I'm no metallurgical expert, but I have always heard about the need to
heat treat aluminum welds when strength or reliability are issues. My
background is in semiconductors and even evaporated aluminum layers on
silicon are heat treated after deposition to relieve stresses.
A bit of internet sleuthing, including at the Aluminum Association web
page (www.aluminum.org), reveals that aluminum alloys only provide (or
recover) their rated strength through proper heat treating ... without
it they are not much stronger than generic aluminum. In addition,
aluminum welds are said to be subject to greater shrinkage during
cooling (compared with steel welds) and also hydrogen embrittlement if
the atmosphere is not well controlled. Even the self-proclaimed "bible"
of aluminum welding published by the Aluminum Association includes
information on post-weld heat treating to address the above issues.
_Welding Aluminum: Theory and Practice_
The problem with brittle welds is that they don't conform well to any
semi-linear strength predictions such as you might find on a
stress/strain graph. A brittle weld is likely to fail catastrophically
in a very non-linear fashion and without warning.
The Aluminum Association book, by the way, also discusses the need to
properly design the joint to be welded if optimum strength is to be
achieved. I suspect that lapping a Z-brace to a tube is not what they
had in mind.
> The lower strength of welded aluminum is recognized in the Aluminum
> Association Specifications standard, and the new Heights series used the
> lower value in the design and analysis of the tower sections. That standard
> makes no mention of 'baking' a weld to regain the lost strength.
> Keep in mind that the 'allowable' strength of any material is the limit
> value used in any design. It is arrived at by discounting, with a safety
> factor for the material in question, the minimum specified limits of the
> yield & ultimate stress for the various types of stress (tensile, shear, &
> bearing). Generally the discounted yield strength is the value used for
> design purposes. Thus the design has an inherent 'safety factor against
> failure, the main purpose of which is to have some margin against
> imperfections in the manufacturing process or other variables.
> Of course, if the welding is defective, all bets are off.
> I do know that Heights is using the TIG welding process in the new series,
> but that still leaves operator skill & technique as a significant variable.
> I do not know the methods used to manufacture the 'old' series.
> Bob, W5LT
TowerTalk mailing list