I think many of the factors involved in the fogging have been identified.
- conductivity (and hence temperature) of different materials.
- condensed water on camera
- water vapor in air
- relative volume to surface area ratio of different housings
- sacrificial condensation (great term!)
- did I miss anything?
But we still haven't really solved the heating up factor I don't think.
I have another variable to throw in there if Joewr is still interested, regarding fogging patterns in my P&S. I shoot it in video mode, so it is running alot, but no flash. It gets hot. You can feel it being hot when you take it out of the housing. The hottest parts, predictably, are the battery compartment and the LCD (which is warmer than I would have expected). This is a fairly thin acrylic housing, as housings go. It is a uniform thickness everywhere except the lense port, which is glass. The most interesting thing: The housing fogs fastest in certain places consistently. Where? The window right next to the LCD first, then the glass (yes, it's glass on this housing) window in front of the lense. But not by the battery compartment. Factoid: Turn camera off within the housing for awhile during the dive, condensation reduces, sometimes allowing for some more video later in the dive before it fogs again.
Theory: less air circulation space in these areas maybe? The back of the housing is right up against the LCD. Not touching, but close. Likewise, the lense is jammed into a tunnel like cylinder, with rubber glare reduction matting on the sides that sort of seals it in almost.
Problem with theory: Wouldn't the nice hot LCD prevent the air right next to it from getting down to the dew point, even if the cold housing were below the dewpoint? Also, this doesn't explain why the heat seems to increase the condensation (or rather, reducing heat by turning off the camera seems to reduce the condensation).
I've been around and around with this question Joe - it's really irritating to not be able to explain it using physics and chemstry! (Mech E background myself).
Can you make anything out of this extra data? (if you're not thoroughly sick of the topic by now).