Hacker News Reader

Comment by londons_explore | Hacker News Reader

Even a modern air source heat pump typically doesn't beat gas for $/kwh. Even in europe where gas prices are sky high, electricity still costs more when you look at the seasonal efficiency of heat pumps, which is typically advertised around 4.4, but various studies show that in most real world scenarios you typically won't reach the quoted lab numbers, so expect to get more like 2.5-3.0.

Study: https://www.sciencedirect.com/science/article/pii/S037877882...

View on HN

Claims like that really aren’t helpful without numbers backing them up

the cited paper gives the efficiency of heat pump but we’re missing multiple crucial values to get from there to a comparison of heating costs

Edit: https://great-home.co.uk/air-source-heat-pump-running-costs-...

At current average British prices of 28p for electricity and 7p for gas, and using averages for both it looks like the commenter was mostly correct.

Not 100% if the spf Gives the realistic cop

In a new build that opted to forgo any gas connection, I’d be interested in seeing the costs from the substantial infra savings and hookup savings.

From that paper, the average COP measured was 3.06 (averaged between 2 and 7 degree outdoor temperatures, typical of the UK).

The UK electricity price is currently fixed at 34p/kwh for electricity and 10.3p for gas.

A new gas boiler has a COP(efficiency) of 1.054 (it can manage efficiency higher than 100% because gas is metered by the 'lower heating value', which assumes the exhaust gas escapes as steam, but the boilers actually condense most of that steam to water, getting additional energy out).

So. Total price is: Gas: 10.3/1.054 = 9.77p/kwh in your home.

Total ASHP price is: 34p/3.06 = 11.1p/kwh in your home.

And this analysis ignores the fact that ASHP's typically have much worse efficiency making hot shower water (which a gas boiler doesn't), and obviously also have considerably higher upfront costs too.

However, the ASHP can also be used to cool your home in the summer. In the past this has been a dubious benefit for most of northern Europe, but heat waves have been getting stronger, last longer, and happen more frequently as time goes on. This is turning into a serious consideration.

Also, you can theoretically power a ASHP with renewable energy, while there are few if any carbon neutral replacements for natural gas.

Most boilers purchased today allow use with a Hydrogen mix, and some under development allow 100% hydrogen. Before the widespread extraction of natural gas, towns were powered with 'town gas', which is ~50% Hydrogen, so this is very much proven tech.

There are a bunch of potential ways to make green hydrogen too.

So, there very much is a path to green with a gas boiler.

Green Hydrogen has not worked out so far. It is not clear that it even has a path forward. There are a lot of people hopeful that it will be a solution in the future, but as of today it is so uneconomical that even people willing to spend more to be green don't use it.

But making green hydrogen has pretty bad efficiency. It could never compete with a heat pump using the same electricity source, even if it had an efficiency of 100%, since the heat pump has an efficiency of 200-400%.

Yeah, these can make ASHP a good idea anyway, but the grandparent was nevertheless correct saying that they don’t beat gas for heat in terms of cost in normal (I.e. not current) circumstances.

I read somewhere there’s market failure too. Electricity price is largely proxy of gas price. While uk gov subsidises has they don’t electricity…

Thus the solar panels. So long as the installation proves to make financial sense, the PV production offsets the heat pump efficiency, even in winter months.

Exactly. Also, the top commenter noted that only electricity was available and currently was relying on massively inefficient baseboard heaters. Solar panels and heat pumps also have great rebate programs and eventually gas will be phased out. Some US cities are now banning gas on new construction in the near future.

The point is that custom solar/geothermal installs seem neat and efficient —- waste not, want not — but probably have a hard time competing with the economies of scale and low maintenance of PV panels and air source heat pumps.

This may work in mild climates, but the areas where winter heating is most needed will have the solar panels under a bed of snow during the winter months.

Additionally, winter months are usually the cloudy ones. It's not very uncommon to have just a couple of sunny days per month in European winters, driving down solar gains even if there's no snow cover.

It's not as bad as this.

In climates where you will have a bed of snow during the winter months, your optimum tilt angle for a fixed solar panel is something like 30° off of vertical or steeper. I find that on mine the snow falls off since it is a south facing, steep, dark surface.

