According to ASTM E8/E8M-13a, a value between 1.15 and 11.5 MPa/s is correct. In my experience, a value between 9 and 10 MPa/s works well. Depending on your machine you may need to convert that value to a force / time value, which can be obtained following the next equation (to convert from MPa/s to kN/min): (strain rate in MPa/s)(specimen cross section area in mm^2)(60 s/min) / (1000 N/kN); this value is suitable until the yield point. In my experience, after the yield point you can follow the next rule of thumb to estimate the crosshead speed for the final part of the test: if the testing rate value before yielding is less than or equal to 10 kN/min choose 10 mm/min; if the testing rate value before yielding is greater than 10 kN/min choose 15 mm/min.
Thank you for you reply, In my case i found the 15 KN/min so i have to prefer 15mm/min cros head speed for tensile test. But you said that the suitable cross head speeds are different for the before yielding and after yielding. So does it means i have to change the cross head speed after yielding during test ?
If the purpose of tensile testing is to characterize strain hardening behavior, I suggest that you do NOT change strain rate during tensile tetsing. In my tests with cylindrical specimens, I have always used a strain rate of 1-3*10^-3 s^-1. This gives you plenty of data past yield strength to analyze. I have found it better to have more data than what you need, as opposed to less.
Testing conditions depend on what you want to measure. If you only want to estimate yield and tensile strength my suggestions work well; if you want to get numeric data, the more of them is the better. In a lot of machines (some of them old machines) you can have different speeds before and after yielding.