I know everyone has been waiting eagerly for a technical report on 3rd gen cabin air filters (lol), but the wait is now over!

A year or so ago I was lucky enough to snag one of the canister-style cabin air filters from the Russia manufacturer. It's 3D printed over a cylindrical filter and pops into the air box above the fan. The canister fits down into the squirrel fan blades. It's a really nifty design and I've been happy to have it BUT... I have always been annoyed by the sound quality of the fan with the filter installed. There's a distinct high-pitched whine at all fan speeds, getting worse with higher speeds. The air flow is also reduced (which is to be expected). Also, the media is not washable or replaceable.

A fella on BookFace named Tommy recently made a run of flat-style 3D printed filters that feature being able to replace the media. I figured I'd grab one and compare it. He's a super nice guy and I've had fun chatting with him. He also shipped my filter promptly!

Taking cues from Leon @Spiker engineering I thought this would be a great opportunity to gather some real DATA to help inform the community's decision about cabin air filters.

Experimental setup:
- El cheapo aneometer from Amazon ($15)
- DecibelX iPhone app (free, with in-app purchases but I didn't use any)



Here's the Tommy filter







The two halves screw together (props to Tommy for printing some really clean threads). He ships it with a thick but porous foam filter.

I opted to also experiment with some Merv4 charcoal air filter media. This is the stuff you'd wrap around a big paper air filter element on a room air filter; it's basically a pre-filter that you can change more often and get the charcoal odor-absorbing functionality. It's pretty thin which should be good for flow which was important to me. I cut out a circle from the material.





Tommy filter with the Merv4 stuff installed. The two halves clamp it on the perimeter so it stays put even though it's thinner than the cavity in the filter holder.





And now, on to some data!

Quick refresher on the design of the Russia filter





Here's where the air filter plugs in.

Russia filter:



Tommy filter:



I took measurements while sitting in the driver's seat. Vents were set to the top vents only (all open), and I'm focusing the recirculation setting only. Phone was placed on the Spiker Engineering phone mount for audio data consistency. The anemometer was placed on one of the center vents slightly to the side to avoid the little handle in the middle.

I tested four configurations:
1) No filter installed (factory) as control
2) Russia filter
3) Tommy filter with supplied foam media
4) Tommy filter with Merv4 media

I took data for fresh and recirc at all fan speeds but to spare you the gory details here's the punchline with respect to flowrate:



All the filters are very close. In general they knock the flowrate down by about 1/2 of the fan setting. So with the fan at 2 it's like having the fan at 1 without a filter. The Russia filter flows the best but only marginally, and the Merv4 stuff is a teensy bit better than the foam.

Looking at simply total dB(A), which is the overall sound volume to your ear there is a clear correlation between flow rate and sound. This is not looking at any frequency analysis yet. What this means is that for the same flowrate having a filter doesn't make the fan any louder, it's simply a function of how fast the air is moving out of the vents. This is the data for all the configurations I tested.



Here's where it gets more interesting. The DecibelX app lets you see the real-time spectral analysis showing dB(A) vs. frequency. You want to look at the blue line which is the rolling average data. The orange line is the instantaneous and it is slightly higher because it's picking up the sound of me clicking the screenshot buttons LOL. All data is recirc at fan speed 4 (highest).

No filter installed



Tommy filter with foam



Tommy filter with Merv4



Russia filter



So what do we see here? First notice that without a filter the sound spectra is pretty much what you'd call brown or pink noise, since it is a very flat curve that tilts downward from low to high frequency. Pure white noise would be a flat, horizontal curve from low to high. Brown/pink noise prefers the lower frequency content in the spectra.

Both Tommy filter configs are also pretty similar. But the Russia filter has two pronounced peaks at just above 2kHz and around 6kHz. This is definitely the whine that I noticed which bugged the crap out of me. Both peaks are about 10 dB above the overall curve meaning they are distinct tones that you can hear against the background.

I noticed immediately that the Tommy filter has a much nicer sound. After driving around for a few hours my observation was the same. It's a big improvement over the Russia filter.

Summary (or TLR) --
- The Russia filter has a pronounced tonal noise which impacts the sound quality. The Tommy filter, especially with Merv4 filter element, is a much easier sound to take and more like the original sound without any filter
- The Tommy filter flows a tiny bit worse than the Russia filter but it is only marginal and not noticeable. Adding any filter element will reduce airflow by approximately 1/2 the fan setting.

Thanks for reading!

Further, after removing the Russia filter now only the second time two of the retaining tabs broke off



The Tommy filter is a tension fit instead of tabs, and comes out easily while fitting snugly. It will be easy to remove and replace the filter element to keep it clean and renew the charcoal odor absorption.

Reserved

Had no idea this design was a thing. I have the russia one and yea the high pitch noise can be annoying. Going to hit this guy up on facebook and get this one. Thanks for all the info!

Do you have a link to buy the "Tommy Filter"?

Wow nice work. Since the air flow rate is decreased with the cabin air filter installed, have you noticed a difference in cooling efficiency on hot days?

Sure, when the cabin is super hot it takes a bit longer to cool it down since it just can't pump as much cold air in. But after a few minutes it has more than enough power to keep it cool and not even at the highest setting.

I don't have any data to back up this claim though

Had a little fun with foam filter media.

I had some UNI 60ppi (fine) stuff sitting around so I tried it out:





Also got some UNI 30ppi (coarse), which is obviously too coarse to be much more than a big ol' chunk filter



The two foams stacked up vs. the Merv4 media



I just focused on the airflow since they all had about the same sound



The fine foam worked about the same as the Merv4 stuff. The coarse foam actually provided noticeable more airflow but it not a very good filter. This could be a good compromise if someone doesn't really want dust/pollen filtration but just to keep out leaves and stuff.

I'm all out of ideas for now!

I'll weigh in on this from my perspective as a HERS inspector (I inspect homes for energy efficiency measures and a big part of that is the HVAC equipment). The noise peak with the Russia filter is possibly strain being put on the bearing(s) for the blower fan. If speed is dropping off by half that tells me that these things are jacking up the static pressure of the system (basically there is a larger pressure differential between the downstream and upstream sides of the fan) which will decrease the life span of the fan. The flat filter though definitely has an issue with surface area and thus you can't get really useful filtration out of that without also having static pressure issues. The ideal would be the Russian style but with a lower MERV media but something still good enough to reduce dust and other large particulate. The most ideal situation would be some heavy fabrication to create a filter box like you see in most modern vehicles where the filter slot is much larger cross sectional area than the rest of the ductwork but that would probably run into numerous packaging issues. Sucks to see this as I've been wanting a filter for my truck but for some reason never applied my professional knowledge to the situation. The crappy thing though is dust will also kill your fan over time so I'm not sure if the filter is actually possibly killing it slower.

Considering how coarse that 30ppi filter is, I was expecting it would be closer to the no filter readings. Makes me wonder if the Tommy filter sans filter would read...might be that a significant amount of the restriction is due to the design of the filter housing?

Thanks for all the tests + sharing this filter with the community, just got mine in the mail, going to install now!