CIBSE 2016 Weather Files

[Terence]

The new weather files published by CIBSE enable designers to test their building models against a variety of weather scenarios. These new files have been updated for 14 UK locations and include predictions for UK summer temperatures up to the 2080s.

Choose which Weather data files you would like to purchase, all data is from the CIBSE 2016 set, for further information on this please click here. Two sets of data is provided (Current and Future) and both DSY and TRY files are provided for all 14 UK locations. A discount is available for CIBSE Members.

Buy 2016 Weather Data

[RossThompson87]

Hi,

Thanks for this.

I've just noticed in the consultation draft for the new BB101 specifies that

the most up to date and appropriate CIBSE DSY should be used for the thermal comfort assessment

So I guess this will be the trigger for most of us needing to buy these.

Ross

[Terence]

They are a requirement for CIBSE TM49 too.

[jholgate]

Hi,
Does anyone know why you can't get them from CIBSE at the moment? It looks like they have been (temporarily?) withdrawn.
Jeremy

[Terence]

As far as I understand there was a small issue with the London files. CIBSE have corrected the problem so they should hopefully be available again from them soon.

We have replaced the problem files on our site with the correct versions and informed all who purchased them direct from IES to login to their account re-download.

[jerry]

IMO those prices are excessive, fine if you work for a big organisation but out of reach for a one-man operation trying to keep costs under control

[PCully]

Actually it's not great for the big guys either since the licence costs are per user (up to 10, then additional users are free). Remember you can get your discount by entering your CIBSE Membership Number

Phil

[btysoe]

So these weather files are nearly a year old now. I'm just wondering who is using them?

There is a firm requirement to use them from the GLA in the energy planning guidance but what about the rest of the country?

However TM59(overheating in dwellings) now suggests you only need to pass TM52 against dsy1 and dsy2&3 are more for information and a pass is not required. Can you apply this logic to non domestic buildings? It's inherently difficult to pass the weather file with an intense hot spell because of the way TM52 criteria is based on the running mean temp.

BB101 is still a draft as far as I am aware. It's been so long coming it already needs revising with the new weather data set in mind...

The EFA FOS instructs you to use the most relevant design summer year but was written before each location had three weather files. So I assume the moderate dsy1 is the most relevant?

There is plenty of information on what and how the weather files are put together and I have a good understanding on what each weather file represents but at the end of the day a lot (not all) of our clients simply want a pass/fail result.

In the reports I see from our competitors most are still using the old 05 weather files..

it would be good if there where guidance to clear up when and where the files should be used. is there anything I'm not aware of?

[btysoe]

Dates in Weather files.

I made an simple mistake and edited the simulation calendar in a recent sim, nothing serious I just started my year off on a Monday rather using the 'take from weather file' option. I was carrying out an adaptive comfort overheating risk assessment. My little mistake made a significant difference to the results in relation to criteria 2 & 3 the severity of overheating. The weather file in question is London Heathrow 2020 DSY1. The peak day happens on July 22nd which normally falls in the week and editing the calendar bumped the peak day to the unoccupied weekend hence the improvement in my results.

22nd of July 1989 = Saturday (The year the weather file is based on)
22nd of July 2014 = Tuesday (The year the weather file was produced) this seems to be the day IES pulls through when using 'take from weather file'


There in lies the dilemma. which should you use. it can be the difference between a pass/fail. Weather has an element randomness to it anyway but perhaps the peak day is more likely to fall on a weekday 5/7. There are other hot days beside the 22nd in the weather file just not quite as hot.

Everybody likes the sunshine at the weekends?

[RossThompson87]

Hi btysoe,

I frequently have this problem, and have never seen any official guidance.

I even found differences in the same weather file.

For example "LondonDSY05.fwt" (from CIBSE 2005 set) aligns differently to the "London_LHR_DSY1.epw" (from CIBSE 2016 set).
Although both of these were recorded at Heathrow in 1989!

In my opinion, it has all gone too far. I am seeing projects that need to hit London Plan, BREEAM Future Climate Change, EFSA all at the same time.
This can mean testing the same building on 6 different weather files!

From all that weather data, only a few days / weeks are decisive anyway. It's a massive waste of resources.

I don't see why we can't measure the thermal response of the building across some robust standard test conditions.

Ross