Heat loss, why calculate...

[JamesPa]

17 minutes ago, marshian said:

 

Was the heating on constant (24/7) and the house up to temp before you started measuring gas used??

 

If you are heating intermittently or using setback temps then measurement of consumption is not going to give you your heat loss as you have heavy usage in heat up but periods when the heating is not on or it's maintaining a set back temp you will lose less as the house cools down.

 

My calculated heat loss is 4.5 kWh (target temp 20 outside temp -2) however I deviated from the default ACH parameters in the assessment used  - IMO they were way too high - with the suggested ACH it would be 6.1 kWh.

 

I sense checked mine in a slighty different way in the cold snap we had in November last year (temp was between 0 and -2 during this time) as we do heat intermittently and don't have set backs on a weekend heating so on those days I heated the house up and then only measured gas usage for the hours once the house was up to temp - as a result I was only effectively replacing the heat being lost in that period

 

First day averaged 4.1 kWh

Second day average 3.9 kWh

 

As a result I'm comfortable that my heat calcs are reflective of my actual heat loss

 

 

Measurements over a single day are intently suspect, because the heat capacity (sometimes called thermal mass) of a house is often in the same order as the steady state loss for 24hrs or more.

 

So I would personally sense check this against a seasons loss or some other measure.  That's not saying it's wrong, just sounding a note of caution before making a commitment.

Edited February 27 by JamesPa

[marshian]

20 minutes ago, JohnMo said:

I did my first heat loss calculation on degree days. But it was miles out. The degree days are closer based on 10 instead of 15.5 or 16.5 OAT.

 

I was using the std 15.5 to base HDD data to compare if my improvements were making a difference - however using the regression tool and historical consumption I quickly established that 16.5 deg was a much better base line for my house

 

Currently between 2.7 and 3.3 kWh/HDD but if I make a mistake in setting the flow temp (get caught out by the outside temps changing quickly) then I can easily overshoot. (HW requirement is excluded)

 

I don't have weather compensation - a 10kW min on a 24kW boiler range rated to 16kW so manual adjust flow temps to suit

For 10 deg outside I can run as low as 45 deg for closer to zero I need 58 for the warm up phase but then I can drop to 52 to maintain temps (I'm looking forward to a new boiler and weather comp in the spring/summer)

[marshian]

12 minutes ago, JamesPa said:

Measurements over a single day are intently suspect, because the heat capacity (sometimes called thermal mass) of a house is often in the same order as the steady state loss for 24hrs or more.

 

So I would personally sense check this against a seasons loss or some other measure.  That's not saying it's wrong, just sounding a note of caution before making a commitment.

 

Absolutely agree but I was trying to make the same point - if you include the heat up phase you are going to over estimate the steady state losses.

[JamesPa]

2 minutes ago, marshian said:

 

Absolutely agree but I was trying to make the same point - if you include the heat up phase you are going to over estimate the steady state losses.

Agreed.

[JamesPa]

2 hours ago, SimonD said:

Then there is the problem of actually getting a boiler small enough. Try buying a 6-8kW gas boiler. The smallest I can get I think is a 12kW heat only unit. Then there's the combi situation, which is pretty dire because of the dual usage.

It's interesting to cross reference this to the heat pump threads on this forum.  5-6kW heat pumps are not uncommon and sufficient (or more than sufficient) for the space heating of many houses.  People with houses that have a lower demand seem to be forced to make a trade off between efficiency of space heating at low demand, and dhw reheat times (assumes a tank).  

 

Same problem, different technology.  Easier to solve with a lossy house than an efficient one.  I'll still take that problem in preference to my 1930s solid wall build!

[waxingsatirical]

10 hours ago, Iceverge said:

 

A little bit lost at this one as I'm unfamiliar with the conversion factor you're using.  

 

 

10 hours ago, JohnMo said:

The calculation should be more like

 

Units × 1.02264 × calorific value
 ÷ 3.6 

Yes, calorific value is ~40 according to my bill and my factor includes a metric conversion factor of 2.83

[waxingsatirical]

10 hours ago, Iceverge said:

At the moment at 140m2 with 3.3kW of peak heating load that gives a specific heating load  23.6W/m2.

For comparison a passivhaus is 10W/m2.

To get just only over double that on an uninsulated house seems unlikely.

Well, I wouldn't say "uninsulated". It's a semi, so one side is warm anyway, the other side is 00s extension so filled cavity, the front and back are a mix of original unfilled cavity and extension, let's say 50/50 for simplicity. The loft is up to spec... nearly. The doors and windows all seal well. I've installed mechanically closing extractor fans as well.

[OwenF]

As another data point, my boiler (V100 11kW system on weather compensation) tells me stats for last 7 days. Right now that’s been a period of sub-zero nights with around 5deg avg in the day.

