It is an interesting problem and is really to do more with how the system is used and the rate at which the flat uses DHW and looses heat.
If the heat rate loss is very low/slow and the hot water is used in one go, typical for one occupant, then the maximum may be more important than the minimum. Think mornings when the flat is being heated and a shower is taking place.
But if the place is constantly occupied and showers/baths happen at different times then the minimum may be more important.

My house is 2 bed, 1 bathroom and constantly occupied and can probably get away with a 1 kW heater (but uses a 1.5 kW storage heater and a 3 kW water heater on E7)

It's not particularly straightforward to work out, as a fair bit will be installation dependent, and dependent on things like radiator sizing and rate of heat loss from both the flat and from the radiators to the rooms. In general it's better to run a combi at the lowest flow temperature you can, but that needs fairly large radiators in order to get a reasonable flow/return temperature differential.

Larger than normal radiators can help if the boiler is slightly too large, by giving a buffer whilst the boiler stays in the anti-short cycling time delay phase. This then limits the oscillations in the room temperature around the set point that you might otherwise get from an over-size boiler with short cycling protection.

I suspect your needs may well be right at the very bottom end of combi sizing, as you've already found out, and that you may end up with a boiler that's slightly over-size for the heating requirement, just to get acceptable DHW performance. Bear in mind that it's the same burner running CH and DHW, and DHW needs a LOT of power, just a single hot tap running full on can easily need 20 kW or more. There's a limit as to how far down a burner can modulate and remain efficient, hence the relatively high minimum outputs that are quoted.

As JSH suggest you need to size your combi for your DHW demand , ie. what flow rate are you happy with
30kW will generally give you a good 12 litres a minute flow rate for a 35 deg temp. rise . If your satisfied with a lower rate then a smaller boiler will do ( eg. 24kW 9-10 l/m)
This should really be you main concern.
the boiler can then be range rated to suit you CH requirements , down to the boiler minimum, though most are suppose to do this automaticly with a modulating gas valve.
Like JSH mentioned if the miniums a bit above demand, then reduce flow temp. to suit.
this might give the additional bonus of forcing the boiler into condensing mode quicker. ( below 50 ish C)

argh , as CW said , posted at the same time

Posted By: JSHarrisIn general it's better to run a combi at the lowest flow temperature you can, but that needs fairly large radiators in order to get a reasonable flow/return temperature differential.

Larger than normal radiators can help if the boiler is slightly too large, by giving a buffer whilst the boiler stays in the anti-short cycling time delay phase. This then limits the oscillations in the room temperature around the set point that you might otherwise get from an over-size boiler with short cycling protection.

In bedrooms/bath/hall there are 4 existing radiators (2 of which are very small) that I will not change.

However, for the main living/kitchen space, I have the choice of what radiator to install. Am I best to make it oversized?

(Also, would one big radiator in this space be better than two smaller ones with a similar total rating?)

Posted By: JSHarrisMy experience says yes, increase the radiator size and lower the flow temperature setting on the boiler. The result should be better condensing performance (as this is directly proportional to return temperature) and hence higher efficiency from the boiler.

I found that increasing the insulation level of the house (CWI, decent DG, thick loft insulation, improved airtightness) meant that most of my existing radiators were then oversized, so worked well with a much lower flow temperature, even in pretty cold conditions.

Hard to be definitive about one big versus two smaller, but in general I'd opt for two rather than one just to get better heat distribution in the room and to increase the volume of water in the system. Although greater water volume means slower heat up response time, it also acts as a fairly effective buffer, helping to maintain more stable room temperatures. Heat output is proportional to radiator surface area and temperature, rather than volume.

As all my insulation is internal, any kind of temperature buffer can only be a benefit!

I think I might go for one biggish radiator (for practical reasons) possibly combined with some underfloor pipework as described in this thread:

http://www.greenbuildingforum.co.uk/forum114/comments.php?DiscussionID=9427&page=1#Item_9

Thanks for your comments.

Posted By: SteamyTeaDo you have any solid internal walls you can put radiators against?

The closest I've got is the 220mm brick party wall between me and my neighbour. As it happens this is where I'll probably put the new radiator.

Not sure what proportion of heat absorbed into that would be returned to me rather than my neighbour...I guess it depends on the temperature difference between me and them?

I've just had an installer around for a quote; he seems fairly on the ball about energy efficiency etc and has recommended I consider a boiler from Intergas

http://www.intergasheating.co.uk/

Apparently it has a different heat exchanger design from other boilers, somehow separate circuits for DHW and heating meaning no diverter valves, supposedly more efficient and less parts to fail. He says he's found them to be pretty reliable and the heat exchanger less likely to be damaged by dirt in the system.

Anyone else familiar with these?

They seem to claim a higher efficiency for DHW heating too, if I've read it correctly... the claim an efficiency in the 90s whereas most others I have seen have been in the 60s.

Not sure what proportion of energy use is taken by DHW against CH? I guess the proportion increases once you're in a well insulated building.

Posted By: DamonHDFor us last year (so, mild winters at either end) gas consumption ~3936kWh, so with approx baseline (DHW+cooking) ~5kWh/day that leaves 2111kWh for space heating, ie nearly 50:50 in a mild year with improving insulation (etc).

http://www.earth.org.uk/saving-electricity.html#meter2011" >http://www.earth.org.uk/saving-electricity.html#meter2011

Maybe allowing for cooking, DHW is somewhat over 25% of the total, ie 1MWh out of the 4MWh, but we've been cutting both baseline and kWh/HDD numbers.

