over sizing radiators

[jpaul]

I'd happily pay a grown up to come and balance my system, but they just don't seem to exist.

earlier thread had some ideas - but agree seems a bit white art

For anyone thinking of trying to balance their radiators to make their boiler more efficient, here is what I did, how long it took and how much I estimate I'll be saving:

(NOTE - I realise not everyone will have access to some of these tools, but I had them anyway - I didn't buy them especially, except for two more DS18B20 temperature probes @ £4 each).

1. Unlock all lockshields and TRVs
2. Open all windows to keep the ambient reasonable (if doing this in winter, leave windows closed)
3. Start central heating and leave for an hour to hit a reasonable tempterature (monitoring output and return temperatures on the pipes immediately in/out of the boiler with DS18B20 / Raspberry Pi Zero)
4. Measure input and output temperature of each radiator (I did this using an iOS FLIR)
5. Close lockshields down slightly and leave to settle for ~20 mins
6. Find out that the lockshields are terrible and only the last ~ 1/2 turn has any effect!
7. Repeat steps 4-5 until radiator output temps and boiler return temps are where you want them (roughly 70% input temp as output)
8. This process took me 2.5 hours (including the 1 hour initial heating time) for 8 radiators in a small 3 bed detached.
9. EDIT - Realised the pump was on the max setting (70 watts). Setting to lowest (50 watts) and making small adjustments to the TRVs has resolved and added, slightly, to the savings.

Return temps before:

-----------------

OK one example - flow rates.

I see the implication of flow rate on dt
more categoric V on optimal heat pump dT's https://clade-es.com/blog/what-is-temperature-differential-in-heat-pumps/

WHAT AFFECTS TEMPERATURE DIFFERENTIAL​

There is a misconception that heat pumps must have small temperature differentials. This is not true.

You may see low delta T’s in the data sheets and promotional materials released by some heat pump manufacturers – we’ve seen designs based on 5 or 10 degC delta T’s. These values, however, are only correct if the manufacturer is using synthetic refrigerants. With natural refrigerants it’s possible to design a system with much larger delta T’s. Heat pumps which use R744 (CO2) or R290 (propane) can perform very well with a much higher delta T’s.

We’ve written before about why we use CO2 in our heat pumps at Clade. CO2 is actually most efficient at a temperature differential of 30 or 40 decC. That kind of delta is ideal for many heating applications because it delivers the required temperature while maximising whole system efficiency.

Below is an illustration of relative COP (coefficient of performance – we’ve written a blog explaining it here) between CO2 and a synthetic refrigerant. You can see that at useful flow temperatures a wide delta T gives the highest COP.

e: but - why that panasonic install manual does not tell you ?

 

[dirtychinchilla]

The irony is that most of the consultants we work with still spec 5K dTs all over the place because that's what the heat pumps work at. The pump should serve the system, not the other way round. When we're looking at non-domestic projects (which is all we do), we can often knock out several heat pumps by having a hybrid plant room, which makes use of other tech like CHPs and electric boilers. In combination, they all work much better than they would standalone. If you had a CHP, for example, it could provide the electricity to run the heat pump and also put out high grade heat at 80C, which could then be stored, but later mixed down to 60C to create a 60/30 F&R network, which is a hell of a lot better than just running at 45/40.

The problem we see every single day is that manufacturers, such as Mitsubishi, only sell heat pumps. They're very commercial. All they want to do is sell more heat pumps. So it's not in their interest to recommend hybrid systems which operate much better than heat pump-only systems because that reduces their sales.