Building a computer controlled ventilation system with HEPA filter to improve our indoor air quality

Introduction - why do we need this ?
In general, the Netherlands offers a reasonably good quality of life. Residential areas are well looked after, noise pollution is greatly reduced by quiet asphalt on roads, air pollution is fairly well controlled in most of the country (but not everywhere: certainly not for those living near the steelworks on the west coast). But there are three things which regularly cause ludicrous amounts of air pollution in this country: fireworks on New Year's Eve, Easter fires, and the idea that people have that it's especially "gezellig" to have a fireplace or stove burning wood in their living rooms. Luckily, none of our neighbours are especially keen on barbecues so the summer air quality is usually OK.

The government advises people not to burn wood for heating when we have weather conditions which will cause the pollution to accumulate, but it seems that almost no-one who burns wood for heating ever takes any notice of this advice. At the time of writing, the "stookwijzer" website advises people in our area that burning wood will cause nuisance to your neighbours and air pollution. From six o'clock this evening they advise that people don't use their wood burners at all because the smoke produced will cause health problems, especially for those with lung problems. But I expect that no-one will take any notice of this and our outside air will be thick with smoke this evening.

What I find especially absurd is that it's still September and we've not even had any cold days yet. Outdoor temperatures have not yet gone under 8 C. We've not yet used any form of heating at all and the lowest indoor temperature that we've seen so far was 19.5 C, in the middle of the night. Why buy firewood and spend weeks in the garden chopping it up (as some of our neighbours do), and then set fire to it when you could insulate your home instead and it'll be warm without any effort being required ?

Air filters
Initially we used air filters quite successfully to improve indoor air quality. They're also good to remove viral particles which visitors might bring with them. Though it's popular to pretend that Covid is "over", it's actually still killing about 200 people each week in the Netherlands, and it's similarly deadly elsewhere.

We have two Philips 800 series air filters. These adjust their speed and run quite quietly unless the air is full of particles, so you can sleep in a room with one active. Other companies produce good products as well, of course. The light on the top is purple because I opened our windows to let in fresh air and the neighbours were burning wood again.

A couple of years ago I also made this super inexpensive air filter. It looks like a cardboard box with a vacuum cleaner HEPA filter on one side and an old computer fan on the other because that's exactly what it is. It doesn't have the same throughput as the Philips filter, but it does clear smoke from the air in a small room surprisingly quickly. The 12 V fan is run from a 9 V "wall wart" power supply, so it runs slowly and reasonably quietly.

A house ventilation system
Free-standing air filters are good at removing smoke from the air, but they can't do much if you open a window for ventilation. Without ventilation, the CO2 level in a room rises rapidly due to nothing other than the occupants breathing. So we needed something else.

For the last couple of years I've been working on improving our indoor air quality. As we no no longer use gas the air inlet in the upstairs (2nd floor - two floors above ground) boiler room was not required for the heating system any more so I now use it to bring fresh air into our home. It opens into the boiler room, allowing air to mix and warm up if the sun is shining, but the air needs filtering before it comes into the rest of our home.

A fan which pulls air through a HEPA filter in the boiler room is controlled by a small computer. The fan is switched on if the downstairs controller indicates that the CO2 level is high. The computer also measures the temperature in the boiler room as well as on the upstairs landing so it can cool or heat the house (slightly) by bringing in cooler or warmer air when that's possible. It also monitors the temperature of the solar power inverter and switches on an external cooling fan when it gets warm. I already had to repair that inverter and I hope that this will extend its lifespan. The computer also can switch on a small heater in the boiler room if the temperature in the room is too low. This has not yet happened.

The wall outside the boiler room: An Arduino mounted on the wall next to the inlet for the incoming air. The green LEDs indicate when the fan is active (there are two speeds). The yellow LED indicates that the cooling fan for the inverter is active. The red LED indicates that the heater has been switched on. I'll cover it up so that it's less ugly some time soon...

Inside the boiler room. No-one usually comes in here, but I still need to cover up the circuitry of course. The HEPA filter is the same as the type used in the Philips air filter. The fan is an inline "quietline" fan from Hornbach, which seems to be one of the quietest inline fans available. The three relays are used to switch the two speeds of the fan and the small electric heater. I recovered them from the old gas boiler.

