I at one point used some VapoChill units for servers that needed to run 24/7 in space that was not cooled over the weekends in the summer (and would get to an ambient of between 130 and 140 F. on some days). Condensation is a huge problem with lower than room temperature cooling methods, and really really tough to deal with effectively on a system that's going to run for more than a few hours at a time, under all conditions. Whatever you do, make sure that you do adequate testing when the relative humidity and environmental temperatures are near whatever the peak is for your your area (probably 100% at around 90-95 F. unless you live in Arizona or the Arctic circle) - rather than assuming that if it's fine on whatever random day(s) you test, it will always be fine.
What you could easily find is that a solution that might work great at 25% RH when the outdoor temperatures are in the 40's fares poorly at 100% in the 90's- and it only takes a single drop of water in the wrong place to make a MAJOR mess of things. Unless you're planning to do some extreme overclocking (in which case you don't care how long things last - and what you've described would be insufficient for anyway) or are forced to deal with some extreme environments (as I was) - I'd urge against using phase change devices for computer cooling. On the risk reward curve, you're just not likely to gain much over more proven methods, yet both cost and risks are relatively high.
Having said all that, if I understand what you're describing, it's is going to be less likely to have condensation issues if properly sealed than what I had to deal with, so ... it's your parts. I mostly wanted to point out that you should make sure you do the testing when both the relative humidity and ambient temperatures are high (unlikely now, as we roll into the winter months), to avoid the nasty surprises when they are high later and the air is heavily burdened with water vapor.
<S>
