undefined | Better HN

0 pointsbeala1y ago0 comments

I'd love to be wrong about this, but I'm very skeptical that the aquarium chiller pictured in the post can move enough heat to cool a human. As mentioned in the article, it uses thermoelectric coolers which are extremely inefficient.

I see at least one aquarium chiller on amazon that uses a compressor, but then you have to wonder if it's quiet enough to sleep next to.

0 comments

zemvpferreira1y ago

Same, though I've seen thermoelectric chillers of that size moving ~200 Watt and a human produces less than 100 Watt at rest. The ones I saw on Amazon for $150 claimed to move around 70 Watt which is ballpark useful. You wouldn't want to cool down to a very low temperature anyway, just remove the heat you produce yourself.

bealaOP1y ago

Good point. That does sound plausible then. Here's my napkin math after some quick googling:

- A human produces about 40 watts of heat while sleeping.

- Thermoelectric coolers have a coefficient of performance (CoP) between 0.3-0.6. So for every watt consumed, they can move 0.3-0.6 watts of heat.

- The wattage consumed and moved all needs to be dissipated.

This random chiller [0] on amazon consumes 100 watts, so perhaps this could move 60 watts max. CoP drops as the temperature difference increases. And it's unclear if the unit can dissipate 160 watts steady state.

But it could plausibly keep you from heating up on a warm night. It doesn't seem like there's much margin for actually cooling you down tho. If someone wanted to experiment with this, I'd definitely read that post.

[0] https://www.amazon.com/MOQNISE-Aquarium-Circulation-Function...

j / k navigate · click thread line to collapse