Building (And Testing) A DIY Air Purifier

Whether it’s the usual pollution of the city, or the fact that your corner of the globe happens to be on fire currently, poor air quality is a part of daily life for many people. One way of combating this issue is with a high quality HEPA filter in your home, but unfortunately that’s not something that everyone can afford to even has access to.

Which is why [Adam Kelly] decided to design this DIY HEPA air purifier that can be built for less than $100. That might still sound like a lot of money, but compared to the $500 sticker price he was seeing for the models recommended by health officials, it’s certainly a step in the right direction. Of course, it’s only a deal if it actually works, so a big part of the project has also been verifying the design’s ability to filter particles out of the air in a timely manner.

To build his purifier, [Adam] found a HEPA H13 rated replacement filter that was cheap and readily available, and started designing a low-cost way to pulling air through it. He eventually went with a 120 mm computer case fan coupled with a step-up converter that can produce 12 V from a standard USB port. Then he just needed to design a 3D printed “lid” which would position the fan so it draws air through the center of the filter.

In terms of testing, [Adam] wasn’t worried about the purifier’s ability to actually filter out smoke particles; unless the manufacturer lied about the capabilities of the filter itself, that part is a given. But he was curious about how effective the fan would be in terms of circulating air through a room.

By installing a pitot tube from one of his drones into the lid of the purifier, he determined the airflow in the center of the filter to be approximately 160 CFM. By his calculations, that means it should be able to circulate all the air in his 25 cubic meter office around 10 times per hour. That’s a promising start, but [Adam] says he’d still be interested in a more detailed analysis of the design’s performance by anyone who might have the equipment to do so.

As he lives in Australia, this project is more than just a passing fancy for [Adam]. He only has to look out the window to see that the air he’s breathing is filled with smoke from the raging bushfires. They say that necessity is the mother of invention, and breathable air is pretty high up on the list of human necessities. Our hat’s off to anyone who sees their fellow citizens suffering and tries to use their skills to come up with a solution.

64 thoughts on “Building (And Testing) A DIY Air Purifier

    1. This is actually the most health beneficial noncommercial I have ever seen. It out performs many expensive manufactured units vastly. And at a small fraction if the cost.

      Most things sold as “air filters” are only filtering out particulate mater 10 micrometer PM10 or larger. This is the size of dust and it really isn’t that bad for your health.

      Most things sold as “air purifiers” filters down to standard HEPA which is 99.97% of particulate matter at 2.5 um or PM2.5

      This is at the top end of partical size that is hazardous to your health. It will capture most of the particles that directly cause respiratory problems. It will also capture some of the larger invasive toxins that go through the lungs and into the blood system however most invasive toxins are smaller than PM2.5 and its the smaller ones that are most harmful.

      This project uses a filter HEPA H13 that filters out 99.95% of PM0.3

      1. I’m not to savvy with terminology but on the health benefits I agree my gmoms has copd and my mother had breast cancer and it wasn’t the cancer that got her messed up it was the treatment and do too the treatments she has a low Immune system. I was looking into plants to filter but if this works plus the plants I can clear and clean the air for both of them

        I hope this catch people eyes that has value in this field of work and can contribute to the project

  1. I wonder if it would be possible to make an air purifier that uses water as the filtering medium.
    Something like it bubbles the air through some water and the small particles get trapped in the water while letting the cleaned air go out.
    You’ll know the “filter” needs replacing when the water is all brown and it would be as simple as dumping out the water and filling it back up from the tap.
    Somehow I think it might need to create very small bubbles or somehow diffuse it well in the water for this to work.
    Also, first comment :P

    1. Sounds like a rainbow (late 90’s door-to-door) vacuum. It was essentially a shop vac with a water tub. The hose intake was routed under the water, the vacuum created above it. So much cat hair in water….

    2. if it made very small bubbles in the same way those fish tank oxygenators work, I’d expect the small particles would get stuck in the thing used to create small bubbles. I love the idea of bubbling the air through water as I expect it would be a very effective filtration method, but I imagine you wouldn’t get nearly as much airflow as slapping a big HEPA filter on a fan.

      1. Proper design of the nozzle into the water, plus secondary structures to break up the rising smaller bubbles could do that with reduced risk of clogging. A final exit filter could catch anything that got by.

    3. I’ve wondered about that same thing. I thought I remember hearing about something similar – it used disks that were half submerged in water, and they would rotate. In theory, the disks would always be wet, and the particles that would stick to them as air was pumped past would get washed off into the reservoir of water as they turned.

      1. Venta air washer it‘s called.
        I use it almost every day. It rises the air humidity too. Very good in cold seasons. (real) Humidity stays at about 45%. High airflow rate, makes some noise though.

