Ceiling Air Purifier
jefftk.com
jefftk.com
1. https://www.predictiveservices.nifc.gov/outlooks/monthly_sea...
You're trying to reduce the viral load in the air, not build a BSL4 lab.
[1] https://en.wikipedia.org/wiki/Corsi%E2%80%93Rosenthal_Box
I never understood the purpose of the bottom filter in the Corsi-Rosenthal Box, so it felt like a waste of a filter, but I'm a big fan of the Comparetto Cube (no put intended). And 4 packs of filters are readily available -- 20x20x2 is suggested for the Comparetto Cube.
[edit] Also, thank you. I couldn't recall the Corsi-Rosenthal name/link, but thought of the wiki immediately.
[1] https://www.thisoldhouse.com/green-home/22231148/diy-air-fil...
It's for when the box fan is not sitting on the floor - either raised up on feet (which many have), sitting in a window, or hung somewhere. That said, you're right - 4 packs are easier to come by and probably work nearly as well in the real world.
Its amazing to me that these designs are less than two years old when the tech has been available for decades.
One of the biggest problems with the C-R and comparetto is that they are large and hideous to look at.
I have been working on a design that makes the C-R or comparetto or a smaller version than the 20" something that can be attractive enough to hang in the corner of a room as both an air purifier and room lighting.
In theory, if you mounted it fan-down and placed some towel or blankets underneath, you could also dampen a fair amount of the noise coming from the fan.
It's probably more effective and cheaper to get thicker filters. A 20x20x5 filter has five times the filter surface area of a 20x20x1 filter, but costs $36 - about 2-3x as much as a 20x20x1. Two 20x20x5 filters would provide twice the filter surface area.
But...these solutions were all intended mostly for emergency situations where purpose-built air filtration units were in really constrained supply. Folks should really just buy a regular air filtration unit that uses much less electricity and is quieter, especially if it auto-adjusts speed depending upon need.
This post is about how one of these made with a ceiling fan can be much quieter than a regular air filtration unit: "Testing my prototype, it has a CADR of ~180 CFM and is only 33dB. By contrast, the Wirecutter's top-recommended air purifier has a CADR of 233 CFM at 54 dB or 110 CFM at 36 dB. With some tweaks it should be able to match the commercial purifier's performance, without being louder."
> especially if it auto-adjusts speed depending upon need
I see how that works for wildfire smoke, but how would it work for covid?
Vornado used to sell an air filter that took two 10x20” house filters. Before I discovered that model I was trying to build my own using computer case fans (my intent was to hide the loud box under furniture).
Home Depot used to carry those in the 1500 classification which was great, but those are getting harder to come by too, and even the regional chain that had them doesn’t as much anymore and I have to use the 1200 or once in a while the super basic ones. Or buy them from Amazon and fuck Amazon.
Air filters became a rent seeking gambit before the term rent seeking was even invented. We should just be able to use furnace filters. And speaking of furnace filters, if you have central air, do yourself a favor and change the filter religiously. Also helps slow down accumulation of dust in your ductwork, which makes it hard to ever get the house clean, though you can hire someone to come out and vacuum them out. They have something like a softer version of a chimney sweep’s brush with a longer handle. Last place we had them do that on like day 2 to avoid any problems from the previous owners.
Do you have data on this? Adding a fan shroud to the Corsi-R supposedly does not increase energy usage or motor temperature:
https://www.texairfilters.com/how-to-improve-the-efficiency-...
I'm looking for data on the power usage and motor temperature difference from a 20" Hurricane in normal operation, with a 20" merv 14 on the back, a C-R and a Comparetto.
He has an addendum at the bottom where they tested the fans and saw no problems with a single filter. I... don't understand how they come to those conclusions based on the charts in the paper. All of the fans were >10ºC hotter and one model went up by 20. 15C is a lot for cheap, planned obsolescence consumer electronics.
Consider that they are designed for cooling in a window application, which means they must be able to survive operation 35C in direct sunlight. As an air filter, mine operates at 22C in shade. That easily makes up for the temperature differential due to load conditions.
https://www.epa.gov/indoor-air-quality-iaq/should-you-have-a...
They almost always are designed for higher voltages (i.e. 20+VDC) so when combined with a common 12VDC power source it's already been stepped down.
(source: used to work for a photocopier company, and when people bought newer units, we would pickup the old units and usually junk them.)
They're typically 120V single-speed devices, although some newer ones may have variable-speed fans. Not 100% sure how most of them vary the speed; my bet is that they just have some capacitors switched inline with the fan motor to add reactance to the circuit and slow it down. (That would probably be how I would slow down a single-speed one to make it quieter, at any rate.)
