air filter vs fuel economy (tricky question?) 11

[ivymike]

Perhaps this is a tricky question...

Does having a dirty air filter hurt the fuel economy of a modern gasoline-fueled car? (Obviously the parts stores would have you believe that it does)

If so, how exactly?

If it's inlet restriction, then would you say the effect is the same as modifying the throttle actuator such that the throttle opens slightly less for any given pedal input than it would otherwise have opened? If this is equivalent, would you expect to get better or worse fuel economy as a result of such a modification?

 

[ivymike]

messed up the last calc a bit, and I'm too tired to go fix it. I think the downstroke work is more like -21.3J for the crazy events version, which makes it slightly more efficient than the throttle version.

 

[cheapcharlie]

i want some of what ever it is jsteve is on

 

[Lou Scannon]

meetoo!

 

[JSteve2]

Sorry guys. Can't ship it across state lines.

 

[patprimmer]

Jsteve

Most of the guys you are arguing with are degree qualified professional engineers who design engines for a living. They are unanimous in their disagreement with you.

What do you do for a livening.

What have you done to qualify you to continue to disagree with such an overwhelming opposition.

Regards

eng-tips, by professional engineers for professional engineers
Please see FAQ731-376 for tips on how to make the best use of Eng-Tips Fora.

 

[ivymike]

so anyway, greg, I think the text version is that in one case all the expansion work is disippated via friction at the throttle and other inlet restrictions, where in the other case some amount of the expansion work is recovered by the piston. I'll do a PV diagram if I get a chance at the airport this morning (not sure whether they'll have free wifi).

 

[patprimmer]

I always thought that moving the throttle to the valve was a publicity stunt. A throttle is a throttle is a throttle.

The real advantage I saw was the increased effective compression ratio due to earlier intake closing and the extra power stroke from later exhaust opening and reduced overlap at TDC at low "throttle" opening. I personally prefer the way Honda does it with the Ivtec where they run 2 valve, small cam 4 valve then big cam 4 valve

Regards

eng-tips, by professional engineers for professional engineers
Please see FAQ731-376 for tips on how to make the best use of Eng-Tips Fora.

 

[JSteve2]

Patprimmer,
Also a degreed engineer working on engine controls. I spend my days challenging really smart people with tough questions like this. Really smart people - including myself (if I'm in that category) - are wrong a lot or they aren't doing anything difficult.

I'm not sure anyone has described the current system well enough to claim victory here, and we have no data of any value (see ivymike's post citing largely anecdotal evidence on both sides).

I don't see consensus as conclusion - merely that is the direction to move forward until better evidence or models are available. Ivymike's own post of external "evidence" is 5-1 on the fuel economy side, although the fuel economy fans from that post are just as worthless as the one consumer reports test with undisclosed conditions.

So if you're looking for me to concede in the face of overwhelming opposition by smart people - fine I concede. But please don't expect me to write a white paper declaring that we've shown dirty filters just degrade available power but don't affect fuel economy.

 

[Lou Scannon]

JSteve,
Can you show or explain to us on a PV diagram (including pumping loop) what you're trying to say.

Pat, if the intake is closed earlier (vs being used as a throttle) then you avoid some of the pumping loss, since leakage aside, you have an air (vacuum) spring helping the piston back up at the beginning of the compression stroke. I believe this type of intake valve timing is referred to as a modified Atkinson cycle, no?

 

[JSteve2]

hemi,
I would expect a restriction to be reflected as a lower cylinder pressure on the intake stroke of the pumping loop, effectively increasing the PV area of the pumping loop and thereby requiring more work to complete the intake stroke.

 

[ivymike]

Okay, sketched a PV diagram for just intake and part of compression for throttle vs intake valve comparison (Greg's comment). Pink shaded area is "extra" work captured by the piston if the inlet restriction happens due to early IVC. The valve events are, of course, idealized, and again I didn't bother putting any kind of static restriction in the intake system. Red line (starts at higher pressure) is for the intake valve case, blue line (starts at lower pressure) is for the throttle restriction case.