      1. The author is in Australia. Huge swathes of the country haven’t seen rain in a very long time. Record-breaking drought (which has contributed to the record-breaking fire season)

      1. I’m not planning on leaving water in my filter for any great length of time (probably dispose of it every 24 hours) but what household chemicals are available to kill Legionella bacteria? What about UV light?

    4. There are indoor laundry vents that do this. They blow the dryer’s exhaust through a box with water in it, and some portion of the dryer lint precipitates out of the air-stream into the water.
      You change the water every few loads.

    5. The downside is that you are making an effective air moisturer. The purified air that comes out from the water is 100%RH. Also water purification is not that good what i have understood. For example they say that sischa is filled with coal dust even after the water scrubber.

    6. I’ve considered the same thing. Some thoughts, which could be incorrect:

      You’ll need to use a high static pressure fan to ensure the air isn’t ‘stalling’.
      The water filtration should use three stages: one with a full-sized input tube that won’t get plugged with larger debris, one with much smaller tubes that won’t get plugged with smaller debris, and one with air stones to create the smallest possible bubbles.
      Significant amounts of chlorine and/or UV-C lamps could be added to disinfect the water. (This–in theory–should necessitate having the air leaving the system flow through activated charcoal to absorb any gaseous chlorine or ozone.
      If the last column of water contains a super-saturated solution of calcium chloride (i.e. ‘road salt’), it will act as a liquid desiccant, removing the humidity added by the system
      The columns should be as tall as possible, but the width doesn’t matter much. Not only does this mean that you can use inexpensive PVC/etc. pipe, but they won’t take up much floor space. In theory, the whole system could be contained in a single series of alternating straight sections and doubled elbows.

      Does that make sense?

  2. My brother who lives in Canberra which was severely affected by the smoke from the bushfires made a very similar filter but used a much mire substantial centrifugal duct fan. He also was was experimenting with blowing air into a body of water (the bath).
    He claims it did help – he was trying to clean the air for the whole house so required a lot more air flow than a computer fan could muster

    1. should work even better when using a big fan with a filter that pulls air from outside to create a positive pressure zone in the building. That way he can be sure that there is no unwanted draft pulling dust and smoke into the building. Downside would be the high consumption of filters

  3. Not sure I’d rely upon the test method used for various reasons. Probably better to use a volume of air over time. Not too hard; fill a large trash bag with air and clamp it around the top of the filter so the whole element is inside. Measure the bag and do a bit of math to get the volume, make sure to subtract the volume of the filter. Turn it on and time how long it takes the bag to deflate. Still not a perfect test but a lot closer than measuring the velocity of air coming from a fan that would move air forward even if the back was against a solid plate, and the error in the bag test adds to a factor of safety instead of subtracting from it. Better tests could certainly be done with better equipment of course, this test just tells you “it’s at least this good” instead of “it could be as much as this good”.

    1. +1 – I love the idea, if it works, but I’d also have serious concerns about the numbers claimed here and it actually working… Even the premium $42 case fan option listed here only specs ~110 CFM – and while I could be wrong, I’d bet that is unrestricted airflow. So 160cfm claimed in article through filter seems unlikely. HEPA are usually pretty restrictive filters, that’s why you don’t usually see them dropped in as furnace filters, due to even furnace blowers not being able to cope. Seems unlikely <=12Watts is going to get much airflow through a HEPA, especially with this style of fan. I would love to be proven wrong though, and might try one out myself if so.

    1. Mainly I think because 12V fans are everywhere and can be found in any computer or any dc powered device, really, whereas whenever I need a 5V fan, its always a tiny 30-40mm one and I wind up having to order it specifically.

    2. Most of the commerical hepas I have are rated at a few changes per hour. The larger ones are about $2k and are more effective at the medium or lower medium speeds. We bought them when we had a bad interface fire here a few years back and the trick was to be on high for 15-20 minutes then back down to the better settings. I also wondered why a couple of muffin fans and a $3 hepa furnace filter wouldn’t have worked. Glad to see someone trying it.

    3. The larger the fan, the higher the voltage to drive it (generally speaking), or you’re gonna lose too much power heating the device (Ohms law). It’s also why electricity distribution is done at high voltages: lower loss to resistive heating.

  4. A foam wrap-around prefilter would be a good addition – easy to clean and catches much of the dust that clogs the main filter.

    The PMS5003 particle detector is cheap and does a good job for testing this kind of filter. It’s easy to work with and there are several arduino/esp projects out there.

    Well done project.

        1. Hey Guys, I do not understand the technology fully but would be a prefilter using cyclonic separation also be applicable?
          Then you just would require a main filter and empty out the prefilter from time to time..

  5. 1 micron filters are pretty common and inexpensive for water purification. It makes me wonder how well they would fair purifying air perhaps with more than one in series. Maybe some sound absorbing finger foam to keep it quiet.

