Over-sized heat loss = over estimated design temperatures

Alan_Bennett · 27 January 2025 19:57

Thanks again, for you info. Excellent advice.

Stove is a fireline fpi8

I’m waiting for heat geek to design my system. I can fit one 12kw easy. The 2nd ASHP Would be seen through a kitchen window if the heat pump is positioned 1m away from the door. If the ASHP Is 750mm away from the door we don’t see it.

Ive had a quote for two ecodans in cascade. 22kw system from a local company.

Space might be the key factor in what I go with. I don’t mind reducing the window a little. It all depends on the boss.

lwis · 27 January 2025 21:12

I think this is another symptom of oversizing, radiators are undersized for low modulation/flow temperatures, meaning the heat pump has a smaller (higher, inefficient) operating window.

What I regularly see (and have experienced) is that installers would much rather design for a high (50C) flow temperature as it is more likely (guaranteed) to meet demand when it matters most, and requires the fewest changes to a property. Which I think is what often leads to a potentially cheaper installation which operates inefficiently.

Alan_Bennett · 27 January 2025 22:35

I did ask if we could design for a low flow rate system of 45c. With me messing about with boiler setting and looking at radiator temp, I think it could be done but at what minimum outside temp could we achieve 45c and hold temp.

At the moment, I think I’m 50c rad temp 100% of the time below 2c. The sun has a massive impact to gas usage.

Around 5c outside the rads do fluctuate 40,45 once I’m at set temp.

House temp is 20c

Thanks all.

Fingers crossed a heat pump will be installed soon and I can get on with the kitchen. Then I’ll spend my days looking at the heat pump, solar app and smile a little

RJK · 28 January 2025 12:27

Hello,
An interesting discussion. My comments from installing a heat pump and operation over 3 winters.
My property has U/F heating on the ground and first floors. I had previously concluded that the recirculation valves in typical underfloor heating manifolds, lead to some of the floor zones overheating. Due to the properties size I decided split the heating system, so the existing gas boiler continues to supply the HW and a small radiator circuit at ~60 degrees. The ASHP now operates with an exit temperature of 35 degrees, which provides a very even temperature across all the heating zones. Due to the large thermal mass of the floors, I concluded after a bit of a trial, the weather compensation curve wasn’t required. The heat pump supplies a 100 litre tank and a separate circulation pump is used to supply the u/f zones. The property was built in 2009, so is well insulated. But the flaw in a heat loss calculation is it doesn’t deal with the fact that various points in a house will have a different heat loss. In my case the core of the house generally maintains an even temperature irrespective of the prevailing outside temperature. But I have a south facing garden room with a lot of glass that struggles to maintain the set temperature, when the outside temperature is below 2 degrees. So irrespective of the size of the heat pump, the heat output of this particular floor would struggle to deal with the heat loss in this part of the house, when the temperature gets closer to zero. This incidentally was very similar when the zone was originally supplied from the gas boiler at 60 degrees. The room has a small aircon unit that is now used as an ancillary heat source during periods when the temperature is <2 degrees.
So in conclusion a whole house heat loss calculation, can only provide a partial answer as what size heat pump is required. Equally I view it is a little pointless designing an ASHP system to deal with a few weeks a year when the temperature is below zero and just accept the fact that parts of a property may need additional heat source at certain times of the year. Or as someone else has commented “put a jumper on”.
For information. The original design heat loss calculation for the property required a 35kW boiler. The ASHP fitted is a 16 kW unit and I appreciate this does not supply the HW requirements and the radiator circuit is just a couple of towel radiators.

Alan_Bennett · 28 January 2025 14:03

I’m always telling the kids to put a jumper on. They walk around in their pj’s saying its cold.

Since I’ve started measuring house temp, now I say, "the house is 20c and the lounge is 19.7c, boiler set to 50c. Rad are 45c.so it isn’t “cold”

Back in the day I remember my grandma’s house having ice on the glass, inside! I don’t remember saying “it’s cold” back then.

My thoughts exactly about having to design a system for -3 that might be used for 4 weeks of the year.

I’ll light the fire if it gets a little chilly. Sorted.

