OpenStudio VersionUpdater: incorrect EffectiveLeakageAreas

asked 2016-06-15 14:55:42 -0600

updated 2016-06-15 14:59:52 -0600

When opening an OpenStudio 1.4.0 version file in OpenStudio 1.11.0, I have observed that the inputs for

SpaceInfiltration:EffectiveLeakageArea

have been modified by a factor of 10000, i.e. if the original input was e.g. 20 cm2 then after updating to version 1.11.0 now we have 20000 cm2 (!). Note that in version 1.4.0 there was obviously a typo regarding the units, which read m(m2) instead of cm2, but it was still treated as cm2. This can also be verified by looking at the corresponding idf files exported by the respective OpenStudio versions.

Is this bug already known ?

edit retag flag offensive close merge delete

add a comment

answered 2016-06-15 16:37:16 -0600

Adam Hilton

3465 ●1 ●15

There wasn't a typo, m(m2) means milli * meters^2 so 1 m^2 = 1000 m(m^2), it was being displayed as intended.

OS was using m^2 up until 1.9.0 when it was changed to cm^2 to align with the E+ idd. That prompted the display change from m2 to m(m2). I'm fairly certain that OS 1.4 was using m2 as the units, so 20000 m(m2) would be the correct conversion as we are increasing by a factor of 1,000. See this bug.

Now when you look at this value in OS 1.11.0 the displayed units are cm2, but in actuality they are still m(m2). The value would have to be multiplied by a factor of 10 to be correctly displaying in cm^2. This would point to a potential version translation bug, but not to the effect that you're describing.

There is a bug addressed in OS right now with respect to the unit conversion from in^2 to cm^2 (see here). This was grandfathered in from the change from m^2 to cm^2 when the display units were m(m2). The units were being displayed as intended but they weren't being changed by a factor of 10 when the idf was generated to convert to cm^2. When the display was changed to cm2 the value was now correctly translated to the idf but the conversion to in^2 was now incorrect.

In the end, the idf has not been correctly generated for a while as the conversion from 'm(m2)' (10^-3 m^2) to 'cm2' (10^-4 m^2) was not being accounted for. This shouldn't be an issue anymore as the display units are cm2, there will be a version translation issue that will cause values to be off by a factor of 10, but not by 10,000.

edit flag offensive delete link

Comments

I don't believe that this is correct because under earlier version of OpenStudio and EnergyPlus simulation results were OK. It was only after updating to OS 1.11.0 that I discovered that totally different simulation results were due to much higher air change rates. After correcting EffectiveLeakageAreas the old simulation results could be reproduced. It may well have been that the intention was to use mill*meters², but apparently it was not correctly implemented.

OS-user-AT ( 2016-06-16 00:21:57 -0600 )edit

Did you read through the bug reports?

Adam Hilton ( 2016-06-16 08:35:19 -0600 )edit

m(m2) - how odd. Definitely not an SI unit at least up here in SI land (Canada - except for Alberta where you still see lots of drawings in Imperial units, but Alberta has always been different - last jurisdiction to implement mandatory seat belts - that kind of thing). m(m2) would be 1/1000 of a meter squared or 10 cm2. mm - milli meter - 1000th of a meter. SI drawings include mm and m - no cm. There are no 2.5 cm φ pipes, only 25 mm φ pipes. It would be much better if we were all on the same plain measurement wise.