Originally Posted by dnewton3
Here are my comments on the concept of the K&N filters, and these UOAs in particular ...
K&N products are not inherently bad. But they present a challenge to get them set up properly. Too much oil on the media and they will often weep the oil downstream to the MAF sensor, which then in turn causes CEL faults, etc; this has been seen a bazillion times. Or, you don't get enough oil on it, and it does not filter nearly as well as it should; often the cause of higher Si counts we sometimes see. The K&N filters do a very good job WHEN THEY ARE PROPERLY CLEANED AND CONDITIONED AND INSTALLED IN A MANNER THAT WILL NOT AFFECT THE ELECTRONIC CONTROL SYSTEM(S). However, that's much harder to achieve for some folks, especially given the American way of thinking that if some oil is good, then more must always be better! As a generalization, it's much easier for Joe Average to install a common dry cellulose filter, because there's far less for him to screw up. K&N filters get a bad rap not because they are bad, but because they are finicky to set up properly and people blame their errors on the filters; unfair or not, that's the reality.
Now as to these UOAs, you cannot actually attribute the Si control to the K&N until you know the presence of upstream intake Si. The low Si count may be because the K&N is doing a good job, or it might just be because there's very little airborne Si in your general area. Example from a different perspective: We used to have outside cats around our house in the rural area I live in. There are no black or brown bears where I live. Is it because the cats keep the bears away, or is it simply because no bears are typical in my area? So, until you did several back-to-back tests, using only the air filtration as the controlled variable, you really cannot claim any proof that the K&N is actually the cause of the low Si count in the UOA.
Good comments - appreciate the discussion! A few more thoughts on some of your points:
I agree 100% with you on the "present a challenge to get them set up properly" - this is why the OEMs don't use filters like this first and foremost - they use paper b/c it's set and forget. The cleaning/oiling is the key. K&N says 50K miles; I've read where people are doing this annually and each time you do it you run the risk of f'ing it up. I will say this w/r to the overoiling/MAF sensor issues, I'm not so sure I believe it. I don't have a MAF in my MK7 Golf so not an issue anyway but do run one in my Atlas which has a MAF and as expected, no issues. This is a great video on their testing of this (over oiling etc.), worth a watch @ 17 min long. Yes, I know, it was put out by K&N but regardless, the testing methods look solid and conclusions make sense to me. I had 2 MAFs go on my MK4 Jetta with standard air filters. They guarantee if you have a MAF issue they will cover it:
https://www.youtube.com/watch?v=gE6moItrZNg
Much like I can't attribute the Si control to the K&N without upstream SiO2 numbers, you can't also say that high-Si in your oil is a result of the K&N without having a back-to-back test with a standard paper filter where you may get lower SiO2 numbers. This goes both ways and my point of the whole post is to debunk the blanket statement that "you get high SiO2 when you use K&Ns" that is almost BITOG lore at this point. I always say in my posts "under my operating conditions" to point out that I drive in "clean" "normal" conditions and these results may look different is say you are in a desert or other dusty area. I live on the East Coast and DD my car mostly on the highway, hardly the kind of conditions that will cause excessive dirt/dust into your intake that would overwhelm the filter's capacity to clean the air adequately.
Using a K&N (correctly) doesn't automatically equate to dirty oil/poor filtration as many on BITOG preach - my UOA results continue to show it. Will be interesting to see how they change over the next year as I approach the 50K mile cleaning interval, clean it, then get data post-cleaning. I suspect I will see zero difference.