 

[drwebb]

Will the dirty filter heat up any due to increased friction of the air flowing faster through it? That could thermodynamicly account for the reduced engine efficiency.

 

[JSteve2]

drwebb,
"the reduced engine efficiency" is of course a point of dispute. I assume you're going along a bit trying to figure out how it could occur if at all.

I would expect the effect you suggest to be extremely minor and easily measured. But again we're freewheeling without data.

My current thought is that if there is lower efficiency at all, it is dominated by the PV area effects due to overall higher intake restriction at higher air flow rates.

 

[cheapcharlie]

I am not an engineer and may have no buisness here. I come here to learn not teach.As hard has as I try I cannot see where jsteve is coming from.The original post was for modern gas engines.Older carburated engines with poorly enginered, adjusted or malfunctioning vents were sensitive too dirty air filters, modern speed density fuel inj. engs are not.Power output would only be affected if restriction was in excess of other factors such as valve timing intake design etc. Correct me if I,m wrong

 

[Lou Scannon]

Correct me if I'm wrong: the only one who thinks you are wrong is JSteve2.

JSteve2, how is the "restriction to be reflected as a lower cylinder pressure on the intake stroke of the pumping loop" any different whether it comes from the throttle or the air filter?

 

[JSteve2]

hemi, this will just summarize what I've already posted. First, if the IMP is the same, there can be no difference as far as I can tell.

This is how I'm thinking right now:
We're fixing MAF. Filters and throttles have different MAF v. pressure drop curves. At some value, those curves will cross - say at 3 lbm/min in a given engine. At points away from that MAF (most interestingly above it), one of them (filter or throttle) will have a higher pressure drop value, and therefore a lower IMP. Since the piston is capable of applying variable work, there is nothing in the system that can enforce a uniform IMP, as the driver is closing the loop with the throttle on MAF, not on IMP.

Whichever case has the lower IMP will see a lower pressure in the cylinder during intake, and therefore the area of the pumping loop on the PV will be larger (this area is work input by the crankshaft, as opposed to the combustion loop which is work output to the crankshaft).

The idea that IMP must be equal at equal MAF is understood, but I don't see how it can be correct. Again - the proposed restrictions have different curves. If the driver keeps opening the throttle to achieve the same MAF (ie power), the engine can achieve that through an increase in engine speed just as well as equalizing pressure drop. We have two levers with different response relationships, so although MAF-IMP may match for some range of points, they cannot match for all operating points.

I understand that no one has been convinced. However, neither has anyone expressed an understanding of the above view while refuting it.

 

[GregLocock]

I strongly suspect the last sentence is very true.

Thought experiment, written in English:

Imagine an intake duct with no air filter, but two butterfly valves in series. Set the first butterfly valve to the same pressure drop as the dirty air filter, use the second one as a throttle. For a given power at an engine speed the pressure after the throttle body will be "P"

Now adjust the first butterfly valve to the same pressure drop as clean filter, and set the throttle to give the same power at the same speed.

What is the pressure after the throttle body?



Cheers

Greg Locock

SIGlease see FAQ731-376 for tips on how to make the best use of Eng-Tips.

 

[Thecardoc]

Ohh this is just too good to let alone. I've been really busy in the shop lately and even today had to fly through most of the responses gaining just enough information to attempt to catch the highlights and make some comments.