The other half of the system is downstairs. There's a controller on the wall in the living room where our central heating thermostat used to be. This measures the CO2 concentration and switches on the extractor fan in the kitchen as well as the incoming air fan upstairs (to attempt to create a balanced flow of air). Communication between the two devices works via a serial link using the same cable as linked the old thermostat to the gas boiler.

The downstairs controller measures CO2 concentration and will automatically switch on ventilation if the level is too high. Buttons allow selection of continuous low fan mode, 'visitor mode' which attempt sot keep an even lower CO2 concentration, and off mode which turns off the downstairs and upstairs fans. The latter is intended to be used if we find that the system is somehow overloaded and is bringing in smoke. The controller is also linked to the cooker extractor in the kitchen so that it knows to switch off the house ventilator fan when that is operating, and even if it was not otherwise required it will switch on the upstairs fan to attempt to approximately balance with the cooker extractor. The downstairs controller also has a light sensor so that it can automatically switch to nighttime mode when we go to bed.

We exhaust air from the kitchen on the ground floor, using the same duct as the extractor fan over our oven. Only one fan can operate at a time. Having air travel slowly from upstairs down to the kitchen helps keep cooking smells and condensation out of the rest of our home even after we've stopped cooking

Downstairs controller. This is also Arduino based, using a Nano this time to make the whole thing more compact. Sensors measure CO2 concentration, temperature and humidity. They agree quite well with my Aranet 4. The buttons on the front allow selection of constant low fan mode, visitor mode and no fan mode. The green LED indicates low fan setting, yellow indicates high and red indicates that the fan is currently off because we have a low enough CO2 concentration. The Aranet 4 made me realise that the concentration of CO2 in a house rises quite rapidly just through breathing. We saw the reading rise quite quickly to over 2000 ppm on winter evenings with the windows closed, but that no longer happens due to the ventilation system.

In the kitchen I put a a t-junction in the ducting from the cooker hood on the left of this picture so that an inline fan under the wooden cover on the right can also send air into the duct. It feeds from another duct on the right side of the photo which reaches nearly to the floor next to the refrigerator so that we take air out from a relatively low level in the room where it is cooler. Non return valves are used to prevent one fan blowing air back out the other. It's a bit ugly and I still need to cover the silver tube up. In this location I used a 125 mm version of the same inline fan as upstairs.

How well does it work ?
We now have fresh smelling air indoors even when the neighbours are doing their best to pollute. That was the intention and it's worked out well.

Though the design of the system as a whole takes advantage of latent heat in the boiler room, the upstairs landing and even of heat leaked out by our water heater, it makes no attempt at all to recover heat from the air which we extract from downstairs. This probably results in a slightly higher heating bill than we would have in a completely closed house, but the air quality difference is enormous. In any case, our heating bill is absolutely tiny so there isn't much to save. I may end up building a heat exchanger downstairs as that seems like a fun thing to try to do, but I doubt we'll be able to measure much of an effect.

We went through last winter with an incomplete prototype of the system. I initially had an old computer fan for the air inlet upstairs, but it wasn't adequate to keep up with the downstairs fan. It also initially just switched on and off with a timer and there was no attempt to even try to balance flow with the downstairs fan. A roughly balanced flow of air does seem to create a better result overall.

Something that surprised me is how poorly CO2 travels between rooms. With no open windows but the door between our bedroom and the hallway open, the CO2 level can reach over 1300 ppm in our bedroom despite the hallway being around 700 ppm. I was surprised that diffusion didn't do a better job of lowering the CO2 concentration in our bedroom. Running one of the Philips air filters in the room moves air around enough to lower the concentration by about 200 ppm. But to achieve a really low CO2 concentration around our heads I think we would need a fan blowing air through the doorway.

Of course in the past no-one had a CO2 meter so we could never measure these things. When we first had the Aranet 4 but we didn't have any kind of ventilation system we'd see the CO2 ppm race up and beyond 2000 ppm in the evening in our living room just from our breath. It also headed in the same direction in the bedroom overnight. I suspect that we've unknowingly slept with very high levels of CO2 in past years without realising it and I think the same is true for most people. CO2 rises very quickly in any room with closed windows and doors, and using candles also rapidly increases the CO2 concentration.

I think a ventilation system has much to recommend it. It's a step beyond what can be achieved by air filters on their own. This system works. If there's interest I'll put the Arduino code somewhere that it can be downloaded.