I would like to know if extensometer is necessary to determine yield strength of nickel 718 alloy? Or using cross head movement to estimate the sample strain is sufficient?
In my experience, an extensometer is NOT required but makes it a whole lot easier. When the tensile test first begins, there is a short period during which the teeth of the specimen holder are sinking into the specimen. Although the crossheads show extension, the specimen is not really deforming. Once the teeth have sunk sufficientl into the specimen, then the true deformation starts. If you are using crosshead movement, you need to clean out the first part manually in Excel, shift all strain data, etc. When you use an extensometer, this is not necessary. Moreover your data will be more reliable. Good luck!
If you do not want to use extensometer, then you need to perform elastic correction on the stress-strain curve obtained from UTM. After elastic correction you can fit a straight line at 0.2% offset to obtain the correct yield strength of the material.
Yes it needs extensiometer. Otherwise you may have "order of magnitude scale errors". The errors roots in the frame compliance, loose mechanical connections, grip sliding etc... The errors becomes bigger if you have stiffer sample of low gauge length (bigger cross section, stronger materials, short samples: e.g. in compressive tests.
The bad news is that the extra deformations are not linear with load as previously precisely investigated by Kalidindi etal, experimental mechanics, 37(2)210-215.
An extensometer always makes the work easy. It can be easily integrated with stress-strain analysis software, where you can separate the plastic part and the elastic part of the strain. Also issues related to machine compliance/stiffness, are eliminated as you have lots of data. In an UTM which is servo-mechanically controlled, the above integration becomes much easier.