Just had the regular weekly e-mail from i-measure. The moderator commented that for his home, using Heathrow as the local weather station, HDD this winter were 75% of last winter. He/she is expecting to see a matching reduction in energy used.

Almost as interesting as the kWh/HDD is the base temperature you need for your HDD calculations to match up as it gives a good idea of the length of your heating season. Care to talk about the details of how you do that, Damon?

Posted By: tonyis there a standard year or is everything comparative etc?

You can use historical data for your location as the basis for normalising your actual consumption. You can get the data from http://www.vesma.com/ddd/. Basically if you know the actual degree days then your normalised consumption would be:

actual consumption X actual degree days / average degree days

It's worth taking such a figure with a pinch of salt, but it's a useful yardstick for measuring your consumption if there's been an unusually sever or mild winter IMO. A major source of error will be if you use the same fuel for DHW as space heating, so it helps if you know how much DHW you use (since it should be pretty constant year round and unaffected by degree days).

Related question:
how are people compensating HDD data for local / site factors?

My nearest source of HDD is an airport 10 miles away.
But:
It's by the coast and I'm 10 miles inland -> I get colder in winter, warmer in summer
It's near sea level and I'm 180m up a hill -> I get colder all year
It gets the sea fog in summer, i dont
I'm on a hill top so dont get such sharp frosts
etc

How do I correct this? (Without building my own weather station)

I did a preliminary DD calculation based on info from Herne Airport (Bournemouth) some 10 miles away. I took my base temp as the average below which I needed heating on. (12-13C ish). That was last November. I now have reinstated weather station which gives data logged PC info and will be able to be more accurate. When heating season ends, hope to be able to say with reasonable precision how efficient house is, having weighed too many logs over the winter.

Only fly in the ointment is solar gain, which I might be able to remove from figures with a bit of work.

Some interesting graphs that verge on nerdiness will be forthcoming.

Posted By: WillInAberdeenRelated question:
how are people compensating HDD data for local / site factors?

My nearest source of HDD is an airport 10 miles away.
But:
It's by the coast and I'm 10 miles inland -> I get colder in winter, warmer in summer
It's near sea level and I'm 180m up a hill -> I get colder all year
It gets the sea fog in summer, i dont
I'm on a hill top so dont get such sharp frosts
etc

How do I correct this? (Without building my own weather station)

Tbh there isn't much you can do IMO. The only way to compensate accurately would be to collect local data to establish possible correction factors, and if you're doing that why bother using the remote data? And why bother using accurate remote data if you're just going to apply guestimated correction?

I'm not 100% happy with the distance between me and the nearest station with DD data, but sometimes you've gotta take what you can get.

Have we not discussed this elsewhere, can't remember which thread now.
I came to the conclusion that there are two HDDs, one for the area and one for the house and you have to distinguish between them.

May be time to sort out this HDD issue once and for all.

The 'R' is the coefficient of determination and shows the fraction of data captured and should really be used along with the standard error to be meaningful. It does not, in itself, tell you where the signal to noise ratio is astray though, so you would not know if you are over heating when it is hot or under heating when it is cold, over heating when cold or under heating when cold, over heating when cold and over heating when hot, or even under heating when cold and over heating when cold.

Correlations, handle with care and know the physics.

If you start to calculate your own HDDs, what does it actually show, except that you have matched your heating load to some weather data. I thought the whole idea of HDDs was so that you can compare different properties in different regions on a level playing field

Posted By: SteamyTeaIf you start to calculate your own HDDs, what does it actually show, except that you have matched your heating load to some weather data. I thought the whole idea of HDDs was so that you can compare different properties in different regions on a level playing field

This was the discussion we had before. The point is that HDDs help you normalise the amount of energy used to maintain a certain internal temperature adjusted for weather. So if you use x energy one year (cold), improve your insulation then use y energy the next (mild); if you do not adjust your data for HDD you will think the insulation gave more of a saving than it would have had the winter been as cold as the one before.

(edit) You cannot compare different properties in different areas reliably as there are too many other factors. AIUI HDDs are best used for comparing what is happening to a single property in different years not different properties in the same year.

However, HDDs are best calculated over a longer period I think although if you compare the energy you used v outside temp over several years this should help you normalise the energy consumption for your property.

Posted By: docmartinI find that some of the 'fun' has gone from imeasure following the recent changes. The average data for the communities now seems to be nonsense & of little use compared with the old Carbon Clubs. Access to HDD data analysis is the only reason I have persisted with imeasure.

Indeed, except for the value of sharing data, if I could be bothered to recreate the HDD-vs-power graphs, I'd probably stop.

My house is a detached 1930 chalet bungalow

So you lose on external walls -vs- volume and win on new-fangled things like cavity walls? Oh and being 120 miles south ... tho' some might think that a lose ;-)

(Visited Jesmond a few times when one of Daughters lodged in Grosvnor Gdns for a while when doing a degree at Sunderland Glass Centre).
Martin.

It's even more bed-sit-land now -- whole streets of houses too big for families so in flats or HMO, some being marketed as 'Party Streets', Osborne Road lined with bars ... :-((

Posted By: DamonHDimeasure currently reports R = 0.804, but it was 0.942 until I entered all my historical data back as far as imeasure would take it: since the behaviour of the house (and us) has changed significantly over that time the correlations are lower:

I'm in the same boat, some of my historical data is based on bills rather than a weekly reading, so there are some terrible correlations. I'm not overly bothered about that, as I'd rather the bulk data was in there for other comparisons. Also, since making some insulation and air tightness improvements over the last year or so my current measurements all hit substantially below the regression line. That's bad for R-value, but I'm more than happy! I'm far more interested in the slope of that line than I am in R-value.

I'm currently at R=0.846 Baseload=80. Slope is 7.44 and dropping nicely. 3 bed 1947 semi in Kent.

Weather station can be bought from Charlie Chaplins with data logging to pc, me old Cockney chums.

Tony: that's like saying there is too much freedom in maths to test any theorem you like, or too much freedom to use woodworking tools with sharp edges.

HDD is a tool. It's quite a handy one for normalising against actual weather. Like all tools it can be misused, though nothing I've seen in this thread looks like misuse.

Rgds

Damon

Just hit against the Celsius/Fahrenheit problem........... yuk!