If you have space for a commercial lab hood, the cost for used/auction is low, if you’re slightly patient. (This is partly because of the bulk, of course. Lots of people might bid on the small, benchtop equipment, but the giant equipment often goes for the minimum bid.)

It's actually really common to let them exposed that long. Not when using them, but to make them: once the agar medium is melted and poured in the petri dishes, closing the lid during cooling causes condensation. Having the petri dish full of water when using it is more difficult and annoying, so it's better to let them cool down with the lid open (a space-efficient way to do so is to have a pyramid of plates where each lid rests on two plates and can support one)

Also, two contaminations out of 4 plates sounds really really bad, but then the blog doesn't say how the agar plates were prepared (how many plates without exposures were contaminated?) and how long they were incubated (sometimes something starts growing after a week or two, if you're culturing a fast-growing bacteria then it's mostly irrelevant).

* HEPA is the extreme edge of a much larger scale called MERV, which rates filtration performance.

* You'd be surprised how few contamination threat models _mandate_ HEPA. If you're doing petri dish work, contaminants you care about are probably in the 1 to 3 micron range or larger.

* This means you can get adequate results from lesser filters, which has massive implications for cost (because filters are cheaper) and ergonomics (because the same dollar buys a bigger filter, which is less cramped to work in).

* Laminar filters are only laminar under specific circumstances, which include both pressure and distance. Too much or too little force and the output will be turbulent (non-laminar). Fans need be adjusted to get this pressure right, across the entire face of the filter.

* As you get farther away from the filter, air falls back into too-low speeds and becomes turbulent again. The lids in the "clean" test for this post are farther away than I'd put them.

* As the filter traps more particles, it becomes harder to push air through, so optimal fan pressure varies for every filter across time.

* The pressures we're talking about here are very small. It's normal for correctly pressurized airflow to be undetectable to a bare hand. Easy to overshoot.

* In theory, you can test this with an anemometer. In practice, I've not found a cheap one that will so much as _turn the prop_ under such a light load.

* A cheap but inexact test of laminar flow is a vertically held lighter's flame being pushed to a flickerless ~45deg angle.

* When working in front of a hood, movement technique matters. If you put a dirty hand in between your hood and your dish, you're still blowing contaminants onto the workpiece, you're just doing it in a straight line.

* Or a turbulent line: obstacles like the end of the acrylic sheet can cause air to whip around the edge and become turbulent.

* Gordotek has a video that's probably of interest here[0], where he shows a cheap filter blowing zero PPM, as measured by a swanky lab particle meter that costs more than most people's rent. He received an enormous amount of heat from keyboard warriors that said this is BS and doesn't work. Notably, the number of haters who came with particle sensor readings of their own is also zero PPM.

* A laser PM2.5 sensor with 99%+ accuracy in the ranges we care about (1+ micron) can be had for under $50. This moves your measurement -> result interval from days to seconds, making it feasible to test more parts of the filter.

* Owning such a sensor, I've seen a duct fan taped to a trash bag taped to an unfinned MERV16 blow every bit as clean as a real-deal university lab HEPA filter in the 1+ micron range.

[0] https://www.youtube.com/watch?v=lInfdAVvBts

Being used mostly by medical (a small market with relatively rich companies) is the reason. It's a lot of testing to make proper one for not too massive market so the price is "eh, they can't be arsed to try to DIY that" basically