E -- | \/ >> 50kWsec = 0.833 KWminutes=0.0138 KWhrs=0.1 cents per >> cooldown >> Also, as you say, I doubt that all of the air fully >> exchanges in a >> typical door opening. The thermal mass of all the other >> stuff in there >> will do two things: 1) make it so that there is less air >> 2) make the >> compressor have to turn on and off less. Other than >> that, it will not >> change the average amount of energy needed per cooldown, >> as whatever >> cooling effect you get from the cold food has to be >> recuperated by the >> compressor. | \/ > So the problem is much worse for empty refrigerator? > One more reason to keep it always full of beer :) Not really. Most refrigerators would have only a small proportion occupied by food, or beer. A dedicated packing with bottled beer would probably not reach 50% fill. Cans stacked diameter to diameter can approach Pi/4 packing ~= 78% but only if the cooled space is an exact multiple of can dimensions in all 3 dimensions. If you close pack the cans in staggered rows max fill increases to - ah - I knew there was a reason they discussed packing in Chemistry all those years ago ... aaagh ... Pi/6 x sqrt(3) =~ 90.7%. Round down somewhat as above. Probably really lucky to get more than about 80%. Note that the 90.69...% is optimum for an infinite sheet of beer bo... er circles. Use a mix of bottle sizes and the packing may be lower and will never be higher. Store your beer in plastic bags and you could do better. You'll be sorry ... http://mathworld.wolfram.com/CirclePacking.html ______ Also If you maintain a high beer fill ratio this is liable to promote a high churn and the penalty for cooling large delta mass of beer is far worse than cooling the same volume of air. Filling it with beer only works if it doesn't get imbibed. To save money adopt the double effective strategy of storing only emergency beer supplies in the fridge and drinking someone else's beer whenever possible. Russell -- http://www.piclist.com PIC/SX FAQ & list archive View/change your membership options at http://mailman.mit.edu/mailman/listinfo/piclist