Split system heat pump hot water heaters.

[Brian26]

This is a good read on split system heat pump water heaters that I didn't even know existed. Japan has had millions in use for over a decade and in Australia they even put the tank and compressor outside. They also us C02 as refrigerant which has a global warming potential rating as 1 compared to R134a which is rated at 1,430

http://www.greenbuildingadvisor.com/articles/dept/musings/split-system-heat-pump-water-heaters

Looks like Sanden has one on the market in the U.S now. What I find interesting is there are no refrigerant connections. The bridge between the tank and compressor is done with water lines. So the compressor is factory sealed with refrigerant, installed outside, and water lines are connected between it to the tank.

It also looks like they use about half the power of current conventional units available here.

(broken link removed to https://www.sandenwaterheater.com/products/)

Wonder if we will start seeing more of these here.

 

[peirhead]

I think the Air to Water system will be the product of choice in the future...see link to Nordic Heat Pumps, designed specifically for chilly Northern climates!! (broken link removed to http://www.nordicghp.com/2015/01/say-hello-new-air-to-water-heat-pump/)

 

[peakbagger]

I took a course last year on heating and these were mentioned. Tominmaine a member on this forum has had some practical experience with these which were less than good but I believe he is trialing a newer design this winter. In theory it has lot of potential for homes equipped with low temperature emitters for heating

 

[maple1]

What the heck GBA?

I tried to look at that article.

Had to sign up for a free trial in their 'Prime' membership.

Couldn't do that because it then said the pay type selected is invalid for my region, even after having to give them my name & address info - there was no 'pay type selection even available.

I used to think kind of highly of Green Building Advisor - not any more I guess.

 

[woodgeek]

GBA has paywalled all their best new content (including all of Martin Holladay) for over a year now. Sad.

Martin's book was also supposed to come out last year, but did not.

 

[Highbeam]

The pressures used in the CO2 systems are crazy high as I recall. I've known and griped for years that other countries get some superior technology for water heating with heat pumps. I believe it was daiken that has the one product that heats water with a split unit.

What's it gonna take to get these here? What conspiracy theory can you apply to this withholding of technology from our market? Is it the HVAC industry? Energy lobby?

On one hand I like the bugs being worked out before release in the US but on the other hand it's been many years now.

Regular old low temp minisplits that use refrigerant lines can be hooked to a heat exchanger to heat water NOW. Instead they only get hooked to interior air exchangers. HSPF of over 10!

 

[maple1]

GBA has paywalled all their best new content (including all of Martin Holladay) for over a year now. Sad.

Martin's book was also supposed to come out last year, but did not.

I sent them a concise email after posting that, making mention of data mining. No reply yet...

 

[Marshy]

Subscribed!

@maple1, weren't we just talking about these on another site?

 

[maple1]

Sort of, yes - but these are on another level. Way more capable with the CO2 refrigerant - or they're supposed to be.

I'm in the same boat as Highbeam - the delay in tech introduction on this side of the pond is very frustrating sometimes. Went through the same thing when researching my boiler.

 

[woodgeek]

I need to see more data before I jump on the split HPWH bandwagon....I'm still agnostic. It seems like I will get better COP from a semi-conditioned space than an outdoor space (requiring defrost 4-5 mos a year)...you just need a suitable 'semi-conditioned' space from which to recover a little heat that would otherwise be lost. They seem pretty common in older US homes.

As for why we lag new HVAC tech...look at the haters going after the geospring and other HPWHs we DO have.

I had two out of three plumbers not only refuse to install a HPWH for me (4 years ago), but stay on the phone to tell me what terrible tech they were, that I'd regret it, come back to them when I wanted to remove it, etc. Unreal.

I have also always been suspicious of the online reviews against them....a small percentage of owners that just say their geosprings died and they couldn't be satisfied by a warranty claim. I don't doubt there were some lemons, but really?

And then there were prominent blog posts that centered on the heat stealing issue....saying that you would have to pay twice for every BTU, so it was actually more expensive than a conventional electric.