There are a lot of projects online that you can find, showing how to build fairly big workshop air cleaners with a furnace blower and some cheap plywood/MDF/whatever. Basically you just build a 5-sided box around the blower with a hole for the outgoing air, and then mount a furnace filter on the open face of the box.
And, of course, it's not really practical in that most people really don't want to be sticking stuff on their ceiling, but it could be an interesting experiment with more details provided.
OP will also want to track performance over time. How long does it take the filters to load.
This is basically the same question as "how long does a filter cube last". See https://www.texairfilters.com/testing-the-efficiency-of-merv... where they found performance was still good at 6m, though decreasing at 10m.
I believe that it is quite possible the majority of the air is not going through the filters. It seems that much of the air might spill out down the insides of the filters, rather than being forced through. In that case, the graph could be equally explained by the fan causing outside air to be sucked in.
Thinking about 3D printing, you could probably print a nice corner bit, that would hold the corner of 3 filters together and square to each other. That would make the whole assembly easier to put together and seal up with tape.
- The fan is not designed for high static pressure. It won't do well pulling air through filters.
- It's ugly. It will get uglier as filters get dirty.
- It's expensive to replace filters.
- Etc.
But it's not a finished, industrial project. It's a prototype. Those sorts of issues are universal for prototypes. They're easy to address:
- Switch to a rotor design with higher static pressure
- Provide prefilters. Ideally, there's a nice cover, followed by a washable pre-filter, followed by a carbon filter, followed by the fancy high-filtration filter.
With an appropriate frame, it could be beautiful -- nicer than a normal fan. The filters could actually reduce noise. The frame could guide air (much as in a Vornado), increasing airflow.
Good job OP. It's a clever idea and a good design. It's a proof-of-concept implementation, but that's the point of a proof-of-concept.
(Unless you like the DIY aesthetic, which some do)
A couple problems. One, space - this thing was bigger than a dorm fridge. Two, availability - no way you were finding five filters during a wildfire. Three, etiquette - if you found five filters that meant two other people weren’t getting any at all. Four, diminishing returns. Two filters reduce the pressure by half. Another only reduces it by a third, and so on.
Filters last a long time unopened: buy them in advance and store them until you need them, rotating stock.
Edit: To your point though, help out the distribution networks by buying yourself an extra filter or two now, and put it on the shopping list every time you use one.
Yes! Socially we need to make a distinction between buying extras of things after an emergency has started (harmful) and doing it in advance (helpful)
The ecobee has a “minimum fan time per hour” setting - it can also help to balance the air temperature.
You can't just swap filters, the 4" thick one usually go in right before HVAC air handler intake rather than at returns. I have had an AprilAire installed for over 10 years in my HVAC when it was replaced last time, recommended by HVAC installer, its has worked very well and I do have my blower fan cycle constantly regardless of need for heat/cool to turn over and filter air.
1. https://www.aprilaire.com/whole-house-products/air-purifiers...
1. If there is not enough air moving over the air-conditioner coils there is the possibility that they will ice over. This has a cascading effect where there is less transfer through the coils and more ice builds up until the whole thing is a block of ice. This most often happens though when people buy condensers that are too big for their system (the thinking being the bigger the better the more cooling, but it is a balance within some tolerance) but can also be done on the other side by reducing airflow too much.
2. Less airflow can also mean too much condensation in the air which can end up sitting in the ductwork and causing it to rust out. This happened to my parents when they installed extra filters in their HVAC system. I guess this is similar cause as #1, just different potential effects.
Yeah, it will also shorten the lifespan of the motor since it has to work harder.
But is the cleaner air trade-off worth it? I think so.
Plus, your "motor" isn't an engine, it's not variable speed, it only has one "hardness" that it works at. It might push less air with the thicker filter, but it's not going to ramp up and consume more wattage or anything like that.
This is incorrect - electric motors draw more current when under load. I would suggest researching this or playing with some toy electric motors run from a battery through a current meter.
Sure, some motors have characteristics that can be tweaked to run at different work loads either with PWM, or allowing them to pull more amps to drive higher loads. Hell, you can even overvolt them and make them actually work harder.
That's not how HVACs work though. Have you ever opened one up and did a repair? The circuitry is dead simple, there's no current limiting setup or PWM to control how much the motor is pulling or spinning, there's no CFM measuring device to give the motor more volts or a higher amp limit.
