsd
  3. Jul 7, 2012 #2


    User Avatar
    Science Advisor
    Gold Member

    The original shape you mean? Ease of testing (e.g. attaching to machine), uniform stress state at the section of interest, size for that section which for typical materials yields a representative (size independent) result to name a few. One of those things that has just evolved out of practical needs.
  4. Jul 8, 2012 #3
    yes i mean the shape exactly . . . are there some mechanical considerations for this ? well there are also some dependent properties which are not intrinsic, if it was for properties which regard for size exclusion it might not pose a problem.
    Thanks for ur support, im looking forward for something more.
  5. Oct 18, 2012 #4
    very thin specimen may come under plane stress condition and thus retard the growth of crack and on the other hand, thick specimen operate under plane strain condition and may thus lead to accelerated crack growth. hence by experiments an optimum size is found to avoid plane stress and plane strain consequences.
    The shape of dogbone, localizes stress in the smaller section and hence ensures fracture within the gage section.
  6. Oct 18, 2012 #5
    To test the material you must apply a known stress, and since tensile stress needs to hold the part, which concentrates the stress there, the part MUST be wider at the ends.

    So unless you can make the part narrower or thinner in the middle, you can't make a sensible tensile test. Sorry for that.
  7. Oct 18, 2012 #6
    I don't see why this follows.

    But see also the date of the original discussion.
  8. Oct 19, 2012 #7
    Because the centre must break, not the fastenings.
  9. Oct 19, 2012 #8
    Enthalpy, in post 5 you offer one reason in post 7 what appears to be a different reason for stating that the ends must be wider than the middle.

    I don't see that either prove your case.

    Pukb did offer a sensible statement.
  10. Oct 21, 2012 #9
    Same reason in #5 and #7. Because fastenings weaken the ends, but the centre ust break an not the ends, the ends must be wider.
  11. Oct 21, 2012 #10
    So how do you account for the shape of the speciment in the brazilian tensile test?

    This test appears in ASTM, BS and Euronorm format.

    There is considerably more to good test design than you are making out.

    One particular issue is encapsulated in the title of this thread


    Creating this stress condition as far as possible, definable at a known section, is one of several important reasons for the dumbell shape.
  12. Oct 25, 2012 #11
    Alright dog bone specimen is not an issue guys, what i was trying to ask is why DEFORMED BARS donot have the same specimen for tensile testing ?
    Perhaps anyone over here knows this
  13. Oct 25, 2012 #12
    Well if you were to write in proper English, and include a full description (for instance references to the ASTM you mentioned) you might get something more.

    As I already said there are good engineering reasons behind testing procedures, but to understand why we need better initial information.

    The round bar test is testing the material in particular.

    Given that a worked bar will have started off as a round bar (and the material tested as such), what do you think is being tested?
    Last edited: Oct 25, 2012
Share this great discussion with others via Reddit, Google+, Twitter, or Facebook