I can't wait to hear what they come up with on these CO2 split systems...the CO2 line can rupture, leak the CO2 out all at once and asphyxiate your family? Explode? The CO2 is made from burning coal, so its not really green? I dunno. They are always more creative with this stuff than I am.

 

[maple1]

I did find some specs on the Sanden linked above - there's some PR on there. Operates down to -20°, COP=5, 15,400 BTU/hr (Gen 3), temp setting 130-175. Pretty impressive sounding - that could provide quite a bit of space heating. Even if that particular unit doesn't turn out to be 'all that', it certainly shows that there should be big potential. IMO.

 

[peakbagger]

Most of the super energy efficient technology is designed or at least made affordable in Japan as power prices are crazy expensive, the Japanese government also invests in new technology as a means of supporting industry. Many of the technologies are not cost effective until the volume gets cranked up substantially and they offshore production. The distribution channels also work against them. Both Mitsubishi and Fujitsu have fairly closed distribution processes and there are a couple of middleman in most sales. Like I and others have figured out we can buy a mini split and install it for about half of what a local franchised dealer charges.

With thermal storage and low temp emitters the potential for this technology is pretty attractive. I question the maximum temperature output as most heat pump hot water heaters are known to have limitations on output temp. The COP is a dangerous number as usually its established at a given outdoor temp which is quite high possibly 50 degrees. Real world studies of mini splits which use the same technology just different heating media shows the heat output and efficiency tanking as the temps drop to the point where its not much better than electric resistance heat. That has been my experience, yes my mini split puts out warm air down to -15 but if there is any moisture in the outdoor air I am going to have to put up with very frequent defrost cycles.

Yes CO2 based refrigerant systems will add new technology and require new tools. The professionals have already figured it out as its already being used in some commercial installations. The unit price of the equipment will go up and reliability may suffer until the teething issues are worked through.

 

[maple1]

I've been on the watch since I did our boiler swap 5 years ago for what I will do next, for when I can't or don't want to do wood anymore. Or if something happens to me - never know. Right now it's our backup electric boiler - which costs arms & legs to run any amount of time. This stuff caught my eye a couple years ago when I first read about it - I'm really hoping it gets itself going before a whole lot longer. Right now I'm thinking about mini-splits. Checked out TOD with the electric boiler, don't think so. Even putting an oil burner back in is back on my radar (shudder). Would much rather go this way though if it ever makes it.

 

[woodgeek]

I did find some specs on the Sanden linked above - there's some PR on there. Operates down to -20°, COP=5, 15,400 BTU/hr (Gen 3), temp setting 130-175. Pretty impressive sounding - that could provide quite a bit of space heating. Even if that particular unit doesn't turn out to be 'all that', it certainly shows that there should be big potential. IMO.

My guess is all these specs are true, but not at the same time.

I can believe COP=5 at low lift (like 90°F to 120°F in the summer, no defrost), but not at -20°.

That said, if CO2 systems can be made cheaply at get COP=5 at reasonable conditions, then make me a non-split HPWH with it.

 

[iamlucky13]

The pressures used in the CO2 systems are crazy high as I recall. I've known and griped for years that other countries get some superior technology for water heating with heat pumps. I believe it was daiken that has the one product that heats water with a split unit.

What's it gonna take to get these here? What conspiracy theory can you apply to this withholding of technology from our market? Is it the HVAC industry? Energy lobby?

On one hand I like the bugs being worked out before release in the US but on the other hand it's been many years now.

Regular old low temp minisplits that use refrigerant lines can be hooked to a heat exchanger to heat water NOW. Instead they only get hooked to interior air exchangers. HSPF of over 10!

It will take either one of the existing foreign manufacturers investing in distributing the products and training and supporting installers (which includes investing in local spare parts distribution so installers are able to properly support their customers), or it will take one of the domestic manufacturers taking the risk to develop their own version. If either of those happen, and the advantages are real, the rest of the market will respond in kind like we're seeing with variable speed compressors.

My gut instinct is to prefer single, well made, high performance unit for both space heating and domestic hot water, but whatever the choices, but I will see what's available the next time it becomes relevant to me. With a little bit of luck and another 5-10 years of diligent savings, my hope is the relevant event will be building a custom home where I can fit whatever heating system looks like the best long term option into the design, but that's a ways off still.

