Stock AirBox VS. Cone Filter

[livestrong14]

Which is better? and why? I plan on doing the coffee cup mod either way.

Your thoughts?

[555BCTurbo]

Well...as I am sure you have read...Josh (Legacy777) pointed out that the stock airbox has a small velocity stack coming into the intake, which speeds the air coming in and reduces turbulence pre-maf...

Unless you build an enclosure, a cone filter will just allow hot air to come into the intake, effectively losing power.

The coffee cup mod is a great idea, I have done it, and it sounds great (that's about all it does, but it doesn't hurt anything).

Also, my airbox was modified with the removed snorkus and a 3" ABS bellmouth coming into it, and a K & N drop in filter...

This setup seems to work well for me...and any others who are using a similar idea

I personally think that cone filters are ricer, unless they are installed in the correct way (isolated from the engine compartment), which happens like .5% of the time...

So, my $.02 is to leave your stock airbox and just do the mods that I suggested. You've ridden in my car Devan, so you know it works

[wallgood358]

The only way to have a cone filter work properly is if it is either heat shielded or you route it into the fender wall where the snokus was. But there is still the velocity issue. I am not fully sure as what the effects of high vs low velocity intakes but it is said that the higher velocity can hurt Hp. I would recommend getting a later model legacy air box where the Maf comes from the rear instead of the side so you can get smoother flow and drop in the K&N. Still sounds just as throaty as the cone filter as long as the snorkus is removed

[farfrumwork]

My '93 SS came with a shitty weapon-r cone filter hanging off the MAF housing. It sounded cool, but I ditched it within a few weeks (luckily he had the stock box as well)

I had MORE power with the stock box, especially low and mid rpm. My mileage also went up from 19mpg mixed to 21-22 mixed. I'm sure it was freaking the MAF out and throwing more fuel than neccesary.

Just de-snorkle, and keep a clean panel filter in there. K&N's are fine and all, but a clean paper panel will be on-par (you just need to replace them often - I do mine every other oil change). I also have cool air at the air-box from the fog light opening - I cut out the opening and routed 3" steel tubing from there to the upper part of the fender well (pointed at the entrance to the stock air box). My inlet tubing is COLD.

[Soul Shinobi]

I've desnorkled and cut the plastic placeholder from the fog lamp socket, but am yet to work some tubing in. I'm a bit worried that it might rain hard and I like driving fast through puddles, so I'm trying to think of a way to work in a fool-proof water escape.

[ramathorn]

im pro-cone for a few reasons.

cleans up engine bay
makes more room for fmic piping
more surface area for air to come into
i like the sound :rolls eyes:

[555BCTurbo]

more surface area for air to come into

Complete BS...

[biggreen96]

more surface area for air to come into

Complete BS...

I dunno, my cone looks alot larger than a stock filter.

[ramathorn]

more surface area for air to come into

Complete BS...

im not a math guru or anything thing and i honestly dont believe what i said is bs. that being said i went outside and took some measurements of just the air filter area (only the part where the pleats are) on my k&n cone filter and the the factory panel filter. plese correct me if im wrong on finding the surface sq. in. because like i said im not a math guru and i THINK what im looking for is comparing the square inch of the two filters.

the factory air filter is 5.5"x9.5" = 52.25 sq. in

my open element cone filter is 4.5" (height) x 18.5" radius = 83.25 sq. in

did i do my math incorrectly?

[Soul Shinobi]

It would be more complicated calculating the cone filter if it is indeed in a cone shape, as it would have a different radii at each end. But if it's a cylinder then, well you'd still be wrong because 18.5" definitely sounds like the circumference, not the radius. You'd need to convert that. If you post the height (measuring with the measuring tool flat against the cone) and the top and bottom circumference (measuring around) I will calculate it for you.

And after that, the results would only be a ballpark idea, as we haven't taken into account the depth of the pleats.

EDIT: Here's a page with the formula if you wish to take a look. Post your measurements so we can double check.

http://mathworld.wolfram.com/ConicalFrustum.html

[ramathorn]

okay i got a couple terms mixed up so the cirumfrence is 18.5" which is irrelivant from what you are saying, sorry for the confusion.

okay just because im bored and had nothing elese to do i did go out and measure the depth of the pleats on each filter and counted the pleats

OEM pleat depth: 3/4"
K&N pleat depth: 9/16"

OEM pleat count: 49
K&N pleat count: 60

i dont really think the radius is the right demisnion were needing in this instance i think that the diameter would be the correct one.

K&N radius 2 3/4"
K&N diameter 5 1/2"

[Legacy777]

The easiest and most conservative way to calculate the surface area of the cone filter, is to take the radius/diameter (doesn't really matter) of the smaller end of the cone, as well as the height, and input it into the equation for the surface area of a cylinder
http://www.math.com/tables/geometry/surfareas.htm

It will be a little conservative, but will still indicate that there is more surface area then the panel filter based on a flat surface. Once you start bringing pleat counts in.....that may change things, but I don't think it will be too drastic.

[Soul Shinobi]

Okay, before I can crunch numbers I need to make sure I've got these values right.



