Cold air intake, myth busters

[2totalled2months]

So today I was looking at adding a cold air intake and pe as well. While searching through youtube for sound clips I ran across this interesting video. I know its a video, probably rigged from the beginning but a few thoughts did come to my mind, one, nobody is giving them a kickback and they are not promoting a product, two, I have looked at afe, on their site they say 27 hp gains, hmm, but another video using their COI and PE and a throttle spacer didn't even come close on the dyno to 27. So, here it is, thought Id put it to the guys that have already been through the stages, done the dynos and don't get paid except when the push the right pedal, please comment away, my knowledge is slim.

[Vigilante375]

CAI's really don't do much unless you're car is making a buttload of power or you're taking it in that direction. It does free up some power but you have to be making a decent amount of power to see the real gains when the engine is wanting to breathe in big amounts of air.

I'm sure other will disagree but we'll see what they have to say.

Oh and nothing beats no air filter......that'll make the most power! If you dare to do so.....

[tracer bullet]

No surprises. They can free up power but only if it's there to be freed to start with. Some factory systems are pretty rottenly restricted with small inlets and a lot of bends, some start off fine. This would be a car that starts off fine. Not to mention the turbo, by the time the air has gone through that and an intercooler you're not seeing much more than just the standard variation in the dyno itself (note the different readings when no changes were made). Do this on an N/A car and you may see different results.

[BEAR-AvHistory]

Not a fan of aftermarket CAI. The stock CAI an provide more than enough air up into the 400+whp range.

[GeeRam]

Originally Posted by Ian///M

CAI makes far more sense on a NA car than a turbo car.

This.


Great mod for a n/a car.........relative waste of money on a turbo.

More efficent, better, bigger intercooler is where you should put your money, not an intake.

[Zeros and ones]

fair enough. Results are results and can not dispute. however, what about CAI system design for the car that you are driving by the actual manufacturer.

Yes, there is a large caveat of, well the manufacturer just wants to make money. But as an example, BMW make a cold air intake for the 1 series. states that it will increase HP by 3HP and torque by 3 ft/lbs.

Is this something that needs to be dis-regarded based on what this video shows?

This is not like a third party saying something like this. This is the actual manufacturer. I would tend to agree more with the manufacturer when they sell me a CAI that is produces more power than a third party such as AFe or even dinan.

But I can for sure tell you this. I put a the BMW Performance CAI in my 1 series. Not only does my but dyno tell me it is quicker but pretty much every person that has had a before and after ride in my car stated it was quicker.

In my Dinan 5, with Dinan CAI, I can say that I lost low end torque off the line but the car pulls much harder in the higher range.

Never the less... good video

[CrustyNoodle]

Great thread!

There are really only two valid reasons why you would want to change the factory fitted intake:

1. To reduce intake pressure loss,
2. To reduce intake air temperatures.

Before you go making changes, you really should first understand if there is any potential for improvement. This would mean taking some measurements of what you currently have. If you do decide that there is potential for improvement and you make a change then you should re-do the measurements to ensure you actually get the improvements you are after.
Here are some articles that explain it all in very digestible detail.

http://www.autospeed.com/cms/article.html?&A=0629

http://www.autospeed.com/cms/A_107824/article.html

[SRT-What?]

it all comes down to surface area more than "hot air".

I have this book at home ("how to make horsepower" or something like that) and the author provides a empirical formula about how much filtering surface area is required for different power levels.

[Stan.mac]

Everyone has good points but I have yet seen someone mention the most important thing besides the true 3-7hp gained by a CAI... What about the 2000hp gain at the ear drum? OMG, the sound of the turbos spooling that is only produced by the aftermarket CAI's are unbelievable! This is the only reason I've purchased a CAI.
Another point to make about CAI's, off the topic of HP gains are, the cost savings of not having to buy a new filter every couple of months, depending on your driving habits and local conditions. Having the ability to make the purchase of one filter and having the ability to clean it when it get dirty is rather appealing; but there's always the K&N drop in.