 

[iamlucky13]

I need to see more data before I jump on the split HPWH bandwagon....I'm still agnostic. It seems like I will get better COP from a semi-conditioned space than an outdoor space (requiring defrost 4-5 mos a year)...you just need a suitable 'semi-conditioned' space from which to recover a little heat that would otherwise be lost. They seem pretty common in older US homes.

In a way, a HPWH in an unconditioned basement is sort of like a hybrid geothermal system that also scavenges part of your heat loss through your floor insulation.

The flip side of defrosts for an outdoor unit in the winter is that in the summer it benefits from the warmer outside air compared to your basement. There's pros and cons to either of these options, and they depend in part on your climate.

And then there were prominent blog posts that centered on the heat stealing issue....saying that you would have to pay twice for every BTU, so it was actually more expensive than a conventional electric.

You do pay twice in the winter, but only once in the summer (potentially less, see examples), and even in the winter, depending on your source of space heat, you might not be paying anywhere near full price either time.

Here's a bunch of napkin math comparisons, assuming 70 deg F (39 deg C) temperature change to your water, where I refer to electric resistance heat as "full price":

- Winter or Summer, electric resistance DHW and electric resistance space heat: 0.171 kWh / gal water + 0 kWh makeup space heat = 0.171 kWh / gal total

- *Winter, HPWH (2.5 EF) and electric resistance space heat: 0.069 kWh / gal water + 0.171 kWh makeup space heat - 0.069 kWh waste heat = 0.171 kWh / gal total

- *Summer, HPWH (2.5 EF) and no heat: 0.069 kWh / gal water + 0 kWh makeup space heat = 0.069 kWh / gal total

- **Summer, HPWH (2.5 EF) and air conditioning (10 SEER): 0.069 kWh / gal water - 0.054 kWh makeup cooling = 0.015 kWh / gal total

- Winter, HPWH (2.5 EF) and heat pump (9 HSPF): 0.069 kWh / gal water + 0.065 kWh makeup space heat = 0.134 kWh / gal total

- ***Winter, HPWH (2.5 EF) and wood stove: 0.069 kWh / gal water + 0.071 "kWh" makeup space heat = 0.140 kWh / gal total

* Even if your effective heating season is fully half the year, you still come out slightly ahead averaging these two (0.120 kWh/gal year round), but probably not enough to justify the higher cost. Milder climates do better. Warm climates do great. Edited per Woodgeek's point that the HPWH waste heat either goes into the water, or into the space.

** Take note of the minus sign. We paid less than full price to heat the water and paid less than full price to air condition the house, and now we're at 1/10th the summer baseline cost

*** Wood usage compared to electricity usage based on cost @ $200/cord, 20mmBTU/cord burned at 75% efficiency, $0.11/kWh

If I also calculated an example with natural gas heat, it would work out in a similar ballpark to the wood heat example, but if you have natural gas space heat, you probably also have a natural gas water heater, and it will be harder to achieve a cost savings by switching to a HPWH because natural gas is relatively cheap as a heat source.

 

[DickRussell]

My guess is all these specs are true, but not at the same time.

I can believe COP=5 at low lift (like 90°F to 120°F in the summer, no defrost), but not at -20°.....

There is a theoretical limit to COP for any heat pump. All such devices obey physical laws (https://en.wikipedia.org/wiki/Coefficient_of_performance). The maximum theoretical COP, with no efficiencies due to friction or the compression not being isentropic, and with zero temperature approaches at both heat absorption and heat rejection ends of the process, is given by:

COP = T(hot) / [T(hot)-T(cold)]

where the hot and cold temperatures are absolute (add 459.67 to Fahrenheit or 273.15 to Celcius). With no heat losses, you get out all the heat absorbed at the lower temperature plus the energy input by the compressor working on the fluid. Cranking in some numbers shows how the maximum COP of a HPWH must drop as ambient air temperature drops. Assuming heat is rejected to water at 120 F (579.67 R):

