Backfiring under hard acceleration
Exhaust leak (was Backfiring under hard acceleration)
EDIT (2019-01-02): I think this may be an exhaust leak after all. I've posted an update at https://www.myg37.com/forums/intake-...ml#post4226282.
I have a 2010 G37 sedan with the Sport Package. No mods at all.
When I accelerate hard, especially when I am at highway speeds, I get lots of backfiring. Never have gotten an SES light, and I have no other problems.
At least, I think this is backfiring. It sounds really close to the sound of an intake manifold backfire on an old carbureted car I had--not a loud bang but a series of dull but powerful thuds--until I let off the accelerator.
I'm not clear where to start with this. What should I check or do?
This has been going on for months.
I have a 2010 G37 sedan with the Sport Package. No mods at all.
When I accelerate hard, especially when I am at highway speeds, I get lots of backfiring. Never have gotten an SES light, and I have no other problems.
At least, I think this is backfiring. It sounds really close to the sound of an intake manifold backfire on an old carbureted car I had--not a loud bang but a series of dull but powerful thuds--until I let off the accelerator.
I'm not clear where to start with this. What should I check or do?
This has been going on for months.
Last edited by AuthorizedUser; Jan 2, 2019 at 12:49 PM. Reason: new discovery
Yeah, that's not supposed to happen...
First thing to do would be to have the codes read. Even though there is no light, there could be a code there.
Besides that, I dunno. maybe there is an exhaust leak somewhere?
First thing to do would be to have the codes read. Even though there is no light, there could be a code there.
Besides that, I dunno. maybe there is an exhaust leak somewhere?
First, question: does a backfire on a stock 2010 G37 sound like this?
If the backfire is still loud on the stock G37, then this isn't backfire.
I was at freeway speeds, paddle shifted down to 4th, and played with the gas. I could still hear clear, dull thudding sounds at hard, high RPM acceleration, but the engine itself sounded wonderful. The thuds remind me of backfires through the intake (at speed) in my old car.
Well, that car in the video sounds heavily modified. It sounds like it's turobcharged, and it's obviously got a full exhaust (and probably no cats).
Here is my car - With no mufflers. When i lift off the gas you can hear lots of little crackles and pops.. is that what you hear?
With my stock exhaust i never heard anything like that, but now with my aftermarket mufflers i occasionally hear a few little pops. A stock one shouldn't be making any noises like that.
Here is my car - With no mufflers. When i lift off the gas you can hear lots of little crackles and pops.. is that what you hear?
With my stock exhaust i never heard anything like that, but now with my aftermarket mufflers i occasionally hear a few little pops. A stock one shouldn't be making any noises like that.
Thanks to both of you for your comments. I have been out of town for a few days, and my normal drive is on side streets and doesn't give opportunities to test things out. I'll try it again when I get to a freeway and report back.
As for the sound in the above video, I do think I hear an effect like the trailing off pops after you let off the accelerator. The sounds here are less than during hard acceleration. Still, I need to drive it more and see if I can make a recording.
Whatever it is, it's nothing near the gunshot sound of a traditional backfire. If the gunshot sound is what would still come out of a stock exhaust, then it's not backfiring. I doubt people outside the car would notice much different noise. It again reminds me of the sound of a intake manifold backfire of an older carbureted engine: it's a series of muffled thuds.
I do not think there is any loss of power. I'll take a video when I can.
Thanks again for the responses.
As for the sound in the above video, I do think I hear an effect like the trailing off pops after you let off the accelerator. The sounds here are less than during hard acceleration. Still, I need to drive it more and see if I can make a recording.
Whatever it is, it's nothing near the gunshot sound of a traditional backfire. If the gunshot sound is what would still come out of a stock exhaust, then it's not backfiring. I doubt people outside the car would notice much different noise. It again reminds me of the sound of a intake manifold backfire of an older carbureted engine: it's a series of muffled thuds.
I do not think there is any loss of power. I'll take a video when I can.
Thanks again for the responses.
Have some new data. It was cold for my area this morning, in the mid-20s. When I took off cold, I was getting this alleged backfiring all over the place at modest acceleration at any speed. It was weird; never had that before. After it was fully warmed up, it had gone back to normal behavior.
I was able to get a video of the normal behavior: https://www.youtube.com/watch?v=Gppg...ature=youtu.be. Listen for when the RPMs increase over 3000. You'll hear dull thudding in the background. This isn't road noise, it's something with the engine. It's more prominent at freeway speeds when I accelerate hard.
I was able to get a video of the normal behavior: https://www.youtube.com/watch?v=Gppg...ature=youtu.be. Listen for when the RPMs increase over 3000. You'll hear dull thudding in the background. This isn't road noise, it's something with the engine. It's more prominent at freeway speeds when I accelerate hard.
Trending Topics
The more I think about this, I suspect a lean condition. This is only happening when the engine should be running its richest: when the block is cold in very cold weather (35-ish degrees or lower) or when it's at high RPMs and heavy load (hard acceleration but only after the downshift).
You may say, "Idiot, your oxygen sensor would tell you if it's too lean!" I surmise that under these conditions, the tolerance that the computer accepts from the oxygen sensor is pretty high. It's unlike warm engine, steady cruise speeds, where the O2 sensor might be expected to have a narrow band of conditions.
I took it to a local, trusted shop just to see what they recommend. He said I should do fuel injector cleaner. I put a can of AutoZone FI cleaner in 3/4 tank of gas, and next tank I'll use Sea Foam. Hopefully that will help this.
By the way, if it's running lean, that would be a misfire, not a backfire.
You may say, "Idiot, your oxygen sensor would tell you if it's too lean!" I surmise that under these conditions, the tolerance that the computer accepts from the oxygen sensor is pretty high. It's unlike warm engine, steady cruise speeds, where the O2 sensor might be expected to have a narrow band of conditions.
I took it to a local, trusted shop just to see what they recommend. He said I should do fuel injector cleaner. I put a can of AutoZone FI cleaner in 3/4 tank of gas, and next tank I'll use Sea Foam. Hopefully that will help this.
By the way, if it's running lean, that would be a misfire, not a backfire.
Pick up a cheap ELM327 OBD dongle and get the Torque or PBD Fusion app for your phone.
Then you can look at your fuel trims and see if the ECU thinks the car is running lean. You can also look at real time data from your air/fuel ratio sensor and see what it is reporting under acceleration. The is also a "misfire" counter (by cylinder) that I think you may want to look at
For about $30 you will have access to a wealth of information.
Then you can look at your fuel trims and see if the ECU thinks the car is running lean. You can also look at real time data from your air/fuel ratio sensor and see what it is reporting under acceleration. The is also a "misfire" counter (by cylinder) that I think you may want to look at

