Central AC heat pump?

[lml999]

We're moving into a traditional colonial on Cape Cod with no central air. We've lived with AC for the past 26 years and will probably want to add it to our new house.We will be living there year round.

The house has 4 bedrooms up, traditional first floor and a mostly finished basements. Individual window units may provide temporary coverage for the master and maybe the den/kitchen area, but are not a long term option. Similarly, mini splits probably won't cover enough of the house to make sense.

We are putting solar panels on the house and will have excess capacity. The house is heated by natural gas, with five zones of baseboard hot water. I'll be adding insulation in the attic and doing a full envelope analysis and sealing where appropriate. Windows are 22 year old double pane...some have blown seals and I'll be replacing the glass over time with higher performing inserts (disassembling the wood frame, installing new glass).

So...

I'm starting to read about central AC heat pumps.

I'd appreciate some input from those with experience...what are the pros/cons of central AC heat pumps versus standard central AC units? One benefit, clearly, is that I can use a heat pump to heat in the shoulder seasons when the wood stove isn't fired up and the baseboard heat isn't as efficient (or maybe that's just with our current high efficiency oil Buderus).

Anyway...thoughts on heat pumps?

Thanks!

 

[woodgeek]

The cost increment for a heat pump versus a central AC is small, if you keep the tonnage the same....it is basically the same hardware plus a couple valves for reversing the flow.

Option #1:
The cost comes in IF you want to have all your winter BTUs come from the HP you would need to upsize the tonnage. E.g. a 2-ton AC could be good in my house at 100°F outdoors, but a 4 ton HP only heats me down to 20°F. If you are doing a cost compare, the HP (and ductwork, wiring, etc) all look a lot bigger for the 4 ton HP than the 2 ton AC, and so it IS more expensive. Moreover, if it is 'single speed', the compressor just turns on and off, then the 4 ton HP does a poor job dehumidifying in AC mode, since it is oversized....it runs 5 mins every hour.

I was ditching expensive oil heat, so I went this route, and get 90% of my winter BTUs from a HP, and have an oversized AC.

In your (4°F average colder) climate, the HP will struggle a little more. More importantly, the economics are shifted since your elec is more expensive and your gas is cheaper than my oil. This option likely doesn't pencil out.

Option #2:
Another option is to say that you want 2 ton AC, and to just get a 2 ton HP instead. It will do a great job heating your house down to about 40°F outside, the 'shoulder season', and will do so efficiently, with a COP of 3.5 or higher. You then get a smart stat that locks out the HP below that temp, and runs the gas when it is colder...so-called 'dual fuel' operation. We have quite a few systems like this down here.

In this case the additional costs (HP versus AC) are small. And so are your 'savings', which you need to compute.....

To compute the cost per million BTUs, for natgas its easy: 1 MBTU = 10 therms....and you can read the therm cost on your bill. To be precise, figure 11 or 12 therms input if you have low eff equipment to get 10 therms output.

For HP, figure that 1 MBTU = 1000/3.414 = 293 kWh if you used a straight electric space heater. At $0.10/kWh, that would be $29.30 per MBTU from a space heater. IF you HP COP = 3.5, it means you get 3.5x the BTUs of a space heater, so 1 MBTU only takes 293/3.5 = 84 kWh. Compute your cost for 84 kWh on your elec bill. The difference is your savings (hopefully) for 1 MBTU from the HP versus gas.

Say you save $5/ MBTU....you prob use 50-60 MBTU in a season, but only 1/3rd or less of that is when it is >40°F outside. I'd figure no more than 20 MBTUs during the shoulder season....so for a crude estimate multiply your savings per MBTU by 20 or something for your annual savings.

AS for carbon, compare the amount of elec required to make 20 MBTU from the HP....20*84 = 1700 kWh/yr....with your projected excess.

Every MBTU you generate by natgas makes 14-17 kg of CO2, so if you offset 20 MBTUs with your solar-HP, you prevent 20x15 = 300 kgs of CO2 per year.

This might (or might not) be more important to you than the $$ savings.

 

[lml999]

Thanks woodgeek, this is very useful. Note -- the solar panels will have excess capacity, so the incremental summer/shoulder season electricity is essentially free. Shoulder season heat at no cost.

 

[woodgeek]

I got that....in this situation I would go HP in a heartbeat....ACs in NewEngland get so few run hours they die of old age before they wear out. Might as well use the compressor to do some good in this world. That said, I would not expect the upcharge on the hardware to be very much....if it was, I would push back.

You can compare MSRP on HP and AC compressors from the same manufacturer....they are quite close. You would also make it clear to the installer that you do NOT want heating strips (this, with wiring, will add expense). They will say that without strips, then defrost will lead to 'cold blasts'....but this is never a problem if you don't run below 40°F.