You engineer to make sure you have enough energy captured each month to meet that months needs. That might lead you to a tilting mount, or just a fixed angle and having surplus energy most months.

In my case, panels were much more expensive when I designed my system, I initially roof mounted them, but when replacing the roof under them I moved them to a pole and went with a tilting mount and manually move the panels twice a year. My load is much higher in the summer, but I still need some power in the winter. And northern winters can be cloudy a lot and have limited sun even on a clear day. At my location there is about a factor of 5 difference in solar energy per square meter per month. So you design for each month and then pick a solution that is best. In my case the summer load is so high that even though I get 5 times the energy, it is still the driving force on sizing the system.

Interesting. Last winter my parents' rooftop solar had zero output until April, as it produces nothing until all the panels are clear of snow. Granted their installation is at a fixed angle that matches the roof, and this is in the Nordics.

During summer months the production mostly covers and partly exceeds their use, but the sell price is so much lower that it doesn't even begin to pay for the rest of the year. But that of course then relates to the installed capacity.

I wonder if it makes sense to build some simple resisting heating elements into the panels, to allow the snow to slide off. Shouldn’t use all that much energy to run it occasionally.

You can actually push a bit if current backwards through the cell to heat it up. Some people have tried this but not sure about what prevents this from being more wide spread

https://www.pv-magazine.com/2020/03/18/heating-solar-panels-...

So much of this discussion depends on where you are. In the US, a typical solar installation is on the pitched roof of a house. When the sun comes out, the dark roof and PV panels heat up, and the snow slides off. Rowhomes or apartment buildings in a city might have flat roofs with panels mounted on racks, but most US cities won't get enough snow often enough for this to be a problem.

My parents do have a traditional pitched roof. Winter sun is not warm enough to warm up anything, even if dark, until late winter. Plus the color of course is white if it's covered in snow :) Yeah very much location dependant.

They are on a pole on your roof or you moved them to the ground?

I moved them to the ground. The roof pitch is about 1:1 and I’m not feeling like trying to stick to it any more. The pole puts everything at a comfortable working height.

Selling PV electricity back to the grid is almost always many times worse, depending on state incentives, than consuming the electricity. For example, my utility sells electricity to me at 9 cents/KWh, but only buys from me at 2.5 cents, and charged me a fixed monthly fee for the meter to boot.

If the commenter’s house doesn’t have gas, then it doesn’t seem to make sense to install it, and the price differential in the US isn’t as great as in the UK (how’s that Brexit thing working out?).

Huh. We have a 1:1 credit system here in Washington.

Considering solar owners are all pushing electricity onto the grid at the same time, when it is needed the least, such a system doesn't make sense. There are real costs associated with getting rid of all that unwanted energy.

> when it is needed the least

I don’t think this is true in many areas. In our area, during the summer the neighborhood is running A/C when the sun is at its peak. During the winter, heating. Also the transport loss from a power station is nothing to sneeze at - my understanding is that locally produced power often just results in reduced demand on the larger grid.

I think solar makes sense for most who use air conditioning. That's a load matched pretty closely to the timing of PV generation.

Here in the Sierra foothills, it's been a blazing hot summer. About 90% of our PV generation powered our home air conditioning. We shipped very little energy to the grid on hot days. And it was awesome to have a comfortable environment without sucking grid power on the days when it hit 113°F.

This is an enormous subsidy to residential PV. Electricity wholesalers would love to be able to sell it to the grid at retail prices, like you are. Instead, they sell it for something like a quarter of retail price. This huge subsidy is only sustainable for so long as the PV penetration is low in residential market.

Yes. Even just being able to connect to the grid and only pay for the energy consumed is a subsidy. There is a large fixed cost to provide the connection and guarantee power will be available on it when demanded.

And Massachusetts.

It would make even more sense to sell that power back to the grid, and then spend the $$$ earned on gas, which would work out cheaper overall.

Obviously in many parts of the world, market distortion means the buy price and sell price for electricity is very different, and in that case a heat pump might make sense to combine with PV.