RJK · 29 January 2025 14:08

Hello Alan Bennett,
I have re-read your previous comments and my property is quite new so it is obviously well insulated. I have explored the options of going in the opposite direction with an older house with solid walls. My original aim was to use an MCS installer for the Heat Pump installation, but even including the Government grant the prices I received were uneconomic. I eventually decided on a self install, the biggest benefit is the heating design can then suit the property and not having to fit the MCS rules. Bi-valent designs have really fallen out of favour because of the political decision to decarbonise domestic heating. I’m sure the MCS will totally disagree with this opinion, but as the Government grant is dependent on MCS accreditation, it creates a market that leads to high installation prices. From a very simple perspective the price of my 16kW Samsung ASHP was around £3k and then the costs to complete the install, which in my case was an additional £3k and note this included the VAT, which as a self install can’t be recovered. The eventual cost will of course be heavily influenced by the heating design that is currently installed in a particular property. I appreciate the benefits the Government grant provides, but access to it means an MCS accreditation and you then lose the commercial benefits of using local tradesmen.
My view longer term is the ASHP grant will eventually become financially untenable for the Government, similar to the outcome where the solar FIT scheme required MCS accreditation and this also distorted the installation market.
My solution evolved into a bi-valent heating system, with the ASHP providing the main heat load from October to March, the Air Con units providing additional heat in the winter period and then providing the main heat requirements in the shoulder months between the winter and summer. The gas boiler is mainly off from May to September and is just used as a HW top up if there is little sun. The Samsung unit is very effective at modulating the heat output from a maximum of 4.7kWe down to 0.1kWe when it is idling. But even operating it with an outlet temperature of 35 degrees, the average COP is only around 3. I have also operated the unit at higher outlet temperatures and as expected the COP is considerably lower. But Air to Air Heat pumps are by design far more efficient and operate ~5.
I am huge fan of underfloor heating and the ASHP is a far better match to the design characteristics of U/F heating, which are usually designed on a Delta T of 10 degrees.
But looking at the design challenges of an older property, my current view is if U/F heating is a practical option then that is the route I would choose. But otherwise a bi-valent system using an Air to Air system is an efficient, economic and practical solution, with an existing radiator system remaining as an ancillary heating system. Also I notice you use Octopus Agile, so are used to dealing with HH tariff’s. Having access to winter heating using both gas and electricity, means you can manage the electricity price fluctuations which often range between minus 0p/kWhr to 99p/kWhr. Similar to your property I have a large solar array and 15kWhr of batteries, which are really crucial to operating on a HH tariff. A final point it is important using a bivalent design where gas and electricity are both used for heat, to have an effective heating control system which can be automated to manage electricity fluctuations. This in the end proved a trickier problem to manage than the ASHP install, but price competitive solutions are available which enable a bivalent system to be controlled and automated to manage the daily peak electricity prices between 4 and 7pm, then fine tuning the operation depending on day ahead electricity prices. Controlled using an iphone and far easier that persuading the kids to go and get some more wood for the fire!
Good luck with your installation.

Alan_Bennett · 29 January 2025 21:56

Thanks for your comments.

I’m waiting to see if heat geeks will give me a yes/no using my gas meter data.

If it’s a no from heat geek a local company designed a 22kw mitsubishi Cascade system with a SCOP of 3.5.@ 50c The company hasn’t done a heat loss cal yet. That would be interesting.

I can see the advantages of a cascade system as I’ll be mostly running on one Heat pump (11.2kw) High loads will be split between both heat pumps as and when I need more heat. Cost of that system is 13k without BUS grant. I need a new boiler and hotwater tank anyway, cost around 6k

Ideally I’d go with heat geeks even if it was slightly more expensive.

Robert,
I don’t know if you’ve heard of predbat. Pradbat controls my battery charge, discharge, hold charge and export. Predbat calculates every 10min and changes the plan accordingly.
Uses octopus api and solcast api. Also reads my house load data.
Predbat is very clever. Takes a few days to setup but once it’s running you just leave predbat to it.
I’m trying to run the system as efficiently as I can.
Predbat can also control a ASHP and a car charger, but I’ll leave that for another day.

My lasts months p/kwh 20.63 on Agile. Not too bad as we’ve had 99p peak a few times over the last week or two
Predbat runs via homeassistant. I’m using a pi5 with 128gb ssd.

I’m keeping my eye on bidirectional chargers. Once the bugs are ironed out on the chargers I’ll probably get a Nissan leaf.
I cycle more miles than I drive so the car will be parked up most of the time.

If my plans all pan out I’ll be on octopus intelligent go 7p for 6 hours and have around 50kwh of battery available. 40kwh from car and 9.7kwh from my solaredge battery. Thats the idea. The wife knows what I want to do and she’s on board.

I think it’s a balancing act with ASHP, installed solar system and right tarrif. I think with a bit of planning and care I can find that sweet spot.

In January 2024 I produce 221kwh. Best day of production was 20.8kwh on Jan 17th.
Jan 2025 solaredge hasn’t been recording production properly. (server upgrading) so far I’ve produced 171kwh, I think.

Fingers crossed I get a yes from heat geek and it isn’t too expensive.

Regards

Alan.

RJK · 30 January 2025 14:47

Hello Alan,

Thank you for your information about Predbat, which I will investigate.

I am also closely watching V2G, bi-directional chargers and compatible vehicles. Apart from the current Nissan Leaf which is soon to be dropped, a solution seems further away than it appeared two years ago.

If you haven’t come across the following problem yet, please note the following.

