The SHPB can be used for tensile and torsion testing as well as compression.Tension testing in a Split Hopkinson Pressure Bar (SHPB) is more complex due to a variation of loading methods and specimen attachment to the incident and transmission bar. The first tension bar was designed and tested by Harding et al. in 1960; the design involved using a hollow weight bar that was connected to a yoke and threaded specimen inside of the weight bar. A tensile wave was created by impacting the weight bar with a ram and having the initial compression wave reflect as a tensile wave off the free end. Many researchers took up this work and made some advancements in SHPB testing.
Following references would be helpful
1) J. Harding, E. O. Wood and J. D. Campbell, Tensile Testing of Materials at Impact Rates of Strain, Journal of Mechanical Engineering Science 2 (1960) 88-96
2) K. Ogawa, Impact-tension Compression Test by Using a Split-Hopkinson Bar, Exp. Mech. 24 (1984) 81–86
The SHPB can be used for tensile and torsion testing as well as compression.Tension testing in a Split Hopkinson Pressure Bar (SHPB) is more complex due to a variation of loading methods and specimen attachment to the incident and transmission bar. The first tension bar was designed and tested by Harding et al. in 1960; the design involved using a hollow weight bar that was connected to a yoke and threaded specimen inside of the weight bar. A tensile wave was created by impacting the weight bar with a ram and having the initial compression wave reflect as a tensile wave off the free end. Many researchers took up this work and made some advancements in SHPB testing.
Following references would be helpful
1) J. Harding, E. O. Wood and J. D. Campbell, Tensile Testing of Materials at Impact Rates of Strain, Journal of Mechanical Engineering Science 2 (1960) 88-96
2) K. Ogawa, Impact-tension Compression Test by Using a Split-Hopkinson Bar, Exp. Mech. 24 (1984) 81–86
If you are interested in stress-strain curve before maximal stress (f.t) I would use LVDT-(multiple use) or resistance strain gauges (one time use). Foil resistance strain gauges are quite expensive, but in some countries you can still find resistance strain gauges 'in paper'.
You can also try to find someone who do measurements of displacement/strain by visual methodes some times they are called: Computer Vision (multi-scale digital image correlation).
If you want to see postpeak behaviour (what happen if stress achieve f.t and crack appears) - this will allow you to see how crack propagetes and how is its width, you should use special LVDT or other gauges which can communicate with testing machine. It's nessesery because the speed of testing should be dependent from the results from this gauge (for example displacement near crack).
More information you can find in: RILEM TC 187-SOC, Planas, J., 2007. Experimental Determination of the Stress-Crack Opening Curve for Concrete in Tension.