@tobyg The Samsung does have a Quiet mode and I believe this changes the frequency (modulation?) and it runs quieter. I assume it’s perhaps less efficient?

Mine is set to Quiet Mode between 10pm and 4am, but is rarely on as I use the Samsung Controller to lower the room temperature down to 18.5C. At 4am it kicks it back up to 20C and as Ian says, has a +1C hysteresis, meaning it overshoots to 21C before switching off.

I mainly hear this very low frequency compressor modulation noise when it’s ticking over at 1KW electrical input, but not cycling (rarely seen this), towards the end of a long cycle.

The only other time is when it’s heating the DHW to 50C and the full 12 KW is being produced…..it sure can heat the tank fast 😁, leaving more time to heat the house on the cheap tariff windows.

In the winter, my neighbours barely hear it at all, and we are out in the sticks. In the summer, it runs for 20 mins a day doing the DHW, around 1pm when the electricity is cheapest the the COP is highest (mid day ambient is normally much higher than the middle of the night).

I think the latest Gen 7 Samsung 12KW is a single fan capable of higher outlet flow temps. It is also said to be quieter than the Gen6.

@tobyg as Ian suggests, radiators are spec’ed at a Delta T (mean radiator flow temp) to ambient room temp, so 70-20C = delta T of 50.

A heat pump flow temp of 50C (return 45) will give you an average rad temp of 47.5 and a delta t (to room) of 22.5C. Roughly if the rads were sized for continuous running of your gas boiler when the ambient is -2C, they would need to be twice the output at these lower flow temps. With lower temps comes greater efficiency and lower running costs.

My system was designed for 50C flow, 47.5 mean, when -2C outside. Given its max is 53C, when -7C the house got cold, but those are rare conditions. So I upsized quite a few of the rads (especially in the living areas and bathrooms, as they had higher room temp targets), I was able to reduce the flow temps, get closer to the 12KW rated output, improve the SCOP and reduce the running costs.

Good points there @mikefl especially the low modulation of the Grant. I very much like that on my 10kW myself. We switch the Grant off overnight, not a clever move. (Her who shall be obeyed, etc.) but for all the sluggish pre-heat in the morning we've found that to be more efficient and cost-effective than having the ASHP short cycle on and off overnight on a room thermostat set back. Oh that we could simply offset the WC curve by a programmable parameter to drop the LWT down 2-3degC. I doubt the new Grant Smart Controller can do that either, oh that it could! Grant are not that clued up on controls to be honest, you can tell that simply by the absolute dogs dinner they've made of redacting the Chofu heating and cooling installation manual for their UK branded Grant heat only product - and the mistakes they've made in the process.

My reservation is whether the Grant can deliver nameplate rating

a) at all (the jury is out with me)

b) with it's very frequent defrosts - A = 'no'.

But it does the job it was designed to do efficiently and well in all other respects. I'm a little disadvantaged by having a unit a little on the small side, but on the flip side the modulation is superb.

I don't think the LLH cripples performance, the Grant unit is one of the taller LLHs, and if balanced the thermal losses seem quite minimal. Of course there's a 2nd circulating pump to run, but for most reasonably sized home heating circuits with typical 28/22/15mm CU pipe I doubt the built-in Grant pump would provide sufficient flow without a 2nd circulator in tandem.

Fortunately, much of the time and as was the case was for my house, many of the rads were already oversized, some twice as big (and all K2’s) otherwise the total install costs would have been much more. I still have some updates to full balance particular room loses to radiator outputs,  as I have old parts and modern standard (insulation).

It seems from your heat loss calculations that the Grant or Samsung should both work, the higher output Samsung giving a bit more headroom against really cold condition, faster to heat the DHW. But the Grant will give the best efficiency if it runs lower output in mild conditions and you can run it more often (or continuous) when at its highest output in really cold conditions. But maybe the Samsung Gen 7 12KW would be better than both these options :).

@allyfish Not to de-rail this topic (so I'll probably post results - if any - on a new topic) but I've just started trialling a scheduled 2°C drop of the WC values to coincide with my overnight setback temp target on the room stat. Nothing exciting has been revealed so far, but early days.

Thank you everyone for your comments, which I'm busy digesting.

Posted by: @peterrsaltings

↑

Mine is set to Quiet Mode between 10pm and 4am, but is rarely on as I use the Samsung Controller to lower the room temperature down to 18.5C. At 4am it kicks it back up to 20C and as Ian says, has a +1C hysteresis, meaning it overshoots to 21C before switching off.

On this, thanks @peterrsaltings. I have to admit, I am still confused on this point, and sorry to come back to it (your long cycling method seems crucial if one's to be able to manage an oversized Samsung). Your graphic shows three on-off cycles in 24 hours, each time with the IAT swinging from roughly a minimum of 18.5 to a maximum of 20.5 – that’s possible with a Samsung controller with its +1C -0C limitation as a thermostat?

Posted by: @peterrsaltings

↑

-- Attachment is not available --

@mikefl  thanks a lot for these thoughts; if you don't mind I'd be interested to know what sort of house you have (approx size and type and insulation level), and what level your heat loss calc is at? It sounds like it might be a good parallel for us.

@tobyg Hi,

Yes, when I want the heat pump on during the off-peak periods, I set, using timers, the desired room temperature high, say 20C. The heatpump comes on, and the rads start heating up the house. As Ian states, in the time period I set, (3 hours), it can overshoot by +1C and if it does, the heat pump switches off. Otherwise it keeps going. After each 3 hour off peak period, I again use a timer to reduce the room stat temp, to say 18.5C. If it’s really cold during the day, it may turn back on but I noticed at lower temps the overshoot is only 0.5C. So it does not run for long, maybe an hour. In the evening I set a timer at 10:30pm to 18C, and if the room stat reached 21C, it takes quite a few hours to drop back this low, by which time the next off peak cycle starts. I don’t think it’s that complicated. I just use off peak periods to the max, overshoot to compensate  for one peak period I switch off the heat and DHW demand completely, and let it come on as it pleases in the standard tariff periods.

