Considering a mid latitude (e.g. in Virginia), south facing, single family house with two stories, AC cooling and solar power on the roof:

What would be the ideal level / amount of shade for the house in summer?

An additional condition is that the entire house is shaded, so the question is not about which part of the house should be shaded, but to which degree the entire house should be shaded.

As a concrete example, there's a tree in front of the house. Is it better to cut the tree down or not?

The interior of the house is cooled by an AC unit that consumes electricity. Solar cells on the roof produce electricity from sunlight, which partially offsets the electricity consumption of the AC unit. However, direct sunlight is also heating the house, increasing the power consumption of the AC unit again.

Additional shade shields the house from the heat of the sun, but also may reduce the power from the solar panels.

So the question would be if there is an optimum point there. Shade could come from trees in front of the house, other buildings, or e.g. solar sails.

The answer could also be "None", or "as much shade as possible".

  • $\begingroup$ Do you have any variables or constants that describe this situation? Are we to take into account the latitude of the house? Perhaps the amount of hours in a day and peak solar output as a function of time of day? You might need to provide more information or focus to your question as there are still a lot of unknown assumptions. $\endgroup$
    – Triatticus
    Commented Jan 25, 2022 at 22:14
  • $\begingroup$ In physics, it is the opposite which is measurable (quantifiable),i.e, the sunshine of the house and which influences much on the energy performances of a house: en.wikipedia.org/wiki/Energy_performance_certificate $\endgroup$
    – The Tiler
    Commented Jan 26, 2022 at 17:14
  • $\begingroup$ Since the range of parameters is so broad, I would be content with any special case already. Since this is a problem of practical relevance to many home owners, it's hard to imagine that no one has studied this problem before. I will also update the question to not allow the trivial case that all of the house is shaded except for the solar panels. $\endgroup$ Commented Jan 26, 2022 at 21:07

2 Answers 2


I would expect that the ideal situation would be to shade all of the house except the solar panels (which are presumably on the roof). These would actually shade the house themselves, and would benefit from as much sunlight as possible. Actually arranging this in practice is probably difficult though.


The answer could also be "None", or "as much shade as possible".

The obvious answer would be "get as much shade as possible for the house, without shading the panels on the roof", but if that was possible you would not have asked this question in the first place. For anything more specific than this, the answer is: it is complicated. Have in mind that even small shade on photovoltaic panels significantly decreases their conversion efficiency!

We are missing many parameters here, and I am afraid even if you give more details there is no simple answer. Depending on the house parameters, maybe you can "afford" letting more sunshine in such that you do not shade photovoltaic panels and get better conversion efficiency. But it is usually very complicated to model these systems.

If you truly want to optimize the geometry of shading, I will just give here some guidelines how to approach this problem:

  1. Get data for which you could do simulations. This includes some historical air temperatures and preferably solar irradiance. The former should not be difficult to get, but the latter might be a challenge!

  2. Setup a model that describes your situation with clear distinction to control and output variables: (i) control variables are the ones over which you have control such as geometry of shading, and (ii) output variables are the ones that you optimize (indoor temperature, electricity consumption etc.)

  3. Setup a criterium function which will give some numerical value from the outputs such that you can compare the quality of a solution. A solution is one particular set of control values.

  4. Employ some optimization algorithm to find the best solution. This could be a basic brute-force method in which you try many different combinations for input variables, or some smarter optimization technique.

  • $\begingroup$ Yes. I had hoped someone had already done such a calculation, even if it's just for a special case. $\endgroup$ Commented Jan 26, 2022 at 21:03

Not the answer you're looking for? Browse other questions tagged or ask your own question.