Air to Water Heat Pumps for Radiant Heating And Cooling

I have an air-to-water heat pump in my own home, that I use for heating and cooling. I believe that air-to-water has a number of fundamental advantages, which I outlined in this thread:

https://www.greenbuildingadvisor.com/question/are-air-to-water-heat-pumps-about-to-boom

But there are advantages and disadvantages compared to other technologies. The key is to maximize the advantages and minimize the disadvantages.

First, I've never been able to get "radiant" cooling to work, I need lots of dehumidification in my climate and that comes from air being blown over a cool surface, so I use fan coil units for cooling.

The principal advantage is that you can deliver small, targeted amounts of cooling, which means superior comfort. You can put a small fan coil in a small room, like a bedroom, and run it at very low capacity. My heat pump is a Chiltrix, but they sell the same PHNIX fan coil units as Arctic, rebadged. The smallest one available is rated 3,000 BTU/hr — about a third of the smallest minisplit head typically available. You can run it on the ultra-low fan speed which might produce 10% of rated output, you can have it cycle on and off for even less. There's just no way to do that with a minisplit, let alone a traditional ducted system.

Another advantage is higher efficiency. I encourage anyone interested to read this article:

https://arstechnica.com/science/2024/07/the-hunt-for-the-most-efficient-heat-pump-in-the-world/

Basically when you're really searching for high COP nothing compares to air-to-water.

I would say the biggest drawbacks center around the fan coil units, and the availability of better ones. The PHNIX ones seem to be the only residentially-sized units broadly available in the US. They are great for quiet comfortable operation, both in heating and cooling. But — they're ugly. No too ways about it. They're hard to install. Real hard. They're not really made for the US market, and everything about installing them is a challenge. They are dimensioned for US framing, they don't use US plumbing fittings. They're quite inflexible in the positioning of the plumbing. And I wonder about their durability, like most minisplits they don't have a disposable filter but rather a "nylon stocking" filter that seems to let a lot of dust through.

They're also quite expensive. The Chiltrix website has pricing, they start at about $800 for the smallest unit.

Q: what specifically can make an air to water heat pump more efficient?

I can name low temps for radiant floor/wall heating. I also see distribution efficiency via a circulator vs a blower, although I think that’s pretty minor since my ducted blower uses probably 50w. Is it negated by the high head loss through the heat pump? Both also apply to water to water of course. Outside air temps can be higher than ground temps, so I certainly see that. But that applies to air to air too. What else is there?

Most people like radiant floors.

Where is the evidence for this? If >50% of Americans like them, why do ~1-5% have them? That’s a huge gap!

I can think of lots of reasons, @Hot_water_fan , and I dare say — from previous conversations — that we would agree on most of them! For instance — installing a radiant floor in existing construction is… difficult. Not always impossible, but difficult. And most folks own or rent in existing construction. Or… some people like rugs and carpets. They don't play well with radiant floors.

It's not that they aren't quite good — like all other ways of staying warm or cool, there are applications where they are wonderful and applications where… not so much. Nor is it that they are too novel — after all, Roman villas in Britain had radiant floors 2,000 years ago which worked just fine, and my "home" cathedral — Durham — has radiant floors which were installed about 1,000 years ago and are still in use.

As always — when you have a project, identify what you want to do, scan the technology available, pick the best fit for the project and the budget, and go for it.

I run Chevys. Used Chevys. I would love to have a fancy new car, or even a less fancy one. But at 3 times my annual income? Perhaps not. I'd love to have another chunk of Cedric's home have a heat pump for summer cooling and shoulder season heat. Can I afford it? I cannot.

Funny thing. In this country, if you make something people want and can afford, they'll buy it. If they can't afford it they won't. If the only product legally available is too expensive, they'll go without.

"As far as radiant cooling goes, the best way to control that is with a dewpoint sensor operating a three way valve. The dewpoint sensor reads humidity levels and adjusts the water temps to stay above the dewpoint. That way you don't get any condensation issues."

I see two problems with that approach.

First, humidity can change quickly. As I noted above, I monitor temperature and humidity in my house, and humidity typically peaks around 9AM when everyone has just showered. It peaks again in the evening when everyone is home and dinner is cooked. The biggest source of indoor humidity is occupant activity. I've seen the indoor humidity move by 20 percentage points in a few hours.

Changing the temperature of the water isn't going to change the temperature of the surface that quickly. The response time of the surface is going to be somewhere between hours and days depending on how it's constructed.