The weather compensation is set precisely to adjust heat demand vs losses, so it’s pretty predictable and consistent. The Samsung controller set up is very versatile, setting a start temp and any slope you want, matching outside ambient to house heat losses to heat demand. If it gets too warm, it switches off. If it gets too cold, it switches on. And the house stays between 18C and 21C throughout the day. Warmer when we have want it the most, cooler when we are sleeping.

Posted by: @tobyg

↑

@mikef  if you don't mind I'd be interested to know what sort of house you have (approx size and type and insulation level), and what level your heat loss calc is at? It sounds like it might be a good parallel for us.

House is a converted bungalow, downstairs being around 100sqm; upstairs 40sqm (just 2 bedrooms up there). Downstairs is mainly open plan, with living area having front door direct to outside; and large, north-facing sliding door 2.5m wide; and stairs in it (i.e. it's a large area with lots of heat-loss options). This area has 3 large K3 radiators in it now. The house is zoned (up/down) which is tricky for the HP to cope with, given the imbalance of size. Downstairs also has a bedroom in a flat roof extension (former garage) which is a good heat drain too.

Insulation in loft (as it's a conversion, the 'loft' is the eaves areas to either end of the upstairs bedrooms - hope that makes sense) was added to during installation. Not sure but I think they added 300mm on top of what was there (which was fairly old/compressed fibre about 70mm?) It's deep enough now that it fills and flows over the rafters now it's had time to expand fully.

Cavity walls insulated in 2005. Not sure by how much (I've only recently moved in) but the guarantee mentions 'Knauf' which is a decent manufacturer. House is externally rendered too, so think heat loss through walls pretty decent.

Heat loss calculations by installer I was never privy to(!) but I did perform two calculations myself; one based on the gas used in previous full year; second based on my own room measurements (window sizes, type of room, external walls, etc etc).

The first one used Heating Degree Days, Heat Transfer Coefficients and so on - think I worked this out from reading 'Protons for Breakfast' blog. I took into account the boiler was a 30-yr old LPG which was only 70% efficient at manufacture, and so on. This came out at 8.5kW if I used the entire year's gas usage as a base; or 6.5kW if I extrapolated from my own usage (I hadn't lived in the house for a full year, and a comparison indicated I would probably use less gas than the previous owners who I did have a full year of data for).

Second calculation (which I hope mirrored the installer's work) came out at just under 7.5kW.

Given first calculation wouldn't have accounted for better insulation, and I'm going to address some of the obvious heat loss elements of the building, I thought the Grant 10kW was a decent choice. I've seen it delivering 8.5-9kW, by my calculations.

Hello again @iancalderbank, @editor, @allyfish, @peterrsaltings, @mikefl and @hughf. After all the help from people here this is to say that our decision on this was to go with the Grant 10kW. While the Samsung 12 kW obviously has strong points in its favour, on balance it sounds like the Grant is more compatible with our house, and the way we’d want to manage our heating (much of what you were saying @editor).

A lot of it comes down to the different estimates of heat loss, and I’m not convinced by the MCS type of calculation as the defining yardstick, based on heating the house constantly to 21c, which would be overdoing it for us. At the other end of the scale, I’m sure that our gas boiler method calculation at 3.2kW is an underestimate, not least because it doesn’t reflect our use of supplementary heating. However, if I substantially adjust the figure, while also factoring the new internal wall insulation, it’s still hard to believe it would be more than 7kW or 8kW, expect in what would be (for Gloucestershire) very unusual conditions. I’m hearing that that is within the scope of the Grant.

It does seem highly likely that the Samsung would be oversized for us (as @iancalderbank has counselled from the start). It’s been helpful to hear how @peterrsaltings and others have found ways to handle this (and even make a virtue out of it, on the cost front). However, it sounds like getting this right depends on other variables that are contingent on the design, including radiator size, and I don't yet know in this ECO4 context how far we’ll be able to get these to line up optimally to make ‘long cycling’ work. Given that the Grant should just be able to meet our needs more simply, most of the time, it feels like it’s not necessary to get into this territory with the Samsung. (I am though chasing for an answer from the installers about the radiator sizes.)

Also (and this relates to a comment by @mikefl), intuitively it doesn’t feel in the spirt of all this to have something that’s significantly overpowered for our needs, that we’re underutilizing and somehow needing to rein in a lot of the time!  

It’s unfortunate that the Grant controls are inferior to the Samsung’s, but I’m hoping this would be mitigated by having less need to calibrate what the Grant delivers.

Obviously it is a concern that the Grant could struggle to keep the house warm enough in very cold weather (and also it’s not ideal to depart from the installers’ recommendation); with that in mind it’s taken quite a bit of figuring out, using the advice in this thread to try to judge likelihood and risk.  

As I’ve said before, I’ve been hugely impressed by the support here, and the thoughtfulness and quality of input, and the effort people have made to help. We’d have been on our own with this otherwise, without much of a clue, and the help here has really made a big difference. I am very grateful.

One other question on a detail please: the heat pump is going to go on a brick-paved sort of patio area, which is pretty flat though with some slight uneven-ness. The installers have said they wouldn’t necessarily create a concrete base here to stand it on; I’m wondering how much that matters (e.g. if it could be noisier on something that’s not completely level)? Thank you.