Strange 2.5T Engine misfire problem w/multiple choice answer game. Come play.

[Tchman2016]

Good evening all,

Here's an interesting one I thought I'd post to get a consensus on an engine misfire on all cylinders issue on my 2020 CX-9 GT 2.5T. 36k miles.

Yesterday, I noticed the engine running rough, lower idle than usual with the car in drive and my foot on the brake. Engine sounded wierd and was stumbling and felt like it was close to stalling. If I remove my foot from the brake, the idle went back up by at least 100 rpm and the engine smoothed out. Press again and idle lowers and engine stumbles.

This does not happen if the car is in park idling, even with foot on brake. It only happens when in drive or reverse with foot on brake.

Broke out my TopDon BT scan tool to look at live data stream and confirmed that there were many misfires on all cylinders occurring with numbers slowly rising, however NO check engine light was on or had been triggered thus far. This was confirmed over multiple trips. Incidentally, if I leave the car to cool off and come back to start it a couple hours later, the engine has trouble starting for a few seconds it cranks and stumbles and finally gets going. Almost as if there was bad plugs or coils, but that is not possible because the misfires are across all cylinders.

I looked under the engine cover and there is no damage to the wires to the coils or injectors from any animals.

Further, the plugs were replaced with genuine NGK at 30k miles as preventative maintenance even though 40k miles are the normal time to replace.

So, before I take it to the dealer(I have a mazda oem extended contract) I thought I'd post and play a game of what could be wrong and who votes the most on the multiple choice answers to the problem below.

Possible choices are:
A) Bad brake booster with torn diaphragm/vacuum leak letting unmetered air into engine causing misfires and idle fluctuations when brake pedal applied under engine load

B) Dirty or stuck EGR valve

C) Cracked Cylinder head causing coolant intrusion and misfiring in all cylinders,also causing hard starting since coolant can pool in the cylinders and is not combustible like gas.

D) One or more cylinders have low compression due to bad rings or for unknown reasons.

E) Combination of any of the above

F) None of the above.

Lets go!

 

[SparkYellow]

I don't know enough to play but am eager to learn. It sounds like you have eliminated battery/alternator as possible culprits. May I ask why?

I was chasing a persistent misfiring in #2 cylinder on my R56. A MINI mechanic went to great lengths and determined that it must be a hairline crack in the block. A decision was made to stop at that point.

 

[Tchman2016]

I don't know enough to play but am eager to learn. It sounds like you have eliminated battery/alternator as possible culprits. May I ask why?

I was chasing a persistent misfiring in #2 cylinder on my R56. A MINI mechanic went to great lengths and determined that it must be a hairline crack in the block. A decision was made to stop at that point.

Hi, thanks for your input. Your situation definitely sucks. I certainly hope its not that in my case, rather one of the other choice scenarios.

Battery is new from 12/2025 and is AGM with plenty of amps and voltage. Alternator provides 14.2V charging when engine running. No issues with charging system or starting.

 

[SparkYellow]

Hi, thanks for your input. Your situation definitely sucks. I certainly hope its not that in my case, rather one of the other choice scenarios.

Battery is new from 12/2025 and is AGM with plenty of amps and voltage. Alternator provides 14.2V charging when engine running. No issues with charging system or starting.

Thank you for your reply! Since no one else is playing, I will. My lay-person guess is it's F, fuel delivery issues.

 

[N7turbo]

If there is a coolant issue, the plugs could be fouled. Or, they are counterfeit.

It's probably time to send an oil sample for analysis and maybe send a camera into one of the cylinders just to make sure.

The cylinder head crack will be evident externally, so it should be easy to confirm if you have that issue.

If not, can you verify fuel pressure or injector duty cycle?

 

[fordin888]

I'm putting my money on the ignition coils or a hairline crack in one of the spark plug insulators. These turbo engines are notoriously sensitive to the smallest gap issues under boost. Just swap the plugs first since it’s the cheapest "guess" before diving into the sensor data.

 

[Tchman2016]

If there is a coolant issue, the plugs could be fouled. Or, they are counterfeit.

It's probably time to send an oil sample for analysis and maybe send a camera into one of the cylinders just to make sure.

The cylinder head crack will be evident externally, so it should be easy to confirm if you have that issue.

If not, can you verify fuel pressure or injector duty cycle?

Thanks for the input!

The plugs were bought from Rockauto and were installed 6k miles ago.

I verified the high pressure fuel pump putting out over 500psi and each injector I cycled on and off with the scantool so they are all firing with a change to the engine idling each injector test.

I was going to do a cooling system pressure test with pump to see if there was any drop over time which would indicate a leak of coolant into 1 or more cylinders if the head is cracked.

