.25 BTC BOUNTY for the best answer

[DontMineMe]

Here is your answer: Northern Wisconsin  

[NewLiberty]

This is also location dependent. Some places it is much warmer in the ground than above it.  
In Malibu, the well water is warmer than the air.  

Where did you come up with that?  In Malibu? surely you jest.  

The earths temp is pretty consistent once you get below a certain level.  I don't think it's that deep either, but I don't remember the exact depth.  It's a constant somewhere around 50 degrees Fahrenheit.   A quick search on "underground temperature constant" will give you plenty of good reliable info on this subject.

Not a jest, I came up with it by measuring... so yes I could search the web... or I could use a thermometer on my well water.
Geothermal heat is highly variably in localized areas, certainly you have heard of hot springs and geysers?
There are some hot springs a few miles away.  In this area of California, the ground moves enough that deep heat does not dissipate evenly as the subterranean rocks turn.  The temperature changes depending on how long I run the water due to the temperature at the different depths.

[guigui371]

Lets hop back on topic.....

OP, Can you tell us what the square footage of this datacenter is planned to be and also let us know what kind of budget you are looking to work with....

Woodser

1500 SF.......I would like to keep cooling to $5,000 or less.  I am looking for out of the box ideas here.  This will not be your typical data center!

For out of the box + not typical...Sell the heat. Have your cooling system pay you instead of paying for it.

Run a jacuzzi / steam room / sauna business next to your data center
Heat a pool
Run a laundramat
Distillery, brewery, bakery
Lots of uses for extra heat if you can control it.

The cost of moving the air where it is more valuable from where it is not wanted on your chips is paid by the receiver rather than the sender.
This is also a marketing press opportunity.  

Or as I said earlier try geothermal cooling. Pros: runs almost by itself, won't affect electricity bill, causes no moisture problems and doesn't require maintenance. Cons: you need to dig a hole in the ground (bad idea if you're just renting the place).

I am civil engineer and i can garantee you that this is not a viable option !
for your information :  to get some "calorie" from the ground to AC a house (we are not talking about 40 000w but much less) in an area that is around 30° in summer the set up cost around 10 000€ (13k$).
It's simple/quick explaination, but as english isn't my main lenguage it's kind of hard to developpe more.
Geothermical is not possible in THIS case, it might work for some bigger set up where the budget is bigger too.

I don't know if in winter you need to heat your house (what is average temperature at your place ?) but you can use the heat to lower your heating bills.

As for the cooling part in summer, you can  use some heat to warm up your pool or someone else (also in winter by the way) but the only DIY solution i can imagine is : a large amount of new air from the outside (don't forget to put a filter to get rid of all kind of bugs, leaves...) directly on the hot spot and extraction of the air on the roof.

This would be a not so expensive solution and with the heat in winter you can warm houses or swiming pool !! think about it  you can make an extra $ selling it to neighboorhs.

The picture below explain what is a thermical exchange (basic stuff).
Usually you get the heat from the outside to have even more heat on the outside (work up to -15° that is why the snow men). But you can set it opposite, use your hot zone (garage with the rigs) to warm up the outside (or a swiming pool or anything)... it will result of the heat being transfered outside.
 

[U1TRA_L0RD]

I highly doubt anyone is going to get the bounty, OP is just stealing your ideas, Unless this comment makes him reply in rage.

[crazyates]

I highly doubt anyone is going to get the bounty, OP is just stealing your ideas, Unless this comment makes him reply in rage.

You must be new here. Let me check.... yep, less than 2 months. Yoch has been around forever, and is very professional. He's probably one of the more successful miners on this forum.

In the past, he has posted bounties for questions. When they're answered, he pays.

[NewLiberty]

I highly doubt anyone is going to get the bounty, OP is just stealing your ideas, Unless this comment makes him reply in rage.

You must be new here. Let me check.... yep, less than 2 months. Yoch has been around forever, and is very professional. He's probably one of the more successful miners on this forum.

In the past, he has posted bounties for questions. When they're answered, he pays.

Frankly, I'm more interested in a uniquely interesting success than a 0.25BTC bounty.

As I slowly inch myself closer to the solar powered mining array that heats my pool and jacuzzi and sauna, and dry's my towels, I can finally aspire to that noble profession: towel-boy.
Yoch's efforts can only benefit this result.

