Should you ever change your air filter?

[Jim Allen]

Originally Posted By: Ken2
The filter may deteriorate with heat and age and come apart. Bad problem.


Define the limits. If we are talking 20 years maybe. If we are talking 3-6 years with a good quality filter, I'll disagree.

Originally Posted By: Ken2

Dirt particles do work their way through filters over a long time. I've changed many that are dirty on the downstream side.


Again, you are overgeneralizing. Define the limits. What you describe does happen with some filters but commonly only when excessive restriction is allowed to continue over a long period. By "some" filters I am speaking about poor quality filters. Also, if the filter is defective or damaged you gan get pass-thru. Pinholes are not always readily visible so unless you really do a detailed inspection of a filter in which you see dirt on the clean side or in the intake tract, you really don't know what happened. I have observed air filters being tested in a lab and even if the DP is allowed to rise well past the rated maximum, good quality filters had little if any migration thru the filter until the structural limits of the element was reached.

Originally Posted By: Ken2

Restriction gauges work best on a diesel due to their lack of a throttle, and even then the engine needs to be running hard with the turbocharger pulling a lot of air in. On a gasoline engine you'd need to watch the gauge during a full throttle, high rpm run.


Perhaps your comment is just inarticulately worded but, if not, it sounds like you have a basic misunderstanding of how restriction gauges work. Restriction gauge are nothing more than recording vacuum gauges. They record the maximum amount of restriction encountered. The instructions for the gauges I know about have the operator to do a WOT run and then observe the recorded maximum restriction with a new/clean filter. From there you periodically observe or reset the gauge and do another WOT observing the increase. Either the engine manufacturer will have a restriction limit or you can use some of the generic limits available. They work equally well on diesels or gasoline engines. Here is Donaldon's info sheet on them Info Sheet and here is Baldwin's Info 2 which has a great filtration "do and don't" on page three. Also look at the Filter Minder site for good info Filter Minder. The only engines that can be problematic, so I am told buy one engineering source, are carbureted gasoline engines. More and more I question that but won't go into it now.

 

[cheesepuffs]

I would never run a filter for an extremely long time even if the filtering efficiency does go up slightly over time. Because I don't know about you guys, but I only use regular paper filters and I sure as heck can tell the difference after I've put a new filter in. My vehicles are noticeably more responsive with a clean filter I'd rather have a little more pep than have a ever so slightly better efficiency when the OEMs know that even a fresh filter still filters enough to be safe.

 

[Jim Allen]

Originally Posted By: cheesepuffs
I would never run a filter for an extremely long time even if the filtering efficiency does go up slightly over time. Because I don't know about you guys, but I only use regular paper filters and I sure as heck can tell the difference after I've put a new filter in. My vehicles are noticeably more responsive with a clean filter I'd rather have a little more pep than have a ever so slightly better efficiency when the OEMs know that even a fresh filter still filters enough to be safe.


If you are talking a lightly loaded filter, I'm sorry to tell you that you are very likely a victim of the placebo effect along with many others. I've done and seen too much objective testing (dyno work) to believe a lightly loaded filter has a noticeable-to-driver effect on performance... in the vast majority of cases anyway. Any differences are very, very subtle if they exist. Plus, only an inadequate and poorly designed intake system would give those types of indications.

As to filtration, "slightly" is not an accurate statement. What you say about initial efficiency of a filter being "safe" is true, in the short term anyway, but I assure you the OEM know about filter loading and its effect on efficiency. How much does efficiency improve? It will vary somewhat from filter to filter, but I have some industry averages compiled from various engineering sources:

Average improvement over lifetime initial to final- 2-3%
Average improvement after 1st 10% of useful life- 1-2%


How much contamination does that equate to:

Code:
For Every 10 Pounds of Dust Drawn Into the Air Filter Inlet:



EFFICIENCY OF FILTER DUST INTO ENGINE



99.95% 0.005 lbs.

99% 0.10 lbs.

98% 0.20 lbs.

97% 0.30 lbs.

96% 0.40 lbs.

95% 0.50 lbs.

90% 1.0 lbs.


Let's say you change your filter so often that you reach only a small portion of the efficiency improvement and a good part of that time you are below that. The 10 pounds of dust above is close to a good average for a large truck in a dirty environment but not a car. I have seen some averages but they vary wildly and aren't from sources I trust, so I will say about a pound for a car in a city environment because I have observed that much.

