A crazy idea for air-to-water heat pumps
In this thread:
we talked a lot about how the output and COP of various heat pumps declines as the outside temperature drops. None of the air-to-water heat pumps do as good a job of retaining their capacity as state-of-the-art air-to-air units with vapor injection. The performance is determined entirely by the construction of the compressor, which got me thinking that wouldn't it be nice if you could somehow use one of those compressors to get heated water as your output. That got me thinking, imagine if you had a multi-head minisplit, and one of the heads had a water output.
That isn't by itself a crazy idea, that's exactly what the LG HydroSplit is, a head that goes on a multi-head system usually used for air-to-air. Of course the issue with the HydroSplit is that it doesn't seem to be actually available for sale.
So the crazy idea is, what if I built my own head? In heating mode, the heads for a heat pump are really simple: hot, pressurized refrigerant is run through a coil and heat is extracted. I've read about people doing it with just a coil of copper tubing. To produce hot water, you'd probably want to use a flat plate heat exchanger, with a circulator pump to move water through and take the heat away. You'd need some sort of controller, which would assess how much heat is needed, send a signal to the compressor to generate that much heat, and modulate the circulator to remove that much heat. I know a lot about programming microprocessors so the hardest part would probably be figuring out how to communicate with the compressor.
Thoughts?
Comments
-
The com. protocol is a hill too steep for me. I had an off the shelf VFD drive idea for an older city multi unit I have but It's low on the priority list. It's not vapor injected so it's no longer peak tech either. I thought it would be easier to write code for an Arduino than reverse engineer oem it but that's an opinion based limited ability not fact. I wanted to heat my radiant slab, buffer tank with an 3/8 R410a copper coil HX in it. DHW has better efficiency potential with cold bottom tank temps. than it is with return slab water temps. There is a void in our market. Some years ago I worked on the Daikin Altherma that worked ok. I thought it was the start of something but they pulled it.
0 -
I was also was thinking maybe you could use an air head unit control board and spoof sensors to alter it's demands to the inverter. You did say crazy in the title.
0 -
There's hydronic CitiMulti outdoor units. And this thread has some pertinent data re communications:
0 -
a friend of mine has a multi-head Mitsubishi system and wired ESP32s to control all of the heads - I think it was pretty straightforward.
0 -
These systems MUST Communicate constantly between the indoor and outdoor computers. Fan commands and actual readings, several temperature readings, vain positions there is a lot of data going back and forth…..
Thats going to be tough to achieve.
0 -
I think the hydrosplit is a European model as they seem to be 50hz units. If i'm reading it correctly though it is an entirely air-to-water system, it's just a split A2W instead of what they have available here which is a monobloc A2W
0 -
I think this might be it:
0 -
This is an excellent idea. Spoofing temperature sensors is easy.
I've always been intrigued by the CoolBot, which allows you to use a regular window-shaker air conditioner to build a walk-in fridge:
That's exactly how the CoolBot works.
0 -
Looking at the wiring diagram for the Mitsubishi MSZ-GL09 (9000 BTU indoor head), they only have three sensors:
Indoor air temperature
Incoming coil temperature
Outgoing coil temperature
It makes me wonder if you could just move the coil temperature sensors to the heat exchanger and spoof the room temperature.
0 -
Hi DCContrarian,
Arctic Heat Pumps makes a great Air to Water Heat Pump with Enhanced Vapor Injection that works great down to -22F in some applications. They are from Winnipeg Canada and are very efficient in colder conditions.
John
I am the walking Deadman
Hydronics Designer
Hydronics is the most comfortable and energy efficient HVAC system.0 -
So if you look at the original posting I quoted, Arctic was included in the original comparison. Yes, they have the best cold weather capacity retention of any of the air-to-water heat pumps in the comparison, but still not as good as the Mitsubishi M-series air to air.
They also had the skimpiest technical information of any manufacturer. More datapoints might make for a more robust comparison.
1 -
Can you get me the source of the study you are quoting and I will look into it.
Thanks,
John
I am the walking Deadman
Hydronics Designer
Hydronics is the most comfortable and energy efficient HVAC system.0 -
See my post about defrosting with liquid refrigerant.
0 -
I thank DCC for starting this tangent from my original posting (i.e., frustrated in VA). As a retired mechanical engineer with plenty of exposure to commercial HVAC through facilities management, I understand all the basics of these systems, large and smaller. That does not make me expert and I'm not savvy with programming or controls, whether PLCs or microcontrollers. Basic relay ladder logic I can follow. Since this thread is of the "crazy" sort, I'll venture some crazy ideas later. Meanwhile, having exhausted myself trying to interest local contractors (and my wife) in A2W hydronic heat pump systems I'm left with a DIY hydronic option or just getting on with retrofit using air-to0-air (most likely the Mitsubishi HyperHeat ducted option because it holds up capacity best in cold temperatures). Otherwise, the A2W options seem all to be too big for cooling while too weak for design temp heating (12F). Design point heating about 51k BTUH; design point cooling less than 30k BTUH. I would need two outdoor A2W units, one for cooling and tandem for cold weather heating (or of course one smaller HP and backup boiler which seems to defeat the promise of "cold climate" HP). Electricity costs are jumping 5 to 9 percent every year. Propane is on site for a heating stove presently.
0 -
In the very first post in this thread I include a link to another thread. In that thread we talked about five different air-to-water heat pumps that are available in the US. I ended up making a chart that compared capacity retention in cold weather and COP vs temperature for all five, as well as the Mitsubishi M-series air-to-air heat pump for comparison.
In the thread are links to the manufacturer's pages where the data came from. Of the air-to-water heat pumps, Arctic had by far the skimpiest information available on their website: they only had COP and output numbers for three outdoor temperatures, and only at one water temperature. Most of the other manufacturers had at least five outdoor temperatures and five water temperatures.