The motor is simply pushing air. Air isn't something like a solid load where a motor might lift something or move a lever or gear, it's fluid. The motor is going to run at max and be done.
Back pressure doesn't impact the load of a fan motor, at least not as the scale we see in HVACs, we're not making vacuum chambers.
Anyway at the end of the day I agree with the original characterization of wear and tear - motor and blower will probably outlast most other components. In the case of the one I mentioned this seemed to be true. You just may end up paying a little more for the electricity to run it.
Apologies for the tone!
> Have you ever opened one up and did a repair?
Well funny you should ask, but yes! Two times. One time, a power relay on the control module shorted and blew a hole in the circuit board. Had to replace the control module there. Another time, the start capacitor needed to be replaced on the giant 2KW squirrel cage motor. I removed the blower, disassembled it, lubricated moving parts, and of course tested it out on a stand. That much blowing power is quite impressive when right next to your face.
Which is a true statement but incomplete and therefore misleading. A fan motor is under the highest load when it is doing the most work. That is, moving the MOST air. If you restrict airflow (e.g. with a dirty filter), the motor is under less load and draws LESS current. A lot of people get this wrong because it's counter-intuitive on the surface of it.
You can test this very easily with a box fan and a kill-a-watt. Turn the fan on High in the center of the room and read the power. Now move it against the wall. You will hear the fan get louder because it is spinning faster, because it's doing LESS work. The meter will also show less power being drawn.
It is true that a dirty filter restricts airflow through the whole HVAC system. This extends the system's "on" cycle, which reduces the whole system's efficiency. But the popular claim that a dirty filter will burn out the motor is bunk.
The problem is (especially with older furnaces; modern ones have safety features to prevent this) that if you have too much back-pressure you don't get enough airflow over the heat exchanger (or air conditioner coils) and it can crack - allowing dangerous exhaust gasses into the airflow (or freezing the coils for the AC which isn't as bad).
Once the heat exchanger is gone the furnace is basically trash and has to be replaced (you can replace the heat exchanger but it's rarely worth it).
An everyday way in which people induce this scenario is by shutting the heating vents in particular rooms because they get too hot. Even on the newer furnaces, this results in the automatic controls shutting off the burner when the pressure and temperature gets too high, and then the fan starts pushing around cold air.
A secondary negative effect of this is that it pressurizes the ducts causes them to leak more, resulting in reduced efficiency, and also quicker failure.
A well designed system which has been configured to deliver the correct amount of heat to each room doesn't experience the same issues. Unfortunately, most older homes and even newer production built homes have poorly designed HVAC systems.
Heat pumps don't experience as many of these issues because they just don't get air as hot, and instead rely on higher throughput of lower temperature air to heat spaces, but that makes them far more reliant on good duct system design.
Many newer AC motors are variable speed or at least multi-speed.
Less air over the coils can inhibit the cooling properties, this was conveyed to me as high "head" compression pressure that can lead to premature compressor failure in the outdoor AC part, the expensive part. This is more important as the higher the SEER level, generally newer the system, the more sensitive they are to this issue. HVAC tech sold me with high head pressure expect failure of compressor in a few years.
Also you are running as efficient since the AC has to work longer burning electricity.
Comes with full warranty from the manufacturer, so I don't think it's going to shorten the lifetype of the system.
I wouldn't put a regular flat MERV-13 in an existing HVAC system, but the pleated ones have so much more surface area that they seem safer.
A thicker filter and/or one with more pleats can filter better and be less restrictive at the same time.
The one thing it can do is cause your equipment to run longer (because of the lower efficiency) thus wearing things out quicker. But, most HVAC systems are not nearly as "fine tuned" to the size of your house as many might think. Most traditional HVAC systems basically have 2-3 speed settings for the fan depending on home size. There is a lot of play between those rough settings where a bit more filtration is not going to put it out of recommended ranges.
I have a very high Merv rating filter on mine plus a charcoal filter, but I change it out every 3 months or so.
Compared to the standard, cheap, barely there filters, the run time and energy usage difference is not enough to notice.
https://www.youtube.com/channel/UC3_AWXcf2K3l9ILVuQe-XwQ/vid...
Consider you need to create a fair pressure gradient between the two sides of the filter and it needs to be higher the more the filter clogs up. AFAIK radial fans are more efficient for that purpose. Downside is the noise.
If you are running the ceiling fan anyway just for the turbulence it creates on a hot day, it's probably fine even if just a small percentage of the airflow is pulled through the filters, your data certainly looks promising.
