Hacker News Reader

Love the idea. Curious if there's any maximal theoretical efficiency. It seems to me that if you had a wide pipe with a narrow pipe inside (each going different directions) that you could theoretically have a system that transfers heat with no loss.

View on HN

You can in theory extend the length of a counter flow heat exchanger to get whatever efficiency you want. In practice, such a long heat exchanger would transfer significant amounts of heat to/from its environment. So then you are limited by how well you can insulate things.

This is approximately correct, except for one big problem with counterflow exchangers, which is frost. Even the membrane types still frost up. What you really want is a regenerator, either a thermal wheel or a fixed matrix regenerator, if you really want maximal efficiency. And it should be what they call the total sorption type, so it has some adsorbent in it. Even then the latent efficiency could be improved with a better adsorbent, the industry standard is zeolite 3A. It's good because it lasts. Silica gel becomes slowly inactive as it adsorbs other stuff that clogs the adsorptions sites basically.

But the adsorbent only starts adsorbing at about 60% RH. So in some cases that actually leads to efficiency loss, compared to some kind of ideal adsorbent that adsorbed at any differential.

Reply on news.ycombinator.com