Option #3:
If you DO want to run below 40°F....to 'burn up' more excess solar electrons (than the 1500-2000 kWh of Option #2), and save even more carbon, then you will have to have defrost cycles, which you might want to have heat strips for....and this will add the cost of wiring and labor to install the wiring. The strips themselves cost like $50 and mount in the air plenum just downstream of the air handler.

If you have that many electrons....and aren't planning to buy an EV, then I could see #3 making sense. If you were planning to buy an EV (likely IMO in the next few year when they get really cool), didn't have that much excess, and/or plan on a lot of wood heat or cheap gas....I would cheap out and go with option #2 and save the money for a downpayment on a Tesla 3 in 2020 or a Bolt in 2018.

After all, the EV saves more CO2 per solar kWh than offsetting natgas or wood heat BTUs, I think.

 

[lml999]

Thanks again...Option 2 seems to make the most sense -- size the system for AC, use it for heat in shoulder seasons, burn wood and the occasional NG molecule in the winter.

I've got 25 years of guaranteed production with the solar panels (SunPower warranty)...I assume that at some point before they tire out, I'll have an EV in the driveway.

 

[woodgeek]

Of course, its 6 of one half a dozen of the other....

I would still favor #2, since in my experience, the 'cold blasts' from defrost in mild weather are no big deal....the heat in the ductwork absorbs most of the coldness....its only a problem if the outdoor unit has heavy ice on it (i.e. from actual freezing rain or snow)....so if your smart stat switches over at 34°F...your system might runssome defrost cycles between 34 and 40, and you might notice them, or might not without strips.

I lock out my strips during defrost in mild weather to save energy, using a custom built circuit. Not a problem at all.

 

[iamlucky13]

Not having ducting currently installed, a "mini-split" or "ductless" heat pump should also be considered. The cost of doing that could potentially be lower than installing a central forced air heat pump plus ducts. The efficiency should also be slightly higher. These work best with relatively open floor plans, as the house would likely have 2-4 indoor units and depend mostly on air circulation for heat to reach rooms without a unit in them. The indoor units are usually wall mounted up high. I've seen a couple intended to mount more out of the way in the ceiling, but I expect the cost on these is higher.

An installer can probably give you quotes for both a central forced air heat pump and a ductless heat pump.

A small ductless system could be complemented by the baseboards. A larger ductless system would let you remove the baseboards. Going with ducted, you'd most likely remove the baseboards and include a natural gas furnace in-line with the unit regardless of size.

 

[iamlucky13]

IF you HP COP = 3.5, it means you get 3.5x the BTUs of a space heater, so 1 MBTU only takes 293/3 = 84 kWh.

I think 3.5 COP would be a bit high if that's taken simply from a COP spec. My heat pump is specified for 4.2 COP at the design temperature, but that is the steady state performance once it has run long enough to warm up, and ignores defrost cycles. The HSPF rating attempts to factor those in (approximately), but the units are weird because they were picked for the convenience of HVAC designers. Multiply the HSPF rating by 0.293 to get an approximation of seasonal COP. For my personal calcs, I also factor in 20% duct loss.

The end result is I estimate my 9.5 HSPF heat pump, although it is rated for 4.2 COP at steady state, probably has an effective seasonal COP of 2.8 when ignoring duct losses and 2.2 when factoring in duct losses. That is still worthwhile compared to electric resistance heat, but may be less so compared to natural gas. Then again, you've got excess solar paying for at least part of your heat pump cost.

 

[peakbagger]

Hard to beat mini splits, they tend to be higher efficiency then conventional heat pumps and generally better at putting out heat at lower temps. There is a big savings in duct loss as its heck of a lot easier to insulate a bundle of tubing than a supply and return duct.

 

[woodgeek]

I think 3.5 COP would be a bit high if that's taken simply from a COP spec. My heat pump is specified for 4.2 COP at the design temperature, but that is the steady state performance once it has run long enough to warm up, and ignores defrost cycles. The HSPF rating attempts to factor those in (approximately), but the units are weird because they were picked for the convenience of HVAC designers. Multiply the HSPF rating by 0.293 to get an approximation of seasonal COP. For my personal calcs, I also factor in 20% duct loss.

My COP = 3.5 estimate is assuming that the HP shuts down below 35-40°F and never defrosts. In the undersize limit, it would not be short cycling at 40°F, with option #2 sizing. The tabulated figures for min HSPF units I have seen are well over 3 at 40°F. Nice point about the duct loss in older construction.

 

[maple1]

Not having ducting currently installed, a "mini-split" or "ductless" heat pump should also be considered. The cost of doing that could potentially be lower than installing a central forced air heat pump plus ducts. The efficiency should also be slightly higher. These work best with relatively open floor plans, as the house would likely have 2-4 indoor units and depend mostly on air circulation for heat to reach rooms without a unit in them. The indoor units are usually wall mounted up high. I've seen a couple intended to mount more out of the way in the ceiling, but I expect the cost on these is higher.

An installer can probably give you quotes for both a central forced air heat pump and a ductless heat pump.

