ÀSHP heat loss - ho...
 
Notifications
Clear all

ÀSHP heat loss - how close to perfect can we get?

104 Posts
10 Users
6 Reactions
1,443 Views
cathodeRay
(@cathoderay)
Famed Member Moderator
7806 kWhs
Joined: 3 years ago
Posts: 1553
 

Posted by: @jamespa

Many gas boilers do or can support WC.

However the way we do dhw means that you may have to jump hoops to implement it.

If many gas boilers can do true WC, ie no room stat interference, even less excuse not to do a trail in WC mode!

For the record, how do gas boilers do DHW, such that they need a workaround to run WC? Isn't the three port two way valve, boiler output goes to either space heating or DHW, still the standard set up? I'm asking so that gas boiler owners know what they need to do to run the tests.  

Midea 14kW (for now...) ASHP heating both building and DHW


   
ReplyQuote
(@jamespa)
Famed Member Moderator
7135 kWhs
Joined: 2 years ago
Posts: 1213
 

Posted by: @cathoderay

Posted by: @jamespa

Many gas boilers do or can support WC.

However the way we do dhw means that you may have to jump hoops to implement it.

If many gas boilers can do true WC, ie no room stat interference, even less excuse not to do a trail in WC mode!

For the record, how do gas boilers do DHW, such that they need a workaround to run WC? Isn't the three port two way valve, boiler output goes to either space heating or DHW, still the standard set up? I'm asking so that gas boiler owners know what they need to do to run the tests.  

Perhaps I should have said that many boilers can do WC but, to the best of my understanding, few systems in the UK are set up to take advantage of the feature.

I know the detail about two boilers/systems, my Worcester Bosch and a friend's atag.  Both of these have an optional, expensive, internal diverter valve.  If you don't fit it then the boiler supports only one flow temperature which can be adjusted by WC.  If you do fit it then the boiler supports two flow temps, one fixed for dhw and the other either fixed or WC controlled.  It switches between the two according to the diverter valve setting, which is selected according to either boiler controls or externally triggered demand for heat acting on boiler contacts.

In both cases the installer didn't install the optional diverter valve, used two external two port valves controlled by an external controller, so the boiler doesn't know to switch flow temp, and of course they didn't enable WC because it's not the norm in this country.

In my case I haven't enabled WC because my rather old boiler requires an add on piece of analog hardware to do so.  However on my friends rather new boiler it's just a control setting which he has enabled, and of course then came across the flow temp problem (the dhw/Ch switch is still external)

It's quite likely that he could instead have used the boiler control also to do the ch/dhw switching and picked up the signal to the non existent internal diverter valve to switch his external valves, but he didn't want to do this to avoid possible guarantee problems.  Hence he implemented an external work around.

I don't know how or if other makes of boiler do something similar, but the essential problem is the relationship between external controls (which are frequently fitted in the UK) and internal controls.  Also many central heating systems, including mine, are designed in the plumbing so they can heat both space and dhw simultaneously which of course precludes WC. 

Given that, for most ch systems in the UK, it is unlikely WC has been implemented (even if it is technically feasible on the boiler) I think it is not worth going down the rat hole.  Unfortunately it's just not the plumbing practice in this country and so, even with boilers that support it, the external controls/plumbing likely make it difficult to turn on unless you are prepared to do work arounds.  Also the WC parameters need to be set up according to the house and they won't be on boilers installed in the UK because the installer didn't need to do so.  Furthermore the external temperature sensor wont be present where WC has not been enabled.

The thermodynamics are sufficiently robust to do the loss determination without, although it would be interesting to compare for a couple of examples to prove or disprove my theory that the scatter in oat vs energy is going to be much greater for a room controlled system than a WC controlled system.

It also occurs to me that combis may be different as the separation between ch and dhw happens in the boiler, so it may well be that WC capable combis only need an add on external sensor to operate WC, and no external plumbing changes.  I don't know what proportion of combis sold in the UK have latent WC capability, it may be manufacturers dumb them down for our dumb heating market.

