Question: I am following ASTM D638-10, and my 'strain at break' results are nearly half of what other labs are reporting for the same material. What is wrong?
Reply: If your material exhibits necking or inhomogeneous strain, ASTM D638-10 specifies that the extensometer needs to be removed at specimen yield. Once the extensometer is removed, nominal strain is calculated to specimen failure. The standard specifies this to avoid tests where the specimen begins to neck outside of the gauge length. A common misinterpretation is that if your specimen necks within the extensometer gauge length that the extensometer can be left on until failure. However, to be compliant with ASTM D638-10, the extensometer must be removed at specimen yield regardless of where the necking occurs, and nominal strain must be calculated to specimen failure.
When the extensometer is left on the specimen until failure, strain at break is calculated by the change in extensometer gauge length divided by the original gauge length. The majority of specimen elongation will occur in the narrow section of the specimen where the extensometer is attached. When removing the extensometer at specimen yield, nominal strain is calculated by crosshead extension divided by the initial grip separation. This method is how ASTM D638-10 specifies strain at break to be calculated. The majority of specimen elongation will still occur in the narrow section of the specimen; however, the entire grip separation is now being used as the gauge length. Thus, this causes lower strain at break results. We have found in our lab that strain at break results are almost doubled when leaving the extensometer on until failure.
We recommended having the other labs remove their extensometers at specimen yield and check their method to make sure they are using nominal strain to calculate strain at break.