2024 CX-5 Turbo Air Filter Box Test - Yikes

I installed a 2.5" stub on the DAP, dirty air plenum, today. I had intended to install a varriable flow damper but couldn't get it to fit in the wheel well.

So I tested with the stub, knowing I could find away to mount the varriable flow damper.

I noticed a drop in low end torque right off the bat. This didn't surprise me since we see that on the Cummins.

Wot tests, NO movement on the pressure gauge. So the added inlet did the trick. Max air flow was ever so slightly more. I need to plot some flow values and compare.

Fuel mileage was 1 mpg less. Again the varrible flow damper could fix this and more.

So I removed the test DAP and installed an unmolested DAP.

I'll get back into this next week.
0727241320.webp
0727241319.webp

0727241304_HDR.webp
0727241300.webp
 
Hah, I was thinking about using a spring-loaded valve of some sort to just get more air at WOT but otherwise it shall stay closed so as to not hurt drivability. Looks like you beat me to it. This looks tricky to get right.
 
Hah, I was thinking about using a spring-loaded valve of some sort to just get more air at WOT but otherwise it shall stay closed so as to not hurt drivability. Looks like you beat me to it. This looks tricky to get right.
Using a 3" tube and 4" backflow damper on my truck was much easier. We have so much more room to work and the air box has a flat bottom to sub into. I was able to adjust the valve by restricting the main air inlet. It was easy to video when the valve opened and track the boost on a digital dash. The valve starts to open at 4 PSI. I don't see an easy way to do the same on our CX 5s.

We could put a micro switch on our accelerator peddle. The switch is tripped at WOT or whenever and opens a valve. Micro switches have been used forever to operate a nitrous kit on street cars.

An exhaust cut out valve would work nicely.

And there is a vacuum operated valve I've see work to allow more air into the air filter box too.

This mod is not all that important to me. I might take another stab at using a 3" backflow damper. I'm really happy how this little turbo/engine is responding at partial throttle. WOT is not all that important.

I'm about at the end of modding this car.
1722150141042.webp
 
Rev 1.0, changed to a 2.5" to 4" 90 degree reducer wit 4" back flow damper. It still a tight fit.

Initial test drive, lowend torque didn't change. Delta P guage showed pressure drop at WOT.

I'll get a log latter today and compare to logs without this device.

CAI damper.webp

0807240910.webp
0807240910a.webp
0807240921a.webp
0807240921b.webp
 
Last edited:
I finally got an airbox, CAI to fit under the driver side fender. Tight fit is an understatement! It JUST FITS.

I installed a 90 deg, 2.5 to 4", reducer today with a 4" back draft damper. I logged WOT with and without the airbox CAI.

Mazda Edit software generated two nice "Dyno" Curves. It was interesting to see the CAI peaked a few RPM lower than without the CAI and picked had a nice bump in torque, 33 and just 7 HP gain.

BUT at 3000 RPM the Air Box CAI picked up 20 HP over stock, 229 vs 209. THAT'S a homerun where I come from...

Equally important, the HP curve is much stepper from about 2500 to peak. THEN it doesn't fall off as much as without the CAI. This is intuitive, the CAI air box is not as restrictive than without. IF the Airbox CAI car was racing the stock airbox, it would walk away from the stock setup. There is much more area under the curve or more power available.

As previous tests pointed to, the stock airbox starves the turbo engine at WOT.

Important note, it was 96 degree F, 55%, 29.7 Hg afternoon. Correction to SAE factor is 1.04 or 260 HP CAI and 252 without. Correcting to sea level, like the manufactures: 395' to 0': 265 HP and 255HP.

I have found these types of "dynos" are very good at showing the difference and power curves. BUT are off on the HP and Torque. In this case, I believe these are conservative numbers.

The delta P gauge was at 7" WC with the CAI and 15" WC without. I think as the back draft damper gets a few more cycles it will open sooner. My first WOT it showed 15", the same as without a CAI. I'll test it in a few weeks again.

This mod requires three items, 2.5" offset dust port, 2.5 to 4" 90 degree reducer and 4" back draft damper, clamps. All sourced from Amazon. Cost: $70.00. This is a dirt cheap HP gain.

The first graph is of the Airbox CAI test.

MAZDA CAI dyno 3000 rpm.webp

MAZDA Dyno NO CAI.webp
 
Last edited:
I'd be interested to see 0-60 and quarter mile time comparison.

Considering stock is about 300ftlb at 2000rpm, I have no confidence in those 'dyno' graphs.
What are those graphs showing, 180ftlb torque at 2000rpm?

The torque curve is not representative of a stock or stock-ish 2.5T.
 
I like your idea of doing a zero to 60 comparison.

What these this shows is how the stock air filter box is restrictive vs. this modified air box the Air Filter Box CAI made more power. I have looked at other areas of the logs and they point to the same conclusion.

Further, the Filter Minder, delta P gauge I installed showed the stock air box IS restrictive to the tune of -15 wc. With the change it's less restrictive.