Remember this, "I'm not an engineer, just a technician that fixes the things automotive engineers create"

A DIRTY AIR FILTER WILL NOT AFFECT NORMAL DRIVING FUEL ECONOMY ON TODAY'S MAF (Mass Air Flow) sensored vehicles. There was a lot of disscussion, but little spelled out for Jsteve to go and test to prove the result. One of the tests that techs like myself use regularly when in the process of diagnosing a vehicle performance issue is to calculate the "reported VE" (Volumetric Efficiency) of the engine at several WOT engine speeds. We use this calculation combined with fuel trim data, (long and short term) to help analyze MAF sensor performance, which is easily confirmed to be accurate or not based on VE variance, combined with an associated fuel trim correction. In short, if VE is low say 60-70%, and fuel trim is making a correction to add fuel, that indicates unmeasured air is entering the engine. The air could be pirate air getting in through a torn air inlet tube after the MAF, or could be a faulty MAF. If VE calculates low, and there is NO fuel trim correction, now we look for a mechanical reason that the engine isn't breathing correctly. That could be a restricted exhaust, engine mechanical issue, or in fact an air filter issue!

In all of my years as a technician (30+) the number of times that I diagnosed a bad air filter as a performance problem accounts for only a handful. Most of them were related to rodents storing food in what appeared (to them) to be prime real estate. VBG. Almost every one of the rest were related to an exhaust manifold or gasket failure, where the thermostatic air cleaner system was picking up the leaking exhaust gasses and coated the air filter with them.

The fact that an air filter that is truly dirty is a diagnoseable condition without lifting the air filter lid and looking is important. That also means that no matter how dirty an air filter looks when inspected, proper testing can prove beyond any doubt as to whether it actually needs replacement.

From this point here is your test Jsteve. If you don't have access to this formula e-mail me and I can send you a version of it. You will need a good scan tool that will give you data as fast as possible. The faster that the data scrolls, the more accurate that the RPM, VS the Airflow reading in grams per second will be.

Go ahead and take your first VE measurement with whatever air filter you have in the car. Take note of long term and short term fuel trims under normal driving conditions at varying vehicle speeds, and throttle openings. Also take note of the calculated spark timing, and if there is spark retard occuring from a knock sensor input. (MAF vehicles use the MAF to assist in not only fuel mapping but spark timing mapping as well) Then take the air filter out for testing purposes and repeat the test! Make sure that your air inlet assembly is clean of any deris etc.

Now the fun starts. Take a can of spray paint and a cheap replacement air filter. Paint 1/4 of the filter to completely block it of airflow. Install the filter after it dries and re-test. Use an enamel paint to make sure it seals the filter. If you have access to an O.E. level scan tool such as an IDS on a ford you could go ahead and drive some type of a course and measure actual fuel economy. In fact backing up you could have started out this test that way!

Now paint another 25% of the filter. You now know without question that you have a filter that is 50% restricted, right? See how this measures out now.

Now go another 25%, so your filter is definately 75% restricted. Do you think you have ever driven a car with a filter that was THAT dirty?

You now have a filter that only 25% of it has airflow. Paint half of what is left and repeat the test, with only 12.5% of the filter allowing airflow. Depending on circumstances I'll wager this to be the first time you see a significant difference in MAX VE. However, I'll also wager you will not yet see a loss of fuel economy, and you could in fact lend this car to another person and they could drive it and not notice anything wrong!

The reason for all of this is gasoline engines are of course an air pump. The speed of the engine is set by the airflow that it recieves, and it does NOT matter where the restriction to that airflow is. The greatest restriction will always be the limiting factor. If you would take my Ford Explorer and restrict the filter to about 12.5% area, you would still have 80% of the cross-sectional area of the throttle body and therefore almost no noted restriction.

Now one of the other responses was asking about variable cam timing. We should start a thread specifically about that, instead of doing it here. Suffice it to say at the moment that scavenging of the cylinder gasses is only the tip of the iceberg. The engines actually effectively control displacement, eliminate the need for EGR, since the cam can do that job, and much more!

 

[JSteve2]

Thecardoc,
If you had put this up a while ago, you could have saved me a beating.

That's good enough for me - I'm going to accept that and yield.

 

[Lou Scannon]

A virtual beer for JSteve and the rest of the crew for sticking with it!

Cheers,

Geoff.

P.S. And a virtual star for you, cardoc!