(if you're having trouble seeing that picture then look at the black and white drawing on this page)

So, you can see the necessary values are h (height), R2 (top radius), and R1 (bottom radius). Also, measuring s (slant height) would be helpful, but not necessary because it can be calculated using the straight height and radii.

So we need:
h = 4.5 (unless you measured the slant height, then this would be s and you'd have to measure h)
R2 = 2.75
R1 = ?
s = ? (unless this is what you gave me for height, this measurement is optional because it can be calculated using height and the two radii)

If you have trouble getting the radius of either the top or bottom (maybe you don't feel like taking it off to measure it) then you can give me the circumference and I can calculate the radius from that.

I should be able to work in the pleat count and depth after I get the general surface area.

[evolutionmovement]

Couldn't you just figure out what the cone would be if the length continued into a true cone and then subtract the volume of the missing part from the total?

[ramathorn]

i forgot to mention that my filter is not a cone. it is a cylinder.

also just a side note on the discussion of cone/ cylinder/ open element filter. when i was taking measurements i noticed that the factory filter had 1 large dark spot on it from the incoming air being forced into that 1 spot ill take a pic of that as well. that would deffinatly contribute to the flow being concentrated into one spot. since its being forced into that one spot its obvious that the surface area isnt much of an issue until the filter becomes dirty.

also i hear alot of people talking about a velocity stack in the facory airbox and how it it makes a huge difference on how airflow comes in. from looking at my factory airbox the velocity stock is just before the maf which really isnt helping that much anyhow (it MAY help but, not to the extent everyone is making it out to be. the stack sits about 1/2'" away from the plastic of the air box so isnt it air coming in passing through the filter getting tossed around not a good deal not to mention the top half of the air box after the filter is sectioned up into 3 rather triangles rather than at 1 point. to me it seems that the complete airbox in general creates more turbulance than alot of people think. thats why when you put just an open element filter on your stock piping intake you can literally HEAR the intake doing more as its not being restricted from the ribs, different angles and everything elese from the airbox.

really the only thing the airbox has over and open elemet filter is the heat shielding (which is marginal at best). also which is debateable and im sure will be the next installment of this debate. ill chime in on this in a little bit because its time to eat right now.

[Soul Shinobi]

Couldn't you just figure out what the cone would be if the length continued into a true cone and then subtract the volume of the missing part from the total?

I thought about it, but there wouldn't be enough information.

i forgot to mention that my filter is not a cone. it is a cylinder.

Oh Christ that makes things easier. In fact your original math, multiplying the circumference by the height, was correct.

So as you said before:

factory air filter is 5.5" height x 9.5" length = 52.25 sq. in
open element cone filter is 4.5" height x 18.5" circumference = 83.25 sq. in

I'll use these values.

OEM pleat depth: 3/4" (0.75)
K&N pleat depth: 9/16" (0.5625)

OEM pleat count: 49
K&N pleat count: 60


To get total surface area I'll multiply the height of each filter by the pleat count by the pleat depth twice (because each pleat has two sides). A formula would look something like: H x pleats x pleat depth x 2. The reason I'm ignoring length is because the pleats being scrunched up makes the value useless.

OEM = 5.5 x 49 x 0.75 x 2
K&N = 4.5 x 60 x 0.5625 x 2

Oh man, honestly this looks close. Results:

OEM = 404.25 sq. in
K&N = 303.75 sq. in

Wow, I'm quite surprised. The K&N filter really does have 3/4ths the surface area of the OEM filter; the pleat depth and height of the K&N hurt it. And yes I did check my math, several times, so long as ramathorn's measurements are accurate, these numbers don't lie.

However, this by no means equates to air flow. Bear in mind that the paper element of the OEM filter doesn't flow as much as the oiled cotton suspended in a wire frame that the K&N has. But much more important still is the restriction caused by the shape of even the desnorkled OEM intake. The inadequacy of it can be seen in ramathorn's pictures.

Which is better? Well, I think someone with the time and pocket change to hit the dyno a couple time will have to tell us.

EDIT: A note on ramathorn's pictures: if any of you find your filter half that dirty, replace it NOW.

[555BCTurbo]

I'm thinking of hitting the dyno sometime soon with my modded stock airbox setup, and then rig up some kind of cone filter setup, and truly see what is better

[ramathorn]

if i had an awd dyno id deffinatly do it but i dont. only fwd dyno's in my area .

also just a not that is on a desnorkeled intake system so you guys that are running that setup check your filter often.

math may not lie and ill be the first to admit i was wrong on my statement of the open element filter having more surface area. but (and theres always a but) im still not backing down that a open element is better than the stock air box for flow.

[Soul Shinobi]

If the tubing from your K&N to the throttle body is after market, that would make a bigger difference than most people would suspect. Not counting the snorkus, the cramped stock airbox is surely a bigger restriction that the filter itself.

If any of you haven't, it's worth reading the Eliminating Negative Boost series on AutoSpeed. It's about what really causes intake restrictions and how to test for and eliminate them.

http://autospeed.com/cms/A_0629/article.html

[555BCTurbo]

Good job with the calculations Soul!

