Apparently the planck spacecraft cooled itself (or just some instruments?) down to 0.1K for some period of time.
Presumably one could transfer the heat into a fluid and expel that as reaction mass.
From what I understand, using high abeldo/reflective materials, presumably an artificially cooled object could then be maintained at a temperature much lower than the 2.7K background for quite some time.
The Planck spacecraft had a series of radiative and conductive thermal shields between the spacecraft bus that contained all the power and control systems, and the instruments which were the part that were cooled. The bus kept the instruments in its shadow as well.
The heat of operation of the instruments had to go SOMEWHERE. There were a series of active cooling systems that generated heat while acting as heat pumps, pulling the heat from a constant flow of coolant (already pre-cooled to ~4k on Earth and stored in insulated bottles) and dumping it overboard via radiators to bring it down to 0.1k. This ran for a while until it ran out of helium coolant, which was constantly dumped overboard after sucking away heat from the operating instruments so as to avoid there being warm pipes in proximity to them.
You can’t just get rid of heat. To locally cool something, you have to heat up something else by more than the amount you cool the cold thing such that in the net you are actually heating the universe more.
You can’t just get rid of heat. To locally cool something, you have to heat up something else by more than the amount you cool the cold thing such that in the net you are actually heating the universe more.
Of course—which is why I mentioned expelling a coolant/reaction mass. Today’s computers use a number of elements from the periodic table, but the distribution is very different than the distribution of matter in our solar system. It would be very unusual indeed if the element distributions over optimal computronium exactly matched that of typical solar system.
So when constructing an advanced low-temp arcilect, you could transfer heat to whatever mass is the least useful for computation and then expel it.
Limiting heat flow in and out of a cold object is quite possible. But if its DOING anything it will generate heat.
In theory with advanced reversible computing, there doesn’t seem to be any hard limit on energy efficiency. A big arcilect built on reversible computing could generate extremely low heat even when computing near the maximal possible speed—only that required for occasional permanent bit erasures and error corrections.
Yes—and that is related to my point—the configuration will depend on the matter in the system and the options at hand, and the best development paths are unlikely to turn all of the matter into computronium.
Apparently the planck spacecraft cooled itself (or just some instruments?) down to 0.1K for some period of time.
Presumably one could transfer the heat into a fluid and expel that as reaction mass.
From what I understand, using high abeldo/reflective materials, presumably an artificially cooled object could then be maintained at a temperature much lower than the 2.7K background for quite some time.
The Planck spacecraft had a series of radiative and conductive thermal shields between the spacecraft bus that contained all the power and control systems, and the instruments which were the part that were cooled. The bus kept the instruments in its shadow as well.
The heat of operation of the instruments had to go SOMEWHERE. There were a series of active cooling systems that generated heat while acting as heat pumps, pulling the heat from a constant flow of coolant (already pre-cooled to ~4k on Earth and stored in insulated bottles) and dumping it overboard via radiators to bring it down to 0.1k. This ran for a while until it ran out of helium coolant, which was constantly dumped overboard after sucking away heat from the operating instruments so as to avoid there being warm pipes in proximity to them.
http://sci.esa.int/planck/45498-cooling-system/?fbodylongid=2124
You can’t just get rid of heat. To locally cool something, you have to heat up something else by more than the amount you cool the cold thing such that in the net you are actually heating the universe more.
http://hyperphysics.phy-astr.gsu.edu/hbase/thermo/heatpump.html
Limiting heat flow in and out of a cold object is quite possible. But if its DOING anything it will generate heat.
Of course—which is why I mentioned expelling a coolant/reaction mass. Today’s computers use a number of elements from the periodic table, but the distribution is very different than the distribution of matter in our solar system. It would be very unusual indeed if the element distributions over optimal computronium exactly matched that of typical solar system.
So when constructing an advanced low-temp arcilect, you could transfer heat to whatever mass is the least useful for computation and then expel it.
In theory with advanced reversible computing, there doesn’t seem to be any hard limit on energy efficiency. A big arcilect built on reversible computing could generate extremely low heat even when computing near the maximal possible speed—only that required for occasional permanent bit erasures and error corrections.
But if it were not the optimal computronium, but the easiest to build computroniom, it would be made up of whatever was available in the local area.
Yes—and that is related to my point—the configuration will depend on the matter in the system and the options at hand, and the best development paths are unlikely to turn all of the matter into computronium.