Air Filter Filtration Test

In the top left corner is the heavily advertised K&N. Going in standard English (not Arab) book order (left to right etc.), next is the Amsoil air filter. (Click on images to enlarge.)
After that is the Napa Gold paper air filter.

On the bottom row are the Jackson Racing filter, Racing Beat air filter, and the Fram air filter I removed from the box from the previous owner. That Fram had to rob airflow seeing all the metal in the backing. It has to be blocking 30% or more of the total area for airflow.  The K&N is of course a steel cage with cotton gauge lining inside the cage which is oiled. I have used K&N air filters for years on various cars. I always heard they let in more dirt. It is actually the argument over flow and filtration on the internet that inspired this test.

There are no definitive independent tests. Automotive enthusiasts have to rely on the claims of filter makers, which I feel are misleading. The Amsoil filter is a two stage foam air filter with a backing steel mesh. The first foam is coarse, and the upper foam below the cage is a very fine foam a lot like you see in HEPA filters. The initial foam is very coarse and definitely synthetic like a plastic. Think of the plastic scratch pads for pots and pans. The Jackson racing filter is almost identical in construction to the Amsoil expect it has a rubber gasket around the edges like the K&N and paper filters to fit the stock airbox. The Amsoil is oversized a small amount and just squeezes down into the airbox. The Racing Beat filter is really just one layer of very fine foam with a metal backing. The Napa filter is just their stock Napa Gold offering. It is some sort of paper and fiber combination with a metal frame down the middle and the rubber ring around it.

Everyone says Fram air filters are not that good, but I have no evidence for or against them. I’m not testing one, however I did use a Fram filter to make the 2nd stage filter. I hacked up a new Fram filter to do it. I used some of the screen wire as backing and the paper / synthetic fiber filtration material to make my test filter. This filter is placed behind the test subject to trap any particles (most of them anyway; no filter is 100% efficient) passing through the test subject. I placed the 2nd (made from the Fram) filter with screen (don’t want to suck a piece of paper into your engine or air meter) between the airbox and air meter.

Here is the view of the second filter from the air box:

The next photo shows the pieces apart. You can see the screen used for the 2nd filter. You can see a dirty 2nd filter after a test. You can also see the flapper door air meter (vane air meter), and the airbox top. The main filter sits between the airbox top and bottom.

Pieces and parts:

Did you see the dirty 2nd filter? Well that is the result of experiment #1. The test subject was the K&N air filter. Yes it let many particles pass through. The K&N was tested for 501 miles. For those that are curious, the test was conducted in Western WA state during the last week of November and December. All of the driving was on public paved roads. The climate here is damp during this time of the year and definitely not dirty or dusty. This is probably the best case scenario in all honesty. I was shocked by the look of the 2nd filter. I drive this car 6-8K miles a year. Many have asked if the black left on the filter is dirt or is it possibly oil residue. It is indeed dirt and not oil. The same type of deposits were left using a paper Napa filter which has no oil on it at all. That should be the end of that theory.

K&N air filter 501 miles, test 1:

The filtration test portion for the Amsoil air filter has also been completed. This was my first experience with an Amsoil air filter. I can say without hesitation that it is by far harder to service than the K&N air filter. You have to wash the filter using warm soapy water. Then you have to literally wring the water out, rinse, wring…. And then you have to let it dry which takes a long time. After that you have to impregnate it with the Amsoil filter oil which means squishing it into the foam pores by hand. Then you get to sop up the excess oil with a rag. It was messy and tedious. According to all the information I’ve ever read published by Amsoil and on their web sites their filter is the best at flow and filtration. Well the filtration portion is done, and the pictures don’t lie. The filter was serviced properly. Here is the first photo of the test filter by itself.

Amsoil air filter 503 miles, test 2:

The first photo is using no flash from the digital camera and the 2nd photo is using the flash. The flash tends to really show how thin the dirt layer is, but the 2nd photo shows a better difference in the amount of dirt on the filters. In the Navy we often use color comparison as a valid test in various chemistry analysis procedures. One of the tests in particular has you distinguish between shades of gray and black. The test are accurate, and there is no way this test is false in this regard either. Let’s have a look at those photos.

Comparison photos of the Amsoil test filter next to K&N test filter:

So are you surprised by the results? I was. I figured the Amsoil filter probably would filter better. At best, it is just as good as the K&N. According to the photos it is slightly worse. My wife who is a business major in college and sure has no stake in which one is which stated the Amsoil one was darker too. I put them both in front of her with the hand written sample text covered. She picked the Amsoil filter darkest, and she sure doesn’t care about the results.

