This blog is not intended as advice. It is for educational and research purposes only. Read this disclaimer first if you want to read this post.
Good Industrial Ventilation practice is to have local exhaust ventilation at the source of the toxic emission. For a toilet cubicle, the accepted extraction is a 10cm extraction grille on the ceiling. For Industrial Ventilation experts and Occupational Hygienists this is known as Dilution Ventilation and is totally the wrong approach for controlling COVID-19 bioaersols generated during defecation.
Good control of faecal bioaerosols in public toilets not only reduced the chance of inhalation exposure, but also reduces surface contamination because the cloud of faecal aerosols will drift and coat walls and other objects.
The importance of faecal aerosols in the spread of COVID-19 is unknown, but the paper in Emerging Microbes and Infections Vol 9 2020 (free download) by Zhang el al., “Molecular and serological investigation of 2019-nCoV infected patients: implication of multiple shedding routes” indicates that faecal contamination may emerge before oral swabs test positive and last after they return negative. This alone is of concern and may indicate that oral swabs, while more sociable, will miss infected people.
With diarrhea, a stated symptom of COVID-19 and very common with travelers, the faecel aerosol is a lot finer and the amount massive.
This discussion is aimed at public toilets at airports and train stations, where the toilets are almost always inside and get no natural dilution ventilation.
There are systems available commercially in Australia to extract odour from toilet bowls, but I suspect their fans are too small to extract the COVID-19 bioaerosol effectively.
- LooFloo http://www.ventilatedtoilet.com/
- PureVentilation https://www.pureventilation.com.au/product/in-line-fans/expella-odour-control-kit-12v-fan-with-motion-sensor/
My measurements are mainly SI, to minimise the chance of error in my calculations. For a standard toilet cubicle 2.4 m high, 1.93 wide and 2.8 m long, the volume is 13 m3. It appears the standard mechanical ventilation rate is 10 air changes per hour (ACH) or 130 m3/hr. This is somewhat lower than that for laboratories (some 14 ACH), but higher than expected in a house by natural ventilation of 3 ACH.
Assuming there is enough suction, what would happen if this ventilation flow was connected to the toilet bowl, to give a much more effective “extraction at the source” using the existing water channels?
My loo is is oval shaped with an average internal radius at the seat of 0.157 m and at the water of 0.005 m. The distance from the water to the rim is 0.025 m. Treating the loo as a truncated cone, this gives a volume of the loo as 0.03 m3 and the inside area of the top (much the same as the toilet seat hole area) of 0.078 m2.
For reasonable capture of bioaeosols, the capture velocity below the seal would need to be 0.5 m/s. If the area of the seat hole/top of loo bowl is used, then a flow rate of 0.039 m3/s or 140 m3/hr. This would give an adequate velocity with the person sitting AND standing (see below for an alternate calculation for the person sitting on the seat. Squatting requires a standing up air flow.
This is very close to the dilution ventilation flow rate above. But could this flow rate be achieved through the existing water channels used to flush the loo?
In my loo there are some 24 x 6.7mm diameter holes and one slot 35 mm x 12 mm at the front of the loo. (Picture me with my head down the loo with a micrometer). This gives a total area of 0.00085 m2 and an air velocity of 45.8 m/s. At this velocity not only would the flow though the holes and slot be very noisy, but it would be unlikely that the needed 140 m3/s flow be achieved with the existing exhaust duct. The pressure drop would be far too high.
Another way to estimate the needed flow would be to use the gap between the seat and the porcelain. This would be appropriate for a person sitting and occluding most of the toilet seat hole, and is calculated assuming all the air moves under the seat. For my loo the gap between the seat and the porcelain is 9.5 mm and the circumference of the “slot” is 1m. I have noticed that this gap is much larger for some loos, but if needed the seat supports could be cut down. For a capture velocity of 0.5 m/s the flow under the seat is 17 m3/hr – a LOT lower
This would give an air velocity at the water channels of 5.6m/s – not too noisy and perhaps achievable using the existing extraction ventilation -perhaps with a booster fan.
Footnote
I’m yet to calculate the pressure drops to achieve the required flow, but I would expect the appropriate booster fan could be a couple of hundred watts. The sizing of the water channels is probably designed to give enough flow resistance to evenly distribute the flows over the holes and give a bit more for the front slot water channel.
In reverse, the sizing would probably give a fairly even flow around the bowl with a bit more at the front – which would be good, as unless the person was very fat, some of the downward air flow would be between their legs and not under the toilet seat.
Once the infected person leaves the toilet cubicle, they may wash their hands. However a surgical scrub will be most unlikely and if they use the air dryer, a further infectious bioaersol will be generated. These air driers should be disconnected in public toilets and replaced with paper towels.
