So Clean Air Filters Don’t Help Fuel Mileage?

While cruising the wide world of web recently, I landed on a .gov website with an interesting article that I thought had some interesting results. The premise of the article is that the US Department of Energy wanted to know how vehicle maintenance effected fuel mileage, specifically dirty air filters. So they did an experiment with 4 cars:

  • 2003 Toyota Camry 2.4L
  • 2007 Buick Lucerne 3.8L
  • 2006 Dodge Charger 5.7L
  • 1972 Pontiac Grandville 455ci (aww yea)

Now, I encourage you to read through the whole article, but if 27 pages of government written science experiments doesn’t sound exciting to you, then you are in luck, because I am about to give you the cliff notes in my own words.

Here is a link to the full experiment in PDF form.

Here we go – The people doing the test first set up a series of gauges on the intake of each of the engines so that they could measure the air pressure before the filter, the pressure drop across the filter, and the intake manifold pressure.

Then with each of the 3 fuel injected cars set up this way, they did a series of lab tests and driving tests.  They started by measuring the pressure drops across clean & clogged air filters.  They “clogged” the new filters by wrapping the filters with paper towels. (Note to self: Wrap friends air filter on 4/1/2012).

Then they moved on to the driving tests by measuring the time it took to accelerate from 20 to 80 mph at WOT (wide open throttle). Clearly there is a performance decrease in these engines.  Just thinking about how the engine must feel during this test makes me want to grab an inhaler.   Now these cars know what it feels like to be a Karmann Ghia.

Next up, they measured the fuel economy for each individual car with and without the clogged air filters.  Here is just one example of the results, this one below happens to be the 2003 Toyota Camry.  As you can see, they did 3 different driving tests for each vehicle, and there is not a measurable decrease in fuel economy. The results were the same for all three cars. Strange huh?  It is apparently because if less air goes into the engine, the computer knows it, and offsets fuel to match the smaller amount of air.  Wrap your brain around that one for a moment.

Oh.  Then the big ol’ carbureted 1972 Pontiac Graaaandville entered the air filter party.  It was officially time for the 455 to shine and show the young whipper snappers who’s boss…

A similar set of gauges was hooked up to the air intake and all of the pressures were measure for a clean vs. clogged filter.  As expected, when they did the driving tests in the Grandville, it immediately tried to eat the clogged filter whole and swallow it like a champ.  It was having no part of this “less air” test.  On the graph below, you can see that the lack of air flow greatly hampered its acceleration time.

Amazingly, even with a carburetor, the 20 foot long rolling couch only lost 2 – 2.5% of its fuel economy during these tests.  One of the tests couldn’t actually be performed because the Pontiac was struggling so badly to breath that it overheated every time they tried to run it. Poor 455.

So what exactly can we gain from this data?  Hmmm, well I guess it’s nice to know that somebody is testing these types of things to keep the public informed.  It certainly surprised the heck out of me that a clogged air filter doesn’t seem to effect a fuel injected vehicle’s fuel economy, though I’m not surprised that it kills every single hope of performance.  Does this mean you shouldn’t change your filter? Not really.  It just means that if you changed it recently in your Subaru with hopes of getting better gas mileage, you may be disappointed.  Sorry me.  The good news is that if you change your story a bit and tell yourself that you bought the fresh new air filter for better throttle response (obviously), you instantly feel better about the purchase. Phew.

All images borrowed from:

Jeremy Nutt

Hi, I'm Jeremy.

10 thoughts to “So Clean Air Filters Don’t Help Fuel Mileage?”

  1. Only on engines with an MAF sensor. OBD-1 would suffer.
    NOW let’s do a potato stuffed in the tailpipe test!
    Oh yeah, and the obomo low tire pressure run.

  2. MAF sensors are not specific to OBD1 and I see no reason that an AFM would be exempt either way

      1. Oops, Euros. I just had the oddball older ride…
        Gee million mile oil, forever plugs, and now air filters…there seems no need to open dahood!
        Now about that stinky blue haze…

  3. Well of course it affected acceleration but not economy. The throttle on today’s computerized fuel-injected cars does nothing but admit air. The computer uses the O2 sensor, throttle position sensor and maybe mass air flow sensor to decide how long to open each fuel injector to get the perfect (called stoichiometric) fuel-air mixture. Limiting the air with an actual or simulated clogged air filter is no different than simply limiting the throttle to, say, 80% opening. Makes perfect sense to me.

  4. Try to run with something over your mouth and nose. you get the picture.. now with a potato up your rear i don’t want to go there !

  5. The dirty filters also filer out more dirt, those gauze filters are not very good at filtering dirt, so don’t expect the engine to last very long. No wonder they have a million mile guaranty ?

Leave a Reply

Your email address will not be published. Required fields are marked *