Residential ccASHP Building Electrification Study
AndyBower
| Posted in Energy Efficiency and Durability on
Some interesting results from this preliminary study. For some context, it’s oriented towards the NE USA.
AHSP demand in general is up; ductless still the most popular.
Ductless multi-zone setups with more than 3 heads had quite poor COP, 1.12-1.52.
NYC contractors were more skeptical of AHSP than MA ones.
There are some more interesting results in there, these are the ones that stuck out to me.
What do you make of it?
GBA Detail Library
A collection of one thousand construction details organized by climate and house part
Search and download construction details
Replies
I had to google what "CC" was, cold climate.
From the study:
"Performance issues were uncommon with ASHPs and nearly all
customers reported lower bills.
Customers were highly satisfied with heating and cooling performance (8.5/10 for heating and 9.0/10 for cooling). "
That's good. What would have been good is to know what the satisfaction level pre-installation was.
performance (8.5/10 for heating and 9.0/10 for cooling).
I wish there was more detail around that 8.5/10 number for heating. Does that mean almost everyone rating comfort at an 8 or a 9? Or is it 85% 10 and 15% zero? It makes a difference.
This was interesting too: On page 26 they compared actual heating load with the contractors' calculated heating load during a cold snap (0 to 15F)
"However, the average heating load for whole-home systems was 43% of the contractor’s design load during the extreme cold snap."
This fits with what we know -- contractors often overestimate the heating load, and oversize systems as a result. It makes sense that the actual loads seen were lower than the contractor's estimates.
Bill
What's really interesting about that table is the engineers load calcs are significantly higher than the contractors.
Also at least some of the percentages in that table make no sense. For the "43%" example 21,917 measured load is obviously over 50% of the 43,403 contractors calculation. And that isn't the only discrepancy you don't need a calculator to question.
I think this is a huge takeaway. Often, finding a mechanical engineer is recommended to perform a heat loss. If they're this bad at it, the advice should be to save your money.
Page 30 --
Customers who used a fuel other than natural gas averaged an annual savings of $682. Customers who used natural gas actually saw an increase in annual fuel cost by an average of $153. This is due to the fact that in New England gas is cheap and electricity is expensive.
This seems to square with what we've seen with posters here who installed heat pumps and saw their heating bill go up.
In my area, natural gas runs about 1.09 cents per thousand BTUs (cost of fuel only), electricity runs about 4.98 cents per thousand BTUs. Assuming a typical 80% efficient furnace, the effective cost of natural gas would be about 1.36 cents per thousand BTUs. Assuming electric resistance heat, electricity is about 3.7 times more expensive than natural gas for heating in my area. That means a heat pump would need to run at AT LEAST a COP of 3.7 to break even. Any higher COP than that and the heat pump will be cheaper to run. Any lower and the heat pump will be more expensive to run.
A more fair comparison of heat pumps to traditional heating systems would be to take an entire heating season's worth of heating and compare that, since that allows for the "Less cold" days when the heat pump is likely to be saving you money over a traditional system. Comparing only the coldest days when the heat pump is working hardest isn't really a fair system comparison.
Bill
Might need to account for cooling season efficiency over low SEER units, or if the house didn’t have an air conditioner before you’ll need to give some credit to the new cooling capabilities.
If you recall this thread:
https://www.greenbuildingadvisor.com/question/atwhp-or-electric-boiler
I tried to estimate seasonal efficiency, in that case for Whitehorse. I did a similar analysis for my climate, DC, and got an weighted average seasonal COP of 3.0. So 3.7 is going to take a mild climate.
I'm surprised -- I would have thought the DC area would be plenty warm enough on average to keep heat pumps operating well up into their "cheaper to run" range compared to gas. My climate is quite a bit colder on average, but not as cold as Whitehorse.
Bill
My numbers are similar. My reason for going heat pump only was that it simplified my equipment (and that equipment's footprint) dramatically.
2nd/3rd floor: Mid-static Fujitsu system that fits tightly into the conditioned knee wall it's in. Switching to a traditional fan coil and furnace would have been tricky, space wise, and the smallest typical gas furnace would be at least 3x oversized.
1st floor: Two ductless units very adequately cover the space. A gas fired solution would have either been hydronic (expensive) or forced air taking up mechanical room/shop space and ducts that would have removed genuinely usable living space.
I might spend a couple hundred bucks more on electricity than I would have on gas each winter, but I avoided what would have been more equipment spend and complexity, and reduced interior space.
And my bills are still much lower than when (pre-renovation and addition) I had mostly wall embedded steam radiators in uninsulated walls that were literally throwing 50% of their radiant heat out the building. It still makes me laugh a bit that that used to be "the way".
I was fooling around with the NEEP database -- https://ashp.neep.org/ -- last night, and they have a page that does just that for you. You have to select a heat pump first, but then you put in your zip code and your heating design load and it spits out a graph and a chart showing heat pump capacity, heat pump COP and cumulative heating load for every temperature. Also annual BTU's and hours requiring supplemental heat.
