Backfiring on deceleration

[Cowger]

Hi everyone,

My 2005 roadster has just developed a new behavior: On moderate to heavy deceleration (using engine braking alone), it will do its normal "burble" for about 3 seconds and then POP a pretty significant backfire. Simultaneous with the backfire, the upshift light goes yellow and the speedo drops by about 50%, just for a moment (other gauges may also move). Strange. After the single backfire, it goes to no burble -- just quiet deceleration.

The "normal" behavior that I've always known with the car is that it burbles on deceleration for 10 seconds, then goes to just running smoothly / quietly as long as your foot is off the accelerator.

Any ideas? It's like the "burble" programming is messed up. The car is otherwise running fine (as far as I can tell).

Somewhat related to this, does anyone have an OBD2 plug-in that adapts to an iPad app that they can recommend? I'd like to see what the ECU has to say about all this...

Thanks!
Bryan

[bluesrt]

you need a burble diagnostic aid- , not throwing any codes?

[Slithr]

Bryan,

This is the scan tool I use via Bluetooth, but unfortunately had to buy a Nexus 7 tablet to use it. Maybe an iPhone/iPad variant is now available ... not sure.

http://www.scantool.net/obdlink-mx-b...g4gaArCx8P8HAQ

[slowhatch]

I use Torque Pro app for android.

[Steve M]

The backfire by itself wouldn't be much cause for concern, but coupled with this:

Simultaneous with the backfire, the upshift light goes yellow and the speedo drops by about 50%, just for a moment (other gauges may also move).

...would be cause for concern in my book. That sounds like a misfire of some sort...I'd be checking plugs and wires first, scanning for and diagnostic trouble codes next. If those things check out, I'd be looking at things that control when the fuel and spark gets thrown in the mix, like a crank/cam position sensor.

[Cowger]

Thanks, guys. It doesn't seem like there are any solid solutions out there for a wireless OBDII plug-in and iDevice app, so I think I'll just go down to the auto parts store today and borrow theirs or just buy one. I'll let you know what I find...

[Cowger]

The backfire by itself wouldn't be much cause for concern, but coupled with this:



...would be cause for concern in my book. That sounds like a misfire of some sort...I'd be checking plugs and wires first, scanning for and diagnostic trouble codes next. If those things check out, I'd be looking at things that control when the fuel and spark gets thrown in the mix, like a crank/cam position sensor.

Exactly, Steve, it's very strange behavior, and the suddenness of its onset has me thinking that something has failed. I definitely would like to understand what changed.

So I borrowed a OBD2 code reader today, and found no codes in it. My buddy said "right, unless the check engine light is on, there won't be any codes". Is this true? Would a more sophisticated reader find something that a simple one does not?

[mad prof]

Do you have any air leak in the exhaust system? I had very similar problems and found that the exhaust pipe was not sealing properly with the cat. After taking care of this, no more backfires, just a pleasant rumble on deceleration.

[usmcly]

Check the gap on your sprak plugs, I had the similar issues. Pulled all my plugs out and they were gapped at .32 when i installed them and 8 out of 10 was .38-.40, I replaced the plugs with new ones and gapped at .32 and everything went back to normal and power came back.

[Steve M]

Exactly, Steve, it's very strange behavior, and the suddenness of its onset has me thinking that something has failed. I definitely would like to understand what changed.

So I borrowed a OBD2 code reader today, and found no codes in it. My buddy said "right, unless the check engine light is on, there won't be any codes". Is this true? Would a more sophisticated reader find something that a simple one does not?

I don't know if a dealer scan tool would get more info, but there are plenty of conditions that take multiple (2 usually) events before it sets a code.

[Cowger]

Good suggestions, guys, thanks! I'll check plugs and the exhaust fittings and we'll go from there...

Bryan

[Mr White]

Exactly, Steve, it's very strange behavior, and the suddenness of its onset has me thinking that something has failed. I definitely would like to understand what changed.

So I borrowed a OBD2 code reader today, and found no codes in it. My buddy said "right, unless the check engine light is on, there won't be any codes". Is this true? Would a more sophisticated reader find something that a simple one does not?

That's how code readers work, they just tell you why the dash light is on. They're cheap, usually run around 50 bucks, and generally aren't really helpful except in those cases where the car runs fine and you can't figure out why the light is on.

A scan tool, will give you live data while the car runs. They vary in price, some are slightly better than the 50 dollar scan tool and others are ridiculous. These are useful for diagnosis, and they are sometimes hard to distinguish because advertisements will muddy up the line between code reader and scanner. We could get into a good discussion about scan tools and whatnot.

So the backfire thing... unburnt fuel is making it out of your cylinders. There could be multiple reasons, weak spark, bad injector, something is telling your computer more air is going in than actually is (which could be many parts), and <insert name of hated political figure here>.

