Temp graph

[jamiehamy]

Not very scientific,  but does this look okay? Bearing in mind there is no heating in the building yet and some small areas to be plugged/insulated and we have no plasterwork. 

 

It's glorious sunshine today but as you can see,  there is limited effect on the internal temperature. The south elevation glazing all has solar tint on it. 

 

I'm trying to finalise what we're doing for heating hence my interest in this - but I'll open another thread in a bit specifically for that, 

 

Cheers, . Jamie

 

[gravelld]

What do you mean by ok? If you mean is the house performing to expectations then we'd need to know the design - the heat loss calculations produced beforehand.

[jamiehamy]

3 minutes ago, gravelld said:

What do you mean by ok? If you mean is the house performing to expectations then we'd need to know the design - the heat loss calculations produced beforehand.

Sorry,  that was totally vague! 

 

I just meant in terms of how the house responded over a 24hour period,  does that look 'normal' given where we are at. 

 

I'm trying to get up to speed on heat loss - until now I've not paid much attention. We don't have any heat loss calls - assuming that's because from a BC perspective this is a conversion? I know we don't need an airtightness test. 

 

I've looked at heat loss calculators online but none are fit for purpose for us. 

 

For u values,  floor is 0.14, walls 0.14, roof 0.15 and windows generally 1.0 - but we have huge expanse of glazing. 

 

Is this the point you direct me to a spreadsheet?! 

[gravelld]

LOL, pretty much, at least some form of calculator. Just because it's an extension doesn't mean it cannot be modelled. You'll have to use an assumed figure for air leakage I guess.

 

But maybe someone can also give you a yes or no based on their experience. The absolute temperature looks low to me, but I guess that might be because of works going on, doors open, closed etc.

[IanR]

There're too many unknowns to make any real estimates, but you could take the Tue 18:00 to Wed 06:00 period where there is no heat gain from solar and see that, with an average temp differential of approx 2.5 degrees, your house dropped 0.2 degrees in 12 hours. 

 

Seems "pretty good", but anything more definitive with the unknowns would be a guess.

 

 

 

 

[Dudda]

Looks ok to me as it's not dropping much even if the external temperature is falling. As stated above hard to be more specific than knowing more information about the details, etc.

 

What are you using to record the temperature? I'd like to install something like that.

[ProDave]

What type of insulation do you have?

 

This is the prediction for my house which has a high decriment delay type of insulation

u-wert-berechnung (10).pdf

 

It shows a time lag of 13 hours, which oddly means peak internal temperature will be in the middle of the night, when it's coldest outside.

 

Yours seems to turn the corner a lot quicker as it starts to warm up outside. I presume that's more to do with solar gain?

 

[ragg987]

I'm not sure we can learn a lot from graph - it shows that internal temperature is only a few degrees higher than external, thus leading to much lower heat losses (which the graph show nicely!).

 

Also, impact of (e.g.) solar gain or internal works (people and tools) is going to be magnified compared to if the inside was a nice toasty 20 deg.

[jamiehamy]

1 hour ago, ProDave said:

What type of insulation do you have?

 

This is the prediction for my house which has a high decriment delay type of insulation

u-wert-berechnung (10).pdf

 

It shows a time lag of 13 hours, which oddly means peak internal temperature will be in the middle of the night, when it's coldest outside.

 

Yours seems to turn the corner a lot quicker as it starts to warm up outside. I presume that's more to do with solar gain?

 

Yes,  I think so. Generally the steep external increases are when it's bright sunshine - although that might sound obvious,  it can be cloudy all day when a weather front hits adding 5degrees with no sun. 

 

My friend has rigged up two sensors that somehow record and send to a webber version and are accessible from anywhere. Quite handy for a whole host of purposes! 

 

I used the MYSON  heat loss calculator which allows me to plug in all the u values - but gives me a Nan result! 

 

I'll try again when I get home. Walls are XPS 200mm, roof 150 PIR, floor 150 XPS. 

