Raising Chimney

Live from HomeLab, it’s the Graduate Program! The Thermal Mass and Buoyancy Ventilation Research Project team members are officially Rural Studio master’s students. The team’s summer semester has started off hot with ventilation opening calibration.

Even with the latest ventilation opening adjustment, described in our airflow post, the data from the Concrete Chimney Experiment reveals the airflow is still choked. As you can see in the temperature signal graph below, the thermal mass surface temperature never rises above the interior air temperature as it should in an optimally tuned space. If we then look at the airflow graph below, we see that the updraft, bulk airflow during the night, is nearly double the downdraft, bulk airflow during the day. When the blue line is above zero, the system is in updraft and when below zero it is in downdraft. Both of these graphs allude that the thermal storage cycle and the buoyancy ventilation cycle are out of sync. This is due to a lack of air. Air drives the cycles as it brings warm air into the chimney to be absorbed and offloaded by the thermal mass.

The team examined their previous math for calculating the total area for the ventilation opening. They’ll spare you the gory details, but the predicted bulk air flow rate they were using to calculate the size was too small resulting in a ventilation opening that was too small. Thanks to the airflow sensors they no longer needed to use a predicted air flow rate and instead used the actual average airflow rate coming from the Concrete Chimney Experiment. After this recalculation the ventilation opening nearly doubled from 3/4” to 1 1/8”. The team then let the chimney do her thing for a week.

a graph showing two days of temperature signals, where the air temperature falls below the panel temperature
Black = Exterior Air
Dashed Gray = Interior Air
Orange = Thermal Mass Surface,
Dashed Orange = Thermal Mass Interior

The data is in and it is as hot as the Alabama asphalt. The team, along with their colleagues were correct in their assumption that the flow was being choked AND the new ventilation opening size is allowing the chimney to operate optimally! In the temperature signal graphs, the thermal mass surface temperature and the interior air temperature properly oscillate. Therefore, the thermal mass is absorbing the heat properly allowing it to be warmer than the interior air at times.

a graph showing two days of airflow data, where the downdraft is larger than the udraft
Blue = Bulk Air Flow

As you can see from the airflow graphs, the bulk airflow of the updraft and the down draft has equalized and is becoming more symmetrical. Both outcomes, in temperature and airflow, reveal there is now a proper amount of air moving through the chimney. The downdraft is still a bit more turbulent than the updraft however and the team wondered if this was due to the concrete pad underneath the chimney releasing heat it absorbed throughout the day. To combat this heat, the team jacked up their Concrete Chimney Experiment… literally!

To raise the chimney, in order to give it some more height via cinder blocks, the Thermal Mass and Buoyancy Ventilation Research Project used car jacks. The team will see if this helps with the heat interference and its possible effects on the air flow. 

two small white dogs in a car

As you can see Wolfie is still in town on his summer vacation! He and Copper like to observe the team work. To insure their safety as the chimney was being raised they watched from inside the car. They really love the car. For more science, design, and cute pets, stay tuned!

Calibrate and Graduate

Team is posing with their new outfit

Exciting things have been happening at HomeLab lately! First, the Thermal Mass and Buoyancy Ventilation Research Project (TMBVRP) Team were able to install airflow sensors into the Concrete Chimney Experiment. Second, the chimney has brought in some impressive data. And third, the TMBVRP team participated in an end of the semester presentation and round table discussion with their big sister team, the Mass Timber Breathing Wall Research Project, and a cast of professionals in the architecture and building science research field.

This week the team received their Sensirion differential pressure air flow sensors. The sensors record a difference in dynamic and static pressure which the team uses to calculate bulk flow. Bulk flow is the total airflow at the sensor location. The team installed two sensors into the Concrete Chimney Experiment, one at the bottom and one at the top, to measure updraft and downdraft ventilation created by the thermal mass.

Just to refresh your memory, updraft occurs during the night when the cool, night air is brought in the bottom ventilation opening, warmed by the thermal mass, and exhausted out the top. Downdraft occurs during the day, the warm, exterior air is drawn into the top ventilation opening, is cooled by offloading heat to the thermal mass,  and vents out the bottom.  Being able to measure the direction and amount of ventilation is critical to understand if the Concrete Chimney Experiment is performing as expected.