[BAC5.2]

I was really worried that you guys were forgetting that a cylinder is simply a rectange rolled up.

The numbers are a little off, I think. But not by much. By two pleats at the max, and it's probably the same for both, so that's fine. Where the material attaches to the frame, there is only one side of the pleat. 60 pleats, 2 of which are only one-sided.

Otherwise, your good.

Surface area of a cone isn't really that difficult. Isn't it just (pi)R X rte(R^2H^2)?

If you know the radius of the cone and the height of the cone, that should be it.

Since the cone filter usually has the top lobbed off, you simply subtract the surface area of the cone that isn't there, following the same calculation. You simply need to measure the height of the non-existant cone, which would be simple enough with two rulers.

[ramathorn]

.

Surface area of a cone isn't really that difficult. Isn't it just (pi)R X rte(R^2H^2)?

If you know the radius of the cone and the height of the cone, that should be it.

Since the cone filter usually has the top lobbed off, you simply subtract the surface area of the cone that isn't there, following the same calculation. You simply need to measure the height of the non-existant cone, which would be simple enough with two rulers.

with alot of cone filters the top part that is lobbed off is really flipped/ folded inwards to make another inwards facing cone. by doing this its essentially the same thing as the velocity stock.

[Soul Shinobi]

I was really worried that you guys were forgetting that a cylinder is simply a rectange rolled up.

The numbers are a little off, I think. But not by much. By two pleats at the max, and it's probably the same for both, so that's fine. Where the material attaches to the frame, there is only one side of the pleat. 60 pleats, 2 of which are only one-sided.

I did go out of my way to find the formula specifically for a cone with the top lobbed off, it wasn't too complicated.

I did consider where the pleats attach, but I don't even know if he counted those. I didn't bother asking because the difference would be trivial.


I was going to ask what a velocity stack was, but thankfully Wikipedia had an article. For those of you who don't know, look here:

http://en.wikipedia.org/wiki/Velocity_stack

[555BCTurbo]

by doing this its essentially the same thing as the velocity stock.

A filter most certainly wouldn't get the same swirl effect that a steel or aluminum velocity stack would

[Legacy777]

Some comments....because I like this issue

Soul Shinobi's calcs are interesting. I do feel that a k&n cone filter will probably flow the SAME as stock paper filter, but have less pressure drop across it at a higher flow rate. I say it that way rather then flowing more for the reason that each filter element will have a corresponding flow and pressure drop curve associated for it. The pressure difference won’t be much and is typically measured in inches of water.

Depending on the filter element & style, the curves should be pretty linear, but may not exhibit any substantial pressure difference until the flow rises beyond a certain point. The point I’m trying to make here is that an engine will have a relatively narrow operating window compared to just putting these filters on a flow bench that measures flow & pressure difference. So if your engine has an operating window where both filters have nearly the same flow & pressure difference, then the discussion about which one flows more doesn’t matter.

In regards to the dark spot on the filter, that is obviously where the intake tube dumps into the box and where the dirt particles get trapped on the filter. That does not mean that’s where all the air is traveling through the stock filter. When the air exits the intake tube into the air box, it will expand to fill the air box chamber. When you get a volume change like that, velocity will decrease. Air has a smaller mass then dirt particles, and can slow down much quicker…..so in my mind, that’s why you see the dirt spot in that one spot on the filter.

In regards the velocity stack. I think there may be some misunderstanding as to what its purpose is. Its purpose is to smooth the airflow, and help the flow to be as much laminar as possible. The reason it’s doing this right before the MAF sensor is to provide a flow profile that is more accurately measured so the ECU can more accurately determine the amount of fuel to inject. The velocity stack really isn’t going to do squat to increase flow. It’s there to improve air metering.

So that’s a lot of theory crap. Let’s talk real world. I ran a cone k&n style filter for probably about 3-4 years. In that time the car saw 3 cross-country trips (3000 miles each) and reasonable amount of city & highway miles. My experience was while my car was normally aspirated. I did experience at least what felt like to me an increase in how quickly the engine rev’d in the higher rpm’s. I also experienced some bogging type issues at the low/mid range rpms that I didn’t notice with the stock air box. Also during that time I experienced intermittent engine stalling/sputtering, which was caused by a bad MAF sensor. I contribute some of its failure to the k&n style pod filter. Lastly, when I put the stock airbox back on, I felt a pretty dramatic increase in low-mid range torque. This is something that you almost have to drive the car for a while to become accustom to it, and then switch back to the stock airbox.

My personal opinion on the whole intake thing is to run a modified stock airbox with extra inlets on the bottom, and run a paper filter element. If you look at the intake system the inlet at the bottom of the airbox is the smallest opening, increase that, to equivalent or larger then your intake tube piping, and that point should no longer be your greatest restriction. I just added a second intake tube pipe, and have been very happy with it. I run a paper filter element because of the potential for the oil droplets messing up the MAF sensor. If the paper filters are changed regularly, they flow just as much as a k&n, and filter dirt MUCH better.

That’s my little (which I use the term loosely) comment on this issue.