Air Temp (F) COP(max)
90 ____ 19.3
70 ____ 11.6
50 ____ 8.3
30 ____ 6.4
10 ____ 5.3
5 ____ 5.0
-5 ____ 4.6
-10 ____ 4.4
-20 ____ 4.1

In practice, working heat pump COP values are much less than theoretical, due to inefficiencies of compression, fan power used in moving air past the cold temperature heat exchange surface, pump power for circulating water through the high temperature coil, and the need for at least some temperature difference at either end for heat transfer to occur. For example, if the delta T at each end is 10 F, absorbing heat from air at 10 F requires evaporating refrigerant at 0 F, and rejecting heat to water at 120 F requires condensing refrigerant at 130 F. Without any other inefficiencies or fluid movement power, that reduces the maximum COP from 5.3 (above table) down to 4.5. Actually, superheat resulting from compression to a pressure that results in condensation at the required temperature raises the compressor discharge temperature even more, with a corresponding drop in maximum COP possible, I think (perhaps someone can comment on this). At any rate, COP must drop as cold fluid (air) temperature drops.

 

[woodgeek]

You do pay twice in the winter, but only once in the summer (potentially less, see examples), and even in the winter, depending on your source of space heat, you might not be paying anywhere near full price either time.

Here's a bunch of napkin math comparisons, assuming 70 deg F (39 deg C) temperature change to your water, where I refer to electric resistance heat as "full price":

- Winter or Summer, electric resistance DHW and electric resistance space heat: 0.171 kWh / gal water + 0 kWh makeup space heat = 0.171 kWh / gal total

- *Winter, HPWH (2.5 EF) and electric resistance space heat: 0.069 kWh / gal water + 0.171 kWh makeup space heat = 0.240 kWh / gal total

Sorry, not buying it. Even if you stole all the heat from an electrical resistance heated home, you would still have the same total cost with a HPWH, not more.

All the electrical energy used by the compressor goes into the water as heat. That amount is not stolen, it is created as a plug load, at COP=1. The other heat (the amount going into the water LESS the amount produced by the compressor) would be made up by the resistance space heating system, also at COP=1, for a total COP=1 for the HPWH. IOW, both of these cases are COP=1. In the winter where tank losses offset space heating (in principle) BOTH systems are COP=1.0 and EF=1.0.

That's the worst case, no savings in winter, for a HPWH in a resistance heated fully conditioned space. Not negative savings. For anyone with heat BTUs cheaper than resistance heat (the vast majority of people) HPWHs give energy savings in winter, even in a fully conditioned space.

 

[woodgeek]

If I also calculated an example with natural gas heat, it would work out in a similar ballpark to the wood heat example, but if you have natural gas space heat, you probably also have a natural gas water heater, and it will be harder to achieve a cost savings by switching to a HPWH because natural gas is relatively cheap as a heat source.

Except that naturally vented NG water heaters have sucky EFs and huge stack losses all winter long, stealing your space heat and humidity.

And they can backdraft in well airsealed homes...a safety hazard. Its just a matter of time before these are banned IMO. Made sense 50 years ago in leaky houses, almost impossible to operate in a modern tight and ventilated house.

 

[begreen]

Looking at the Sanden unit I have a couple concerns. The first is the risk of pipe freezing in an extended power outage. The second is that there is no resistance electric backup. Not sure what the recovery rate is at low outside temps either. Secondary concerns would be calcium deposit cleaning on the heat exchanger in areas of hard water and snow protection.

 

[iamlucky13]

Sorry, not buying it. Even if you stole all the heat from an electrical resistance heated home, you would still have the same total cost with a HPWH, not more.

All the electrical energy used by the compressor goes into the water as heat. That amount is not stolen, it is created as a plug load, at COP=1. The other heat (the amount going into the water LESS the amount produced by the compressor) would be made up by the resistance space heating system, also at COP=1, for a total COP=1 for the HPWH. IOW, both of these cases are COP=1. In the winter where tank losses offset space heating (in principle) BOTH systems are COP=1.0 and EF=1.0.

D'oh! Yep, I forgot for a moment where the boundaries of my system were. I'll edit the post.