Here are the answers to your questions as has been my experience.
1. I have heard that there are a lot of invalid tests that occur when testing the HAZ. Is there anything that can be done during welding to minimize this?
Invalid specimens result from uneven crack growth during the fatiguing portion of the test process. Uneven crack growth is caused by varying degrees of mechanical properties across the plane being tested. As an example, visualize a 45 degree butt weld being sampled through the HAZ. The notch in the specimen would intersect weld metal, HAZ and base metal. This sets up the possibility that the crack front under the notch will not propagate evenly as it intersects the different areas of the weldment. The main reason for this occurrence is variation in yield strengths. Specimen pre-compression and repositioning of the specimen can be the difference between a valid or invalid crack front. The biggest problems are found in smaller weldments because the specification tolerances provide less latitude. Many company procedures now require the weldment to have a straight side on the groove so that the notch can be mostly contained in the HAZ. Also I see invalid specimens in weld metal CTOD specimens due to weld defects interfering with the fatigue crack growth.
2. Do you use the B x 2B x W, or compact specimens? Which is easier, or cheaper to use?
B x 2B is the preferred size. It is easier and less expensive, especially when testing at low temperatures. B x B specimens are normally only used when material is sparse or a surface notch is requested. Compact tension specimens have more complex machining and testing set-up requirements. They are also difficult to use for thicker weldments.
3. Any other pitfalls, or common mistakes for first time testers.
Be sure and choose a laboratory that is experienced with, and accredited for CTOD testing. The tests are too expensive and time consuming not to be handled properly first time. Always anticipate the possibility of invalid specimens. The lab can only do so much to produce a valid test. Do all the NDE you can on the weld to make sure the sampling area is clean-clean-clean.
I hope this helps and good luck with your tests.
Regards,
Mike Mason
Bodycote Materials Testing
Email: MMason@BodycoteUSA.com <