It's a pity we didn't embrace WC in the UK as it would both have reduced our heating costs (because our condensing boilers would actually condense, at least some of the time) and eased the transition to heat pumps. Others did.

 

This post was modified 4 weeks ago 4 times by JamesPa

   
ReplyQuote
(@ianmk13)
Reputable Member Member
2223 kWhs
Joined: 2 years ago
Posts: 141
 
image

Here is a plot I created when I first began the analysis of my gas-fired heating system around 18 months ago. Since we are discussing how someone such as myself could best determine the required ASHP capacity, could I please address another related important issue.  I have performed my own detailed heat-loss survey and concluded that although all the emitters are pretty-much right sized for Δt50, the usual guidance would dictate that I should replace all my radiators and, perhaps, one or two sections of pipe. Does this hold true with R290, where higher LWTs are achievable should the ASHP control system determine they are necessary (I assume this is how the controls operate). COP may be low in such situations, but they seldom occur in practice, any extra costs being offset by the considerable savings on replacement radiators, etc.


   
ReplyQuote



(@jamespa)
Famed Member Moderator
7135 kWhs
Joined: 2 years ago
Posts: 1213
 

Posted by: @ianmk13

.  I have performed my own detailed heat-loss survey and concluded that although all the emitters are pretty-much right sized for Δt50, the usual guidance would dictate that I should replace all my radiators and, perhaps, one or two sections of pipe. Does this hold true with R290, where higher LWTs are achievable should the ASHP control system determine they are necessary (I assume this is how the controls operate). COP may be low in such situations, but they seldom occur in practice, any extra costs being offset by the considerable savings on replacement radiators, etc.

Rads are actually pretty cheap, however, funny you should ask as only a couple of days ago i knocked up a quick spreadsheet to work out the business case for replacing radiators (mostly to convince myself retrospectively that I had made the right choice when deciding to do so).

Somewhat to my surprise I got a payback period of 10 years or less for the parameters I explored, which I am happy with.  However yours will be different.  That said if your base case is an ft of 73, I think it may be a no brainer particularly if you can get down to 45 or less.

I'm away at present but back tomorrow pm and will post the spreadsheet.

The BUS grant requires a scop of 2.8 or above which might swing the decision anyway.

 

Your degree day plot has a similar appearance to mine but higher loss of about 10kW at -2.  How does it compare with your fabric loss calculation?

This post was modified 4 weeks ago 2 times by JamesPa

   
ReplyQuote
(@ianmk13)
Reputable Member Member
2223 kWhs
Joined: 2 years ago
Posts: 141
 

My fabric analysis came out at 11.7kW but I recall erring on the side of pessimism with measurements, U values, etc.

I'm assuming (to be verified) that the COP/SCOP will be similar to R32-based ASHPs most of the time, only being exceptionally low on a small number of days per year.  If it prevent access to the BUS grant, then it will definitely be another gas boiler 😕 

A major issue for me is that Mrs MK13 doesn't like the look of larger radiators and doesn't want the disruption and mess of having them replaced.  I've not even mentioned the outdoor unit yet.


   
ReplyQuote
(@jamespa)
Famed Member Moderator
7135 kWhs
Joined: 2 years ago
Posts: 1213
 

Posted by: @ianmk13

I'm assuming (to be verified) that the COP/SCOP will be similar to R32-based ASHPs most of the time, only being exceptionally low on a small number of days per year.  If it prevent access to the BUS grant, then it will definitely be another gas boiler 😕 

R290 scops appear to be broadly similar to R32, there may be a small systematic difference which someone else can doubtless comment on.

The only reason I mention the 2.8 thing is your suggestion of running at ft73, not because you mention r290.  That may well push the scop below 2.8.  BTW you may have to work out your own estimate of scop from cop unless scop at 70+ is published.  I just used cop at 7C as a proxy for scop, but wasn't looking at such an extreme case so wasn't too bothered if it was only roughly following a similar trend.

