Several Solar Enclosures - How to determine parent rooms?

Hi all,

I’m a bit stuck. In my model are 3 solar enclosures. I tried to figure out which rooms belong to each of them.

I searched all files in the /run folder but couldn’t find sufficient information.

Is there way to get this information either graphically or out of a file?

Hi @Martin6 ,

The HB Color Rooms component will allow you to get a list of solar values that are aligned with the output rooms.

As you probably know, the logic for how the solar load is distributed between the rooms of the enclosure can be found here.

@chris when I follow your advise it raises even more questions.

The output of the HB Color Rooms component looks like this.

In case of this enclosure it may not best practice to have an airwall at the ground?
I did it because it was easier to deal with the construction this way.
Out of curiosity and to learn something I decided to remodel the whole model anyway.

Still locking at he results of the model what bothers me most is the discrepancy between the high solar gain indicated for the NW facing room 209.90 kWh/m2

and the low gain in the yearly display where it’s very reasonable late afternoon exposure only 1/10th of the amount compared to the other SW and SE Solar Enclosures

And the large Atrium in the middle isn’t getting much solar gain.
This doesn’t make much sense to me at all ?

Here is the solar allocation within the building.

Hi @Martin6 ,

I’m sorry that I just can’t tell from your screenshots what is unexpected about the results. Is the issue just that the distribution of the solar load between the rooms using their exterior window area not realistic? Or do you think it’s a deeper issue in that the rooms are being matched to the wrong solar enclosure?

A minimal sample file that labels the unexpected behavior would probably be a big help.

Hi @chris I think this is for sure a more complex case involving a couple of parameters like insufficient modeling and, when I now recall some of the observations during the development of the file, problems to calculate the solar gain properly when airwalls are involved.

It works somehow when the model is altered.

Here is a picture with the result when the model is build differently

Way different loads and the distribution of the solar load between the rooms is there. Also if related rooms have no windows it seems like they get nothing.

Well I can still not determine out of the model how they are related?
But when I take a look at the hourly plots, they look pretty reasonable to me regarding room orientation and daytime to receive solar energy.
It looks more like the HB Color Rooms component is having difficulties to calculate proper results

Thanks @Martin6 ,

A sample file would be really helpful for understanding whether there’s an issue in the component that should be fixed.

Hi @Martin6 ,

Thanks for bringing up this issue. I was able to create a simple case where the solar enclosure data was not matched with the correct Room. So your hunch was right and there is a bug for the case where you have multiple solar enclosures (aka. multiple groups of rooms that are continuously connected by air boundaries).

For some reason, the E+ and OpenStudio developers made it extraordinarily difficult to get the solar enclosure info out of the EnergyPlus calculation and there’s currently no way to do it with the OpenStudio SDK since you have to add this object to the IDF, which is not supported by OpenStudio:

    ViewFactorInfo;          !- Report Type

However, it seems like the E+ developers built a “back door” way to figure out the solar enclosures, which is that they always seem to sort the rooms alphabetically by the Zone name before they group and assign an enclosure number to them. I hate having to rely on something that isn’t documented and is really just a guess on my part but I have tested this with more than 20 combinations of Rooms and it always seems to match the Rooms with the correct enclosure.

So I pushed a fix for the Color Rooms component that simply does this alphabetical sorting:

When you get the chance, can you run the LB Versioner and test the fix with your model? Thanks again for all of your help here.

1 Like

Hi @chris
I can confirm what you found out seems to be working exactly like you think. The results make sense.
Thanks a lot for your fix. It’s a great improvement.

1 Like