When I fitted my first solar system, a Solar iboost was used to divert any exported electricity into the HW tank. It works by monitoring any export using a CT on the incoming tails of the supply to the consumer unit. This worked well for a few years, but when I fitted a second solar system with batteries this required a second CT to be fitted. The iboost then started diverting stored electricity from the batteries to feed the immersion heater. The problem is caused by the sampling frequency of the inverter and the iboost which are both about 4 minutes. The iboost and the inverter both often see excess electricity being exported and then attempt to divert the excess to the immersion heater and the batteries. If the output of the solar array then reduces as the sun goes behind a cloud, the two control systems become in conflict. The solution in the end was to remove the iboost and install a Powerflow unit which has a sampling frequency of 20 msec and set it to ignore any excess export up to 600W. In addition the car charger also has a CT, but this is less of an issue as my current charger only operates at 7kW. But any future replacement will be designed to load follow and divert to the car battery and this will create a further source of conflict. But as long as the problem is understood it can of course be managed.

Good luck again with your design choices and I would be interested in knowing your eventual solution, as we are currently thinking about moving to a new area where we will most likely end up with an older property…….9 inch solid walls etc!

Best regards

Robert

Alan_Bennett · 13 February 2025 13:10

Hi all.

Heat geek won’t use my gas usage to design a ASHP. I understand that they have to be accurate and have confidence that my gas usage data is correct.

My local heat geek has proposed that we monitor the house using HTC testing. He’s waiting if head office will be OK using that method.

Any members had a HTC test?

Heat loss survey was 30kw
I use 34,000kwh of gas.

Thanks.

Chris_Hill · 13 February 2025 14:13

they wont use your gas usage for a heat loss calc , there is no guarantee your gas usage pattern matches a heat pump pattern,

Andre_K · 13 February 2025 14:19

Why wouldn’t it? If you keep the house at the same temperature, you’ll need the same heat, irrespective if it comes from gas or a heat pump. My gas usage heat loss estimates were quite spot on to my measured heat input from the heat pump. I’d say a properly done heat loss estimate based on gas use and outside temperature is much more accurate that any heat loss survey with guesstimates for all involved parameters.

Chris_Hill · 13 February 2025 14:25

@Andre_K yes you are completely correct ,
variables are
is the house at same temp ?
outside temp will vary
flow rates
but most heat pumps dont have the peak power outputs a heat pump does and most importantly MCS rules

Alan_Bennett · 13 February 2025 14:36

Hi.
I’ve been running my boiler at 13.6kw 24/7 since 22nd jan

Nov 2024 boiler has been running 24/7 but boiler was still at 40kw but rad max was 50c

My boiler is a worcester 40cdi. Rad max temp 50c

House temp set to 20c.

Running 24/7 and comparing previous gas usage there isn’t that much difference when I was running on amd off method.

Alan_Bennett · 13 February 2025 14:42

I’ve posted in this thread about my highest gas usage was at -5.

Looks like I’ll have to wait to see if heat geek will do a HTC test.

Regards.

Andre_K · 13 February 2025 15:18

Ah, the universal constant “rules before brains”.

Alan_Bennett · 13 February 2025 15:51

Hahahah.

Perfect.

If heat geeks won’t use HTC method I have a local company that can install a mitsubishi Cascade system 22kw.
Advertising SCOP of 3.5 at 50c

Its been hard to find info about SCOPS and COPS of a mitsubishi cascade system.

Regards

Timbones · 13 February 2025 17:21

There’s a little bit of info on this topic: Converting Ecodan Systems to Cascade

System is also listed here on HeatpumpMonitor.org

HydroSam · 13 February 2025 22:51

I wrote about my experience using the HTC method here, https://forum.ovoenergy.com/home-and-heating-138/correctly-sizing-air-source-heat-pumps-ashp-bigger-is-not-always-better-14428. After four years of running with monitoring kit, the heat loss is almost spot on with the HTC estimate.

It’ll give you a bit of background with some links to articles that will explain how to go about running the calculations.

It’s a bit of a long read, but at the time is was all part of the narrative of trying to convince OVO (as part of the electrification of heat trial) that the heat pump they’d installed was well over-sized, it didn’t work, but let’s not take over this thread with that story .

toadhall · 14 February 2025 10:36

I was part of the beta-trial of the Build Test Solutions smartHTC product with Octopus a few years ago. It took measurements (inside temperature, outside temperature and energy input used within the heated envelope of the house) over 3 weeks and produced an HTC value. The value correlated well with a calculation just done on gas usage, so it seems to work.

I was thinking I could use this value to do a much better job of WC (weather compensation) since with WC you are trying to match the heat loss by tweaking a fixed curve by trial and error, whereas using HTC you should be spot on first time.

Alan_Bennett · 14 February 2025 12:05

Thanks all for your input.

My local heat geek is trying to see if a 12kw ASHP will be enough to cover my needs.

At the start of this journey, a HTC test was mentioned but its talking a while for heat geek to make the decision. Understandable as therw putting their name on the install.

On cold days I do light the stove and it’s ticking over for a few hours in the evening.

The boiler isn’t working flat out when the outside temp is over 4c outside.

On this day -3c over night. House holding 20c.

Plan B will be a mitsubishi 22kw Cascade system.

I’ll have the flexibility of running only one ASHP on low loads.

Over the years I’ve learnt that somethings in life can’t be rushed.
I’ll be adding ASHP on that list.

Regards