Approach
Air changes per hour (ACH) is a critical parameter to know when modeling indoor disease transmission. The ACH of a room tells us how many times, on average, the air is changed out in that room in one hour. For example, a room with an ACH of 4 bring in an air volume equal to 4 times the volume of the room in 1 hour. The higher the ACH, the faster contaminated air is removed from a room. This decreases the exposure to contaminated air for those in the room. However, you do have be careful in interpreting ACH. Many rooms are not what engineers term well-mixed. In a well-mixed room the concentration of contaminant (e.g., aerosol droplets containing viable SARS-CoV-2 virus) is the same everyone. Many rooms have dead spots – areas where the air does not circulate as well. These areas would no be flushed as rapidly as the overall ACH would suggest. Still, ACH is a critical parameter to know in modeling airborne disease transmission in an indoor space.
When we started our modeling project the College of Engineering had estimates of ACH for some of their rooms. It was unclear how accurate these estimates were. We decided to estimate ACH for classrooms. To date (8/27/21) we have about one third of the College of engineering classrooms done. We use the carbon dioxide decay method. There is a good description of this on slide 14 of this slide deck. The summary of the process is:
- We use dry ice to bring the carbon dioxide concentration in the room up to about 2,000 parts per million.
- We remove the dry ice from the room, make sure the HVAC is on the same settings as when the room will be occupied, leave the room so that we are not carbon dioxide sources, and monitor the decrease in carbon dioxide over time until it reaches about 800 parts per million.
- From there directly calculate the ACH (see for example the equation on slide 19).
There are challenges in rooms with high ACHs that necessity coordinating with facilities to turn the HVAC system off while we increase the carbon dioxide concentration and then turn the HVAC back on. Some rooms have been closed due to construction or other activities. We continue to work to measure ACH in additional rooms.
Important Caveat
Our measurements were taken during a period when many of the classes were operating under enhanced airflow due to the ongoing COVID pandemic. If these conditions are changed (e.g., if HVAC settings are returned to pre-COVID settings) the ACH values would decrease, in some cases potentially substantially.
Results
To estimated ACH values for the rooms we have done to date are below.
| Building | Room Number | Estimated ACH |
| Beyster | 1670 | 2.6 |
| Beyster | 1690 | 2.0 |
| Dow | 1005 | 3.7 |
| Dow | 1006 | 3.1 |
| Dow | 1010 | 3.7 |
| Dow | 1013 | 3.9 |
| Dow | 1014 | 3.5 |
| Dow | 1017 | 4.0 |
| Dow | 1018 | 4.4 |
| Dow | 2150 | 3.5 |
| Dow | 2166 | 4.7 |
| Dow | 3150 | 5.5 |
| EECS | 1003 | 5.6 |
| EECS | 1005 | Needs to be redone – HVAC inconsistent |
| EECS | 1008 | 3.5 |
| EECS | 1012 | 3.2 |
| EECS | 1200 | 3.1 |
| EECS | 1311 | 6.8 |
| IOE | 1680 | 3.5-7 (inconsistent – depends on HVAC cycle) |