If you're talking about the aluminum extrusion sold by mtm scientific, they're small enough you can put one end in a vise and twist each blade by hand.
Aluminum will die quickly in low cycle fatigue (stresses in excess of yield, fully reversing) so twist one blade, then "sneak up" the same amount of twist on the other blade(s). You don't want to have to twist them the other direction.
The difference between twisted and non-twisted blades is HUGE. It's like the diference between a ceiling fan and an airplane propeller. The twist keeps each portion of the blade at the same angle of attack with respect to the relative wind for your designed wind speed. The calculated twist is not linear along the blade, but a simple twist is far better than none at all.
Taper is not as important, as long as the root of the blade can handle the stresses. If blades aren't tapered, the load on the blade perpendicular to the plane of rotation becomes much higher, due to the small amount of twist (relative to the plane of rotation) at the tip.
One other important point. Aluminum has no range of "infinite life fatigue stress" like steel does. For example, in general, if cyclic stresses are below 20% - 30% of yield, steel will never fail in fatigue. Any cyclic stress on aluminum will cause fatigue failure. Very low stresses might be on the order of decades, while higher stresses might be on the order of minutes or hours. But just because they ran fine for a week, doesn't necessarily tell you anything about tomorrow.
-Dan M