I was considering doing a compression test for each cylinder.

Is it normal for the RPM to go down ONLY when the car is in drive or reverse WITH foot on brake? If I put the parking brake on with the car in drive but no foot on the brake, the engine rpm is higher around 725 and seems to idle ok. Once I put my foot on the brake, the rpm drops to about 550 and then I get the shaking and feeling like it wants to stall but doesn't. It just runs like crap.


I guess I'll pull a couple plugs and look for damage or fouling to the plug and also do a coolant pressure test.

I believe the firing order is 1,3,4,2 from crank side if i'm not mistaken?

 

[Tchman2016]

I'm putting my money on the ignition coils or a hairline crack in one of the spark plug insulators. These turbo engines are notoriously sensitive to the smallest gap issues under boost. Just swap the plugs first since it’s the cheapest "guess" before diving into the sensor data.

Thanks for the guess. It would be unlikely that all 4 ignition coils are bad since the misfire is in all cylinders.

The car drives fine under normal operation except when putting foot on the brake with car in gear and idling. Then it turns to s***.

 

[Tchman2016]

Can anyone confirm the idle drop with car in gear, parking brake on and then pressing the brake pedal?

 

[N7turbo]

Is it normal for the RPM to go down ONLY when the car is in drive or reverse WITH foot on brake?

So Mazda has something called Neutral Idle Control, but I don't think that explains what you are experiencing.

I can test the EPB scenario later when I go out, but I normally don't use it.

 

[Tchman2016]

 

[Tchman2016]

Those are the plugs I used 6k miles ago from Rockauto.

 

[Tchman2016]

So Mazda has something called Neutral Idle Control, but I don't think that explains what you are experiencing.

I can test the EPB scenario later when I go out, but I normally don't use it.

Thank you! I normally do not use the EPB either but I thought it was convenient to use just to keep the car stationary in drive without having my foot on the brake and then be able to test the idle fluctuations when applying and releasing the brake pedal.

 

[Tchman2016]

Pulled the plugs in order 1,3,4,2 and found cylinder 3 plug covered in soot, threads dark and, slightly wet and smells like gas. Cylinder 2 plug also looks a little funky.

What does that suggest? Where should I investigate from here?

Thanks all!

 

[N7turbo]

There is the exhaust valve stem steal leak issue. That caused oil consumption and was covered under an extended warranty:

But I thought it only applied to the 2021 model year. Check to be sure.

If you suspect a coolant leak caused by the cylinder head crack, you should notice an external leak on the exhaust side of the engine. An Oil Analysis wouldn't be a bad idea to look for traces of coolant in the oil: https://www.amsoil.com/c/oil-analysis/102/

I use KIT01 and take a sample when I change my oil, but you can get the pump to take a sample at any time.

 

[Tchman2016]

There is the exhaust valve stem steal leak issue. That caused oil consumption and was covered under an extended warranty:

But I thought it only applied to the 2021 model year. Check to be sure.

If you suspect a coolant leak caused by the cylinder head crack, you should notice an external leak on the exhaust side of the engine. An Oil Analysis wouldn't be a bad idea to look for traces of coolant in the oil: https://www.amsoil.com/c/oil-analysis/102/

I use KIT01 and take a sample when I change my oil, but you can get the pump to take a sample at any time.

Yes the exhaust valve stem seal was an issue on my wife's 2021 cx-5 but not my model year. The head crack coolant leak issue is my model year 2020 but not hers. Crazy.

I'm going to do a cooling system pressure test now and see if there is a leak there.

Did you see my plugs above? Cylinder number 3 (second one from the top)is the plug that is all carboned out and smelled like gas when I removed it. Cylinder 2 plug is a little suspect as well.(Last one on bottom)

I will look into the oil analysis thanks for the link!

 

[N7turbo]

Yes the exhaust valve stem seal was an issue on my wife's 2021 cx-5 but not my model year. The head crack coolant leak issue is my model year 2020 but not hers. Crazy.

If your 2021 is really a turbo, both issues could apply. Fortunately, the head crack on the turbo seems to be rare.

 

[Tchman2016]

If your 2021 is really a turbo, both issues could apply. Fortunately, the head crack on the turbo seems to be rare.

Mine is the 2020 cx-9 2.5T with this current misfire problem. Hers is the 2021 cx-5 2.5T which had the exhaust stem seals replaced and is good now.
My 2020 is not burning nor losing oil.

 

[SparkYellow]

Here are some old-school clues from Haynes Manuals (NA/1998 and scooter/2014). Is there a new set of photos for plugs off Skyactiv engines?