Bob D

[Ranger]

Bob,

Not long ago we had a discussion about a dirty air filter. I said it would not affect MPG and you disagreed. Then you came back and said that I was right after doing some research. You found the post where Guru had explained the theory (carburated vs fuel injected). Do you remember the post where you found it? I have searched and searched but am unable to locate it.

[Bruce Nunnally]

You probably already see this, but note that one of the gadgets on the search window lets you select which posts to search through, and I think defaults to posts from the last 30 days. Change that setting to posts from Any Date to search through a larger index. I have good luck with using and / or as appropriate for linking keywords, so filter and injected for example, and putting phrases in "double quotes" like "air filter" if I want that exact phrase. Partial strings work for member name, so Guru works there for that author.

[Ranger]

Bruce,

Thanks, I actually don't need the post any longer but still would like to find it if I could. I have not had much luck with the archives. I have used the 30 day switch with success but the "name" thing never works for me. I was unaware of the "and/or" or the quotes. I'll give it a try next time.

[JasonA]

Larry, are you sure it was posted to this forum, and not another or not a personal email? I remember you guys' discussion about it, but don't remember the subsequent postings from Bobynski and Bob D. It was explained that a dirty air filter DOESN'T have an effect on gas mileage? I'd love to see that post. As we all know, air filter manufacturers market to the contrary...but I wouldn't be surprised at all if their advertising was incorrect...it's happened before with air filter companies.

Thanks,

[JasonA]

Bill, thanks once again for this great information!

[an01sts]

I cannot say one way or the other. I will say that it's an intereseing view on the matter. There is one thing, however, that still leaves a gray area. You say that the proof is with a vacume gage. You don't mention whether the engine is under load or without a load because there is a huge difference in engine operation under load, compared to standing.

(Now I'm just throwing this up for thought, so I'm not trying to defend it, other than it may be something to consider.) Even with the test done with the engine under load, the vacume gage as proof can still be questioned: Because it's a gradiant change--rather than a flap, shutting off flow--the change in vacume will be in small gradients, so slight that the vacume gage cannot detect it. This ever so slight change, which cannot be detected by a vacume gage, may affect mileage.

Another factor to consider is the duration of the test: Let's say that with a clean fileter, you get 17 mpg @ 60 mph. That means that it takes one hour to burn one gallon of gas. And let's say that the test with the vacume gage has been conducted while the engine is under load. Is the test conducted for one hour under load? If not, there is signifigant question as to whether or not it works on paper, compared to actual operating conditions. After all, the proof has to come from actual mpg tests, not a bench job.

[JasonA]

There is one thing, however, that still leaves a gray area. You say that the proof is with a vacume gage. You don't mention whether the engine is under load or without a load because there is a huge difference in engine operation under load, compared to standing.

I believe he's talking about putting the vacuum gauge in the intake system, between the throttle butterfly and the filter. Plumbed to the intake tube for instance. The point here is that if you see ANY vacuum, it would demonstrate that either the intake system and/or the filter is providing a restriction. If there's no vacuum in the intake system (upstream of the throttle body), there is no restriction.

[CadiKing]

Another factor to consider is the duration of the test:  Let's say that with a clean fileter, you get 17 mpg @ 60 mph.  That means that it takes one hour to burn one gallon of gas. 

an01sts,

I cannot follow your thought. If you go 60 mph for 1 hour, you will have traveled 60 miles in that hour. If you are getting 17 mpg, you would have burned one gallon of gas in the first 17 miles...No?? It takes 17 minutes to go 17 miles and you would burn one gallon of gas in doing so...

my .02

[an01sts]

Along with spelling, I don't do well with numbers, but the concept should be clear. My thoughts are that the only way that I wouldn't question results is if actual tests were conducted, meaning that with a dirty filter, the engine is ran under load for x-amount of time, consuming y-amount of feul. Likewise the same test is done with a clean filter.

One thing about the vacume test has to do with ambient pressure: Even at an idle, the induction system always maintains a lower pressure than ambient. You can say that the engine sucks in air. The thing is that the engine isn't sucking in air because the laws of pressure say that air is pushed into the engine as ambient pressure flows to the less-than-ambient pressure area. (As with cold attracts heat, high pressure migrages to low pressure.) In other words, the ambient air is filling a vacume; therefore, the induction system is always under a vacume, regardless to what a guage claims.

Again, I have no clue as to electronic feul delivery, so it's hard to argue about something that which one doesn't have the knowledge on which to base an argument. The whole point that I want to make is that I would have to see actual mpg test with a clean and dirty filter because, based on simple engineering logic, that type of test the conclusive proof, providing the final answer.

Like I said, I cannot say one way or the other; rather, I only question the method in determing the affect on mpg. I'm sure that someone has done the actual test, rather than making conclusions based on a bench tets because bench tests and actual tests can have different results.

The thing is that I find this an interesting concept, and it has nothing to do with right and wrong; rather, I'm a hard sell witout absolute proof, so I'll look around and see if I can find actual tests.