And the results are in, our initial measurements from the airflow sensors do show that during the day the chimney is operating in downdraft and during the night it operates in updraft. This gives us proof of concept, that thermal mass is able to alter the atmosphere inside the chimney so that it goes against the exterior environment.

graph showing airflow in the test chimney

The GreenTeg temperature sensors have also brought in proof of concept data, showing that the thermal mass is having a damping effect on the interior air. It is important that the temperatures of the thermal mass and interior air cycle with the daily swing in temperature so that heat is absorbed by the mass during the day and offloaded during the night. This shows that the internal thermal mass is effectively moderating the temperature in the chimney and causing continuous ventilation. We are continuing our testing to further calibrate the amount of ventilation to achieve the most efficient and effective heat transfer between the internal thermal mass and air.

Temperature signal graph comparison

To wrap up our undergraduate work, we had a roundtable presentation via Zoom to give an update on where our work is and share our exciting results with Auburn, our collaborators at McGill, and professionals in the architecture and building science research field.  This panel included Billie Faircloth, a partner and research director at the architecture firm Kieran Timberlake in Philadelphia, PA.  Second, we were joined by Jonathan Grinham, who is a Lecturer in Architecture and Research Associate at the Harvard University Graduate School of Design.  Last but not least, is Z Smith.  Z is a Principal and the Director of Sustainability & Performance at Eskew Dumez Ripple in New Orleans, LA.  

It was a privilege to be able to present and have a productive discussion with such esteemed professionals.  We gained valuable insight on how to best relay the work we are doing do both those in the research field and the common person. In addition, their backgrounds led to an intriguing discussion on how The Optimal Tuning Strategy could be implemented at the building scale. It was especially awesome to discuss the successful data the team recently got form the Concrete Chimney Experiment. Both the data and the discussion gave the Thermal Mass and Buoyancy Ventilation Research Project Team a boost of confidence and pride in their work. It not always easy for these architecture students to wrap their heads around the science, but the hard work paid off. Thank you to Rural Studio, Salmaan Craig, Kiel Moe, David Kennedy, and the reviewers for a positive end of the undergraduate phase of the Thermal Mass and Buoyancy Ventilation Research Project.

Final shout out to the incredible Mass Timber Breathing Wall Research Project Team. As they complete the paper on their research and graduate from the Master’s program they still had time to do something very sweet for their little sister team. They passed along their Rural Studio lab coats, crossing out their names and writing the names of the TMBVRP team members. Their work, dedication, and attitude could not be a better example for the TMBVRP team to emulate. From one research project team to the other, thank you for helping us whenever we needed and being the best big sister team imaginable. We hope to live up the legacy! Well, everyone, stay tuned (optimally tuned) this summer for the start of the graduate program at HomeLab.

If You Know, You Airflow

Ready for some more math? Well, you’re in luck! Today’s post is dedicated to calibrating the size of the ventilation openings on the Concrete Chimney Experiment.The Thermal Mass and Buoyancy Ventilation Research Project (TMBVRP) team has been researching equations for the “effective” opening.

diagram showing the exploded axon of the chimney test, with ventilation openings highlighted
exploded axon of the Concrete Chimney Experiment

The effective opening size differs from the total opening size because it accounts for friction. For example, 1’ x 1’ window has a total opening of 1 square foot, but due to friction caused by airflow around the edges of the window the effective opening may only be 0.9 square feet. With that concept in mind, we can look into why and how the TMBVRP team has been improving their experiment through trial and error.

diagrams showing changes to ventilation strategies
section through the concrete chimney showing the insulation and ventilation openings.