[2totalled2months]

Wow, great responses, wish I could quote everyone. I have done some research since the post and everyone here seems to be right on. I noticed on the Nissan that it was turbo but more importantly it had an aftermarket catback exhaust and the factory scoop for the intake. Point being that Nissan had already made an effort to get the most air in and the owner had dealt with releasing back pressure to let the air out. So putting another COI didn't change much. I think with the 128 version that BMW left room for improvement and offers an upgrade up front, with the 135 they don't offer COI but they do offer a pretty good PE. This kinda makes me wonder if on the 135 if a COI is worth it or not, other than the awesome sound described by some. I know **********s has a dyno video where they install on a 135, PE, COI and throttle body spacer by afe, if you add up afe's claimed gains of each one they would total around 40hp, I believe they dyno around 15 after they installed them. My guess is without having anywhere for the air from the intake to go,,,ie dp and exhaust then a COI wouldn't make much gain. I can see one of the posters points though, you have to have the power to need that much air...

[Freon]

Colder air is denser. For NA cars, the density increase will result in almost the same percentage power increase. I'm not going to bother with the math right now, but you can look up the ideal gas law. Pulling a number from vague memory, I think something like 30F of temperature drop is around 5% air density. I think it is more or less Airtemp1 dividied by Airtemp2 (in Kelvin). So going from 100F to 60F is (310, 294) = 1.054, or about 5.4% change. I think I'm at least close with that...

Turbo cars have intercoolers which can bring air temperature of hotter air intakes in the engine back back closer to ambient (outside) temperature. Note "can." If the intercooler is already strained it won't help much. It will depend on a lot of variables. Intercoolers have thermal mass, so without a few hours to think about it I couldn't give you a good model, but it would include a LOT of variables. So if you drop intake temp by 30F, actual temperature change after the intercooler into the throttle body may only be 10F less, leading to minor gains. Again, things can be complicated. You can heatsoak your intercooler just sitting idling in traffic if you are pushing hot engine bay air through it from the inside. But the radiator fan may help. Again, very complicated, good luck coming up with a good mathematical model that will work for all cars.

A cold air intake may also just increase power due to a larger, less restrictive air handler. The filter, the box, etc. Usually a larger filter element is less restrictive. So is a fresh, clean airfilter, so swapping intakes and gaining power may just mean you have a clean filter.

The whole thing becomes complicated again with modern turbo cars. They often have load targets like our fancy BMWs. Any gains with a less restrictive intake or colder air may be negated by the computer backing off on boost. It is reading air temperature and pressure in many locations and also knows the VE of the engine, or uses a mass airflow sensor. It is trying to make sure a given mass of air is in the combustion chamber, and that will hide a lot of inconsistencies with physical setup.

I very much doubt NA cars have load targets in the computer, so I would bet they will show the air density gains expected with both colder air and less restrictive intakes (denser due to less pressure drop across the intake system). I doubt on particularly cold days with no air filter that they purposely close the throttle to limit power. Unlike turbo cars, runaway conditions are unlikely, so doing load targets seems silly. The only load control is the throttle plate and cam position. I guess it is possible there are minor tweaks with cam position for air density, but I also would guess they're always trying to maximize airflow and power.

[brusk]

Their test car sucks as a choice as it makes very little power to begin with. Most factory air boxes already pull from a cool source so unless they a restriction I don't see any gains. Had they gotten something with a 2.0 or bigger that would be a better test. Another thing people aren't taking into consideration is every type of car is different. Some with restrictive air boxes some with ECU/PCM's that just do not like change. My beater Corolla will not allow a CAI the computer completely freaks out but others have swapped to a newer year ECU and found decent gains.

[fullboost]

CAI's do very little on most cars nowadays other than just being a noise maker. A sweet, sweet noise maker.

[Zombie1]

Originally Posted by fullboost

CAI's do very little on most cars nowadays other than just being a noise maker. A sweet, sweet noise maker.

Yup

[Freon]

Originally Posted by brusk

...Most factory air boxes already pull from a cool source so unless they a restriction I don't see any gains.
...
Another thing people aren't taking into consideration is every type of car is different. Some with restrictive air boxes some with ECU/PCM's that just do not like change. My beater Corolla will not allow a CAI the computer completely freaks out but others have swapped to a newer year ECU and found decent gains.