My thoughts also. Retrofitting ductwork could be a huge job. Plus (I think) added efficiency.

I would be very curious on hearing all-in quotes for ducted HP vs. ductless mini-splits. I was looking at somewhere in the $10k area, just for ductwork, when I was contemplating geothermal. Heck with that, I said.

 

[woodgeek]

I was assuming that there was existing ductwork associated with the NG furnace....in that case a 2-ton install would be $3-4k in my area.....

Ah, he did say baseboard hot water....it is Massachusetts, the hydronic state.

In that case, he has to decide whether he wants to go mini or central. IMO a big part of this is the Cape Cod location.....in my experience (I vacay there every year), the time was you never needed AC out there or maybe 1 or 2 days a season....so there was very little central AC, just a rarely needed bedroom window unit. In the last few years the climate appears to have changed just enough that it has crossed a line where AC is more needed.

You also get days that are not hot, but the humidity is too high and you feel like you are going to grow mold. Speaking of....CC houses are prone to that not so fresh 'musty' smell from the humidity. A mini with 'dehumidify' would be really nice.

So, I could imagine a single mini covering the place and certainly providing welcome dehumidification. You'd prob have to sleep with the bedroom doors open 5-10 nights a year...no bigs for most people.

In heating, it would burn more electrons (and displace even more NG/wood at higher eff) and still need the wood/NG backup, and the bedrooms would get a little chilly in the shoulder season with the doors closed, but prob not too bad. Hard for me to say. If it gets so cold the balance becomes an issue....you could turn off the mini. The OP also heats with wood.....so is presumably 'down' with the passive heat/coolth distribution thing.

But a single mini would prob cost half as much as a CAC with ductwork. And they are popular on CC, so there should be good installers, and at resale it won't be 'weird'.

My 'wild azzed guess' would be that a single (large size) mini could provide half of the needed winter space heat, and prob use 2000-3000 kWh/yr in the process.

 

[lml999]

Interesting thoughts on the min-splits.

I just added this to the original posting:

The house has 4 bedrooms up, traditional first floor and a mostly finished basements. Individual window units may provide temporary coverage for the master and maybe the den/kitchen area, but are not a long term option. Similarly, mini splits probably won't cover enough of the house to make sense.
​

Given the house is on Cape Cod, we do expect lots of guests/visitors in July and August, when we'd be most likely to need AC.
People scattered all over the house.

Conversely, in the coldest months, my wife I I will probably hibernate in the kitchen/great room (with our insert), the master bedroom and maybe the excercise room in the basement. There are five heat zones in the house, and I just mentioned three of them.

Annual NG usage seems to run around $1400 for heat, DHW and cooking. I expect to cut that by 15-20% with additional insulation, and maybe another 10-20% with the wood stove. So winter heating alternatives isn't critical.

 

[woodgeek]

I am a fan of minisplits in the right application, but continue to think that a ducted solution is generally superior from a comfort perspective, as well as not requiring user attention to doors, etc.

I could see the insert heat working its way upstairs in winter, but if you want 1 mini head to heat and cool, do you put it upstairs or down, right? Once you start getting into two heads, I could see the argument for just going ducted.

My one regret in my install was that I didn't have the installers mastic-seal the ducts. I ended up doing a lot of taping and sealing of my new ducts when I realized how leaky they were.

 

[lml999]

I'd have way more than two heads...we could have 10 to 15 people staying over in August, in four bedrooms, den, living room and basement.

And I prefer the quiet and evenness of the central air system.

With regard to sealing the ducts, that's a great point. This Old House just demonstrated the sealing, and if I remember correctly, the leaks were equivalent to a six inch hole in the ducting!

 

[STIHLY DAN]

They do make conventional HP systems that have inverted compressors and are efficient at 5*. Look into Bryant systems.

 

[begreen]

Good suggestion, I hadn't seen this unit. I'm glad to see this tech spreading to conventional systems. According to the manual it appears that the Bryant inverter compressor switches to backup heat at 10ºF.

 

[iamlucky13]

They do make conventional HP systems that have inverted compressors and are efficient at 5*. Look into Bryant systems.

A very good option, but it still doesn't look like you'd get a COP of 5 except in steady state conditions. Their best looks like it has an HSPF rating of 13, which corresponds roughly to a seasonal 3.8 COP, or 3 COP if including a 20% duct loss.

Which is not a criticism. 3-3.8 is a really good range to be in. I figure my own system probably works out to ~2.2 overall COP, and I'm content with that, at least for the time being, since I wasn't in a position to take on the up-front cost of more efficient unit at the time.

And woodgeek's suggestion for managing the heat pump strictly for mild weather conditions also would help keep it in the more efficient end of its performance range so it can beat those numbers.

Since lml999 mentioned low noise as a desired trait, an inverter-driven model is also likely to have a quieter startup. Mine makes a definite grunt when it kicks on.