If you short cycle the filter, lets say every 10K miles, you suck in .3 ounces of contaminants with a 98% filter. In reality, you will see some improvement in those 10K miles so we will make that .24 oz/10K miles (see below). Over 50K miles that's up to 1.2 oz. of contaminants going into your engine.

Lets's compare that to a guy that runs his filter 50K miles in the same environment as you, but changes his filter according to some objective standard like a restriction gauge that trips at 50K miles. We could objectively say in that case that the filter's useful life was 50K miles in that environment. In the first 5K (10% of life) the engine drew in .16 oz of contaminants. Following the averages, at that point he and you gain a percent at that point so that amount drops to .08 oz for the next 5K for a total of .24 oz in 10K miles. That's where you changed the filter but he keeps going and his efficiency gets better. At 20K miles, he has reached the 2% level and we'll say he stays there thru 40K miles when he gains another percent and his final dust total at 50K is around .54 oz.

So, in this hypothetical example, over 50K miles, your engine ingested 1.2 oz of dirt when you changed the filter every 10K and the other guy's engine sucked in only .54 oz. by leaving his filter in place for 50K. You are willing to allow your engine to ingest more than double the dirt in exchange for a gain in "response" that is probably fractional if it exists at all. Doesn't seem like a good trade to me, even if your engine can "take" it in the short term. Maybe you trade your cars often enough that the long term issues will never be a problem for you.

 

[cheesepuffs]

Jim, I appreciate your very in-depth response but I still don't think it's a problem to not be consciously loading your filter. Even if my hypothetical engine did ingest twice the amount of dirt as yours, it's still not much dirt. Maybe in a lab setting it looks like a big deal but in actuality we don't see regularly maintained (short air filter change intervals going by BITOG standards) vehicles on the road failing due to lower filter efficiency. It's literally a non-issue in the real world. If running too new of filters too many times was a problem then OEMs would start specing higher efficiency filters to begin with and counter it.

As far as the responsiveness goes, I'll start off by saying that I am extremely skeptical by nature, have formal automotive training, and am a very into the technical side of things. But yes, I notice the difference and have on multiple vehicles ranging from sub-two liter four cylinders to large V8s. It's not that it's a feeling of, "wow, my 0-60 time has been cut in HALF!". Not at all. It's more of a willingness to rev more freely that translates into the car feeling "healthier". I was never searching for this difference when I first noticed. I noticed because it took me by surprise.

 

[Jim Allen]

Originally Posted By: cheesepuffs
Jim, I appreciate your very in-depth response but I still don't think it's a problem to not be consciously loading your filter. Even if my hypothetical engine did ingest twice the amount of dirt as yours, it's still not much dirt. Maybe in a lab setting it looks like a big deal but in actuality we don't see regularly maintained (short air filter change intervals going by BITOG standards) vehicles on the road failing due to lower filter efficiency. It's literally a non-issue in the real world. If running too new of filters too many times was a problem then OEMs would start specing higher efficiency filters to begin with and counter it.


To start with, thinking in terms of "how much dirt can my engine tolerate?" is not the best way to look at it. It's like,"How much e coli can I stand on my hamburger?" To knowingly let your engine eat more dirt than you must, especially if you have simple and easy ways to prevent it, does not seem wise.

Back in the days when a good air filter was 80 percent efficient, it was a miracle for an engine to achieve 100K miles. Air filtration was a big part of that and engines operating in dirty air environments like deserts saw radically lower lifetimes than 100K miles.

OE filters have indeed been increasing in efficiency generally but they are almost never at the top end of the available technology. Why is simple... cost cutting. Increase the profit margin and keep the costs lower for the customer. What they install is more than adequate for normal operation, especially if you consider that efficiency improves with time and loading over a normal interval (which is the main thrust of my argument), and they are more than good enough to get them out of warranty. Goodenuf is goodenuf!

On top of that, most people don't operate their vehicles in an environment that severely challenges the air filter or provides the opportunity for the engine to have to eat a lot of dirt. Engine longevity then doesn't become an issue for most people no matter what filter they use. If you know your area has clean air, the engine longevity risks of a lower efficiency filter are not great.

People like me who drive their cars and trucks for decades have a slightly different view than those who trade in every few years. I still have a 2000 Honda bought new and an '86 truck bought nearly new. My attitude towards "how much dirt should my engine eat" might be different than a guy who trades his car in after five years and really doesn't care what happens to it down the road. But this second guy is hard for me to understand if he wastes money on the car by changing the air filter too often.

