Lens hood length - avoiding vignetting

[orto]

The problem with a secret is that it remains so... if George gets hit by a bus what then?

I didn't speak about it with him. Who knows?

[Q.G.]

Dan,

Though i (literally) don't know about George, and though i perhaps would have chosen different words, i agree.

[jd callow]

FYI orto, GoldFart and George are all the same user with different ID's on APUG and have been merged into Orto. Whether the APUG George and the lens shade George are one in the same is something I can't answer. Having multiple IDs are generally not allowed.

[makan]

George calculated all of my lens shades, all working perfectly. I'm horrified his posts got corrupted.

[Uhner]

...

[hoojammyflip]

What an enjoyable thread! I've done some back of the envelope drawings and concur that Helen's elegant simple arithmetic agrees with my drawings.

Simply explained for anyone wondering how to derive it, take half the FOV (call it theta), and HD=2.(HL.tan(theta)+LD/2), where HD is hood dia, HL is hood length and LD is lens dia. [theta is actually equal to arctan(filmDiag/2/focalLength), so I think Helen's calculation is perfect, the trig functions cancel out]

If you want a more efficient design, imagine fitting the rectangular hood inside the circumference of the round hood. For a 2:3 35mm format, the hood will have sides of 0.55xRHD:0.83xRHD, which again comes from simple Pythag stuff.

The rectangular hood will cut flare by nearly half as the area of the rectangle is 0.59 of the round hood.

For a further improvement, I'd be really interested to hear how to adapt this calculation for stopped down lenses!