Cold side ball on left, hot side ball on right.
IcyBall was a name given to two early refrigerators, one made by Australian Sir Edward Hallstrom in 1923, and the other design patented by David Forbes Keith of Toronto, Ontario, Canada (filed 1927, granted 1929), and manufactured by American Powel Crosley Jr., who bought the rights to the device. Both devices were unusual in design in that they did not require the use of electricity for cooling. They ran for a day on about a cup of kerosene allowing rural users lacking electricity to utilise the benefits of refrigeration.
edit] Operation (Crosley Icyball)
The Crosley Icyball was an example of a gas-absorption refrigerator, as can be found today in recreational vehicles or campervans. Unlike most refrigerators, the Icyball had no moving parts, and instead of operating continuously, was manually cycled. Typically it would be charged in the morning, and provide cooling throughout the heat of the day.
Absorption refrigerators and the more common mechanical refrigerators both cool by the evaporation of refrigerant. (Evaporation of a liquid causes cooling, as for example, liquid sweat on the skin evaporating feels cool, and the reverse process releases lots of heat.) In absorption refrigerators, the build up of pressure due to evaporation of refrigerant is relieved not by suction at the inlet of a compressor, but by absorption into an absorptive medium (water in the case of the Icy Ball).
The IcyBall system moves heat from the refrigerated cabinet to the warmer room by using ammonia as the refrigerant. (Note: household cleaning "ammonia" is actually solution of ammonia in water) It consists of two metal balls: a hot ball, which in the fully charged state contains the absorber (water) and a cold ball containing liquid ammonia. These are joined by a pipe in the shape of an inverted U. The pipe allows ammonia gas to move in either direction.
After approximately a day's use (varying depending on load), the IcyBall stops cooling, and needs recharging. The IcyBall is removed from the refrigerated cabinet, and the cold ball, from which all the ammonia has evaporated during the previous cycle, is submerged in cool water. The hot ball is then heated gently to boil off the ammonia dissolved in the water inside it. (The solubility of ammonia in water drops as temperature rises.) The pressure in the system rises to around 250 PSI, and at this temperature, the ammonia readily passes through the u-tube, and condenses in the colder ball, which is kept cool by the water bath.
When the cold ball is fully charged with liquid ammonia (indicated to the user by a whistle), the device is turned around, placing the hot ball in the cool bath. As the hot ball cools, the pressure in the system falls, eventually dropping to the point where the liquid ammonia in the cold ball begins to evaporate (ammonia has a boiling point of -28ºF | -33.4ºC at standard air pressure), and the cold ball begins to freeze. After several minutes it is cool enough for ice to form on its surface. It is then placed on the stabilizer inside the refrigeration cabinet. The stabilizer is filled with an antifreeze solution which both supports the cold ball and provides a large thermal inertia to moderate the cooling. A small hole in the refrigerated cabinet allows the u-tube to pass outside into the room.
The cold ball has a tube into which a special ice-cube tray could be placed, the forerunner of the "freezing compartment" in modern refrigerators.
The actual construction of the Icyball is slightly more complex than described above, to improve the efficiency: The connecting tube runs to the lower part of the warm ball, allowing the ammonia vapor to bubble through the water speeding absorption, and also serving to stir the solution so heat is better transported to the finned walls. This "bubbler" was bypassed by a liquid (no moving parts) check-valve during regeneration, so that only gas, and not liquid solution was transferred to the cold side. The operation of the liquid check valve was somewhat similar to the water seals (J-traps) used in plumbing drains. Mechanical check valves require too much pressure to function properly in this application. To minimize the amount of water transferred to the cold ball during the recharge cycle, trapping structures were placed in the upper part of the connecting tube, allowing only gas to pass, and directing water back to the warm side ball.
In practice, too high a flame and the water will boil, contaminating the ammonia that, alone, should liquefy in the cold ball, and if the water bath is allowed to warm, the ammonia will not fully condense.
While the Crosley Icy ball refrigerator is no longer sold or manufactured, absorption cycle refrigeration is still in use. In addition to RV applications, ammonia cycle refrigerators are still used in undeveloped countries. These are also batch-cycle devices, but incorporate various condensers, check valves, integral kerosene burner, etc, so that the disassembly and tub of water required to regenerate the Icy Ball are no longer needed. Ammonia refrigeration is also used in large industrial applications, where its efficiency more than compensates for the higher initial cost, and associated risk. Though it was once fairly popular for home air conditioning, concerns related to ammonia leakage have caused mechanical refrigeration to dominate that market.
The following text relates to the Canadian/American version of the device.
We had no electricity, and our icehouse was not big enough to supply us with ice all summer. So we used the icyball. There was a daily routine to keep an icyball running. The following procedure was for the very icyball now in the Henry Ford Museum.
The Canadian version of the device came with a cooling chest that looked like a modern freezer, with the door opening upwards. It also required a large tub for water to cool one ball of the device, a mounting bracket to steady the device on the edge of the tub, and a blue-flame kerosene burner mounted in a tray. On one side of the burner tray was the burner and on the other side, connected by a tube, was an upturned cup into which you would fit a little tin of kerosene. The little tin of kerosene had a domed cap that had two small holes in it, so as to allow the kerosene to slowly dribble into the burner cup, tube, and burner. Fastened to the tube that went between the can and the burner was a float level. In the morning when we got up, we would start a fire, and put a kettle on to boil water for coffee. We would be careful to boil more water than we needed for coffee.
After you had poured out the water you needed for coffee, you would remove the icyball from the chest, and upend it, hot ball downward. (The handle had a bracket to support the device in this position.) You would pour a few cups of boiling water over the cold ball. After much gurgling you would carry the device to the water tub, which was about three feet high. First you would immerse the hot ball in the tub for a few seconds, just to fill a small reservoir on the top of the hot ball with water attached to a whistle. You would then reverse the device, hot ball outside and the cold ball in the water, resting the device on the edge of the tank. Then you would tend to the burner.
You would fill the little can with kerosene and screw the lid on. You would then tip the filled can upside down into the waiting burner can, and light the burner, positioning it under the hot ball. (The burner was not adjustable, and had an asbestos wick.) You could then go about your chores for the morning.
Later, when all the kerosene had been consumed, the water in the hot ball reservoir would boil, blowing a whistle to alert you that the burning was done. (Our whistle had been broken years ago, but we could hear the whistle from our neighbour's icyball.)
Then you would switch the position of the hot and cold balls, putting the hot ball in the tub water (hiss!), and in a few minutes the cold ball would be covered in ice, and ready to be returned to the chest for the day. (The cold ball had a hole through it, into which you could put a small metal ice cube tray.)
Once you got used to this routine, the icyball worked beautifully. It was certainly better than burying your food in the ground to try to preserve it.