The original ventilation opening for Concrete Chimney Experiment was a 12″ long PVC pipe with a 3/4″ diameter. After reviewing the temperature data of both the interior space and thermal mass, the team saw that the airflow was being choked. This means the effective area of the opening was not allowing for enough ventilation. This caused kept the thermal mass from fully absorbing or offloading the heat from the air. The length to width ratio of the pipe was too high, creating unwanted friction, and slowing the airflow.

mathematical formulas explaining the change in ventilation hole size

For the next ventilation opening iteration, the team needed to reduce the friction by making the ventilation opening a “sharp opening.”  This means that the length/thickness of the opening is significantly less than the diameter of the opening.  The 1′ thick layer of GeoFoam on the top and bottom of the chimney was preventing the ability to have a “sharp opening.” So, the team carved out the top and bottom insulation in the shape of a cone to negate the friction. The bottom of the funnel was capped with a 6″ square of ½” insulation with a ¾” diameter opening. The ¾” diameter opening is the actual area of the opening, the effective area after we calculated for friction is only about ½” in diameter.

version two of ventilation hole sizing

Third times the charm when it comes to ventilation openings!  The ¾” opening in the ½” insulation had a diameter to thickness ratio of ~0.6.  After further investigation a true sharp opening needs to have a diameter to thickness ratio that is much less.  Due to this finding we replaced the ½” insulation with a 1/16 in acrylic sheet to achieve a ratio of ~0.1.  Even after all these calculations we won’t know for certain if we are achieving sufficient airflow in the chimney until we can measure the exact velocity.

version three of ventilation hole sizing

The Thermal Mass and Buoyancy Ventilation Research Project team is looking into how to install airflow sensors into the Concrete Chimney Experiment. Until then, they will keep on analyzing temperature data and designing their experiment.

At Rural Studio, students learn through construction that the design of a building goes far beyond our architectural drawings. Builders and construction workers are designers. Through the Rural Studio Research Projects students are now learning the complexities of designing experimental methods and scientific instruments. The TMBVRP team has developed a deep appreciation for this avenue of design they may not have considered before.

Another important note from this week; Copper’s brother Wolfie came for a visit! The brothers love chilling at HomeLab and keeping an eye on the Concrete Chimney Experiment. Stay tuned to see what the Thermal Mass and Buoyancy Ventilation Research Project Team learn next!


The Thermal Mass and Buoyancy Ventilation Research Team, like the rest of the world, came back from Spring Break to a less clear future. As everyone is trying to navigate moving work, school, and life online, the team along with Rural Studio and Auburn University faculty are making a plan for the possibility of not returning to campus until Fall 2020 or beyond.

A beautiful Newbern sky from the front porch of the Red Barn before Spring Break

Before Spring Break, the team along with the rest of the 5th-year class at Rural Studio underwent a “Pre-Stress Test.” These presentations by each team update Auburn University faculty members including Margaret Fletcher, Rusty Smith, Christian Dagg, and Justin Miller, on the status of each of the thesis projects. The faculty then advise the teams on how to best proceed in order to be approved for building in the summer. The TMBVRP team received feedback on how to better communicate the scientific concepts on which their research is based. They were also encouraged to not rush their research in order to build bigger. After Pre-Stress Test the students left Hale County for Spring Break. Now they are back, figuring out how to research, design, and build from home.

Fortunately, there are circumstances that are encouraging for continuing the Thermal Mass and Buoyancy Research Project at home. First, teammates 3/4 teammates were on spring break together and already live together. This means no social distancing is needed between 3/4 of the team. Second, the Concrete Chimney Experiment resides in the carport theses student’s home (from now on referred to as HomeLab). Third, Rural Studio was able to bring a tool trailer to the HomeLab with contactless delivery. So now, the TMBVRP team can continue their rigorous testing and experiment calibration from the comforts of a 110-year-old house in Greensboro, AL. You’ve gotta love scientific research Rural Studio style.

After two weeks of self-isolation, the fourth teammate will be able to join the team at the HomeLab where they can continue analyzing data and manipulating the Concrete Chimney Experiment. Their plans for the Summer are to insert airflow sensors into the chimney in order to properly resize the ventilation openings and to get a Wood Chimney Experiment up and running. The TMBVRP team is thankful for all the support from Auburn University, Rural Studio, and their partners at McGill University. Their goal is to keep pushing their experiment while staying safe and sheltering at home. To the amazing Rural Studio students moving on in this difficult time, the team wants to thank you for an incredible year together and for setting a wonderful example of how to work hard with bold hearts.

