Personal computing discussed
Moderators: renee, Starfalcon
Waco wrote:Sometimes I think peltier gets thrown out as an excitement word. I keep finding the suggestion that Programable Variable Temperature (PVT) inlets on GC's can be peltier cooled. I think, "that would be something to see", considering that these inlets can be cycled from -40C to 250C every 15 minutes. But other than as a teaser, I can not find any actual examples. Probably because it would have to be a really brutish power hungry cooler.A smart controller can control condensation and keep the water above the dew point. However, why go through the hassle of a watercooling system and not overclock? If you overclock, those TDPs go flying out the window pretty quickly. It also used to be that cooling closer to ambient would increase clock headroom, but most recent chips aren't thermally limited with halfway decent blocks and a good die interface.
Add in the heat/power usage/complexity of pelts...and I know I can't justify them. I love the idea, but it just isn't going to meaningfully increase performance and can very much worsen the noise profile of the system.
just brew it! wrote:fish are not a heat source
just brew it! wrote:fish are not a heat source
Chrispy_ wrote:I last dabbled with TECs in 2001 with a Thunderbird B Athlon 1.4GHz.
Condensation and massively increased cooling requirements made it a nightmare, and I seem to remember the peltier drawing far more power that required silly-high fan speeds even when the CPU was idle.
Vhalidictes wrote:Chrispy_ wrote:I last dabbled with TECs in 2001 with a Thunderbird B Athlon 1.4GHz.
Condensation and massively increased cooling requirements made it a nightmare, and I seem to remember the peltier drawing far more power that required silly-high fan speeds even when the CPU was idle.
When you have a really nasty heat problem (where you would look for help from peltier cooling), that is the precise scenario where peltier cooling doesn't work well. The cooler can't sink it's own heat (generated by the peltier effect).
I did once have the idea of combining a peltier element inside a water cooler, so that the extra heat had a efficient sink, but I don't know that this would work well. I also haven't seen this kind of solution even as a homebrew, although it's possible that it would be too expensive/complicated to bother with.
Captain Ned wrote:just brew it! wrote:fish are not a heat source
No, but they can be a voltage source.
just brew it! wrote:Tiny correction, most aquariums also have a heat load from the work put into the tank by circulation pumps. My 55 gallon uses two power heads and a 1/10 HP circulation pump for the fluidized bed filter as well as a UV sterilizer.As already noted, unless you get one that is sized to cool a really large aquarium, it is not going to keep up. Being cold-blooded, fish are not a heat source; so the chillers are only designed to remove heat which diffuses into the tank through its walls.