The second problem is that with a heat pump efficiency is all about temperature difference, if you're cooling water and then mixing it to warm it back up again you're giving away a lot of efficiency. As I said earlier, the key to getting the most out of any technology is playing to its advantages and minimizing the disadvantages. With a heat pump, if you're tempering, you're doing it wrong. Also true with heating.

Fundamentally heat pump efficiency is driven by temperature delta, so yeah, low temperature delta is what makes heat pumps more efficient. Specifically, that it's practical to run at lower temperature deltas with an air-to-water than with other technologies.

For example, imagine a house with a 40K BTU/hr heating load. You could heat it with water leaving at 90F and returning at 80F, you'd need 8 GPM or a 1" pipe. That's doable. It would be a challenge to get 40K BTU/hr from water with an average temp of 85F, you'd need a lot of surface of radiation, but that's basically what the guys who do the extreme efficiency challenge do.

Now imagine trying to heat the same house with an air-to-air heat pump producing 90F air. It would be 20F above room temperature, you'd need about 1900 CFM of air flow. You'd need about a 20" round duct to provide that or maybe a 10x34. You start running into practical issues finding room for ducts that size and it's easier just to bump the heat pump temperature up.

I will concede that with cooling it gets more complicated because your choice of temperature gets driven by the need for dehumidification.

"Overall, I’m looking for efficiency advantages unique to air to water, not something that is also true of water to water."

The big advantage of water-to-water is that water supplies tend to be more stable in temperature than the outside air.

One of the things you see across the heat pump business is that technical innovation tends to show up in the most popular product types first and in the less popular types later. Right now new technologies typically are introduced in minisplits first, then air-to-water and water-to-water even later.

So the most advanced air-to-air give air-to-water a run for their money and in some ways surpass them. Water-to-water seems to be mostly running 20 year old technology and a lot of the theoretical advantage dissipates.

It's also clear that the technology is still evolving rather rapidly.

EVI is useless if you’re ground source yeah? You don’t need low temp performance because the lowest temp it would see is ~32F.

All else equal, doesn’t EVI lead to lower efficiency vs an equivalent ASHP? It looks that way for the Mitsubishi’s I’m familiar with. You trade low temp performance for COP.

That doesn’t seem correct based on what Mitsubishi has published. Anyway, I’ll wait for your long answer.

I wonder if it's time for my challenge again. It's still out there. National Register Landmark house, so no change to appearance inside or out permitted, except in the basement. One outside compressor or equipment unit permitted. Noise level less than 40 dB outside or inside. Known peak heat loss 275,000 BTUh. Known minimum outdoor temperature -15 F. Design indoor temperature 75 F.

The challenge: You install a heat pump system capable of meeting the heating requirement. If, at the end of 20 years, the all up cost, including operation, maintenance, capital investment, replacement if end of life, and lost capital opportunity, is less than the all up cost of the existing oil fired steam system, and if the maximum down time for repair or maintenance has been 2 hours or less, we'll buy it from you at that time. If it isn't, you will bear all costs involved in restoring the building including existing heating system and grounds to the present condition.

There's the table stakes, gentlemen. Meet 'em or fold.

yeah. i know. I just like to poke the bear now and them!

https://forum.heatinghelp.com/discussion/comment/1812300#Comment_1812300

At least on min splits within Mitsu lineup, EVI does not increase COP. It does increase maintenance capacity by great deal. You can look at two similar units one EVI:

EVI:

https://ashp.neep.org/#!/product/31992/7/25000/95/7500/0///0

non-EVI:

https://ashp.neep.org/#!/product/107511/7/25000/95/7500/0///0

Even at low modulation, the non-evi COP is higher. The unit does fall flat at 5F but still well above on COP of the cold climate unit. The one big issue with looking COP only, is the non-EVI unit will blow warm air which is definitely uncomfortable.

@DCContrarian ,

In a system with radiant cooling, rather than using mixing valves to raise the water temperature, why couldn't we either:

• run the whole system at a higher water temperature, and use dedicated dehumidifiers (either several small units throughout the building, or a central one integrated to the HRV)
• run the system at water temperature low enough to get dehumidification, but plumb the fan coils in series with the radiant panels, and control the flow so the water leaving the fan coils is above the dewpoint. Optionally their could be a bypass around the panels to allow control of the SHR.

Luke