Wood Wise

An important part of the Thermal Mass and Buoyancy Ventilation Research Project is investigating whether, if properly sized, mass timber can be used as a thermal mass. To do so, the team will build a Wood Chimney Experiment to run in parallel with the Concrete Chimney Experiment. As they work to compare the thermal mass performance of concrete and softwood, they must learn more about the thermal and structural properties of the materials. A material’s anatomy affects how it absorbs, transfers, and offloads heat. Concrete has been used as a thermal mass for ages and therefore has widely known and defined thermal properties. However, there is far less information on the thermal properties of wood. Let’s take a closer look at the composition of timber as taught to the team by colleague David Kennedy, a self-proclaimed, “Wood Anatomist Fanboy.”

The first consideration for wood is the difference between isotropic and anisotropic materials. Isotropic materials, like concrete, have consistent properties in all directions. Anisotropic materials, like wood, have orientation dependant properties. Wood has an anisotropic, cellular structure that provides its strength and its the ability to move water and minerals to the outer reaches of the branches. A tree’s main job is to transfer water and nutrients from the ground to the sky and vice versa. Therefore, wood cells can be visualized as bundles of straws, with some acting as pipelines while others store nutrients. These cells consist of tracheids, parenchyma, and epithelial cells. Tracheids are actually dead cells that function as water transporters. Parenchyma store starches that give nutrients for the tree. Epithelial cells take on the job of building the tree. These cells act vertically, while ray tracheids, ray parenchyma, and ray epithelial cells work in the horizontal direction.

As trees grow their structural properties change based on what they need. This is why different parts of the tree are composed of different types of wood depending on their age. If a cross-section is examined, two important types of wood will be present, juvenile wood and heartwood. Juvenile wood is a weaker less dense outer ring and is usually distinguished by a lighter color. This wood makes up the early growth and is not evenly distributed throughout the tree. Often, the tops of trees will be 100% juvenile wood, while the lower parts of the tree are more mature heartwood. Heartwood, the mature, older wood, is produced by converting stored chemicals that are produced by the dying parenchyma cells that have been sent to the center of the tree. This area of the tree is typically darker and denser. These different cells and wood types can result in different characteristics throughout the same tree, which must be carefully considered. 

Timber mapping diagram
Different cuts of wood result in different grain condition which cause different construction systems

The performance of thermal mass is heavily influenced by the thermal conductivity of materials. The thermal conductivity of materials is heavily influenced by the density of materials. With the variety of conditions that are possible all within the same tree, it is important to consider how the material is cut and oriented. This is also important because of the grain directions present in the wood. Just like water flows through the tree along cellular pathways, heat will follow the grain of the wood. Therefore, the cut of the wood must be selected so that the grain in the same direction as the preferred heat transfer.

How heat transfer is affected by grain orientation and construction.

Now that we have learned more about the structure of wood and how that affects its heat transfer we can understand how the wood should be cut and oriented. Initially, the team’s wood test panels utilized the timber end grain. This end grain direction would be parallel to heat transfer. However, depending on the size of the tree and how blocks were cut, the types of wood in the panels could become inconsistent. This meaning they could be made of different mixtures of juvenile wood and heartwood. They were also pretty inefficient to make.

To find a better way, the team investigated the different ways timber is milled into boards. The TMBVRP team found that different cuts of wood could produce different grain directions. The most efficient grain direction for our panels would be to have the grain running perpendicular to the wide face of the boards. This allows the board to cover the area most efficiently, and the board thickness can be controlled to work with the app results. This grain direction can result from several milling, but we believe quarter sawing will produce the most favorable boards with the least waste. The team then made their own version of these boards, but as panels for their wood chimney.

Wood is a far more complex than the Thermal Mass and Buoyancy Ventilation Research Project Team realized! They too are now “Wood Anatomist Fanboys” and hope you’re on you’re way to becoming one. Now, Give your brain a break and look at Rosemary and Dijon being mischievious! Thank you again to David Kennedy for all the help and as usual stay tuned to see what theTMBVRP Team learns next!