It is likely to raise your power bill significanly - like by a large factor.

The power required by an air conditioner is proportional to the amount of heat it pumps times the temperature difference across which it pumps it.

Suppose your air conditioner is putting out 70 degree air and the outside temperature is 95.  You'll be pumping across a 25 degree differential.

Now suppose your water heater is set at 120 - the lowest it should EVER be when the house is occupied.  Now you're pumping the same amount of heat across a 50 degree temperature difference - and paying twice the electric bill to do it.  (Note that the air conditioner is probably more than half your electric bill, too.)

Suppose your air conditioner is putting out 70 degree air and the outside temperature is 95.  You'll be pumping across a 25 degree differential."

I didn't know that. I thought it was more like the power to run the fan and to pump a pressure through a compresser, tubing, and radiators. Since you are still runing the same motor and pressure though the same hardware I always figured the only extra cost to be pumping the water across the heated radiator to extract as much heat as possable.

So kinda what your saying is if your house is 100 degrees and the coils outside are in the shade (cooler) you use more electric then if the coils are in the sun (hotter)? And you would use the most electric if you bury the coils in the ground which would cool them even more?

You might be right, but to me that doesn't make any sense at all, then what is the purpose of heat pumps? Wouldn't it just be cheaper to use all electric heat then following that theory and toss out the compresser and extra motor? Then mount the airconditonair indirect sun so it stays hot?

Now my house gets over 100f in the summer at times, so to cool it to 80f I should use a HOT air conditioner mounted in the sun and I use alot less power to get that 20f difference than if I cool it and mount it in the shade, or even bury the radiator in a tank of cold watter?

One thing to consider though, perhaps is the missing factor, the unit runs a shorter time to produce the same temp difference, or it does more total work in the same amount of time. Let's say I want my house at 80f and it is 100f now. I need 20f in cooling capacity total. SO if I have a unit cooling at 40f in theory I think it should run half as long (roughly) as a unit cooling at only 20f which I would think should run almost constantly. Either should drop the inside temps 20f total.

Or I can get more work from the colder unit and drop the temps to 60f using it constanatly at the 40f max, whereas I would never get below 80f with the 20f unit.

So I think in the long run I would be better off either getting more work from the same unit, or I would use less power because it only runs half as long.

I am planning to cool my air conditioner this year, either using water or building a custom unit and moving the radiators. But I have no data to compare it too for any power usage differances. Although I actually need more cooling from the same unit, not doing it to lower power usage which would be nice too though.

.
nothing to lose

To heat the water just put all the hot water lines on the exhaust side of the raditator! You don't heat the raditor, it still gets ambient temp air, the hotter air leaving the radiator flows over the water lines to heat those. Don't block the air flow!

Don't put the hot water lines on the intake side of the radiator, you may heat the air hotter than ambient temps which would result in lower cooling effcientcy of the radiator.

I think I misread the post earlier, SORRY! I was thinking cooling the radiator and the post was reffering to heating it I think. I do agree if you heat it that is bad!

.
.
nothing to lose

By the way, an air to air heat pump or air conditioner cools more efficiently when the external unit is in the shade (cooler).

Any opening of a refrigeration system is subject to federal regulation.  

Tube in Shell heat exchangers have been used since the 1930's to supply hot water from air-conditioning systems in TEXAS.  I'm sure they have been used elsewhere.  I think Phillip Hobson 'patened' such a device in the last 15 years or so. Please don't try to bring up  a 'prior art' discussion.

You don't cool 95 degree air down to 75 degrees.  The exchanger temperature is normally 40* or more lower than what you are trying to have as a 'room temperature'. The temperature at the head of the compressor will go well over values needed to heat water.  It will also LOWER energy consumption for cooling.

Installers of air-conditioners are normally quite IGNORANT of such devices.  AND, new stuff will always be slow to be adapted.  ANY problem to a newly installed system will be attributed to the UNKNOWN device (read that as 'I ain't seen that before).  The installer doesn't care about about your electricity bill, he cares about getting paid and not having 'call backs'.

I was fortunate in having a friend who learned refrigeration before WWII.  His father was in the refrigeration also.  There is very little 'new' in the world of energy transfer. Good research will yirld a lot of information.