new AEM CAI

[Darth Vader]

Well, I bought an AEM CAI for my car with my tax return. The Cobb SRI that came with the car will make its way to Ebay as soon as I can find time to take it off and install the CAI.

It's what I always wanted for the car so, I just had to do it. I'll let everyone know what I think of it compared to the Cobb, direct comparison. Both are in primo new or like new condition so, it should be valid enough.

 

[MM BOOST]

how much you selling the cobb for?

 

[Sport6]

It would really be interesting if you can feel a difference between the two, other than sound. Though I have heard the top end gains are a little better on the CAI.

 

[MSMS3]

No way to "feel" the difference. Butt dyno responds to changes in sound more than anything else. Our ears fool us. What sounds louder usually is perceived as being quicker, whether it is or not. It takes a true 20-30 hp gain to objectively and consistently "feel" the difference in two mods. Gains that pick up a consistent tenth of a mile in the quarter or even to 60 mph (significant) can't be felt at all.

A much better and still cheap way to compare mods is with a simple stopwatch, measuring changes in acceleration times. You have to get up high enough in speed to dial the variations in launch out of the equation. So, measuring say the time between 30 and 100 mph or 40 and 100 mph on the same road under the same weather conditions, will tell a lot about whether a mod is making more power, especially if the measurements are repeatable and consistent.

Drag strip runs or use of an accelerometer are even better, but take more time and money.

To be truly accurate, you would also have to reset the ECU after each mod and give the car two or three days of driving to "learn" the changes in intake air flow.

 

[Darth Vader]

Top end is very strong on the Cobb. That's its strength. It takes a touch off the bottom to achieve this but, it is hella strong up top.

I expect the longer runner length of the CAI and it greater plenum volume before the throttle plate will make the car as much or more torquey than stock. The longer runner length may not help top end as much as the short ram might. That's my idea at the moment, given testing so far with my homemade CAI. The better constructed AEM piece may mask this but, I still expect it to be largely true. I will let everyone know, as I said before.

 

[mazdaspeedster3]

We have been down this road to many times to do it again. They are similar if not identical gains just at different RPM ranges. They sound different and look dif. It all boils down to preference from one to the other... CAI/SRI

 

[Darth Vader]

Well, let's put it this way. I got ragged for not doing back to back tests on the same car before stating a conclusion in another thread. I'm correcting that "error" in this case.

You're very likely correct in your statement, though.

 

[MSMS3]

I'm with Darth on this. Best way is to comparison test, with an interim ECU reset. Getting identical testing conditions after a reset and ECU "learn" will be tough, but it can be done and there are objective ways to make the measurements -- dragstrip, manual stopwatch, accelerometer, etc.

What, if anything, does this chart say about top end strength with a CAI? I say hitting 100 mph in 11.5 seconds is pretty strong. 60-100 mph was only 6.1 seconds. True, I have a catless DP/RP too, but the point is that the intake, under full boost and once in gears that will let the engine produce full power, is pretty damn good, IMHO. Please excuse crappy launch!

Note: My accelerometer will not hold date and time when unplugged. The run was made yesterday morning.

 

[Darth Vader]

Ok, I've done the installation of the AEM CAI, the running around and testing. Here's my impressions:

Ultimate top end power favours the Cobb, at least on a cooler day. We are in the 15 degree Centigrade range here, now. The power hit on the Cobb is stronger, making the car feel more ferocious.

The AEM has better light throttle driveability, is smoother in its power delivery and may be down a pony or two way up top. It has a broader spread of useable power for a street or otherwise unmodified car.

The Cobb is way noisier. If you're looking for a stealth fighter, this intake ain't it. The AEM is still noticeable in terms of noise over a stocker but, really, what you hear loudest is the stock blowoff valve psshing away.

 

[MSMS3]

Darth, did you ECU reset? What was your ECU relearn period? Do you have objective measurements, such as simple stop watch timing? Or is your report more of a butt dyno sense?

Subjective "feel" on throttle response may be just about the only way to make that comparison. But power generation comparison is objectively measurable.

 

[Darth Vader]

Yeah, I did ECU reset and then drove the car for 30 or 40 through all operational parameters from idle to WOT, to train the ECU.

I am reporting butt-dyno but, my butt-dyno is calibrated by years of being a drag racer so, I can definitely differentiate noise and power. I also did some distance and speed testing, covering a fixed distance and looking at the speedo when arriving at a point.

I'll try to get some G-tech data later, for what that's worth.

 

[Darth Vader]

See, before we had little gadgets like the G-tech or an available dragstrip, we used to test mods just this way: WOT hit, and check speed at particular rpms. The higher the speed, the higher the power. Not too reliable for measuring exact amount of change but, adequate to determine if it was an improvement or not.

 

[MSMS3]

See, before we had little gadgets like the G-tech or an available dragstrip, we used to test mods just this way: WOT hit, and check speed at particular rpms. The higher the speed, the higher the power. Not too reliable for measuring exact amount of change but, adequate to determine if it was an improvement or not.

Yes, speed change at WOT between two fixed objects is usable and you don't need any device except the speedo. Will not help much with small changes in power (like our CAI v. SRI issue is likely to be) but can be a good way to know if you are making progress with mods.

Say, if you have a GTech (the RR can download the data and print charts) I'd be interested in what it's data shows on your comparison, and also in just how close your runs compare to your strip times. That little electronic bastard's cold hard data will call my ass out when I shift early or late or lift throttle during a shift, or launch with too much mustard on the go pedal. Very perceptive of the little stuff when the charts are generated after download to the laptop.

I think there might be something to your theory about how the CAI and SRI produce their best power gains in different parts of the power band. Your logic makes sense, assuming that the SRI can get air of the same density as the CAI.

 

[240ka]

top end gains are better with the CAI.

my friend and i race equally modded CAI to SRI, and i began to pul on him around 80 beyond. multiple times.

when i switched to SRI.

we were equal the entire way. everytime. it was a drivers race not to misshift

 

[matsuda]

Well, I bought an AEM CAI for my car with my tax return. The Cobb SRI that came with the car will make its way to Ebay as soon as I can find time to take it off and install the CAI.

It's what I always wanted for the car so, I just had to do it. I'll let everyone know what I think of it compared to the Cobb, direct comparison. Both are in primo new or like new condition so, it should be valid enough.

Are you able to monitor LTFT's?

 

[Darth Vader]

Nope.