Originally Posted By: cheesepuffs
As far as the responsiveness goes, I'll start off by saying that I am extremely skeptical by nature, have formal automotive training, and am a very into the technical side of things. But yes, I notice the difference and have on multiple vehicles ranging from sub-two liter four cylinders to large V8s. It's not that it's a feeling of, "wow, my 0-60 time has been cut in HALF!". Not at all. It's more of a willingness to rev more freely that translates into the car feeling "healthier". I was never searching for this difference when I first noticed. I noticed because it took me by surprise.


I also have formal automotive training, going on 50 years of experience, and part of my job of the last 23 years is to road test and modify vehicles. I've done many before and after performance evaluations for magazine stories and been able to compare what the dyno (both chassis and engine) show vs what the "butt dyno" says out on the road and I can say with some authority that I can't "feel" a 5 hp difference on a V8. Many times I can barely "feel" 10 hp. After all these years, my butt dyno is pretty well calibrated I think and after literally hundreds of cars and trucks tested, you'll have to forgive my skepticism of claims like yours where people say they can "feel" the difference between a clean filter and one loaded to 40% of capacity (or some other lightly loaded point).

Of course you are going to feel a highly loaded filter but that's where the restriction gauges come in. If performance is your bag, objectively determine the restriction level where performance is degraded and mark that as your point to change the filter. Otherwise, you run the filter to the maximum rated restriction, which is adequate for those of us who seldom see the high revs.

 

[Clevy]

Wow Jim. Your responses in this thread are over the top informative and I'm pretty much on board with everything you've written.
I especially like your comment on dirt ingestion and comparing it to E COLI. When an engine is ingesting enough sand that a used oil analysis shows increased silica in the oil I see that as being equal to 200 grit abrasive moving up and down inside the cylinder which under no uncertain temps is grinding down the rings and widening the bores and will most certainly shorten the time between requiring a ring job and possibly opening the bores enough that a larger piston is required.
I've been a gear head all my life though never any formal training other than countless hours wrenching on my own and others rides and I can honestly say I've never felt any differences between high flow air filters vs paper/cellulose and even bolting on a cai vs the stock intake tract. There just isn't enough of a difference for me to notice,unless the engine is far from stock and the stock intake tract is just too narrow now,but a just basic bolt on engine doesn't suck enough air for there to be a restriction.
In fact I've learned that when going too large on the intake side the air velocity slows down which costs driveability in the way of toe in throttle and mid-range and only benefits at screaming high rpms and even then it's barely 10hp tops.
Because of the velocity loss the 1/4 mile times end up slower because from idle to 4000 rpm the engines not sucking well because the velocity is slower and therefore less power,and those few extra ponies up top just don't compensate enough to offset the loss down low.
Anyways I just wanted to say thanks Jim. Your posts are always enlightening and I look forward to them.

 

[Jim Allen]

Aww, shucks!

This is one of the topics in which I am most interested and I have studied it alot. On top of that, I have had access to engineers from several different companies and have gleefully abused their time schedules to ask every silly question I can come up with.

Modern engine generate so little inherent internal contamination that air filtration becomes the number one input for wear-inducing contamination. If you are thinking long term with regards to engine life, it pays to think in terms of engine air filtration efficiency. It really doesn't take a lot of moola or effort to improve it either. You've won half the battle by making sure the filter seals well in the housing and that the housing is well sealed to the intake. There are some higher efficiency filter elements available but even if you just go with a premium off the shelf filter, it may be little better than what came stock. After than, letting the filter load up and increase efficiency as you drive is the second way, and it's no cost. A restriction gauge will tell you EXACTLY the right time to replace the filter and if you are performance oriented, you can simply change it at a slightly lower restriction. Either way, you have changed the filter according to an objective standard in your operational situation, rather than a one-size-fits-all-worst-case-scenario OEM interval. Or a WAG on your part.

 

[gaijinnv]

Originally Posted By: Jim Allen
A restriction gauge will tell you EXACTLY the right time to replace the filter and if you are performance oriented, you can simply change it at a slightly lower restriction.


As always, great advice, sir!

It's such an easy mod, I am always amazed at the irrationality of the resistance offered by the naysayers:

 

[Malo83]

Originally Posted By: MuzzleFlash40
I change mine once a year, They cost $6 at walmart. I know they could go longer but I feel better doing it my way.