I was rolling into the throttle at 2000 RPM so the trans wouldn't downshift. Still it doesn't mirror an engine dyno curve. AND take a moment to read this article that compares an Engine Dyno to a Chassis Dyno. My data would qualify as a chassis dyno....it WILL look different from an engine dyno curve.

https://www.motortrend.com/how-to/testing-ford-393-cleveland-on-engine-and-chassis-dynos/photos/

Back when I raced street cars, I had similar software to Mazda Edit. I used to see how a new tune or modification worked. I had to roll into the throttle on these cars or the tires would spin. These cars would break the tires loose at 60 MPH. They were really not safe to operate on the street.

When I put street car racers on a chassis dyno, the changes were almost the same. So close, I stopped renting chassis dynos to confirm the change. We confirmed the changes at the track. As the saying goes, DYNOs LIE, the TRACK DON'T. Even Mazda's advertised engine dyno. I'd love to SEE the setup Mazda used to get these numbers, intake or not, exhaust system and on and on...

My friends who engine dyno "race", Engine Masters competition, learned how to trick the dyno all the competitors were using and won several events using this trick. What I'm saying, Mazda's dyno is a good starting point but not gospel.


1723144096011.webp
 
Last edited:
HAD to know what the difference in 0 to 60 times.

Conditions: 91F, 63% humidity, 29.7" Hg, AC ON. Two car seats, tools, rubber mats and other stuff in the car. I could have gotten 100 pounds out if it but I only wanted a comparison....

I didn't let the car cool down on the Air Box CAI, which makes a difference. If I was scientific about this, I would let the car cool down 30 min or more between tests.

The Stock Air Box, the car had a short cool down with the hood open, about 10 minutes as I changed the air box back to stock.

I used this calculator on the Manual mode, FWD, 4100 pounds to calculate HP.

https://www.gregraven.org/hotwater/calculators/0-60.html

0 to 60 times at the above conditions, same spot on the road:
Air Box CAI: 7.106 sec
Stock Air Box: 7.235 sec
Published: 6.1 sec

I used Torque Pro app and a blue tooth OBD adaptor so it's calculating time and car speed from the Mazda ECU. Times are WAY off published 0 to 60. I'll do a 0 to 60 test at 60 degrees some time and take all the extra weight out of the car....

HP per the above conditions per the calculator:
Air Box CAI: 262.5 HP
Stock air box: 256.5
or 6 HP difference

If any of you have ever Hot Lapped a car at a drag strip, you will know as it gets hotter, it slows down each lap. I did several 06 to 60 tests and each time it went slower as it got hotter. Maybe I'll let the car cool down 45 minutes between tests if I test again.

Filter Minder, delta P gauge:
Air Box CAI: O" wc, yeah!
Stock air box: 15" wc

The back draft damper has loosened up! The car still has excellent low end torque.
 
IF we correct the above HP numbers using Density Altitude to SAE corrections used on dynos, 77f, 0 humidity, 29.73 at see level we get get a 13% increase in HP. That dropped the 0 to 60 to 6.5 seconds....

On a cold winter day, maybe my CX 5 will run the published 0 to 60, even with the mods....interesting.

I wonder if anyone has run the published 0 to 60 times NA or Turbo?
 
Don’t know if it would even fit, but the airbox on my old mazda 6 2004 had exactly the kind of trap door you mentioned that would open at higher rpm to increase the airflow. It was right under the filter and would suck in warm air but it was still better than the restrictive air intake designed to reduce noise over performance. One of the common mod was to force that trap door open and also remove the resonator. (airbox mod???)

If you find a mazda 6 of that generation with the 4 cylinder you might be able to retrofit the airbox or the mechanism.

Also, as per your rant - those « cold air intake » that suck air from inside the engine compartment should really be called short ram intake. But marketing popularized the term CAI so now everyone is calling any intake mod a CAI, even though as you correctly said they are not cold air. The stock airbox on the modern Mazda’s actually are CAI, because the intake is directly outside of the engine compartment (although a very restrictive one).

 
Don’t know if it would even fit, but the airbox on my old mazda 6 2004 had exactly the kind of trap door you mentioned that would open at higher rpm to increase the airflow. It was right under the filter and would suck in warm air but it was still better than the restrictive air intake designed to reduce noise over performance. One of the common mod was to force that trap door open and also remove the resonator. (airbox mod???)

If you find a mazda 6 of that generation with the 4 cylinder you might be able to retrofit the airbox or the mechanism.

Also, as per your rant - those « cold air intake » that suck air from inside the engine compartment should really be called short ram intake. But marketing popularized the term CAI so now everyone is calling any intake mod a CAI, even though as you correctly said they are not cold air. The stock airbox on the modern Mazda’s actually are CAI, because the intake is directly outside of the engine compartment (although a very restrictive one).

It seems like the CX5 draws air up the the grill via a port in the area where the hood latch is. I made sure that area is as open as possible.
 
Don’t know if it would even fit, but the airbox on my old mazda 6 2004 had exactly the kind of trap door you mentioned that would open at higher rpm to increase the airflow. It was right under the filter and would suck in warm air but it was still better than the restrictive air intake designed to reduce noise over performance. One of the common mod was to force that trap door open and also remove the resonator. (airbox mod???)