Mostly my rad upgrades have significantly improved the visuals according to the wife, the old rads looked very tired, the new ones very sharp.  I made one mistake namely fitting a type 33 which is a bit too deep visually, I should have gone for a 22 which is very slightly undersized but not by much.  I still have a fan radiator to go in, which the wife is happy with because it's smaller and neater than the existing radiator and has some technical advantages useful in the particular location.  I also have one where I have added a bunch of computer fans because the deficit isn't that much and there wasn't a particularly attractive upgrade route.

I spent some time on a couple thinking about the options before making a decision.  Rads are furniture as well as functional so it's worth taking visuals seriously.  Rooms aren't totally isolated so I doubt (and my installer is convinced) that a deficit of eg 100W in one room makes much difference if the adjacent room has a surplus.

I did over half of my rad changes myself at the rate of 2 per quite relaxed day.  I didn't bother taking up floorboards, as long as the new rads are at least as long as the old, cranked tails look fine.  Don't make them shorter though; cranked tails where the cranking is 'inwards' look awful.

 I presume 'Mrs MK13' is not number 13!

This post was modified 4 weeks ago 4 times by JamesPa

   
ReplyQuote
cathodeRay
(@cathoderay)
Famed Member Moderator
7806 kWhs
Joined: 3 years ago
Posts: 1553
 

Posted by: @ianmk13

Here is a plot I created when I first began the analysis of my gas-fired heating system around 18 months ago.

Thanks for posting this. R squared 0.92, pretty good. Looks like around 240kWh at around -3 degrees C OAT, or 10kW power rating.  

Posted by: @ianmk13

My fabric analysis came out at 11.7kW but I recall erring on the side of pessimism with measurements, U values, etc.

Probably just me being thick, but when you say erring on the side of pessimism, does that mean erring towards increased losses? Either way an empirical 10kW vs a model based 11.7kW is close-ish, and from both @jamespa's  and my experience it appears empirical estimates to produce lower values than model based values.

Aesthetics are important. A K2 radiator just about manages to look in proportion most of the time, while K3s definitely look boxy, especially in smaller sizes, but for a given output, they take up far less wall space. For this reason, I accepted mostly K3s because had I gone for K2s, some rooms would have been dominated by a K2, making my house look like a radiator showroom, and in some rooms, the low window sill heights would have meant very long radiators. Using a K3 mostly meant they had the same or a slightly larger wall footprint, and could sit under windows and in that sense look normal, if a little boxy.      

Midea 14kW (for now...) ASHP heating both building and DHW


   
ReplyQuote
(@jamespa)
Famed Member Moderator
7135 kWhs
Joined: 2 years ago
Posts: 1213
 

Posted by: @cathoderay

Either way an empirical 10kW vs a model based 11.7kW is close-ish, and from both @jamespa's  and my experience it appears empirical estimates to produce lower values than model based values.

It would be interesting to get to the bottom of this but I suspect needs quite a lot of houses and certainly is not a priority.  The average house consumes 4MWh of electricity per annum, that's about 400W, most of which will end up as heat.  Human beings are about 100W at rest, more of they are active.  Obviously the fabric method ignores these contributions to the heating.  Solar gain might disturb the average demand in a low loss house with lots of windows.  There is also a question of how accurate U values actually are.

We are now at n=3!

 


   
ReplyQuote
(@ianmk13)
Reputable Member Member
2223 kWhs
Joined: 2 years ago
Posts: 141
 

Posted by: @cathoderay

Probably just me being thick, but when you say erring on the side of pessimism, does that mean erring towards increased losses?

Basically, yes. I've got an unusual shaped home; it's not a regular rectangular box. It also incorporates several different construction methods. Some rooms have a mix of room above and insulated sloping ceiling or roof void.  Some rooms have walls with a diagonal change: heated room below the diagonal and outside space above. Walls are mainly brick and block cavity (larger cavity in a recent extension) but there are sections of insulated timber frame with weatherproofed cladding. If the rooms had each consisted of 6 rectangular areas punctuated with a few windows it would have been a doddle 😉.  In some cases I've guesstimated a material U value. At the time I took the measurements, my thinking was that I should overestimate the heat loss so that I did not underspecify the required heat pump output.  I've since discovered that over-specifying the output also presents issues and that I can't rely on the output being as stated on the rating plate when it's cold.