 

This is the latest weekly graph BTW. On Monday we had a 3kw heater on in our working area. The sensor is quite away away from the heater. 

[jamiehamy]

1 hour ago, ProDave said:

 

 

This is the prediction for my house which has a high decriment delay type of insulation

u-wert-berechnung (10).pdf

 

 

Thanks for this - will try tonight

[SteamyTea]

It looks OK, but as others have said, it all depends on what you were trying to achieve.

 

Another way to look at the data is to plot the internal temperature against the external temperature, then time slice it to find the effects of solar gain.  Then you can look at the temp differences and variations, standard deviations and generally play about with things when it is dark or daylight.

One other thing that is really helpful is wind speed and direction.  This can give an indication of cloud cover as we tend to have a warm, but cloudy SW wind that is quite fierce, compared to a cold, but clear NE wind that is quite gentle.

 

Is your data collected as a text file, if so, you can post it up and I can have a look.

8 hours ago, Dudda said:

What are you using to record the temperature? I'd like to install something like that.

You can make your own quite easily and cheaply with a few 1Wire sensors (or some DHT22s) and a Raspberry Pi.

All really depends what you want to log and for what reason.  There is usually a WeatherUnderground weather station nearby which you can grab data from as a text file.  Some of the weather stations are pretty good bits of kit and have a solar power meter on them.

Edited January 4, 2017 by SteamyTea

[jamiehamy]

13 hours ago, SteamyTea said:

Is your data collected as a text file, if so, you can post it up and I can have a look.

You can make your own quite easily and cheaply with a few 1Wire sensors (or some DHT22s) and a Raspberry Pi.

All really depends what you want to log and for what reason.  There is usually a WeatherUnderground weather station nearby which you can grab data from as a text file.  Some of the weather stations are pretty good bits of kit and have a solar power meter on them.

 

I think the data will be collected as a text file, but the problem is we have been using a gas heater at weekends (until we ran out of gas over Christmas!), so the longer term trend is distorted somewhat - we've not been using the gas for a couple of weeks now. 

 

I think my friend used the same combination - I know all the onsite cams run on one or more Pi's and there are two wire sensors. 

 

@ProDave

I've used the link and this is a revelation! Spend the night plugging in all the values for the components and now have some good figures to go on. Quite pleased and I think this really knocks out bothering with an LPG gas central heating system completely and the number of electric heater that were planned. Considering now going for an inline heater with the MHRV system - will speak to BPC in the morning to look at options here. Thank you for sharing that - it's probably been discussed ad nauseum before but I've only really gotten down to looking seriously now we are ready for putting in some form of heating. I got there eventually!

 

Thanks everyone for the inputs

[TerryE]

@jamiehamy, Jamie you won't get Passive-class performance with this spec, but you should comfortably exceed 2013 Breg performance.  Peter mentioned decrement delay and it's worth reading up on this.  This is a time constant which reflects the speed with which external heat changes propagate through the wall fabric to be experienced internally.  I also did a blog post discussing this.  It relates to the ratio of the U-value of the wall divided by its thermal capacity.  My profile which is 300mm cellulosic filler with an external 125mm stone skin has a delay measured in days.  On the other hand whilst XPS and PIR has excellent R-values, it's thermal capacity is very low, so the decrement delay is quite poor.  You can see this by the way the lag between the external temperature and the measured internal is only 3 hours or so.   Also your house has lost ~0.7 °C in response to a ramping delta 1-8°C external to internal.

Also large areas of glass are really bad news from a thermal perspective.  A U-value of 1 might seem good but that is 8× what I lose through my walls and 6× what you lose -- and the decrement delay of radiant loss is zero, of course. 

 

The main impact of this level of heat loss is that the number of heating days that you have will be quite high and the heat input will require a full solution such as UFH, plus you will probably need extra heating on the first floor to maintain a comfortable temperature in the colder months.