I started this process by importing the geolocation topography of Devils Lake, WI, from SketchUp Pro into Revit. I chose a location on the north shore of the lake, facing south, to optimize solar heat collection and photovoltaic potential.
Conducting a weather analysis using Climate Consultant, I developed a Psychometric Chart and modified the parameters to understand the comfort constraints. The location was set to the Madison, WI weather system as this is close to Devils Lake. Not to much surprise, the largest constraint is the very cold outdoor temperatures that are experienced during the summer. The following three charts show the comfort when different settings are prioritized:
Without a HVAC system, it is very hard to maintain a comfortable level:
A heating system alone can cover much of the comfort needs. Air conditioning/cooling isn’t a major necessity:
By adding some of the other design strategies, while still refraining from implementing air conditioning/cooling, an acceptable comfort can be achieved year round:
I then created three mass models on this terrain with a floor area between 30,000 to 35,000 feet. It is staggered by its floors to better fit the sloped terrain and optimize light exposure. I then created a solar study to view how light hits the buildings throughout the day. Overall, there is good exposure; however, later in the day, there are shadows cast on the building due to the hills nearby. I may have to revaluate my site location as this could prove to be a greater challenge in capturing solar energy.
Mass Model 1 (Cube design with flat roof and large south face):
Mass Model 2 (Cubed design with sloped roofs and side wings):
Mass Model 3 (Staggered bone shapes with slanted roof):
Possible Model 3 idea sketch:
This idea brings in the general staggered concepts of Mass Model 3 but makes them more architecturally intriguing. The roofs would be slanted to increase PV potential. There is also the potential to include flowing water through the center for hydroelectric power. Unfortunately, these isn’t a river in the current topography so this would be difficult to implement. I want to create a mass model similar to this but feel limited in the current block types to make this accurately.
When comparing the three mass models in Insight with a set of baseline scenario design characteristics, the following energy levels are given, where each mock-up number corresponds to each mass model number:
The specifics for each Mass Model’s Insight analysis are:
Mass Model 1:
Mass Model 2:
Mass Model 3:
Solar Insolation Analysis:
Mass Model 1:
Mass Model 2:
Mass Model 3:
While analyzing the three mass models, mass model 1 stands out as being most efficient according to Insight. On the contrary, mass model 3 was best when it came to PV potential and payback. This design has more optimal roof space and solar capture. I decided to continue with mass model 3 as it showed promising results within the PV stage. I continued to try and optimize its mass design and went to Insight for further refinement. Within Insight, mass model 3 says that Architecture 2030 standards are outside of possible EUI, however, I was able to modify the design constraints to bring the average down. I believe there could be some form of insulation loss within the model as infiltration was the leading energy user. I will continue to investigate this going forward and hopefully bring Architecture 2030 back into scope.
When further investigating mass model 3, infiltration, efficient HVAC incorporation, and southern wall WWR had the greatest effects on EUI. At first, I wanted to maximize southern wall glazing, however, it showed better performance with less. This could be due to overheating during the summer months that may require extra cooling. I also modified the mass model, as shown below, to include sloping on the two front wings.
This seemed to work opposite as I had hoped and increased the range. This is shown by the jump from a minimum of 31 to 37. Continuing forward, I will further refine this design to hopefully reach the Architecture 2030 standard.