 

Heating: 8.1m3 / day

DHW: 1.2m3 / day

 

Quick online calculator tells me for heating that means roughly

 

90kWh / day

= 3.74kW

 

My calculated heatloss using Heat Engineer software was around 9kW. (This includes ongoing insulation upgrades, but I checked existing and rather sadly it wasn’t much different!)

 

I was sceptical of the ACH even for a 1980s build. I suspect this is hugely overestimating losses.

What’s scary is that 2/3rds of my downstairs is unheated and drafty at the moment and apparent measured heatloss is 1/3 of calculated before I’ve even completed the insulation upgrades included in the calc!

 

Further data my boiler gives me is %modulation, burner hours and boiler starts.

I’ve averaged approx. 100minutes burn time per start over 680hours (including DHW firing), which I understand is not too bad.

 

%modulation has usually been around 10-15% in steady state conditions. (Can be 40% first thing after setback on a cold night)

I don’t understand this figure tbh as the 11kW boiler has a minimum modulation of 3.3kW.

10% of 11 is 1.1, so I don’t know what calc it’s doing

 

though with a ‘measured’ heat loss of 3.74kW on an average winter day, I expect my fancy low modulation boiler will still be cycling when it warms up! 

should have got a heat pump (cry)

 

[JamesPa]

1 hour ago, OwenF said:

I was sceptical of the ACH even for a 1980s build. I suspect this is hugely overestimating losses.

 

That's certainly what the only guy who got my loss correct (in the sense calculated = measured) claims.  He says that the MCS assumptions of ACH almost invariably lead to an overestimate of loss.  He's done a fair few installs so presumably had evidence for this. 

 

My 1930s house, 30% extended, insulation part upgraded, double glazing, ACH=0.5 to get to calculated heat loss = measured heat loss, vs MCS assumption of 2-3ACH.  I cant find any other explanation for the difference so I reckon that must be it. 

 

I have got mostly solid flooring over the original suspended flooring downstairs, the DG windows have no vents, and the only 'remaining' chimney is 'sealed' with a wood burner, the house is almost a perfect cube in shape, so there are a few reasons to think that ACH might be less than average and thus 0.5 ACH is at least plausible.  

Edited February 28 by JamesPa

[SimonD]

11 hours ago, JamesPa said:

Same problem, different technology.  Easier to solve with a lossy house than an efficient one.  I'll still take that problem in preference to my 1930s solid wall build!

 

I don't think it is easier to solve in a lossy house. It's our perspective and design criteria plus a cultural perspective on heating. Lots of other countries that have far better and therefore less lossy housing don't appear to have the same issues. Take Sweden for example. Building regulations there changed to specify heating system temps to 55C in the mid 80s, our regs only changed to this the last couple of years. Also, heat pumps are not designed to deal with full load in Sweden - this is managed using auxiliary heat source.

 

From a cultural perspective, we've also essentially moved in the wrong direction for the last 30 years by throwing out hot water cylinders in preference of dreaded combi boiler.

 

For me one of the major problems is that in the process of sizing whatever heat source we use, we have contradictory requirements. We want low running costs, high efficiency, comfortable warm space immediately, and we want immediate hot water supplies. Now, in theory it's perfectly possible to achieve this through careful and proper whole system design, installation, and commissioning, but there is a real reticence to accept in part the real-world upfront costs of this on a commercial basis. The discussions on this forum aren't really helpful or representative of the larger majority out there in terms of tendency to understand and tinker with systems, which is what leads to a scewed view of the time and costs of putting a system in that works properly for the long term. This forum also presents lots of single personal system design and setup, which isn't really a very secure foundation for determining a reliable methodology for sizing systems, it just gives us an idea of various approaches used but rarely to we get the full picture.

 

1 hour ago, JamesPa said:

MCS assumption of 2-3ACH

 

This is not a MCS assumption. This kind of figure, if used consistently throughout the heat loss calc is down to the person completing the heat loss calc. Most input from the MCS (based on the CIBSE design guide) is 0.5 - 1.5ACH so what you've seen is a fault with the installer not the underlying calculation method.

 

In truth, like I suggested earlier, we need a more balanced approach to calculating heat loss and system loads, which will inevitably require amixture of real world data combined with some theoretical calculations to arrive at solutions that are good enough and then use the learning from the good enough to interatively improve the process, but for that we need lots of data and time since we're so behind the curve.

Edited February 28 by SimonD

[sharpener]

14 hours ago, JamesPa said:

Does this point to resistance electric, which suffers from none of these problems, as the eventual 'destination heating' for the most efficient houses?

 

Possibly. MVHR systems could do this easily, mine has an inline 2.2kW heater (which I have disconnected as otherwise it would come on whenever the internal temp is "too low" which is most of the time. It is made in Norway where electricity is cheap).

 

Edited February 28 by sharpener

[sharpener]

12 hours ago, Iceverge said:

Would it be sensible to size boilers/heat pumps for the modal demand rather than the maximum given supplementary heating can be supplied by cheap fan heaters for the few days a year they're needed. 