Do mine annually as well, you would think reading all these posts it was rocket science

 

[cheesepuffs]

Jim, I understand where you're coming from. But you said it yourself when you mentioned how most cars don't see extremely harsh air quality in the first place so there for normal modern efficiencies are enough. I agree, and that's why I don't lose sleep over this kind of thing. I mean this with all due respect, but many members of this forum simply obsess over non-issues. If you replaced my air filter once a week I wouldn't be concerned in the least.

Also for the record I do not trade in vehicles often. My household has two vehicles. One was purchased new by us in 2001 and the other is the truck in my signature that I haven't had all that long but is 22 years old and I will be keeping around. I know the previous owner and the air filter was replaced in normal intervals, not longer "loading" intervals. The truck runs fine, has strong compression, and does not burn a drop ofb oil which means that the filter changes have not caused excessive ring wear.

 

[Jim Allen]

I guess I have to explain further.

For every one like you, cheespuff, there are 10 reading this exchange who are in complete ignorance of the topic and have no idea of the science behind it. Is it better to teach them the whole shebang properly or just say, "don't worry about it?" If you have the knowledge then you can evaluate your own situation and make the right choices. Everything I said is factual and based on sound automotive science.

I also think you are missing another point. I am not advocating longer filter intervals just so it will get dirty and be more efficient. I am trying to help people not obsess over a slightly dirty air filter and change it too often, unnecessarily. Especially for some unquantifiable and unmeasurable non-goal as "responsiveness." Or thinking it's somehow "better" or "beneficial" for their car to do so. Clearly it is NOT beneficial to change the filter more often than necessary and it's a waste of money. All I am doing here is giving people a metric for the RIGHT time to change the filter.

I don't lose sleep over people changing their filters too often but I am often gobsmacked over people who argue in favor of wasteful and counter-productive maintenance practices.

 

[Errtt]

Curious, not time, but how many miles do you go generally go on a air filter.

Note: I do not check my air filter every oil change like most folks I know. I leave it alone as long as I can stand it.
Approaching 20k on current filter, going to check it next oil change. Have a new one in box on hand, so will see if I change it or not. If not, I might reverse clean it a little with low pressure air and hold it up against light (sky) to eyeball it.

 

[Shannow]

Jim,
really good, and well reasoned posts and information.

We had the big fires through here earlier this year, and I checked my filters when the weeks of smoke had passed, and they were utterly putrid.

Paper one in the Caprice I changed, and the AFE cotton on the Nissan I washed...and was appalled at how little "stuff" came out of it...lots of bugs, a bit of grey, but not much "stuff"

 

[Jim Allen]

Originally Posted By: Errtt
Curious, not time, but how many miles do you go generally go on a air filter.

Note: I do not check my air filter every oil change like most folks I know. I leave it alone as long as I can stand it.
Approaching 20k on current filter, going to check it next oil change. Have a new one in box on hand, so will see if I change it or not. If not, I might reverse clean it a little with low pressure air and hold it up against light (sky) to eyeball it.


Not sure to whom you question is directed but I guess I'll take a shot.

First don't clean it. This is big. There are some robust filters out there capable of being cleaned. They are mainly in the HD realms. Car and LD truck are more vulnerable to damage. You take a change of damaging it every time you blow thru, even if careful, and the damage might not be obvious. The only time I might take a chance is if I lived in a very dusty area where I loaded the filter up really fast. If you do it be VERY careful. You can displace fibers and make pinholes and not even see them. Over time these pinholes will let a lot of dirt thru but more importantly, they create channels where the air can take the path of least resistance and further enlarge the hole. I've had this explained in detail to me my the people who design air filters and they strongly caution against cleaning.

My suggestions are these:

1) Install a restriction gauge and change according to that.

2) If you don't install a restriction gauge:

a) Evaluate your environment. Is the air where you normally very dusty most of the time, modertately dusty, occasionally dusty or clean?

b) If you are in the clean realm I wouldn't touch the filter until it reaches the OEM FCI. Then check it.

c) If you live in a dirty environment, you may need to use the severe OEM schedule (not all mfrs. list that) or check often until you learn how often you must service the filter.Once you make that determination, avoid molesting the filter as much as possible.

d) Avoid molesting the filter as much as possible and be very very careful to make sure it seals well in the housing when you do (new or used filter). Every time you R&R the filter you risk a sealing issue that will allow dirty air to bypass the filter. That's why filter grease is a good thing. If you use it from the start, you make it easier to R&R the filter safely.

e) Some filter and housing designs are difficult to make seal well, others are well designed and fit like they should. Sometimes a particular filter brand will fit and seal better than another. You'll figure out which you have pretty quick if you pay attention.