For about $30 you will have access to a wealth of information.
Pick up a cheap ELM327 OBD dongle and get the Torque or PBD Fusion app for your phone.
Then you can look at your fuel trims and see if the ECU thinks the car is running lean. You can also look at real time data from your air/fuel ratio sensor and see what it is reporting under acceleration. The is also a "misfire" counter (by cylinder) that I think you may want to look at
For about $30 you will have access to a wealth of information.
Then you can look at your fuel trims and see if the ECU thinks the car is running lean. You can also look at real time data from your air/fuel ratio sensor and see what it is reporting under acceleration. The is also a "misfire" counter (by cylinder) that I think you may want to look at

For about $30 you will have access to a wealth of information.
Here's two more videos:
In both, misfires obvious around 4000 RPM. Actual outdoor temperature was around 25 degrees (car is garaged, so it starts out at a little warmer temp than ambient).
The misfires can easily be felt on the floorboard.
In both, misfires obvious around 4000 RPM. Actual outdoor temperature was around 25 degrees (car is garaged, so it starts out at a little warmer temp than ambient).
The misfires can easily be felt on the floorboard.
Pick up a cheap ELM327 OBD dongle and get the Torque or PBD Fusion app for your phone.
Then you can look at your fuel trims and see if the ECU thinks the car is running lean. You can also look at real time data from your air/fuel ratio sensor and see what it is reporting under acceleration. The is also a "misfire" counter (by cylinder) that I think you may want to look at
For about $30 you will have access to a wealth of information.
Then you can look at your fuel trims and see if the ECU thinks the car is running lean. You can also look at real time data from your air/fuel ratio sensor and see what it is reporting under acceleration. The is also a "misfire" counter (by cylinder) that I think you may want to look at

For about $30 you will have access to a wealth of information.
The misfire counter is towards the end and look for "MID:$a2 TID:$0b"
Here is a sample from my car:
MID:$a3 TID:$0c
Misfire counts for last/current driving cycles(calculated) - Misfire Cylinder 2 Data
Max: 65,535Count Min: 0Count
Test result value: 0 Count
Good luck!
Start torque, select "Test Results" it will take a while to scan. When complete you can view on screen or email the results to yourself.
The misfire counter is towards the end and look for "MID:$a2 TID:$0b"
Here is a sample from my car:
MID:$a3 TID:$0c
Misfire counts for last/current driving cycles(calculated) - Misfire Cylinder 2 Data
Max: 65,535Count Min: 0Count
Test result value: 0 Count
Good luck!
The misfire counter is towards the end and look for "MID:$a2 TID:$0b"
Here is a sample from my car:
MID:$a3 TID:$0c
Misfire counts for last/current driving cycles(calculated) - Misfire Cylinder 2 Data
Max: 65,535Count Min: 0Count
Test result value: 0 Count
Good luck!






