Solar Gains query

[L Headley]

Hi,

Apologies if I am repeating a thread but the search functionality in these forums is pretty poor, this is a separate conversation altogether!

I have a model with multiple floors and is completely surrounded by adjacent buildings. The ground floor is smaller than the other floors so effectively there is an overhang surrounding it and the adjacent building sit directly next to this overhang and go all the way to ground level.

Therefore the ground floor is completely shaded and does not see direct sunlight which I have checked using the sun path function in Model Viewer 2.

However within the results the rooms on the ground floor are still receiving a solar gain, why is this?

Any help with this would be much appreciated.

Thanks

Luke

[bootsam]

Reflected direct radiant from the external buildings and diffuse solar would be my guestimate without inspecting the model.

[L Headley]

I would be able to see that if the buildings were not directly adjacent to each other.

The only gap between the building and adjacent buildings is at ground level due to the overhang, the levels above are "butted" against the adjacent buildings so there is no path for sunlight to take to reach the ground level. I'm not even sure rain would get through after finishing.

So I'm not really sure why the ground floor rooms are registering solar gain, granted there will be a little from some diffuse solar radiation but not enough to have such a drastic effect on the cooling loads as they currently are.

[bootsam]

Hmmm, yes. That does seem odd. Plus you can check whether diffuse or direct from detailed room analysis if memory serves. However it sounds as if the calc isnt taking the adjacent building into account. Change it from adj building into a 'room' and see what happens is my only suggestion. Sorry.

Yes these forums arent the best. I'd expect them to be patrolled by IES devs offering tips, tricks and suggestions. But they dont make use of them, foolish imho.

Failing that, cab it and send it onto support. The support isnt too bad.

[Wasted Energy]

LH

Diffuse irradiation can be very significant in open ground assuming a hemispherical view of the sky.

However this is, as I understand it, far from the case here.

I do wonder what adjustments SunCast makes to the diffuse irradiation due to shading.

Looking at the results in Vista wont necessarily help because it only provides data on the net solar gain.

However you could try running the model twice - once with the obstructions and once without (simply turn SunCast on/off) and look at the differences in solar gain. This will at least check that the obstructions are having an affect and give an idea of its magnitude.

W

[farahghanem]

You could also try and change them from adjacent buildings to "shading devices"... wonder if that changes anything with how SunCast works.

Let us know how it works out!

[L Headley]

OK,

So I have run the model in the following ways and have the corresponding results

1. No Adjacent Buildings - Suncast Link on = 1.6kW Solar Gains
2. No Adjacent Buildings - Suncast Link off = 15kW Solar Gains
3. Adjacent buildings as "Adjacent Buildings" - Suncast Link on = 1.6kW Solar Gains
4. Adjacent buildings as "Local Shade" - Suncast Link on = 1.6kW Solar Gains
5. Adjacent buildings as "Adjacent Buildings" - Suncast Link off = 15kW Solar Gains

All of these were run using CIBSE loads.

Why Suncast makes a difference with no adjacent buildings? is an interesting question. It's clear that the adjacent buildings are not making a difference for these steady state results. I guess the purpose of Suncast is to give you a more realistic solar irradiance based on your location and orientation, it doesn't take into account any solar shading. Does anyone know if this is true or not?

I have some results from a dynamic simulation and these are showing no solar gains as expected. However, most senior engineers prefer to have steady state results as they seem to trust these more than dynamic. I can foresee this model needing to be run numerous times soon and in dynamic this just isn't viable due to time constraints.

[farahghanem]

that's quite odd... can you share an image of your model so we get a better understanding of how it looks? Also, are you re-running the suncast calculations after performing each change?

[L Headley]

Could someone point out how to upload images?

Then I can upload some of our model.

Looking in the FAQ it says about having the files on an open server/website.

Does anyone use a specific website so they can upload to here?

Thanks

[farahghanem]

photobucket is easy and free