The results are in for the Napa paper filter. I always heard on the ‘net that paper filters best. It does, but it isn’t as superior as I thought it would be. The K&N doesn’t filter nearly as bad as the horror stories say, and again the Amsoil was a disappointment. Here are two pictures comparing all 3. The first is with no flash, and the 2nd uses the flash in the same exact location and same wattage lighting as in the K&N / Amsoil comparison photos from above. My wife again established an order for filtration based on color not knowing which test filter is which, not caring, but putting up with my silliness anyway. She said the Napa filtered best, followed by the K&N, and last the Amsoil. She again reiterated that the Amsoil and K&N are very close as well. The Napa was clearly the lightest in color with the least particulate deposited.

NAPA air filter 498 miles, test 3:

Also tested were the Jackson Racing two stage foam filter, a Mazda factory replacement, and a Baldwin filter. The Mazda is a fiberous filter, and so is the Baldwin. The Baldwin filter is the red one in the picture below and the Mazda one is white. The Jackson racing filter performed very similar to the Amsoil foam filter, and the other two fiberous filters (Mazda and Baldwin) performed much like the Napa filter. Using the shade of gray comparison test, the Baldwin may have a slight edge in filtration. A longer duration or bench test may show a more significant difference.

The next contenders:

After 3500 miles of filtration testing, I am done. 3,500 miles of horrible power is more than I can bear. The secondary filter really kills flow and power above 4500 rpm, which is when a Miata starts to come to life. I’ve actually been done for a few months now, but haven’t bothered posting the results until now. I tested the Baldwin filter twice and observed the same results both times to verify repeatability. Anyway, here are the final 6 test filters all gathered together with a picture, and another picture using the flash. From top left to bottom right, the test filters were: Jackson Racing, Amsoil, K&N, Mazda, Baldwin, and Napa. The Mazda and Baldwin filters were also tested for 500 miles.


Well there is a clear pattern on filtration ability compared to both flow and the type of filtration media used. The “high performance” cotton gauze and foam filters do not filter as well as some have claimed. I actually received an e-mail from K&N stating their filters filter within 99% of the OEM filters. This may be true, and 1% may not sound like much. I contend that 1% over many miles, may be important. Really, it is up to each individual to decide. The poorer flowing filters remove more particles, and the better flowing filters remove less particles. If you think about it, that conclusion passes any and all common sense tests, so it is not surprising. There are many that will be shocked by the results, though that should not be. I’ve used high performance filters in the past, and I might again in the future. At the same time, I know that the stock OEM type filters perform very well in filtration and don’t inhibit flow nearly as much as some think.

I know it will be asked…. The K&N was properly cleaned and serviced using a K&N filter recharging kit which I personally own (owned quite a few K&Ns). The Amsoil filter was serviced with Amsoil filter oil. The others will be serviced with the appropriate oil as well. The paper air filter will be installed…. I did have a question regarding the deposit on the filter possibly being oil. It is dirt. I took a piece of my 2nd filter test stock and put a few drops of K&N filter oil on it. The oil stays bright red on the filter as well. One person even mentioned the dirt in their area isn’t that black. Well this is Western WA state where everything grows lush green. I live 90 minutes from full fledged rain forest and the soil here is black like topsoil. I guess I could take a picture of my white powder coated wheels turned black from road dirt if I had to. If you have a HEPA filtration air purifier in your home and have ever changed the HEPA filter, you will notice that it too is black. The Napa paper filter tested showed the same black deposits. Paper filters have no oil on them, so the oil hypothesis is a dead issue. In the e-mail I received from K&N, I actually got contradictory information on this point. They claimed that the deposit on my test filter was oil (it isn’t), and then they said oil does not leave the filter and damage mass air sensors. Honestly, you can’t have it both ways. Either oil leaves the filter or it doesn’t. Oil will damage a mass air sensor if it gets on the sensor wiring. This is highly documented by TSBs easily available online.

Some have asked that additional filters such as Wix, Fram, and ITG foam be tested. Only the filters listed above will be tested. I do not have the funds, time or desire to test every possible filter. A clear pattern already exists. Foam is foam no matter who makes it. Fibrous filters are still fibrous filters, and paper is still paper. Sure there are minor differences, but in this test a clear pattern emerged based on basic filtration media rather quickly. I personally do not like the difficulties associated with servicing foam filters. I do enjoy the simple drop in of an OEM style filter. Imagine servicing and changing 5+ filters in one day for pressure drop testing. You begin to dislike serviceable filters quickly.

Tags: ,