Very good points.

I agree factory airboxes usually pick up air from a good source. Fender or front grill. The N54/N55 have really large airboxes with large filters as well, so they don't get dirty as fast and don't create the restriction a smaller filter might.

Often the CAI really screws up the mass airflow sensor reading. I had been tuning Subarus for many years and it was a real crapshoot on trying to calibrate the MAF sensor table depending on what kind of intakes people had. Some were fine, some were a nightmare and never really read consistently. There is a very real possibility there is no way to properly calibrate the MAF sensor if the intake is designed poorly. It may never have a transfer function against the metrics that the ECU takes into account (usually just voltage, maybe a few minor tweaks for barometric pressure and temperature). Air might be swirling around...

I generally told people to keep the stock airbox and just change the filter every 5k instead of every 10k, and to replace the intake pipe after the MAF before the turbo, which was a crappy piece from the factory. But the airbox and MAF housing were usually best left stock for the best overall behavior of the car. Yeah, bigger turbo guys would want to replace it eventually as the filter wasn't that large (maybe 1/2 or 2/3 the size of an N54/N55 filter), but there wasn't much need for a normal WRX.

I have a dual cone intake, but I'm really not diluting myself into thinking it was a huge power gain. Some piping was removed and I probably increased filter area by 30-40%, but it sounds a lot cooler. I logged intake temps and they're really just fine so I don't sweat that.

[2totalled2months]

Originally Posted by Freon

Colder air is denser. For NA cars, the density increase will result in almost the same percentage power increase. I'm not going to bother with the math right now, but you can look up the ideal gas law. Pulling a number from vague memory, I think something like 30F of temperature drop is around 5% air density. I think it is more or less Airtemp1 dividied by Airtemp2 (in Kelvin). So going from 100F to 60F is (310, 294) = 1.054, or about 5.4% change. I think I'm at least close with that...

Turbo cars have intercoolers which can bring air temperature of hotter air intakes in the engine back back closer to ambient (outside) temperature. Note "can." If the intercooler is already strained it won't help much. It will depend on a lot of variables. Intercoolers have thermal mass, so without a few hours to think about it I couldn't give you a good model, but it would include a LOT of variables. So if you drop intake temp by 30F, actual temperature change after the intercooler into the throttle body may only be 10F less, leading to minor gains. Again, things can be complicated. You can heatsoak your intercooler just sitting idling in traffic if you are pushing hot engine bay air through it from the inside. But the radiator fan may help. Again, very complicated, good luck coming up with a good mathematical model that will work for all cars.

A cold air intake may also just increase power due to a larger, less restrictive air handler. The filter, the box, etc. Usually a larger filter element is less restrictive. So is a fresh, clean airfilter, so swapping intakes and gaining power may just mean you have a clean filter.

The whole thing becomes complicated again with modern turbo cars. They often have load targets like our fancy BMWs. Any gains with a less restrictive intake or colder air may be negated by the computer backing off on boost. It is reading air temperature and pressure in many locations and also knows the VE of the engine, or uses a mass airflow sensor. It is trying to make sure a given mass of air is in the combustion chamber, and that will hide a lot of inconsistencies with physical setup.

I very much doubt NA cars have load targets in the computer, so I would bet they will show the air density gains expected with both colder air and less restrictive intakes (denser due to less pressure drop across the intake system). I doubt on particularly cold days with no air filter that they purposely close the throttle to limit power. Unlike turbo cars, runaway conditions are unlikely, so doing load targets seems silly. The only load control is the throttle plate and cam position. I guess it is possible there are minor tweaks with cam position for air density, but I also would guess they're always trying to maximize airflow and power.

Awesome information, I knew the right people would pop out on this topic. ??? then, if the ECU changes turbo pressures to compensate, is that the same with a tune like COBB, my intial plans were to add a PE, COI and COBB, Im considering eliminating the COI if not needed, but if I do a tume like COBB will the COI be beneficial other than sound. Ill prob go with BMW PE so no worries about warranty and hopefully the COBB will reset undetectable by the dealer. Im not wanting a track car just one that will take most street competitors, for now at least until warranty is up. I have the PPK2 on there now.