Ultimate Stress and load orientation on steel 1

[Hyrax]

My work in a metallurgy department is having me look into ultimate stresses of different alloys under different testing methods.

I have cylindrical 1018 steel samples. Why is it that when performing a tensile test on it, the ultimate stress is lower than the ultimate stress when testing the same material in a 3 point bending test? The ultimate bending strength is about 2x the tested tensile stress.

I am pretty sure it has to do with load orientation or grain boundaries, but I can't find anything to cite that. Does anyone here have a book or preferably online source I can be referred to that would explain why the ultimate stress in a tensile test is different from a bending, shear, or torsion test?

Thanks!

 

[rconner]

You have not stated exactly how the cylindrical shapes were formed, but nevertheless ignoring any effects from that it sounds like you are comparing essentially beam bending tests with tensile testing. In this regard, you might want to do a search with the keywords "shape factor(s)" (of beam cross-sections). [Also, could there also be some non-obvious shifting of "neutral axes" in some essentially beam testing at high strain levels, that may not be reflected in your stress calculations?]

 

[Hyrax]

Thanks for the response rconner. That is essentially what I am comparing, 3 point bend tests with tensile testing (or tensile, with torsion, shear...). I am not debating the results of the experiments. I'm assuming the values I recorded were correct. I ran an experimental FEA and that supported my results.

The tensile cylindrical bar and the bar used for bending were the same shape except for the grip areas for the tensile bar, the diameters were the same.

Since both samples tested in tension and bending are cylinders with the same cross-section diameter, would shape factors apply? Are you referring to the moment of inertia?

I'll look into the neutral axis shifting, but the results are not wrong or inconsistent. There is an underlying fundamental I am trying to pinpoint. And it's driving me nuts.

Thanks again for responding!

 

[Hyrax]

The cylindrical bars are solid.

 

[TERIO]

How are calculating the ultimate stress in the 3-point bending test from your load-deflection data?

 

[Hyrax]

I'll get that info for you in a bit. Are you saying that the ultimate stress in bending should be the same as the ultimate stress in tensile for an "all things being equal" sample?

I guess a more clear question would be, why is the ultimate stress in a bending test higher than the ultimate stress in bending? Why does steel behave stronger in different test methods?

 

[TERIO]

In a uniaxial tension test the state of stress is uniform up to the ultimate load (necking begins at ultimate load and the stress is no longer uniform). Since the stress state is uniform you can calculate the stress by dividing the load by the area (initial area for engineering stress, current area for true stress).

In bending you never have a uniform state of stress in the material; there is always a stress gradient. Since the stress is varies at different locations it is hard to associate a given load with one value of stress.

In order to know how stress varies through the material you need to know how stress is related to strain (strain is from the usual beam theory presumption that "plane sections initially normal to the neutral axis remain plane and normal to the neutral axis"). A linear elastic material results in the usual engineering beam formula of My/I. Once the outer fiber of the beam yields this is no longer valid.

 

[desertfox]

Hi Hyrax

Have a look at this link its mainly rectangular beams but it mentions shape factor near the end for circular beams.

http://books.google.co.uk/books?id=...#v=onepage&q=plastic bending in beams&f=false

desertfox