All the existing radiators are 450mm height and either P1 or K1 apart from two column radiators in the refurbished kitchen/diner. They were the biggest available, so I may need a third. I wonder if I should take a closer look at R290 heat pumps to see if I can limit the need for radiator changes.

 


   
ReplyQuote



cathodeRay
(@cathoderay)
Famed Member Moderator
7806 kWhs
Joined: 3 years ago
Posts: 1553
 

Posted by: @jamespa

We are now at n=3!

Indeed, onwards and upwards!

Posted by: @jamespa

There is also a question of how accurate U values actually are.

This is almost certainly a factor in my case, see the SPAB report mentioned earlier. The spreadsheet model used U values that are almost certainly much higher that my actual values, particularly for the walls, which account (on the spreadsheet model) for over 50% of the total loss.

Posted by: @ianmk13

Basically, yes.

Thanks for clarifying. Your home sounds interesting, but equally a case where a spreadsheet model is very likely to struggle, too many unknown unknowns as well as known unknowns. You did well to get close with the model, but if were in your shoes, I would definitely favour the empirical losses over the modelled ones. 

Midea 14kW (for now...) ASHP heating both building and DHW


   
ReplyQuote
cathodeRay
(@cathoderay)
Famed Member Moderator
7806 kWhs
Joined: 3 years ago
Posts: 1553
 

Posted by: @ianmk13

All the existing radiators are 450mm height and either P1 or K1 apart from two column radiators in the refurbished kitchen/diner. They were the biggest available, so I may need a third. I wonder if I should take a closer look at R290 heat pumps to see if I can limit the need for radiator changes.

I should really leave this to others better qualified than me to comment on R290, but I can say that P1 and K1 rads have much lower output than K2s and K3s, and I suspect running at a higher flow temp alone won't be enough, you will still need to upgrade the majority of the rads. I ended up having all K3s except one, and that was with a high design flow temp (on a R32 system) done precisely to limit radiator size as much as possible, because of available space restrictions.   

Midea 14kW (for now...) ASHP heating both building and DHW


   
ReplyQuote
(@jamespa)
Famed Member Moderator
7135 kWhs
Joined: 2 years ago
Posts: 1213
 

Posted by: @ianmk13

All the existing radiators are 450mm height and either P1 or K1 apart from two column radiators in the refurbished kitchen/diner. They were the biggest available, so I may need a third. I wonder if I should take a closer look at R290 heat pumps to see if I can limit the need for radiator changes

If most of your rads are K1s then upgrading to K2 double finned should bring the required flow temp down materially without compromising appearance.  The lower you can get it the better.

Attached as promised by radiator upgrade business case spreadsheet (and a load of other stuff).  Its very crude, basically it uses the figures in the A7COP column to calculate consumption and cost.  Really you want SCOP but since this is quoted only at 35 and 55 I used COP at 7C and the design flow temperature as a proxy.  If you want to be precise you could calculate a SCOP from the COP tables for your heat piump, but a crude indication may be good enough.  My guess is that you will have a sound business case to upgrade, the only question being to what FT.  Realistically the cost of the rads doesnt vary that much with size so probably its the lowest FT you can get to without degrading appearance or doing something extreme.  That will require some playing with figures.  You can find a table of output per metre length for various heights in the Myson or Stelrad catalogue, which makes it fairly easy to play around with options.  Here is a lookup table I use.  Hope that helps.  Output of most radiators is proportional to DT^1.3, so given outputs at DT50 its easy to calculate output at any other DT.

image
This post was modified 3 weeks ago 2 times by JamesPa

   
👍
1
ReplyQuote
Page 8 / 9



Share:

Join Us!

Latest Posts

x  Powerful Protection for WordPress, from Shield Security
This Site Is Protected By
Shield Security