If you don't want to affect inside the ends, you can try to high-strength glue them in  steel tube holders.

You can use the dynamic elastic properties analyser for finding the modulus.

Really, you can make a hole at both ends of the composite tube and pass a bolt through for the fixation in testing machine. If your specimen is enough long, you will obtain the uniform stress state in middle part of specimen for the correct measurement of elastic modulus by extensometers or strain gauges.

Also you can use grip for the coaxial fixation of tubular sample without any destruction.

Nima,

First, Tensile testing is a Destructive test, as the sample would be rendered unusable for it's intended function after the process. Non-Destructive Testing methods are those that cause no harm to the specimen, such as Radiography, Ultrasonic testing or Microscopy. If through the tensile test, you pass the elastic deformation point, the specimen would have been "Destructively" tested. Finding the upper limit of Young's Modulus, yield strength and ultimate tensile strength produce this result. In addition, for tensile testing in general, the sample to be tested often requires modification or machining to be placed into a "testable" format, also rendering the sample unusable for it's intended function.

For an attachment method, there are 2 options that come to mind:

1. You can, as suggested by the contributors above, drill a hole through both walls some distance from the ends and use a "Pin" system, having the force from the MTS applied to the pins rather than the sample itself. This has to be done carefully, as the method used to make the holes in your sample as well as their placement can have dramatic effects on the test result, especially for a carbon fiber matrix and related materials.

2. Second suggestion would be to machine or 3D print inserts that conform to the Inner Diameter of the tube with a tight tolerance, place these inside your sample, then use the tubal holding jaws on the MTS, selected to conform to the radius, to apply gripping pressure to the Outer Diameter of your sample without the potential for a crushing effect.

Either of these methods are in common usage. 

I hope that you find these comments helpful,

J. Cole Zimmermann

      If we confine oneself only by task "to measure sample elastic young modulus" as Nima Ahmadi said, the tensile testing, nevertheless, may be assumed as non-destructive test because corresponding stress level are well below yield limit and tensile strength. In other words, significant structural damages are absent in this case.  All the best

John Cole Zimmermann,

Thanks for your suggestion. I have, however, one concern regarding the first method you explained. Although this method apparently gives more accurate result, the drilled holes might cause stress concentration. I was told before the holes need to be drilled with high precision by skilled technicians.

Is there any name associated with this method so i can research more about it?

Thanks  

Nima,

You are correct that the drilled holes may become stress risers if not done properly. You are further correct that the holes would have to machined by a skilled technician, with a close tolerance to position, size and surface finish of the cut edge. As to the position of the holes, I can't really say for the material you are testing.

Below is a link to a publicly available document, describing some techniques for tensile testing in general. The method of mounting we are discussing as my first option is called "Pinning".

http://www.asminternational.org/documents/10192/3465262/05105G_Chapter_1.pdf/e13396e8-a327-490a-a414-9bd1d2bc2bb8

You may consider: Machine holes for your mounting pins into the tubing, then use grommets to protect the cut edge of the holes, as well as allowing the diameter of the pin, and corresponding hole, to be as large as possible in your sample to further mitigate the concentration of stresses.

I hope that you find this useful and good luck!

J. Cole Zimmermann

I highly recommend you read the article that looks interesting as a solution:

Development of a Hollow Cylinder Tensile Tester to Obtain Mechanical Properties of Bituminous Paving Mixtures

by

William G. Buttlar, Ph.D., P,E. Assistant Professor

Ghazi G. Al-Khateeb, M.S.C.E. Graduate Research Assistant

Diyar Bozkurt, M.S.C.E. Graduate Research Assistant

Submitted for Publication in the Journal of the Association of Asphalt Paving Technologists, December 9, 1998.