Is it possible to put a 400' dx loop directly into a vertical well? 
One contractor is suggesting a pump and dump in the same well solution, but that means that we're using energy to pump that water up and than back.  Wouldn't it be more efficient to put a single Dx loop down deep in the well?  Is there a limit on the length of the Dx loop?  How can I estimate the energy absorption of the water in the well?  We will need approximately 5.5 to 6 tons of cooling capacity. 
Thanks for your input. 
James

For that to work for the tonnage you cite, I think you'd have to be awfully certain that the aquifer has some lateral water movement to carry the heat away. A variation on the theme of a standing column with bleed could work. I have no experience with that but it sounds intriguing...sort of a hybrid between closed and open loop.

From what I've read, DX systems seem to use multiple shallow bores, whereas conventional closed loop can have many shallow bores or fewer deeper ones.

A design constraint on DX systems is the need to maintain refrigerant velocity in vertical tubes high enough to be certain to carry the compressor oil up. The oil can't be allowed to pool at the bottom of DX bores. Using thin tubes guarantees refrigerant velocity, but will come at a price of decreased efficiency owing to friction losses. Other problems with high refrigerant delta-P is a requirement for sufficient subcooling to prevent boiling where it is not desired. Too much subcooling is a disadvantage as well - refrigerant already liquefied consumes condenser area needed for condensing vapor.

I'll kick back now and let the DX experts and others take over from here...

thanks engineer,
I'm concluding that I would have a 15 degree F rise in temp in a standing column of water 400 feet deep per hour at 50,000 BTUH usuage. That is a lot of energy to be carried away. I agree that I would need to have some transfer of the water in the well. The bore is through granite so a lot would depend on the cracks in the rock and I won't know that untill after it is drilled. I'm feeling less and less comfortable with the idea of a long Dx loop.

On the DX loop question... Engineer is right. It has mostly to do with oil return and refrigerant velocities. Oil separators might be able to be used, but then there is the problem of getting the refrigerant to efficiently work in that depth. It adds load to the compressor, which is the main cost of a heat pump. The less the compressor has to work, the higher efficiency it will have. I would definitely not recommend going that deep with any DX well.

Engineer, what's deta-P ?

Delta-P in this instance refers to the drop in pressure across a lengthy tube run with refrigerant at high velocity.

Delta-Anythiing means change in the "Anything" parameter. If return air hits an avaporator at 80 degrees and supply air is at 60 degrees. I describe that as a 20 degree "Delta-T" across  the evaporator. T=temperature

In furnaces and water heaters you might see "Delta" referrd to as "rise", in other words, the number of degrees the air or water is heated during its trip through the furnace or heater.

The reason husky dogs don't lose too much body tempature thru their feet is because of the design of their leg arterys and vains. The leg vains and artery are close so as the warm blood moves down the leg it loses heat to the returning blood. When it reaches the foot pad it is at a low tempature and loses little heat to the frozen ground/snow/ice. On its return trip it gains heat from the warm blood going down the leg, thus returning to the core body very little cooler than it left. Does this not happen on a DX vertical loop? Would it not increase effecancy/effectiveness to insulate the hot side for the trip down? Or am I displaying my ignorance, and this is allready being done.

With EarthLinked DX loops, the top half of the liquid line is insulated. This prevents flashing of the liquid from the heat of the vapor line. Yep, it's already being done.