I can't tell what assumption they use for temperature distribution. The heat pump COP and capacity seems to be calculated the same way I did for the Whitehorse question, straight line between the 47F and 5F data points.
That's a neat extra feature I hadn't noticed before. It's similar to some of the reports you can generate with BEopt, but it's nice for it to be integrated with the equipment data.
Thanks for sharing.
It’s a chicken and egg thing in NE. Electricity, actual electricity, isn’t that high. But delivery charges? Oof. Perhaps as more heating oil people abandon oil, AC becomes more widespread, and plug-ins are more common, annual usage will start to increase enough to cover those costs. I’m not much south of NE and electricity delivery charges are only $.04/kWh since AC and hot summers pay for the wires. The flip side is that gas is crazy high here because the pipes are expensive, winters are milder and people are starting to leave the gas grid.
Delivery charges in my area are just a teensy bit over 7 cents per kWh. I'm on a time of day rate too (by choice), with on-peak charges of a bit over 17 cents/kWh and off-peak charges of just over 7 cents/kWh. That makes my total on-peak cost around 24 cents/kWh from 11 am to 7 pm weekdays, and my total off-peak cost about 14 cents/kWh all the rest of the time. For my example above, I weighted the rate to about 17 cents per kWh, which is probably pretty close to an average, since heating equipment can be expected to run a bit more when it's colder at night than during the day. That's probably offset somewhat by my setpoint going down to 64F at night though, although heat pumps don't really like different day and night temperature setpoints.
NE has issues with gas capacity due to a lack of gas transmission pipeline capacity. That's a problem since most new electrical generation over the past several decades has been natural gas fueled, straining pipeline capacity in that region. People "leaving the gas grid" are going to be using more electricity, most of which is being supplier by... Natural gas... That's the big reason I often say here that you're better off burning the gas directly when using it for heating than you are using electric heating gizmos. The ONLY way around that is to use a heat pump system, such as a heat pump water heater or similar. Electric resistance heat in almost all cases will actually consume MORE fuel than a heater using that fuel directly, since in most areas the electric grid is up around 2/3 supplied by a combination of various combustible fuels.
Bill
Bill, I agree, but surely it goes without saying that we aren't advocating electric resistance heat for people. Using heat pumps will use that gas more efficiently than burning direct and will help with that constraint.
In response to #22.
Agree again. If we're comparing gas on-site vs. off-site, the bar is low to beat a furnace in terms of efficiency. That should help relieve the constraint in the NE, since less gas will be wasted in furnaces/boilers. We can probably assume the utility pays about the same for the gas it consumes and the gas it sells. As I see it, then it becomes a question of what the delivery costs are for gas vs. electricity and what their trends are. Our gas delivery cost is $.75/therm, which should lead to many more economically driven gas grid defections.
Possibly yes, but it comes back to the COP again. If we assume around 60% efficiency for a combined cycle power plant (which probably includes most of the newer units that have been installed), and around 10% losses for transmission and distribution, then the net delivered energy from that natural gas is about 54% of what we started with. Now, to be fair, lets assume that the comparison for new equipment, between a heat pump and a 95% efficient natural gas furnace, which means that a BTU of electric energy "cost" about 1.95 times more than a BTU of gas directly, when considering only the units of natural gas consumed vs useful energy output (heat, in this case).
That works out to a COP of about 1.71 is the minimum needed to make the heat pump consume less natural gas than the 95% efficient furnace would need to burn directly for the same number of BTUs put into the house.
What this all means is that while a COP of >3.7 is needed for a heat pump to be cost effective, a COP of >1.71 is needed for the heat pump to use less input natural gas compared to burning the gas directly in a 95% efficient furnace. If your goal is to reduce your overall emissions, than you need your heat pump to run at an average COP over the heating season of AT LEAST 1.71 to break even. Any higher than that, and your heat pump is responsible for less emissions overall than burning the fuel directly.
In reality, the calculation is more complex, because the electric grid is a mix of sources, including things that put out more emissions (coal fired plants), and things that put out zero emissions (hydro electric, nuclear, wind, etc.). In most regions, the mix is heavily weighted towards natural gas and coal though, with those two sources usually averaging around 2/3 of the total generation. That would have the effect of HELPING the heat pump, in terms of reducing net emissions, and allowing for a lower average COP to result in a net reduction in overall emissions per unit of heat put into the home.
Bill
Natural gas prices tend to fluctuate more than electric rates, since the latter usually are set after the state public utility commission assesses utility rate requests.
Last winter, NG rates in MA were very high, so I suspect the NG advantage over heat pumps may have changed.
It is a definitely a quirky way to price things: electricity uses a 6 month fixed contract here and gas is month-to-month, from the same utility.
Curious to see how more solar panel adoption would impact these numbers.