I had a new furnace/AC installed last year with a 4" filter. $50 CAD for a carbon filter and leaving just the fan on made a vast reduction in smoke smell during a wildfire or two from BC.
I suspect the ideal version is an adaptation of the US central HVAC model. i.e. Push clean air into the various rooms.
...but hard to retrofit.
I don't know if that's really true. Looking at the top 20 or so Air Purifiers on Amazon, almost none of them mention the CADR or flow in the title, and very few of the reviews mention it. Most talk about sound, "HEPA", "ozone-free" etc.
My guess would be that people care much more about physical appearance than maximum flow. I would personally not tape an octagon of dirty air filters to my ceiling because that would be look ridiculous and probably not filter as well as my beautiful, silent $200 Winix box.
One way to see this empirically is that pm2.5 levels don't go up again after turning off the fan
Does this affect the fans cooling performance? The air blows more straight down than diffuse I am assuming due to the shroud?
My main concern would be that the filters limit flow, and thus it might be more likely to pull air up, causing a tight vertical loop (air goes down, turns outwards, and then back up while still inside the filter ring).
I don't really have a solid mental model of how air flow works, though. Some experimenting with the shape (and some colored smoke, perhaps) could be interesting. What if the filters tilted inwards to be slightly narrower than the fan's reach? Or if they were shorter than the ceiling-to-fan distance and just inside? Or does it help to put them further out? (Seems unlikely, but would be a good thing to test.)
Wood for the frame, metal grid sheet to cover the filters & still allow good airflow
Contrast this with the furnace filter, which needs some other mechanism (often, a schedule) to determine when to change the filter since it's not readily visible on a daily basis.
So running the filters a long time before replacing, might be the cost-saving option.
Since the Corsi cube can be hung from the ceiling (it looks weird being a cube hanging up there) but it is ungainly. This is a much more elegant solution.
Aesthetics matter. Now if you could somehow make the blades the filters, that would be badass.
I for one would happily proclaim “my shitty looking box fan contraption that costs $100 outperforms your $700 dyson by a decent margin”
Why lick the boots of brand centric corporations. Maybe you’re one of those “lifestyle brand” people… I think that’s code for no originality or personality.
If you're ok with this contraption taped to your ceiling, while having guests over, more power to you. This of course would be totally fine if you lived in a state dealing with wild fire smoke I will say.
Is there some kind of tracking?
Is there any way to block that?
Founder is very supportive, good product, easy to build
The sensor will display PM2.5, CO2, Temperature and Humidity on the display and can optionally send the data to any server for data logging (e.g. the AirGradient platform or any other cloud backend).
Parts: Wemos D1 Mini USD 2.24 Wemos OLED display USD 2.47 Plantower PMS5003 PM Sensor USD 13.89 Senseair S8 CO2 Sensor USD 28.00 SHT30 or SHT31 Temperature and Humidity Sensor Module USD 2.55
I'm using a Temtop M2000 because it was the cheapest option I found with per-minute data export: https://temtopus.com/collections/temtop-co2-monitor/products...
Black light tape along the latticework, black light at the top of the room. Other lighting is all dim lighting in the lower half (warm lamps).
I dunno. It'll be hard to make this not look weird unless you do something crazy, like a drop ceiling so the fan and filters are inset. If the aesthetics are important, I'd put it on the home office (out of view from Zoom) rather than the living room.
With a little extra effort the filter could be swapped out by removing/lowering the vent intake.
i.e. there's no need to build a filter the size of the fan; you can cover "most" of the airflow by using something similar but extends 50-60% of blade length(pure guess) and the whole arrangement can be above the rotating plane (so easier build?)
—
In the summer of 2019, we purchased a Molekule Air (the flagship model) and tested it. We bought an Air Mini that fall and tested it in February 2020. At the time we tested the Molekule Air, the company claimed that its “scientifically-proven nanotechnology outperforms HEPA filters in every category of pollutant.”
Our tests proved otherwise. And by mid-2020, that language had been withdrawn, after many of the company’s claims were ruled against in a case before the National Advertising Division and upheld in a later appeal before the National Advertising Review Board. The Molekule Air turned in the worst performance on particulates of any purifier, of any size, of any price, that we have tested in the eight years that we have been producing this guide. The Air Mini outperformed it, but that’s not saying much: It still produced the second-worst performance we’ve ever seen.