  6. An easy modification that I like to do with these air purifiers is drop in activated charcoal pellets in the air filter. Its adds a small amount of air restriction but its cheap and easy.

  7. Axial fans are great at moving airflow and you have a wide selection to choose from. I think there is more room for improvement if you went with a smaller and much cheaper air filter and put that money into a centrifugal fan setup. Centrifugal fans are better optimized for higher pressure differentials and are usually quieter. There is a smaller selection but hey, brain power is cheaper than anything. IDK about the efficiencies between the BLDC axial fans and the more commonly dc motor centrifugal fans, I believe there is not much difference for noticeable gains unless someone else can chime in. What you guys think?

  8. A number of years ago while dumpster diving, I scored 4 2 foot by 4 foot HEPA filters (3 were still(are) sealed in plastic-the other was unused).
    I was thinking about using them to make a dust filter for my woodshop.

  9. Built a bigger one of these for getting soot out of the air faster in an aircraft test lab. Had a 7.5kW fan, drawing air through a pre-filter and then HEPA filters. If I remember correctly it flowed the room’s volume every 18 seconds. Due to the very high air speed, it was wickedly turbulent in there, giving excellent mixing of the cleaned air with the remaining and whisking up all the air in the room. Thus, we hypothesised that the half life of soot in the air to be 18 seconds. With 10 half lives giving a 1024 reduction, we would expect it to look pretty clear after 3 minutes.

    This wasn’t quite the case. Certainly a big improvement, but still a little smokey after 3 minutes. So we upped the wait time to 5 minutes which was just enough to have a cup of tea. Quite smoke free by then.

    Note the room also had a permanent extract but that didn’t do a very good job all on its own – but did keep it negative pressure.

  10. “By installing a pitot tube from one of his drones into the lid of the purifier, he determined the airflow in the center of the filter to be approximately 160 CFM.” — i think that would be the airflow circling around inside the case.
    By my experience, such filters require a lot of pressure to push air through and I doubt that 12 cm computer fan can do it.
    If he has access to a differential pressure sensor (probably what is used anyway) i would estimate the flow using the difference in pressure between outside and inside.

    1. Yeah, no way this does anything near 160cfm. Pitot tubes measures static pressure differential, not air flow, so we would absolutely expect a higher pressure inside the filter (totally unrelated to the airflow). If you stick the tip of the pitot tube in a sealed balloon it will think there is a high air flow rate, but obviously the air is not moving. I just built a filter like this with a 12v pc fan and it hardly does anything and I’ve also worked with pitot tubes in wind tunnels.

  11. There is no way that tiny fan makes enough airflow to service a room properly without sounding like an airplane. If you look at the kinds of fans used in even cheap off the shelf purifiers it will be obvious. It’s a cool solution, but a fan that size would have to be very loud to move that much air.
    I’ve looked into doing this exact thing, but for my house (not my lab) and by the time I factored in making a decent case for it, adding some sensors, driving the whole thing, I just got a commercial unit, it was just 30% more expensive for a way nicer looking product.
    Still, you could do it in a pinch, all a purifier does is move air through a filter, all the other stuff is just fluff.

  12. This got lots of people thinking about projects, and that’s good in itself.
    The expensive part of running a HEPA filter, however, isn’t the appliance, it’s the cost of the replacement filters.

  13. How about the through water idea. No expensive filters or if you like then push the scrubbed air through a final hepa filter which should last a whole lot longer. Could start with a 90mm pvc pipe with end caps fitted. Drill this through extensively at one end and cover with kitchen plastic scourer pads so that they touch side to side. Hold these in position with stainless steel mesh aka hardware cloth. Use a small fish tank pump to circulate water over this pad assembly letting it trickle and drip into a sump below. Bore out the end cap opposite the filter pads to suit a fan or a number of serially connected fans to maintain air volume at low – quiet speeds. The idea here is to recycle the room air a number of times through the unit each time removing a percentage of suspended solids. Even if the filtration removes only 10% of particulates each time the room air will over time get cleaner. Maybe a timer could be use to dump the water every hour or day or use some way of shining a led through the sump and dumping it once it gets dirty enough to obstruct the ldr opposite the led. That or just light up an indicator that the water needs changing.

  14. Filter through water? Guys, that’s a water bong, and what you inhale still makes you cough and gets you high. Through the water cools it, adds humidity, but does not come anywhere near filtering it.

    Like others have said, this designer has a good concept but should switch out to a centrifugal fan which creates more static pressure for a given energy input.

Leave a Reply

Please be kind and respectful to help make the comments section excellent. (Comment Policy)

This site uses Akismet to reduce spam. Learn how your comment data is processed.