4 ms·
CO2 heat pumps are an interesting trade-off. I think Sanyo used to make them and they work well but never get the best SCOP values, tho I don't think they are f
by jnsaff2 3y ago
CO2 heat pumps are an interesting trade-off. I think Sanyo used to make them and they work well but never get the best SCOP values, tho I don't think they are far off.
Their major benefit obviously is that the refrigerant has a GWP of 1. Instead of several thousand for the mainstream gases.
So say an air-to-air home heat pump with about a kg of refrigerant could release couple of tons of CO2e in case of a leak or an unknowing/caring worker cutting the pipes without properly collecting the gases during the dismantling.
This could easily negate the lifetime atmospheric CO2 savings of that heat pump.
Whereas in case of a heat pump using CO2 it would be a non-issue.
- pstuart 3y agoNow to get the price down so it crowds the dangerous stuff out ;-)
- samtho 3y agoIt’s not really about price, you can get refrigerant grade CO2 for about $2/lb while other common refrigerants can range from $4-$14/lb. The problem is making sure equipment is certified to handle r744 because it does have a different thermo profile, expansion ratios, and critical point. As you can imagine, manufacturers of more expensive refrigerants are probably not super eager to push a cheaper, easy to capture or manufacture alternative that is effectively not patent-enforceable.
- jnsaff2 3y agoCO2 is acidic so could there be any engineering problems working with CO2?
- jabl 3y agoAFAIU the main challenges is that the thermophysical properties of CO2 are different than for the usual fluorocarbon refrigerants, so it needs a different design taking that into account (among others, needs to handle much higher pressures). The advantage of the fluorocarbon refrigerants is that the current automotive one, R1234yf, is mostly a drop-in replacement of the previous R134a, which in turn was mostly a drop-in replacement for the freon refrigerants (R12) that were banned due to destruction of the ozone layer. So going down this path manufacturers can basically slightly tweak their original designs, potentially originally dating back to the beginnings of the automotive AC era. So saves on R&D costs. The downside is of course that these fluorocarbon refrigerants are very expensive compared to CO2, and the ever tightening F-gas regulations around the world meaning there seems to be a never-ending rat race where a fluorocarbon refrigerant gets banned, necessitating the development of a replacement which is more expensive, and then in turn gets banned in a decade or so. Rinse and repeat.
- citrin_ru 3y agoSufficiently dry CO2 cannot be acidic, only with water we are getting an acid (as far as I remember school chemistry)
- throwawaymaths 3y agoIt's still generally a lewis acid (depends on the reaction ofc) but it's effectively unreactive.
- usrusr 3y agoThe price for the medium in the loop isn't much of a factor in the total cost of a heat pump system.
- samtho 3y agoThe manufacturer of the heat pump system is rarely the maker of the refrigerant but they maintain a close relationship with them. That relationship would be harder to maintain if they suddenly decided to use a very common substance rather than the Dow, Carrier, or Honeywell r410a. It’s likely these companies would be happy to sell r744, but the margins are not going to be as good.
- stephen_g 3y agoThat's not exactly how I understand it - my understanding is that they are generally very good in terms of COP but the trade-off is that they operate at much higher pressures than other refrigerants (including the ones we're phasing out), so material properties, servicing, etc. are more challenging. (of course, it's more complicated than that because COP changes depends on input and output temperature, both in absolute and relative terms, and different refrigerants can be better for colder inputs or hotter outputs etc., but I am talking in general).