Weather Converter statistical calculations HDB 99.6% CDB 1%

asked 2020-11-18 04:23:34 -0500

updated 2020-12-12 11:47:39 -0500

Hi, I'm working on a Custom weather file with a WMO station not included in ASHRAE list. Therefore for design day Weather Converter (version 8.1.0) gives values calculated based on THIS weather file ONLY as stated in AUDIT file.

I know statistical year are most likely not tought to evaluate design conditions (and in effect this is the case), anyway I was trying to get some values about it.

I go to the point: design conditions from Weather Converter (WC) and returned in AUDIT and STAT files do not coincide with values one can easily calculate by spreadsheet over the 8760 hours, e.g.

HDB 99.6%: -6.1 °C from WC vs -4.4 from spreadsheet

CDB 1% 32.6 °C from WC vs 31.0 from spreadsheet.

What calculations does the Weather Converter perform?

I search through AuxiliaryPrograms E+ doc, and I did not find anything strange about what it should do.

I post links to CSV (where there are all the hourly data) and AUDIT files from the Weather Converter run.

Any insight will be appreciate.

Does anyone faced this before?

edit retag flag offensive close merge delete

Comments

Maybe @lklawrie could give me some hint (I've reached her through Help> About of the Weather Converter)

Ag ( 2020-11-20 16:19:41 -0500 )edit

add a comment

answered 2020-11-22 14:26:59 -0500

lklawrie
440 ●14 ●4

updated 2020-12-01 10:24:30 -0500

Sent answer via email. Do not see it here:

I think the probable difference is that the weather converter looks at Summer and Winter seasons rather than the entire year.

For North Latitude, the summer season are the months 4-9, the winter season are months 10-3

UPDATED Answer: (only 1 answer per user allowed -- what a crock). Anyway, I shamelessly copied the code from the DOE2 weather converter for making the "design condition calculations" -- The code puts temperatures, etc into "bins" and then uses the 99%, 1%, etc bins to make the desired temperatures.

That is probably why there is no exact match.

Linda

FIBPSA, FASHRAE http://climate.onebuilding.org - free repository of climate data for building simulation Climate.onebuilding is a FREE service not supported by any outside organization or government agency.

edit flag offensive delete link

Comments

Not even with your suggestion I can match the Weather Converter results:

-5.5 °C (10-3 months) vs -6.1 from WC 99.6% CDB

31.9 °C (4-9 months) vs 32.6 from WC 1.0% HDB

And more:

-6.6 °C for JAN (both coldest month and month where cold peak temp occurs)

34.0 °C for JUN (month with DB peak temp); 31.5 °C fol JUL (hightest mean tdb)

-5.9 °C for central winter (12-1-2)

32.8 °C for central summer (6-7-8)

I've tested many other combinations , but I've never find the match !!?

Ag ( 2020-11-24 05:03:55 -0500 )edit

Comment after your UPDATED answer:

Thank you for your reply. Can you let me know interval lenght and interval step on which data are binned?

i.e. for example:

--- [min DB; max DB] as data lenght (I think it cannot be anything different)

--- 0.1°C as intertal step (into which data are binned)

I really cannot figure how AUDIT data are found by WC (even binnig like above, HDB and CDB result 0.1 °C close the values I've calculated directly on 8760 records)

We are talking +/- 1.5°C (WC vs. Spreadsheet), a difference that cannot be easily neglected in my humble opinion

Ag ( 2020-12-12 11:46:04 -0500 )edit

The DOE2 source code is available, maybe that will help. Go to http://doe2.com/Download/DOE-22/, download and unzip DOE22WeatherUtilities.ZIP, and look at the HTDGDY/CLDGDY methods in SRC/WTH.F.