That's highly dependent on your local electricity and gas prices. A quick google search tells me residential electricity in Germany costs about 2-3x what it does where I live in the northwest US. We're on mostly government owned hydro power and electric prices have been stable for a long time, meanwhile gas keeps going up.

EDIT: Did some quick math using my last power bill, at current prices a heat pump just needs to be about 2.9 average COP to beat gas in cost for me, if gas keeps going up that'll keep dropping!

This seems like a (regulated) pricing error.

There should be no economic way that a high efficiency turbine produces and distributes electricity to heat homes at a greater than 1:1 ratio compared to storing, pressurizing, and delivering, then burning in irregularly maintained consumer homes.

Likely the error is that gas pipes to the house are subsidized (albeit the electrical likely is too, with heat pumps and induction stoves, the gas lines are unnecessarily redundant)

A furnace causes a very large second law loss, converting chemical energy to low grade heat. If that gas is used to drive turbines, and the work produced used to drive heat pumps, much of this entropy generation is avoided.

A home gas-driven heat pump could be a better option from an efficiency standpoint, but those are not widely available, probably for cost and reliability reasons.

"high efficiency" turbines are not all that efficient. Burning gas releases 100% of the available energy as heat. Converting that gas to electricity is << 50% efficient. You also lose ~5% just transmitting the electricity to homes.

Also, electricity from gas is relatively expensive. Other sources like coal or hydro are cheaper, which lowers the average cost of electricity.

If you combine gas turbines with district heating you can recover almost all of the heat energy. If you use the produced electricity for heat pumps you come out ahead. For climate change purposes it is way better since gas pipes to homes leak way more methane than those to central power stations

But gas turbines should be more efficient than a burner under a boiler or a burner in a furnace.

Yes, solar, wind, hydro, nuclear should be cheaper per unit (coal is not usually cheaper than gas in the US)

There's not really a "should be" in thermodynamics. All heat engine cycles have upper bounds of theoretical efficiency, and burning gas in a gas turbine to generate electricity to create resistive heat is never ever going to be more efficient than burning that gas at the point you need the heat. It's simply not possible. There's always going to be losses - the exhaust gas will contain energy, there will be mechanical and transmission losses. There is no physical way we can change that.

The best thermal power plants - combined cycle gas turbines - get about 60% efficiency. Most are closer to 45%.

What we can do is either use the excess heat from the gas turbine in district heating (which combined with resistive electrical heat probably approaches the same efficiency as a local gas boiler), or use that electricity to drive a heat pump, which gives a greater than 1x return on heat where you want it. An efficient combined cycle gas turbine driving a heat pump is going to give you more heating than the same gas being burned in a boiler. - 0.5 * 4 = 200% efficient.

You can also get (although they're much less common) gas powered heat pumps (eg propane fridges in RVs). They might have a "primary energy ratio" of 1.5-2, bringing the total system efficiency of a local gas powered heating system back up to pretty close to that of a remote generation + electrical heat pump system. The electrical system has the benefit that you can slot renewables into the mix as well.

> to create resistive heat

We're talking about heat pumps not resistive heat.

95% efficient gas furnaces are the norm for newer houses[1].

Getting electricity out of gas is fundamentally harder than getting heat out of it.

[1] https://iwae.com/resources/articles/95-afue-gas-furnace-work...

There is a large second law loss even in a "95% efficient" furnace. A gas-fired heat pump would have efficiency much higher than that.

That's kind of disappointing news. I had a skim through the article and was surprised to find that it's all based on models. There wasn't a single empirical measurement of an installation of either kind. Not that I don't believe in modeling and its usages, but I am going to radically discount the findings of this paper because there was no actual experiment performed here, just fiddling with models.

Give it a re-read. They collected data from actual houses (although granted only 6 boiler-years worth of hourly data), then fitted a best fit model to that data, then used the model for their conclusions.

They did that because they needed to compare the manufacturers datasheet lab figures to the real world figures, but there are 10+ variables that affect efficiency, and a direct comparison isn't possible unless all the variables match - hence using a model to act as the 'convertor'.

Reply on news.ycombinator.com