If you find a mazda 6 of that generation with the 4 cylinder you might be able to retrofit the airbox or the mechanism.

Also, as per your rant - those « cold air intake » that suck air from inside the engine compartment should really be called short ram intake. But marketing popularized the term CAI so now everyone is calling any intake mod a CAI, even though as you correctly said they are not cold air. The stock airbox on the modern Mazda’s actually are CAI, because the intake is directly outside of the engine compartment (although a very restrictive one).

Great Info right there! So Mazda had a similar solution!

I don't know if you read, I tested with a 2.5" dia hole in the air box. The engine lost its low end torque. A short ram intake may do the same if it is less restrictive than the stock air box.

I'd call that intake a short RUNNER intake. No ram going on.

We use to remove a head light and run the CAI into the head light. The MAP sensor show a small ram gain. We were class racing and every HP counted. We had high stall toque converters. The engine rpm stayed between 4000 and 7000 rpm on a push rod V8. Today these push rod engines run out to +10,000 rpm. The move a lot of air.

Here is an example of a RAM CAI. :cool:

Cop Car Pic at Steele.webp


Or a little more serious center RAM CAI on a "MORE" Door. WEEE

Front_wheels_up_atco_2016.webp
 
Last edited:
Here is more info on the old mazda 6 Variable intake system:


Fully agree with this below:
I'd call that intake a short RUNNER intake. No ram going on.
 
Another thing to consider is that the engine bay on the CX-5 is a lot more sealed up than on older cars. With all the bottom aero parts and the seals around the hood, I suspect the air in there gets a lot warmer than on older models we may be used too.

And you also have the active grill shutter which reduces the airflow in the engine compartment under specific conditions.

I have a 2018 CX-9 and i am amazed how clean (relatively) the engine bay as remained. But that probably isn’t ideal for short runner intakes. I suspect a lot of the performance gain would evaporate once the temperature raises in the engine compartment.

Have you considered using the stock cold air intake location but removing components and straightening the air path? Something similar to this:
 
Another thing to consider is that the engine bay on the CX-5 is a lot more sealed up than on older cars. With all the bottom aero parts and the seals around the hood, I suspect the air in there gets a lot warmer than on older models we may be used too.

And you also have the active grill shutter which reduces the airflow in the engine compartment under specific conditions.

I have a 2018 CX-9 and i am amazed how clean (relatively) the engine bay as remained. But that probably isn’t ideal for short runner intakes. I suspect a lot of the performance gain would evaporate once the temperature raises in the engine compartment.

Have you considered using the stock cold air intake location but removing components and straightening the air path? Something similar to this:
I made this varriable air flow to my 24 air box.

My tests confirmed it dropped the delta P in the air box -15 wc. $70 for parts and a couple hours to build and install.

0807240910.webp
0807240910a.webp
 
Last edited:
Another thing to consider is that the engine bay on the CX-5 is a lot more sealed up than on older cars. With all the bottom aero parts and the seals around the hood, I suspect the air in there gets a lot warmer than on older models we may be used too.

And you also have the active grill shutter which reduces the airflow in the engine compartment under specific conditions.

I have a 2018 CX-9 and i am amazed how clean (relatively) the engine bay as remained. But that probably isn’t ideal for short runner intakes. I suspect a lot of the performance gain would evaporate once the temperature raises in the engine compartment.

Have you considered using the stock cold air intake location but removing components and straightening the air path? Something similar to this:
I like that. Too bad it's outrageously expensive.
 
Make your own for $10.00

I made my own grill covers on my truck for driving in the winter. We struggle to get enough heat in our Cummins Compression Ignition engines with their huge radiators designed for towing.

I noticed a 2 to 3 MPG highway increase in FE with this mod.

I made the grill covers from vinyl flooring I picked at Home Depot. I used the glue side facing out, after cleaning off the glue with brake cleaner. Then cut and drilled holes in the strips. Zip tied it to the grill. From a few feet away, you can't see the zip ties.

I normally add half the grill covers in the fall and do a full cover when the high are in the 60s F or less.

Us Compression Ignitions guys use radiator covers and block heaters as well. I added 2 magnetic oil pan heaters. What use to take 15 min to get heat, now it heats up in 5 min. Also improves FE and much less wear on the engine.

If I was as OCD as I am with the truck, I would have a block and oil pan heater on the CX 5. I'd preheat every morning of the year....Engine life would definitely increase...FE would improve as well.

I would do a grill cover on the CX 5 if this was my DD. My wife mostly drives it and I'm not going to ask her to watch ECT until I feel comfortable with the set up. I've carried tools in the truck to remove the covers in the beginning. I never had high ECT driving solo. Towing, I removed most of the covers, just to be safe...
 
Last edited:
Our cars have active grille shutters :cool:
I see, at the bottom. What about the rest of the cooling stack? The other 75% of the grill could house some love too.

We will see how long this part lasts. The part is not all that expensive. Looks like the front facial has to be removed...maybe it can be accessed from the bottom...


1723314155119.webp
 

New Threads

Back