 

[Jim Allen]

I wanted to add some pics and couldn't find them and my edit feature timed out. Here is Part 2 to above[
I am nearing 50K and 6 years on my Honda. The Ford gas pickup has an AEM cleanable that is about 4 years old but only has about 40K miles. The Ford diesel has an oiled cotton gauze filter ( : < { which was part of the Banks kit I installed in '87. I serviced it last in '09. All these rigs have restriction gauges and not anywhere near the point where they need changing/service.

e) It's dificult to tell the level of resriction on a fitler visually, even if it appears pretty dirty. I found a dirty filter in the trash at my local Ford dealer that fit my F150 housing. I installed it into into my housing and took the assembly to be flow benched. IIRC, it only flowed about 20% less than a new filter. Below is the pics of the filter test and the linked article has a pic of the filter.

Air Filters

 

[OVERKILL]

Great information Jim

 

[Vikas]

Quote:
IIRC, it only flowed about 20% less than a new filter.
To a layman, that means his engine is producing 20% less power or using 20% more fuel etc! Most do not understand the engineering or logic behind the necessary flow rate for the engine air filter. I know Jim has explained it multiple times but some still miss it. A simple statement such as until the flow rate drops below the maximum required by your engine, don't worry about it is a difficult concept to swallow. Especially for the guys who want to treat their "baby" right. I am talking about the guys who want to feed only the best premium gas that money can buy

 

[Jim Allen]

Originally Posted By: Vikas
Quote:
IIRC, it only flowed about 20% less than a new filter.
To a layman, that means his engine is producing 20% less power or using 20% more fuel etc! Most do not understand the engineering or logic behind the necessary flow rate for the engine air filter. I know Jim has explained it multiple times but some still miss it. A simple statement such as until the flow rate drops below the maximum required by your engine, don't worry about it is a difficult concept to swallow. Especially for the guys who want to treat their "baby" right. I am talking about the guys who want to feed only the best premium gas that money can buy


Perhaps true and more's the pity. Can't help those who won't learn.

 

[Shannow]

"Flow rates" are fairly arbitrary, as to flow, there needs to be a change in pressure.

Take a carb, they are flow tested at an arbitrary 1.5" water pressure drop across the carb...(IIRC, it's been a while)...so your 600 Holley can be ranked against a 750 etc etc. The two barrels are flowed at double the pressure drop.

If your engine doesn't move enough air to create the 1.5" pressure drop, even at wide opn, your 600 may only be flowing 400cfm.

Air filter testing similarly uses Differential pressure and flow...load it up, and the flow at the test DP will drop.

However, given the amount of air that the engine requires, the pressure drop that your engine sees at W.O.T. is likely below the test pressure drop firstly.

Secondly, nearly nothing is driven at W.O.T. on the street, so the throttle plate is already controlling engine load to a point below W.O.T. Throw in an extra inch or two of filter differential pressure on top of a mostly closed throttle plate, and the engine will never notice a dirty filter. (Turbo Diesels with ambient referenced wastegates will simply make up for the loading automatically).

 

[Jim Allen]

Originally Posted By: Shannow
"Flow rates" are fairly arbitrary, as to flow, there needs to be a change in pressure.

Take a carb, they are flow tested at an arbitrary 1.5" water pressure drop across the carb...(IIRC, it's been a while)...so your 600 Holley can be ranked against a 750 etc etc. The two barrels are flowed at double the pressure drop.

If your engine doesn't move enough air to create the 1.5" pressure drop, even at wide opn, your 600 may only be flowing 400cfm.

Well said.

Air filter testing similarly uses Differential pressure and flow...load it up, and the flow at the test DP will drop.

However, given the amount of air that the engine requires, the pressure drop that your engine sees at W.O.T. is likely below the test pressure drop firstly.

Secondly, nearly nothing is driven at W.O.T. on the street, so the throttle plate is already controlling engine load to a point below W.O.T. Throw in an extra inch or two of filter differential pressure on top of a mostly closed throttle plate, and the engine will never notice a dirty filter. (Turbo Diesels with ambient referenced wastegates will simply make up for the loading automatically).


Well Said!