In a “Rammed car tyre foundations” thread

http://www.greenbuildingforum.co.uk/forum114/comments.php?DiscussionID=4490

on Sep 4th 2009

Posted By: biffvernonHappen I was talking to a structural engineer last night who deals with things like dew points in roofs. She looked at ours and thought it would probably be Ok because the dew point was likely to be well short of the impermeable membrane. There will be a temperature gradient across the 45cm thick layer of straw that forms the roof below the plastic. The position of the plastic will be well in the cold, sub-dew-point zone.

I really thought I was misunderstanding this until I read the last sentence. Nope I wasn't, but how can it be right; if the membrane is in the cold, sub-dew-point zone then surely horrible things will happen?

Yes

Good solid, safe and boring engineering practice would say that if you cannot keep water/moisture out of an area, then you have to give it an escape route.

Somewhere on here we have discussed this, all to do with the molecule size of water vapour and the droplet size of liquid water. Two very different animals and have to be treated as such.
The way I see it you are not going to stop water vapour getting into a void/insulation matrix and you are not going to stop the temperature dropping below the dewpoint. What happens then is the crucial bit, how do you get rid of the water droplets (condensation). The thinner you can spread the condensate out the faster it will evaporate when the temperature goes above the dewpoint.
Question is then, what is the best way to do this.