Guide author Tim Heffernan asked Molekule CEO Dilip Goswami why the language was removed. He answered, “The point about ‘in all categories’ is that we see a device that outperforms across all of the categories. Right? So we’re not trying to say that individually, on any particular metric, we would be number one. Right? What we’re saying is, when you look across all the categories, we outperform HEPA. Right? And that’s what we’re attempting to convey with that. And so—it’s fair to say that we needed to re-examine some of the language to make sure that it’s saying what we’re intending to say.”
https://www.nytimes.com/wirecutter/reviews/best-air-purifier...
This is reflected as well in the testing on Molekule's own website: https://molekule.cdn.prismic.io/molekule/4ec92005-d806-4991-...
And, I will say that they have numerous tests by independent labs. They did do some testing with their own lab and with a lab affiliated with them, but this is far from all the testing they have done. So to say that all their tests were tainted or affiliated with them is simply not true. See: https://molekule.com/papers
Specifically about the Molekule - their claims seemed to be empirically false and the device performed worse than a standard HEPA filter (and they were resistant to allowing independent tests at all).
When the tests came out awful their responses were mostly bullshit. [1]
[0]: https://en.m.wikipedia.org/wiki/Brownian_motion
[1]: https://www.nytimes.com/wirecutter/reviews/best-air-purifier...
1) MERV14 will capture 75%-84% of particles in the 0.3-1μ range.
2) A filter which captures 70% of particles with an air flow of 100cfm will capture the same amount of virus as a filter which captures 100% of particles with an air flow of 70cfm. Both will clean the room just as fast. For air filters, lower filtration + higher airflow is usually a better design option. Going from 70% to 95% to 99.9% means you'll have a more expensive, power-hungry, and more noisy product over one which just has a little bit more air velocity [1].
3) The rating is for 0.3μ since that's the hardest size to capture. A filter will actually capture more particles below 0.3μ.
4) COVID19 virus particles are around 0.1-0.2μ, but that's beside the point. They're travelling on water molecules. Those are much bigger.
5) Even if there were a virus particle somehow floating around, a viral load of one virus is very unlikely to get you sick.
From an engineering standpoint, something around MERV14 is almost certainly the sweet spot for a COVID19 room air filter.
[1] High-filtrations makes sense in places like vacuums, face masks, and other places where the goal is to have clean air coming out. Vacuums shouldn't blow up dust. That's a different engineering design goal than a room air filter. If you'd like to see the impact of loading on a fan, put your hand behind one, and hear how much noise goes up. MERV14 has a much lower load than HEPA.
2) It will not clean the room just as fast, it will take longer to clean the room.
4) COVID-19 particles do travel on water molecules, but they are also airborne. The CDC has admitted this and there is a growing body of research proving this to be true as well.
5) This has not been proven.
> It will not clean the room just as fast, it will take longer to clean the room.
Here are two different models:
A. Air moves sequentially. First you filter all of the air once, then you filter all of it another time etc. In this model, a filter with 100% efficacy will get everything in a single pass, and the CFM determines how long that pass takes, while a filter with lower efficacy will never get it all, but will get pretty close after a few passes. In this model you want high filtration.
B. Air moves randomly. At each minute, the purifier selects air from the room at random, filters it, and spits it back out. In this model, a filter with 100% efficacy at 70 CFM is exactly equivalent to a filter with 70% efficacy at 100 CFM, and you will often want to trade off efficacy for flow.
I think real rooms are generally much closer to (B) than (A), though of course somewhere in the middle?
2) The percent of material removed by a filter per unit time is the product of (1) filter efficiency with (2) what percentage of a room's air passes per unit time.
4) "Airborne" is generally via microscopic droplets. The CDC's guidance changed from large droplet transmission (which is relatively short-distance and short-time) to airborne. This doesn't mean individual viruses are floating around without any H20.
5) No one credible believes 1 virus particle is likely to infect you, except by very bad luck. Most citations give claims in the 100-1000 particle range. Low initial infectious dose also /appears/ to correspond to less aggressive infections. This has not been rigorously proven (and it's hard to do), but has a strong theoretical basis:
- One virus particle is unlikely to make it past the mucous layer, unless you're super-unlucky.
- If it does, your innate immune system can usually handle minor infections before they escalate.
- If it can't, your adaptive immune system has more time to respond. You're looking at a few days before it kicks in. With a lower initial infectious dose, you'll still have that much less virus when it kicks in.
If you'd like to contradict any of this, please provide citations. I'll read them. I'm glad to be proven wrong. Perhaps I'll learn something.
The respiratory particles that transmit covid are a range of sizes, but likely mostly a bit larger than that: https://www.jefftk.com/p/how-big-are-covid-particles