The Cold Air Intake: World’s Most Popular Drop In Mod Finally Busted?

By | June 9, 2014

Cold Air Intakes Don't Work

Image credit:

Back when I first graduated from college, I bought a 2001 MX-5 to celebrate my leaving of classrooms behind forever. It was a wonderful car and lived up to every bit of hype that has since made the Miata the so-called “answer to everything”. But just like every other Miata that came before it, the power of the car left quite a lot to be desired.

So off to the forums I went to find the best recommendations for easy mods to give my car that extra bit of oomph I so desperately wanted.* One of the first and most repeated recommendations was a cold air intake (CAI) to replace the stock air box and filter. CAIs are of course available from almost every aftermarket parts manufacturer, some reasonably priced and some that would break the bank.

The concept is simple: taking your stock airbox and traditional air filter out of the car and replacing it with a higher flow cone (or pod) filter that is positioned every so slightly closer to the outskirts of your engine bay will promote a more consistent draw of colder air into the engine. Cold air of course is more dense than warmer air, bringing in more oxygen molecules to the combustion process in the engine and resulting in a more powerful explosion of compressed fuel and air.

Sounds great, right? After all, handfuls of forum dwellers claimed that CAIs added 15+ hp to their Miatas. But 15 more ponies was nearly a 10% increase power… I was skeptical. If this easy upgrade to the car had such a positive impact, why didn’t the manufacturer do it in the first place? Was the power gain actually real or just a desired mental outcome? Where was the proof?

To me, it seemed that moving an air intake a few inches to the side and taking it out of a sectioned off airbox would only give it access to warmer ambient air in the engine bay, which would have the opposite effect of cold air.

Apparently I wasn’t alone. Lots of debate surrounded the claims as to the effectiveness of the CAI, and despite some modders showing their before and after dyno results, I still wanted more proof. Luckily the veteran modders and pro mechanics behind the “Mighty Car Mods” YouTube channel had the same question on their minds and decided to settle things once and for all with several dyno runs of an R34 Skyline featuring different airbox setups. The video is below, which is very well done, but if you don’t want to watch the whole thing, scroll down to see the results.

In the end, each set up of a CAI resulted in less power than the simple stock airbox with the stock filter. Only when the CAI was literally placed outside of the engine was there a very small gain in power. It’s hard to argue with the dyno numbers, which offer definitive proof that the perceived power gains and my 15 extra horsepower for my Miata were too good to be true.

In short, if you want a better noise when you put your foot down and a better looking engine bay, get a CAI. But if you want phenomenal power gains with a bolt-on mod, however, it’s best to leave the CAI option out of the picture.

*Years later, my opinion is that no bolt-on mod is really that effective. There is no replacement for displacement or adding a forced induction.


Dave on June 10, 2014 at 8:28 am.

I used to be an engineer for a company that made intakes. It was not our primary product. But it was one of many that we provided because our customers demanded it. Gains, or lack of gains was very vehicle dependent. In most cases, little to no gain (other than sound) was had. But customers still wanted them for the look and the sound. So we continued to provide intakes.

For one project (2006 350Z) I was instructed to design a SRI. I pulled apart the factory airbox and told the owner “I can make something, but it won’t make any more power, this factory setup is a great design.” I made a design. We made a 3D printed part. Dyno tested it. It made and extra 1HP. That’s it. I was questioned as to why the competitors were claiming 10-15 HP gains? So we tested 3 competitor intakes (1 SRI, 2 CAI, IIRC) including the vehicle manufacturer’s performance division product. Every one of them lost power. I felt vindicated of course as my design proved to be superior, even if it was very marginally superior to the stock setup, at least it was not worse. And in the end I was correct. Even if forum people refused to believe us.

On the other hand, there are cars like the Mazdaspeed3 that do respond extremely well to an intake. Even a SRI with no airbox or anything to block heat out. The factory setup was obviously flawed upon first look. And my first design for that one made over 10 HP and 15 lb-ft increase with the 3D printed part. And the production part was even slightly better than that. However, of all the cars I designed intakes for, none of them responded as well as the Mazdaspeed3 & 6. But at the same time, none of my designs lost power. However, the Mazdas are the only ones I would consider spending money on.

Side notes:
#1- Some of the vehicles did respond well to intakes once turbos were upgraded and the flow got to a point that the factory airbox became a restriction.
#2- Most aftermarket intakes change the diameter of the MAF area thereby making the engine run leaner or richer. The ECU tries to compensate, but usually can’t. This is not safe for the engine.
#3- N/A cars with stock engines are the least responsive to intakes. But they usually love ‘em some exhaust mods…

rossl on June 10, 2014 at 1:02 pm.

Sad that sometimes you can even lose power with a CAI.

Chance Hales on June 11, 2014 at 2:21 pm.

An intake is worthless unless you do exhaust too. If a car can breathe well it you’ll see power gains. If you only do half the mod then you’ll see very little, if any difference at all. I agree with Dave. Intake, exhaust, tune.

Leave Your Comment

Your email will not be published or shared. Required fields are marked *

You may use these HTML tags and attributes: <a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <cite> <code> <del datetime=""> <em> <i> <q cite=""> <strike> <strong>