When my rig is lit by the power of the sun and nothing is wasted from its output, we'll have something worthy. 
Whatever Yoch's ultimate result will be, it is sure to be interesting.

I have a little math yet to do on the surface area of the pool and daily average unheated temperature to see how many BTUs I can pull out of a server room.  Having a little tropical pool zone would be a nice bonus to a mining operation.  Until then I can savor imagining the extra enjoyment of reclining in a jacuzzi knowing that it is heated by the byproduct of securing bitcoin.

[U1TRA_L0RD]

I highly doubt anyone is going to get the bounty, OP is just stealing your ideas, Unless this comment makes him reply in rage.

You must be new here. Let me check.... yep, less than 2 months. Yoch has been around forever, and is very professional. He's probably one of the more successful miners on this forum.

In the past, he has posted bounties for questions. When they're answered, he pays.

You offended or something? 

Why hasent yoch replied for a while?

[jermwerty]

My understanding is that evaporative cooling can pose a risk to electronics, as the air might become saturated with moisture.  I am in search of some expert advice on this subject.  

1)  Is this a feasible option?
2)  How risky is it?
3)  Can the risks be addressed?
4)  If not, what is a better option?

I work for a company that runs a few different datacenters with cheap power in WA state.  We do actually use evaporative coolers in one, pretty neat system.  So let me do my best here:

1)  Yes evaporative coolers are an option for datacenters but it depends on the type.  The cooler you linked to would be a "Direct evaporative cooling" type which basically is just like a regular swamp cooler (see: http://en.wikipedia.org/w/index.php?title=Evaporative_cooler  then the "Direct evaporative cooling" title) and is not suited for datacenter use as the temperature drop is very low and a large amount of moisture is added to the air.  Eventually the air can become saturated with humidity too high for the electronics to survive and then you risk frying electronics simply due to condensation concerns...

2)  Well for a quick test get a regular swamp cooler in a closet, put a computer near the air output and see how long it lasts...  There is obviously a much more technical answer but even then it just depends on a lot of factors.  Most swamp coolers pull in outside air and rely on air entering/leaving the space to prevent the buildup of moisture.  Remember these "evaporative coolers" feel a lot colder to us humans than it actually drops the temp due to the evaporation aspect.   Again if you were to just recirculate the same air over and over eventually it would just be hot muggy air, and you would get condensation and start frying stuff.  Really evaporative cooling could be described as "add water to the air to make it feel cooler".  Again - this only goes so far and only works so much.

3)  Yes by changing the design to "Indirect evaporative cooling" (refer to the wiki link) where you use a heat exchanger this eliminates a large amount of the humidity.  But even this has its limitations and thus the options used for a datacenter are generally a standard AC/evaporative combination which leads us to:

4)  What datacenters use:  http://www.trane.com/datacenter/pdfs/TR_EvaporativeCooling_web.pdf  (problem:  $$$$$)

Really all the core data is written right on the wiki:

DISADVANTAGES:

• The air supplied by the evaporative cooler is typically 80–90% relative humidity;
• High humidity in air accelerates corrosion, particularly in the presence of dust. This can considerably shorten the life of electronic and other equipment.
• High humidity in air may cause condensation of water. This can be a problem for some situations (e.g., electrical equipment, computers, paper, books, old wood).

So how much humidity do you want?  Not this much:

"a problem in the facility's building-management system led to high temperature and low humidity air from the hot aisles being endlessly recirculated though a water-based evaporative cooling system that sought to cool the air down – which meant that when the air came back into the cold aisle for the servers it was so wet it condensed."

Long story short:

"Some servers broke entirely because they had front-facing power supplies and these shorted out. For a few minutes, Parikh says, you could stand in Facebook's data center and hear the pop and fizzle of Facebook's ultra-lean servers obeying the ultra-uncompromising laws of physics."

http://www.theregister.co.uk/2013/06/08/facebook_cloud_versus_cloud/

The problem is any home-grown evaporative cooler solution is going to have that exact same problem.  Miner heats air > Evap cooler adds water (cools a bit) > miner heats air more > evap cooler adds more water (cools even less) > miner heats air > .... rinse repeat until you have a jungle!  ... then POP/FIZZLE...

Why not just leave a window open?