Thread by @Poppendieck

A thread on how I estimated ventilation and filtration effectiveness in a classroom using CADRs and CO2 concentrations. 1/

Before I start there are many, many caveats.

See this great thread by @CathNoakes :

https://twitter.com/CathNoakes/status/1431998165662806019

And

https://twitter.com/Poppendieck/status/1366055136859070468

2/

Specifically, ventilation and filtration prevent the buildup of aerosols that can indicate long range infection risk. They do NOT address the risk of short range infection (masks, spacing are still necessary). These interventions are IMPORTANT! 3/

We have four ways remove aerosols from a classroom: 1) Portable filtration. 2) Ventilation controlled by occupants (windows, fans), 3) HVAC system ventilation, 4) HVAC system filtration. This thread will cover the first three. The last requires building operation knowledge. 4/

We are going to talk about two related things with regards to ventilation: Ventilation rate per person (L/s/person) and Air change rate (ACH). 5/

Ventilation rate per person (L/s/person) is the amount of outside air that should be brought into a space according to national and international standards. These rate are set depending on the space and activity use (e.g. ASHRAE 62.1). 6/

An air change of 3/h means that air equivalent of 3 volumes of the room has entered the room in one hour. Importantly, air entering does not equal aerosol removal due to mixing.

@CathNoakes explains:

https://twitter.com/CathNoakes/status/1431998181404028933

https://twitter.com/Poppendieck/status/1294977516063657985

7/

Pre-pandemic the recommended ventilation rates for classrooms (6-7 L/s/person, depending on age and size for ASHRAE) working out an air change rate of 2-3 ACH. Most classrooms were not ventilated to this rate prior to 2019. Guidelines BEFORE Delta call for 5-6 ACH.
8/

Classroom. The classroom I was interested in is 25 ft x 40 ft x 9 ft. So it has an area of 1000 ft2 and a volume of 9000 ft3. The classroom has 30 kids under 9 years old and 4 adults. Importantly this classroom is a portable/temporary classroom not attached to others.
9/

Portable Filtration. The classroom has two HEPA filters each with a maximum CADR of 300. This means it cleans air wrt aerosols at a rate of 300 ft3/min. We can convert that to an hourly rate: (2 Filters)(300 ft3/min)(60 min/h) = 36000 ft3/h
11/

To get an effective air change rate for aerosols we can divide by the volume of the classroom: (36000 ft3/h)/(9000 ft3) = 4/h or 4 ACH. Now due to noise those units might typically be run at half that speed. So the effective air change rate is probably closer to 2 ACH.
12/

This thread is not about efficacy of various technologies of air treatment. But you now can see why a air cleaner with a maximum CADR of 60 will not be very effective in this room (0.4 effective ACH for this room).
13/

So with two large HEPA filters we are either at 2 or 4 effective ACH. Now what about CO2 and ventilation?
14/

We can use CO2 with two approaches: 1) What is the maximum CO2 in the classroom (assuming steady-state) with all the kids present? 2) How fast does the CO2 decay once they all leave?

15/

CO2 with kids PRESENT. We will start with the first one. To use this method we need to know the number of people in the room, activity level, approximate age, weight and the area of the room.

16/

With this information we can use this tool to estimate what the expected CO2 concentration should be for a given ventilation rate (L/s/person).
pages.nist.gov/CONTAM-apps/we…
17/

I chose user defined inputs. Outside and initial concentration of 420 ppm of CO2. A primary ventilation rate of 15 L/s/person is about the twice the value pre-pandemic ASHRAE rate (7.4 L/s/person for typical 5-9 year old classroom). 18/

The occupant density is on a per 100 m2 basis which is 1080 ft2. Our room is 1000 ft2. So we have (34 people/1000 ft2)(1080ft2/100 m2)=36.7 people/100 m2. So I added three more kids to give the appropriate density. 19/

The met level (kcal/min) is an activity indicator. Watching TV would be a met level of 1.0. Walking would be 3-4. Classroom met levels are probably typically 1.5 for kids sitting. 20/

Then I added occupants with average values for the kids and adults. The age and weight of the occupants impact the amount of CO2 emitted. 21/

The results predict what the maximum CO2 levels in the room should be if we have a ventilation rate of 15 L/s/person. In this case if the end of the day CO2 levels in the classroom are around 1170 mg/m3 we likely have around 15 L/s/person ventilation. 22/

NOTE: The tool reports concentrations in mg/m3 not ppm. To convert mg/m3 to ppm multiply by 0.53 (at 20C and 1 atm). So here (1170 mg/m3)(0.56 ppm/mg/m3) = 655 ppm. 23/

In this case IF at the end of the day CO2 levels in the classroom are around 655 ppm we likely have around 15 L/s/person ventilation. 24/

