crab:
A useful technique, and the technical methodology is not bad, if you have the curves available. The problem is that you often don't know the power required (or available) and so, can't draw the curves and figure out the airspeed for absolute maximum climb angle at that moment. The 35-40 Knots is a compromise, and I'll be the compromise is based more on having a repeatable, reliable airspeed indication than on anything else.
The US Army did some tests years ago with a UH-1D, and they recommended 28-30 knots as the best angle of climb speed. they had used a special airspeed system to arrive at this number. The problem was that no-one could get a repeatable indication off the production system at those speeds.
I checked with my more learned FW brethern on the subject of Vx (max angle of climb airspeed), and rather than power required vs. Airspeed, it's actually based on rate of climb vs. airspeed, and you draw the line where the rate of climb starts to curl back (i.e. less rate of climb as the airspeed reduces). Haven't got that into the syllabus yet...
But, at the end of the day 35 to 40 KIAS works pretty well. It will probably put you into the HV curve, which is why it isn't an 'official' technique, as the FAA requires the climb profile in the FM to stay away from the HV curve. But if you know why you're doing it, in a Part 27 helicopter, fill your boots.