 

Yes in theory. Fan heater good for warming up a bedroom in the evening when you are not there. Less good on a timer first thing, wakes you up prematurely, reminiscent of first flat with no CH. No good at all in living room, OH doesn't like background noise or smell.

 

Edited February 28 by sharpener

[JamesPa]

1 hour ago, SimonD said:

 

In truth, like I suggested earlier, we need a more balanced approach to calculating heat loss and system loads, which will inevitably require amixture of real world data combined with some theoretical calculations to arrive at solutions that are good enough and then use the learning from the good enough to interatively improve the process, but for that we need lots of data and time since we're so behind the curve

I'd go with that.  For this to happen the industry needs to admit that the current method isn't good enough.

 

MCS told me about a year ago that they sort of do accept this, but don't currently have an alternative.

[Roger440]

17 hours ago, JohnMo said:

You short cycle when you are adding more heat than can be taken away and absorbed in to the house or a big enough mass of water.

 

My first attempt at running the boiler was direct to UFH with no mixer and 6 zones. The boiler could modulate down to 6kW, against a max house demand of 3.5kW.  The cycling was on for about 2 mins, off for 10 mins.

 

Open a boiler and just a metal combustion chamber with heat exchanger(s). Most boilers if the run cycle is too short, have a set period lock out from firing again for 10 mins. That is plenty of time for water and metal temp to drop. So it spends most of its time heating and cooling a slud of water and metal. Bad short cycling will not actually heat the house but use lots of energy.

 

Big difference concerns is the energy price - 7p v 30p per kWh

 

That is a much bigger difference that i thought wasa possible.

 

In the context of my solar ponderings elsewhere on the forum, of which fitting a buffer tank comes with a number of costs, if it reduced my oil consumption by 30% would put a different slant on it all.

 

So just to be clear, essentially you inserted a big buffer such that the boiler could run for long periods, and in condensing mode, to achieve that improvement?

[JohnMo]

31 minutes ago, Roger440 said:

 

That is a much bigger difference that i thought wasa possible.

 

In the context of my solar ponderings elsewhere on the forum, of which fitting a buffer tank comes with a number of costs, if it reduced my oil consumption by 30% would put a different slant on it all.

 

So just to be clear, essentially you inserted a big buffer such that the boiler could run for long periods, and in condensing mode, to achieve that improvement?

Those wins only come if your boiler is short cycling badly in the first place. If it's running reasonable run times currently the saving are not really there.

[Roger440]

9 minutes ago, JohnMo said:

Those wins only come if your boiler is short cycling badly in the first place. If it's running reasonable run times currently the saving are not really there.

 

Its not as bad as yours was, but its certainly not good. Ive twiddled it down a fair bit, but it still cycles, especially in milder conditions.

 

I really need to fit a temp probe to the return to see what it actually is before i go further.

 

If i saved 20% then another 20% with solar, plus a few bits to house i could be at circa 50%

 

[Iceverge]

9 hours ago, waxingsatirical said:

 

Yes, calorific value is ~40 according to my bill and my factor includes a metric conversion factor of 2.83

 

 

Can you post your raw data units used at the meter please and the period of time between readings. 

 

 

[SteamyTea]

7 hours ago, SimonD said:

9 hours ago, JamesPa said:

MCS assumption of 2-3ACH

 

This is not a MCS assumption. This kind of figure, if used consistently throughout the heat loss calc is down to the person completing the heat loss calc. Most input from the MCS (based on the CIBSE design guide) is 0.5 - 1.5ACH so what you've seen is a fault with the installer not the underlying calculation method

When we talk of air changes and hour, are we assuming this is at 50Pa. When a house is at close to atmospheric pressures, the infiltration may be a lot lower.

[marshian]

5 hours ago, Roger440 said:

 

Its not as bad as yours was, but its certainly not good. Ive twiddled it down a fair bit, but it still cycles, especially in milder conditions.

 

I really need to fit a temp probe to the return to see what it actually is before i go further.

 

If i saved 20% then another 20% with solar, plus a few bits to house i could be at circa 50%

 

At one stage I was running 10-12 cycles per hour on manufactures stock settings - it was driving me nuts so I decide to deviate from the manufactures base settings

 

I now only hit 6 cycles per hour when the house is in a steady state condition but most of the time it's around 4

 

Manufactures base settings are exactly that "base settings" get familiar with your boiler parameters and understand what they do and what impact they have and you can tweak the settings to reduce cycling without impacting warm up or comfort.

 

Biggest impact is getting the flow temp right (if you don't have weather compensation) - next comes range rating to meet the heating requirements (in my case 16 kW on a 24kW boiler is more than enough)  - then POT (pump over run timer) and then ACT (anti cycle timer) although both the last two have limitations