Rgds

Damon

Seems like a substantial enough chunk to be worth paying some attention to...especially for those who like to linger in the shower...

Going back to the original question.. I'm reminded that most oil boilers can't modulate down. They allways burn flat out or cycle. There is a debate on what that does to efficiency..

http://www.oilfiredup.com/site/news/item/1353

"research commissioned by OFTEC shows there is no evidence that non modulating domestic oil boilers are any less efficient than modulating ones"

I've come across this:

http://www.cleanboiler.org/Eff_Improve/Operations/Johnston_Boiler_Turndown.pdf

which seems to be arguing that it's better to have a boiler cycling on and off near its full output rather than one with a high turndown ratio running constantly at its lowest output... have I understood that correctly?

<blockquote><cite>Posted By: lineweight</cite><blockquote><cite>Posted By: alec</cite>the important thing for reliability, efficiency and comfort is the compensation controllers...not the on-off controls we are so used too</blockquote>

How do I identify a boiler with good compensation controls? Or do you just mean that it's important to actually use the controls.</blockquote>


Compensation controls just shift the internal boiler flow temperature switching point in response to external temperature change as a rule, increasing flow temperature when outside temperatures drop and decreasing it when they increase. It's not linear control, the boiler still just switches on or off as required, but it does tend to make the boiler run at the lowest acceptable flow temperature (for best efficiency) more of the time, if properly set up (which can be an issue).

The latter point is key, as compensation control assumes a fixed linear relationship between outside temperature and heating requirement. This may or may not be the case, depending on the installation. For example, I tried (and failed) to get such a control system to work on the system I installed in our house in Scotland. The problem was that heat loss was far more dependent on local wind speed around the house than it was on outside air temperature. Trying to get wind-chill correction didn't work and I gave up on the idea of compensation control for that application. Similarly compensation control may not work well in attached or terraced houses, and particularly in flats, with limited external walls/roof. In those cases heat loss rate may well depend a lot on what your neighbours do with their heating.

I think we're at cross purposes in the terms "linear" versus "on/off".

I know of no boiler that can linearly change output (or non-linearly change output come to that) in response to a varying signal from any external sensor. Many can modulate output (usually in fixed steps) over a limited range, perhaps from 30% of maximum up to maximum in maybe four or five steps. Most compensation controls seem to just change the switching threshold of the boiler flow temperature, when you get down to it.

This isn't linear control though, it's just switched, albeit switched over a set of varying ranges that gives a close enough approximation to being linear for such a slow response control loop. True linear control would require more complexity in the burner, and as I understand it would impact significantly on burner efficiency, unless some means of changing the physical size of the burner is incorporated. Also, true linear control would make the feedback loop even more sensitive to installation conditions than the pseudo-linear switched threshold controls systems we seem to have are.

The controller I designed used a PID loop to try and maintain tight control of the boiler flow temperature in response to outside temperature, using an electrically driven mixer valve to get true linear control of flow temperature (it was a non-modulating oil boiler) but despite having used complex control systems for years I simply couldn't get a system to work well with the high wind speed and direction related heat loss conditions our old house was subjected to. Had it been in a more sheltered location I'm certain I could have tuned the loop to work well, as it worked fine with no wind blowing.

I suspect that, as you say, there are many UK installers who'd really struggle with this, not to mention the high probability that many users would fiddle and muck up the settings!

<blockquote><cite>Posted By: alec</cite>Y
It would be interesting to see how your install would work with one of these boilers...surely though wind is the issue increased draught proofing and insulation is the solution, not turning the boiler on and off...</blockquote>

The house was pretty well sealed and very well insulated (it was an imported Norwegian timber kit), the losses were just down to the increased heat loss rate from having wind blow over the outside of it, wind chill in effect. Wind makes a big difference to heat loss in exposed locations, and this house was pretty exposed, high up facing the sea on the West Coast of Scotland!

<blockquote><cite>Posted By: alec</cite>under these circumstance with a vaillant I would specify the wireless indoor sensor (located in the coldest room) and TRVs in other rooms...and an outdoor sensor...

I guess us in London dont know what cold is! or wind for that matter!</blockquote>

Close to what I did in principle. I zoned the system (the house was a bungalow) into a bedroom zone and a rest of the house zone, with separate thermostats/programmers for each zone, and fitted TRVs to all the rads bar those in the rooms where the sensors were (master bedroom and hall). Without the attempted weather compensation the system worked very well, it was particularly useful being able to turn the heating off in the bedrooms when they weren't being used, and the same for the rest of the house. We needed to be pretty disciplined about keeping doors shut, but the design of the house was such that all the bedrooms were naturally well isolated from the living areas, so this wasn't difficult.

Having designed and built the controls for this system (it was a fair time ago, before sophisticated µcontroller based heating controls were readily available) I went on and did a better system for a friend, where we fitted electrically operated valves to every radiator and linked them back, with multiple temperature sensors, to a single µcontroller based programmer. Although setting this up was a pain, once set it allowed much better control of individual room temperatures, and added the ability to pre-programme temperature profiles for any room in the house. The chap it was for was a bit of a geek and just loved it................