If there is better data available I would love to include it, just point me to it.
0 -
I have more COP charts available from Arctic. We have them for each HP we sell and they are more detailed.
I am the walking Deadman
Hydronics Designer
Hydronics is the most comfortable and energy efficient HVAC system.0 -
What you want is a Midea hyper heat unit with one of their relay modules. The relay module takes care of all communications, so you can use standard thermostat to drive the outdoor unit.
From there the outdoor unit makes either hot or cold gas which you can pipe through your choice of DIY plate HX or an indoor unit from one of the split R410 air to water heat pumps. Last I checked the indoor units are a reasonable cost, so DIY might not be worth it.
Before going down the monoblock route, I seriously considered this option but too many other projects.
1 -
Thanks. How do you get the outdoor unit to modulate? Does it just look at the temperature of the returning refrigerant and modulate itself? Or do you sacrifice modulation?
0 -
It seems that making cold is a lot more complicated, you have to worry about freezing the heat exchanger or not vaporizing all of the refrigerant and returning liquid to the compressor. Plus there's not really a lot of benefit to air-to-water for cooling. Although it is nice to be able to get cooling in small increments.
0 -
The thermostat module runs the outdoor units as two stage heat/cool. There is also an indoor air temp sensor on the thermostat module but it is not too clear what it does. Maybe they use that to modulate the outdoor unit, not sure.
There is not enough clear information out there on how this all works but for sure you can put a standard A frame coil without an expansion valve onto a Midea outdoor unit with it.
0 -
I think for cold you are better off using a multi split and with one zone for hydronic heat and the other zone a ducted air handler.
Hydronic cooling is not worth the complication.
0 -
Agree, mostly. What's nice about hydronic cooling is that you can get a 3000 BTU/hr air handler and run it at the slowest fan speed, maybe 10% capacity, and get just a teensy bit of cooling. Great for small rooms. There's really no way to get something similar any other way. But it does complicate things.
0 -
The arctic unit or any dedicated split or mono block A2WHP is not what's missing. The multi head A2AHP system with a flat plate kit or buffer tank kit would be great. Excellent a2a cooling, slab heating and DHW from one outdoor unit with no glycol. I'm guessing that compromises to accommodate the different loads would render efficiencies so variable that state approvals would be dang near impossible. Too many ways for it to be used wrong means no one gets to try use it right. Unless…. you can make things and are willing to fail to learn and toss good money at maybe. That definitely thins the heard.
1 -
I'm an old guy with very limited understanding of controls so I get lost in those discussions. I remained a generalist engineer by choice and without regret, but that meant trying to understand broad concepts in many fields of engineering and operations without deep dives into the details. Still, I was able to innovate and improvise when needed, using the expertise of specialists when necessary. Simplicity was always emphasized but not always achieved; BAS expertise was increasingly necessary. My capstone/final project before retirement was a water cooled data center in a university research facility; we used a nano-particle enhanced glycol solution for enhanced heat transfer in the server rack heat exchangers and fluid coolers. Now, with only personal resources, I'm faced with hurdle after hurdle just to get multi-zone comfort and control in renovating a 1960s house. Hydronics seems the ideal way to go but too risky for local HVAC contractors to consider.
So now, crazy thoughts: Assuming controls and acknowledging that each heat exchange has its losses, let's say I, Mr. Cheapskate, could get my hands on inverter outdoor units and indoor coils with decent COP characteristics. What can be done differently and in low-tech fashion to create hot and cold water for hydronics? There are refrigerant to water heat exchangers out there but for some reason they seem to be only viable for cooling (Ref. iDronics #13); why?; difficult to defrost the outdoor unit in heating mode? What if the indoor A-coil were placed in a well insulated "hot box" where the refrigerant coil exchanged heat/cold with air-to-water coils? Better yet, what if the A-coil was immersed in an insulated tank of water-glycol and that liquid was pumped to a buffer tank and back and buffer tank temperature communicated demand heat/cool to the outdoor unit as a surguate for room temperature?
0 -
I think for your load you'll need something like the LG Multi V.
https://ashp.neep.org/#!/product/36406/7/25000/95/7500/0///0
It supports multiple heads, ducted and ductless. There is a hydronic head that is (theoretically) available for it, the Hydrosplit:
However, when I last inquired about it LG was only letting people with factory training install it, and those people were not to be found. That may have changed.
Anyway, it's a system that meets your sizing needs and is at least theoretically upgradable in the future.
0 -
I've been asking the local LG distributor about Therma V Hydrosplit for almost a year and never get an answer about availability. Too bad. The NEEP capacity charts look pretty good.
0 -
Daikin Altherma 3 is just about finished with listings in USA, told by Daikin it ships 1st qtr 2025. The 1st Altherma I hear was ok, this one has higher temp capability. Shall be interesting. They have in Europe already. Still tapping my foot waiting for Mitsubishi!
1
Categories
- All Categories
- 86.2K THE MAIN WALL
- 3.1K A-C, Heat Pumps & Refrigeration
- 53 Biomass
- 422 Carbon Monoxide Awareness
- 90 Chimneys & Flues
- 2K Domestic Hot Water
- 5.4K Gas Heating
- 99 Geothermal
- 156 Indoor-Air Quality
- 3.4K Oil Heating
- 63 Pipe Deterioration
- 915 Plumbing
- 6K Radiant Heating
- 381 Solar
- 14.8K Strictly Steam
- 3.3K Thermostats and Controls
- 53 Water Quality
- 41 Industry Classes
- 47 Job Opportunities
- 17 Recall Announcements