[JasonA]

In other words, the ambient air is filling a vacume; therefore, the induction system is always under a vacume, regardless to what a guage claims.

Isn't the very definition of "vacuum" where high pressure can't fill the low pressure's void quick enough? If the air pressure in the intake system is always at least at ambient (meaning the ambient air can back-fill the low pressure intake system faster than the pressure can drop), then there is, by definition, no vacuum. That's what a vacuum gauge measures. It's true that the air in the intake system is always moving. But the point of a vacuum measurement is to quantify whether ambient air can back-fill the void faster than the void can form. If it cannot...meaning there is a restriction in the intake system, a vacuum is formed...meaning a lower pressure than ambient. If a vacuum is not measured, this means the ambient air can fill the intake at a rate sufficient to always overcome the pressure loss...meaning the intake system is NOT a restriction.

[an01sts]

We're getting a little off focus on the topic because there is a lot of exchange from both of us that don't relate to whether or not dirty air filters reduce gas mileage. My point is that from the information I can gather, dirty filters reduce mileage. I don't know your occupation, but I was a professional mechanic (glass/chassis/body/thef recovery specailist) for over 17 years.

As former Texas Motor Vehicle Inspector #: 13885889, part of my inspections were emission testing, cars up to 1994. It's not like I kept track of the numbers because I knew I'd need the statistics for later on, but my best guess was that 20 to 30% of the failed emission cars, failed because of dirty air filters. Simply, replace the air filter, and it reduced emission and the car passed the test. Of course, the defintion of dirty is subjective: These filters weren't major crusty, and they would probally meet the definition of the level of dirt you would consider as not affecting mileage. If the clean filter reduced emissions, it seems logical that it would increase mileage, and it seems as if it's fairly credible documentation because this was proven time and again, in actual before/after testing of engine operation.

I also tried to find information, using various phrases of clean versus dirty air filters, but the only thing I could come up with was 100% agreement that dyno tests indicate that clean air filters increase mileage up to 10%. I know that people selling filters have vested interest, but there are as many sources who don't have vested interests in air filters making the same claim.

The only problem I have is finding the actual test results. Also, no matter how I phrased it, I couldn't find even one comment that claimed that dirty air filters don't affect mileage.

The main reason I engage in this is because I listen, and I understand the foundation that you stand: I also have a 69 Firebird 400, slush box. The air cleaner lid can be inverted so that it operates with an open element filter, and the open element gives considerable performance gain. The tradeoff is that the gas pours through the car. (I was driving through TN on 30, cruiseing @ 103 to 110 mph with the traffic. I ran 18 galons of gas through the car in little over 60 minutes.) If I flip the cover over so that it draws air through the snorkel, the gas doesn't pour through the car near as fast as with the open element, but I give up performance.

My point is that I fathom the concept that a restricted air filter might increase mileage. I restrict the air filter on the Firedog, and it increases mileage. The huge difference, besides that bad-*smurf* Holly-eating Quadrajet on the car, is that late model cars already have more than their fair share of restriction, and additonal restriction probally decreases mileage. The hole in my frame for the cold air pick-up is smaller than the air hose from the air filter box to the throttle body and/or whatever jargon that ones wants to refer to.

My testing and other testing lends proof that dirty air filters decrease mileage, and you are the only source I have seen that claims that it makes no difference. With all due respect, if what you claim were fact, it would seem that there would be other documentation to make the claim.

I have conducted actual tests that prove that dirty filters increase emissions; thus, it's a safe bet to claim that the clean filter also increased mileage. Trust me on this one, I didn't have vested interest in selling air filtes. One number I'm sure of is that 100% of the air filters I installed were given away. (When I replaced a windsheild, the car had to be reinspected. Therefore, I became a state inspector by default of my job. Because some of my repairs work voided a valid inspection sticker, the customer received free air filters, and I didn't get paid a dime to install them. So an air filter killed me bigtime because an inspection only paid .3 hr. If it wouldn't pass with a clean air filter, it went to tune-up/emissions line. The thing is once I checked off the car, I had no clue as to what it took to pass emission inspection.)

My point is that all documentation contridicts your comments. I like an underdog who goes against the norm--I go against the norm too--and I find your comment interseing and undestand the foundation on which you stand. The only thing I would like to see is actual test results. With my test results, I cannot prove that it increased mileage, but I can assume so because the reduction of emissions in the same x-amount of air volume means a good chance that less feul is being consumed. Also, I cannot find actual test results that proves the norm. I might spend more time searching but I spent almost an hour looking.

If I can find actual tests, I'll post it. Likewise, please do the same. The bottom line is that neiter one of us can produce actual test results--that means an engine on a dyno under load, having a mpg test done. Therefore, we are pretty much at gridlock because neither one of us can produce such documentation.

[Bob D]

Whew! Glad I showed up in time for that....Actually, it looks like you did ok without me..Oh well. LOL

My satellite modem's been down since last Sunday morning. Up here at the ranch , you can forget anything to do with cable, DSL or T1, for that matter. So..that leaves you with dial up or satellite. Hmmm..Thing is VPN's don't do all that great with dial up, especially with large data streams.

