I went to a larger university, and was in the math department for most of my coursework. My math homework was graded by fellow undergraduate students who received a rubric and selected 4 problems per homework set, out of about 8ish (most classes. A few tough classes only assigned 2-3 proofs or so a week) but the TAs would grade the exams and the professors wouldn’t grade anything. Homework was at most 15-20% of one’s grade, exams being the rest. But most of my professors didn’t care about the homework as it was a tool for learning the material that was on the exams.
Those fans will not have sufficient static pressure to be useful. These makeshift purifiers may be great in a pinch (wildfire season and stores are all out), but far from the best.
This is indeed the "conventional wisdom," but if you do the math it turns out this is incorrect.
The trick is to use high-airflow MERV 13 filters (ideally 3M 1900, which has high pleat count and low resistance) instead of high-resistance HEPA filters, so axial PC fans can generate enough pressure.
You want at least MERV 15 so that virus sized particles get filtered in meaningful amounts, but yes the guy above you is wrong and choking on dirty air
Merv-13 means it filters at least 75% of the particles passing through it.
For a standalone air filter which is pushing the filtered air into the same room its drawing from, we can assume reasonably good mixing of the intake and exhaust air. Thus to be equivalent to a HEPA-filter it only needs 1/3rd more airflow... Which really isn't much.
This doesn't follow for a a fancier setup where you're taking dirty air and blowing it into a cleanroom or something... but that's not how we're typically using household air filters.
They only measure down to 0.3 microns in size, and MERV-13 only does 75%. A proper HEPA filter does 99.97% at that size.
And that really does make a huge difference to someone who has allergies.
IQAir does HyperHEPA, and they go down to 0.003 microns (so, a hundred times smaller, down to the size of viruses), and they capture 99.5% at that size. That’s why they’re used in hospitals.
Is it safe to assume all particles are equally as likely to be filtered, or are there particles that HEPA will filter that a MERV-13 will never filter no matter the number of cycles?
75% refers to the Most Penetrating Particle Size (MPPS), so they intentionally measure at the worst-case size so you don't have that problem. All other sizes are filtered at a higher percent.
The linked PC-fan purifier has been tested and shown to purify air faster than the conventional alternatives, and are not makeshift, so I'm not sure what you are referring to.
From the specs it looks like they are probably closer to higher-static-pressure server fans, the CADR is a few hundred CFM and it pulls 30+ watts. And as with many of these designs, you can compensate for low pressure performance by increasing filter surface area... the greater the surface area the lower the pressure drop. Of course, the problem with this is you are just spending more money on filters, which is the expensive part.
The filter cost for the AirFanta 3 doesn't seem particularly high? In fact it seems cheaper than competitors.
All the purifier companies realized they make more money selling overpriced filters, so $60+ for the OEM filters is common nowadays. Despite it just being a standard HEPA filter cut to a specific size.
Airfanta 3 Pro filters (four) - $57
Coway Mighty Filter - $60
Coway Mighty2 Filter - $70-$80
I'd be concerned about the availability of filters after a few years. Some of these companies have been changing their models and filter type every year to add some nebulous feature. Others just go out of business.
Yeah, the retail prices of filters is all over the place because of the popularity of razor/blade model of pricing. I'm not talking about that, I'm talking about the BOM itself -- this device is using 4x the filter media of comparable filters, which would typically have one filter that size.
HEPA media itself is a commodity and aftermarket filters (and the BOM cost to the OEM) tend to level out at whatever rate is minimally profitable per area of filter media.
The upfront cost of this device is high for what you get, because you're mostly buying filter media. The device itself is not much of a polished product. I'm sure it works fine but I wouldn't want one in my living room.
But since you bought mostly media, you didn't get a housing that hides the media, and you get to look at any collected dirt for that longer interval as well...
I'm sure it works but like any engineering decision it has pros and cons.
If that's your biggest worry in life, seems like a high-class problem! Many people are satisfied 'just' protecting their heart and lung health.
However, if you really don't want to look at it, you can always flip most DIY designs inside-out so the fans blow in and dust collects on the inside instead of the outside! This doesn't clean the room quite as efficiently, but if the appearance of dust is a major concern that should be a fix.
Yeah it’s functionally fine but I would opt to buy a more polished product for the above reasons. Probably would be great for a shop or similar, which is where I’ve done DIY purifiers myself. My personal favorite for cutting down on dust during home projects is a 20x20 furnace filter slapped on the intake side of a box fan. It requires zero extra parts or assembly.
Flipping the fans isn’t great because then you have to clean the fans, this is why draw-through is the standard configuration. Also many designs like this rely on negative pressure to seal properly or even stay together. You might just have the lid take off like a helicopter.
Sure, if you want to make a comparison where the price tags aren't the same order of magnitude, then a recent server is obviously going to be powerful. But since the baseline of this comparison is a laptop and several Thunderbolt SSDs, the kind of servers or workstations with 1.5–2TB of RAM that you can reasonably compare against would have to be the really old ones, barely new enough to support that much total RAM.
And despite the theoretically high memory bandwidth of recent EPYC CPUs, approximately nobody who can afford one is doing LLM inference on them.
Claude will do this for you after a few times. But yes, I have a skill called plan-to-epic which creates a Jira epic and ticket per milestone. It helps my agents persist context and, because I’m terrible at competing with my coworkers for “visibility,” means I can point to all my work if asked.
reply