lower u-value with higher cooling energy?

asked 2016-04-30 09:18:46 -0600

Pouya

215 ●1 ●9 https://www.linkedin.c...

updated 2017-06-02 18:23:40 -0600

__AmirRoth__

4436 ●5 ●17 http://bleedinggreenna...

Dear All,

I have simulated two types of construction for a one-story house (1 living room, 1 sleeping room and 1 bathroom) in EnergyPlus. Each room has 1 window and the windows and doors are identical for both constructions. The house is quite small (320 square foot), there is no shading for windows and the location is in the climate zone 3B-dry.

The construction with lower u-value has lower required heating energy that totally makes sense. But this construction has higher required cooling energy than the other one. How can this be justified?

With and without considering natural ventilation and internal gains (home appliances) the results are same. The construction with lower u-value, has lower required heating energy, but higher required cooling energy.

I have rechecked the simulation various times and there seems to be no problem. Are these results justifiable? I appreciate any hint!

Thanks a lot in advance!

edit retag flag offensive close merge delete

add a comment

5

answered 2016-05-02 04:39:33 -0600

Julien Marrec

29706 ●94 ●76 http://www.effibem.com/

Space cooling penalties with lower U-value can make total sense for fairly warm climates. For (very) warm climates, you might actually end up having more of a cooling penalty than heating savings.

This is actually counter intuitive to most people until you start thinking about it, but that's really where energy modeling has a value!

The short story is that by having a lower U-value, you're "trapping heat" in the summer. Your building becomes a lot better at not transmitting this heat that would have been dissipated to the outdoors especially at night when the temperatures outside are lower.

I've actually had a reviewer asking for the same thing on a model I submitted recently, and I gave him relatively the same answer. To illustrate my point, I used the DOE Commercial Prototype Building Model of the same type (multifamily in my case), and simulated it in the same location as my project and varied the window U-value. It exhibited exactly the same behavior.

Now a couple more things:

You should still make sure that the increase in cooling is still reasonable and not way too big to be believable
Another characteristics of windows is the Solar Heat Gain Coefficient. This will have quite a big impact for summer cooling (the lower the better for cooling, the opposite for heating...).

edit flag offensive delete link

Comments

Dear Julien, thanks a lot for your detailed answer! it was one of the things came to my mind, and now I understand! really interesting indeed! The increase is quite reasonable, I will simulate more models with altering the windows U-value and solar heat gain coefficient to confirm the explanation. If you have your papers published would be great to read them! Once again, thanks a lot!

Pouya ( 2016-05-02 05:17:30 -0600 )edit

1

I've seen this in a few models as well.
In some mixed-climates, internal gains can be high enough and heating gas cost low enough to give an "optimum" insulation level at or barely above code level according to the model.

However, there are two ways where the model can get this wrong: (1) operable windows. They often exist, and often aren't modeled as such. This free ventilation takes away the insulation penalty during cooling. (2) (some) internal gains are exhausted with return air and never appear as a load. The model can capture this, but it isn't a default.