Back to the satellite..Word had it by Monday that it was FUBAR (space junk?). Next launch by Intelsat is 12/17 and if they could comandeer that one, there's orbital alignment, transponder alignment, testing, FCC beurocratic stuff and so on, realistiaclly it will be ready by ...maybe March? Not good. Been completely wiped off the internet lately for only six days? Not too fun or productive.

Miraculously, they somehow got the old bird revived, (no one will say what really happened..) I just loaded up some "recovery" software and badabing! It works.

Back to the real world...

[an01sts]

I know that I didn't clarify it, but when I was an inspector, I was working at a Cadillac dealerhip, and some of the cars I was refereing to were powered by the 4.6 Northstar. I also negelected to mention that almost none of the cars I emission tested were caburated, and almost all of them were less than 3 years old.

To side track a little, GM warrenties cars for whowever long for emissions. In fact, all manafactrures have to because there is a federal law that says that all new cars must meet whatever emission specks, for however long. The thing that amazed me was that these cars were state of TX emission exempt for 2 years by defalut from their inspection stickers. Because I destroyed their inspection stickers (well, actually it is illeagle to transfer the sticker from the old windsheild to the new windsheild) the car had to be reinspected. And some of these exempt cars failed emission testing, so how many were within specks? The majority passed, but the ones that failed, 20 or so percent would pass with a new air filter. Of course that means over 80% of the failed emission-exempt cars had other issues.

Most techs there did superior work (none of the techs has NQPI scores less than 90%. You see, our NQPI was posted in the customer's waiting area, and customers could request techs. If you had a tech in their who had 70%--what's considered perfect, but that's another story--he would be starved out. Simply, the customer would say, "Anybody but him." Too bad that everyone doesn't post their techs NQPI, but the majority of GM dealerships don't have a clue as to wtf NQPI is.)

Anyway, we didn't pump cars out, but there is a fine line between getting it done right, and making money. The chunk in an air filter was that fine line. If the car failed emission testing, you would chunk in an air filter because it would solve about 20% of the failures. (Almost everyone was a state inspector. It was done for efficiency. That way, if the car need to be inspected, it got inspected in the system, rather than having an inspection system. The air filter weeded out a small group of failures so that emission repair techs could focus only on failed emission components.)

I always wondered how many emission exempt cars had faluty emission. The thing is that they were exempt for 2 years, and after 2 years, the repairs weren't covered under warrenty. In ohther words, these cars had faulty emissions off the line. Because of the exempt status, the customer got hoodwinked, stuck with whowver much of a repair bill when emission testing time arrived.

The fate of the failed because-of-air-filter cars on their next annual emission test was unknown to me. With the majority of the cars, however, I did the customer a favor. After all, had I not voided their emission exempt sticker, they would have been suck with an emission repair bill after 2 years because most of them would be out of the federal mandated warrenty mileage.

My point is that this lends question to tests and design. According to the bench, these cars were fine. In actual operation, however, the bench proved to be wrong.

Until I see conclusive proof from either side, I'm on the fence on this one. (Ima' performance type of guy, so--for me--dusty air filters go in the trash.) And this is only offered as food for thought.

Again it's beating a dead horse, and I'm glad that we can talk differences without it offending each other, so I just want to point out one other thing as to the vacume test.

We have 2 settings: A car gets 17 mpg @ 60 (17/60). Then it starts to get 16 mpg @ 60 (16/60). In a fractional porportion, the mileage decrease (17/60 - 16/60 =) 1/60th of the unit's value. How is this 1/60th decrease in the unit's value detected with the vacum gage? It seems that such a small change would be impossible to detect on a vacume gage. The reason I ask is that chnage is a good average for increasing mileage by 10% (10% of 17 mpg is 1.7 mpg.)

[Regis]

There is some underlying humor in this discussion and I mean no offense to anyone in particular. Medically speaking it's like the nurse practitioner telling the brain surgeon about the brain. And in doing so a lot of great info surfaced.

This discussion was very informative. I learned that I've been replacing the air filter too often needlessly. I also realized that I'm a creature of habit. I always changed the filter once a year even though I never saw a "performance" gain or loss in doing so. Once I had to change it early because the filter was covered with leaves and twigs. A freak occurrance.... well not entirely. That's what I get for crashing through all those neat leaf piles people rake and push onto the road.

[mike5514]

I have read this thread to the best of my ability and i agree with Guru as to merits of a dirty/clean air filter on a modern engine. It is apparent that the electronic components adjust to the air flow keeping milage relatively stable which is the subject of the first post. In regard to emmisions it becomes real complicated, in the 70s the less emissions a engine created the worst the milage became, when the emmision controls were defeated, milage (for the most part) was likely to increase. Therefor in a modern engine a dirty air cleaner could increase emissions without decreasing gas milage? As to horsepower volmetric efficiency would be the determing factor and there would be little noticeable change under normal driving conditions.? mike