Yield Strength of tubular sections

[benwalker5]

I am testing tubular sections manufactured from S350 steel. The test involves sitting the section between two trestles and applying a central load for one minute (as a simply supported beam). I then measure the permanent deformation (curvature) in the section after the load has been removed (leaving 1 minute for recovery). Why might I be getting similar results to a section manufactured from S250 steel? Am I correct in thinking that the S350 grade should have less permanent deflection?

I have used max stress = My/I and determined that is should be less than 350Mpa. Should there not be a very limited residual deflection on the S350 section?

Any insights would be very helpful!

 

[OHIOMatt]

Are you sure that you have two different grades of steel? Perhaps the lower grade of steel has similar properties as the higher grade or is the same. That is very often the case in the US. Perhaps you can request or perform yourself a test to verify that the steel is what it is presented as.

 

[FishTheStructure]

If neither grade of steel is being stressed (according to stress = My/I) past the yield strength, then neither should exhibit any permanent deflection.

Be careful of how you are applying load. Your loading mechanism may be causing very high local stresses at the point of load application which also coincides with the maximum bending stresses. Typically a solid cylindrical steel member with a suitable diameter is used to apply load.

Great spirits have always encountered violent opposition from mediocre minds - Albert Einstein

 

[JStephen]

If the steel specifications gives a minimum yield strength, the samples may run considerably over that. You should be able to get copies of the mill test reports that would show actual yields.

The yield stress of materials is fairly well defined by various standards. What you're measuring would include not just the yield stress, but the amount of yield at some particular stress level. So a stress-strain curve with a definite yield point might show quite a bit more deformation than one with a smoother curve, even if both had the same yield stress.

Are you sure you're not just measuring the amount of denting in the tube wall at the supports?

 

[dik]

I would take coupons of the material and test for Fy and Es (actual stress-strain curve; hydraulic equipment allows a very precise determination) and that includes the corners as well as the face. In addition, the geometric properties should accurately be measured. Then you have something to compare.

Dik

 

[structSU10]

Could there be some residual stresses from shape fabrication throwing off results?

 

[benwalker5]

Thanks for the feedback.

I've had certificates of conformity from the supplying mill, I have arranged for independent pull tests to assess their mechanical properties.

The load is applied as a distributed load over 100mm, would this minimise local peaks in stress caused by loading as proposed? There is no denting of the beams at the supports with the loads I apply, therefore I'm just measuring the permanent curvature of the beam.

At present I have tested both grades with identical tests and received nearly identical results (in fact the S250 performed slightly better!) - is this indicative that the independent metallurgy results will show they are in fact the same grade? The sections have exactly the same profile and are tested over the same length.

A couple of other questions:
Where would be best to find the stress-strain curve for my particular steel? Would the mill provide that? What is the best way to predict the results theoretically? Should you include a distributed load over the entire length of the sections to represent it's own weight? I get quite a significant amount of sagging over longer (7m+) lengths.

 

[hokie66]

If the bending stress is below yield, there will be no permanent deflection due to bending once the load is removed. However, there could be local distortion due to the method of load application or the support condition. Is the tube round or rectangular?

As to the stress-strain curve for steel, they are all the same...until they reach yield. Young's Modulus, remember?

Yes, the self weight stress and deflection will be additive to any applied load results.

 

[ameyerrenke]

Did you check the curvature in the section before the load was applied? It is typical for tubes to come bowed from the mill.

Otherwise, I agree with everyone else, the 2 tubes may have similar yield strength material.