If we were supplying 7.4 L/s/person (alternate scenario) to the classroom then we would should expect a maximum CO2 concentration at the end of the day of about 890 ppm (1590 mg/m3)(0.56 ppm/mg/m3). Assuming an outdoor concentration of 420 ppm. 25/

Note that these values will change with different activities, age, number of occupants, and room size. 26/

If the ventilation rate is 7.4 L/s/person (we keep the room below 890 ppm) we can estimate what the effective air change rate is from ventilation. (7.4 L/s/person)(34 people)(60 sec/min)(60 min/h)= 905760 L/h. 27/

(905760 L/h) (0.0353 ft3/L)= 31986 ft3/h. Dividing this number by the room volume (9000 ft3) gives us an air change rate 3.5/h. 28/

So for this room (1000 ft3, 34 people) maximum CO2 concentration of 890 ppm = 7.4 L/s/person = 3.5/h. If the maximum concentration is 655 ppm = 15 L/s/person = 7/h. Remember these are estimates very dependent on age/size/activities (met). 29/

Intermission... 30/

CO2 when kids have LEFT the classroom. This method is good for portable classrooms which are NOT attached to other rooms. When classrooms are attached (most schools), the CO2 concentrations in other rooms impact the result and the assumptions are not valid. 31/

For this method we recorded the CO2 concentration over time after kids (and all other CO2 sources are removed from the room). In this case this happens at 3 pm. (Open windows 3-4:30, Windows closed at 6 pm). 32/

Once we have the data, we can apply a mass balance and some math to determine the air change rate. A mass balance is simply a way of counting things. For instance, if we wanted to know the number of people in a store over time we could constantly run around the store and... 33/

try to count everyone in the space. Or we could stand at the door and count the number of people entering and leaving the store over time with the difference being the number of people in the store at any given moment. A mass balance on the carbon dioxide concentration... 34/

in a room works the same way as counting at the door. This spreadsheet uses a mass balance approach to estimate the air change rate in a single room building (like a portable classroom). (The fancy math is in the last tab). 35/ poppendieck.com/IAQ/images/Air…

After solving the mass balance, we can graph a function of the CO2 (natural log of (difference of inside and outside concentration/difference of initial concentration and outside concentration) with time. The slope is the air change rate. 36/

For THIS day the air change rate with some windows OPEN was 1/h (the accuracy of this method is only to one digit). For later in the same day with the windows CLOSED (after 6:30 pm) the air change rate was 0.3/h! This is how this room was normally operated pre-pandemic. 37/

Note that the maximum CO2 concentration for this day was ~1200 ppm and there were only 20 kids and 3 adults in the room. The higher maximum concentration (1200 ppm) correlates with lower air change rates (1/h). 38/

If this analysis was done on other days, with the windows open different fractions and different weather the air change rate could be higher or lower. Air change rates can easily change be a factor of two higher or lower based on weather alone. 39/

WARNING: This mass balance approach is only valid for portable classroom when no other rooms are connected. For classrooms that are attached (schools) a much more detailed analysis must be conducted to determine the air change rate that is beyond this thread. 40/

So what have we learned about this classroom? We can combine the effective air change rates from the portable air filters and ventilation calculations. 41/

Note this does not include any effective particle removal from HVAC system with MERV 11/13 filters (insignificant for this room).

Filtration. The portable filters supply about 2-4 /h effective air change depending on the filter setting. 42/

Ventilation (CO2 WITH kids in the classroom): For THIS room keeping the maximum CO2 level below 890 ppm and the filters on maximum this is likely have another 3.5 /h air change. This is dependent upon the kids activities in the room. 43/

We can add the ventilation to the filtration to estimate the overall effective air change rate. For this classroom if the maximum CO2 concentration is around 890 ppm, then the room is likely getting 5-8/h effective air changes. 44/

We can also use our second ventilation estimate (CO2 WITHOUT kids in the classroom): (a single snapshot in time with windows, weather) the air change was 1 /h. So on this day (max CO2 of 1200 ppm with fewer kids, so windows were likely not... 45/

open enough) the total effective air change was 3-5 /h (1/h+2-4/h from filtration). BUT since that day the teachers have learned to use the CO2 monitor data to open windows sufficient amounts. 46/

If the HEPAs are run on the highest setting (4/h) and the teachers try to keep the CO2 concentration below 655 ppm (likely an exhaust fan is needed), then the room may operate at an effective air change rate of 11/h! 47/

So take home points #1:
1) There is a lot of variability in CO2 analysis (assumptions, weather, etc..).
2) Natural ventilation variation will have a big impact. 3) On average the effective air change rate in this classroom is a reasonable effort at risk reduction. 48/

Take home points #2:
4) A high CO2 concentration does not necessarily indicate that a virus is present.
5) A low CO2 concentration does not necessarily mean a location is safe, just safer than higher values.
/end

Thread seems broken. Here is the rest.

https://twitter.com/Poppendieck/status/1432184373495967751