This is going to sound like an odd one, but I had another Chrysler vehicle that the gauges and lights would go nuts. It wasn't all the time, but it would be really random. They were symptoms of the battery's end of life. The car started fine, every time, which was odd. It's worth taking it to a place that can test it. It could also account for a weak spark on deceleration.

If that checks out, then the real fun starts.

[AZTVR]

Hi everyone,

My 2005 roadster has just developed a new behavior: On moderate to heavy deceleration (using engine braking alone), it will do its normal "burble" for about 3 seconds and then POP a pretty significant backfire. Simultaneous with the backfire, the upshift light goes yellow and the speedo drops by about 50%, just for a moment (other gauges may also move). Strange. After the single backfire, it goes to no burble -- just quiet deceleration.

I am no expert in troubleshooting such problems. However, like some of those above, I will give an educated guess until an expert comes along.

Interesting that the two electrical symptoms that you mention are related to sensors on the transmission. Does that have any significance? If it does, then a point of commonality with those symptoms and a misfire or momentarily excess fuel issue might be the ground to the engine. If it were loose such that it was intermittently not touching in a specific scenario; maybe it could cause all of the symptoms. It seems like such an intermittent would cause more random effects than you are experiencing; but, at least such a loose or corroded/bad connection is not unusual.

[Cowger]

Well, I pulled 4 of the plugs and found them to all be well within spec (which for my car is 0.033 - 0.038"). This didn't necessarily surprise me, since I had replaced them just 10k miles ago.

And if there's an exhaust leak, I can't find one...

I continue to be interested in how the "burble" algorithm works, so if anyone knows, please pipe up. I'm assuming it's controlled by the ECU and looks something like:

If (throttle_position = 0 && [insert other conditions]) then inject a small amount of fuel into [X] cylinders for 10 seconds...

Er something like that. I'm of course just guessing, but I'm hoping that if someone actually knows how this works, it might provide some insight into what all sensors / inputs are involved in that programming....

[Steve M]

I'm guessing the burble you are referring to is DFCO - deceleration fuel cutoff. On the Gen 4 cars, it enables around 1500 RPMs and above as you coast down...the Gen 3 cars are probably similar in how it operates. I'm thinking it also pulls timing, but not sure on that...either way, when enabled, the fuel gets cut (I don't think it does completely though), and whatever is left ignites in the exhaust.

DFCO occurs between RPM set points under deceleration and no throttle input.

[Steve M]

I'm not sure if this will be useful, but these are the tables that control DFCO on a Gen 4...I'm not sure if the PCM can (and does) track the current gear selected, but the tables are set up that way. It looks at pedal position, engine RPM and vehicle speed (VSS = vehicle speed sensor). Above the enable RPM and VSS, DFCO can come into play (provided the engine coolant temp is above the values shown)...below the disable RPM and VSS, DFCO is not in play. I'd assume the Gen 3 cars have similar parameters and set points.

[JonB ~ PartsRack]

This sure sounds like a Cam Sensor to me......

Note to Steve M.: good stuff, I'm impressed!

Hi everyone,

My 2005 roadster has just developed a new behavior: On moderate to heavy deceleration (using engine braking alone), it will do its normal "burble" for about 3 seconds and then POP a pretty significant backfire. Simultaneous with the backfire, the upshift light goes yellow and the speedo drops by about 50%, just for a moment (other gauges may also move). Strange. After the single backfire, it goes to no burble -- just quiet deceleration.

The "normal" behavior that I've always known with the car is that it burbles on deceleration for 10 seconds, then goes to just running smoothly / quietly as long as your foot is off the accelerator.

Any ideas? It's like the "burble" programming is messed up. The car is otherwise running fine (as far as I can tell).

Somewhat related to this, does anyone have an OBD2 plug-in that adapts to an iPad app that they can recommend? I'd like to see what the ECU has to say about all this...

Thanks!
Bryan

[Cowger]

I'm not sure if this will be useful, but these are the tables that control DFCO on a Gen 4...I'm not sure if the PCM can (and does) track the current gear selected, but the tables are set up that way. It looks at pedal position, engine RPM and vehicle speed (VSS = vehicle speed sensor). Above the enable RPM and VSS, DFCO can come into play (provided the engine coolant temp is above the values shown)...below the disable RPM and VSS, DFCO is not in play. I'd assume the Gen 3 cars have similar parameters and set points

Interesting, Steve, thanks! I agree that Gen3 probably has something similar...

[Cowger]

This sure sounds like a Cam Sensor to me......

Okay, you got my attention... Do you suggest I just go ahead and replace mine? As soon as I click "Post" here, I'll be searching for what that involves...

[67PLY]

Slightly off topic, I thought I read something a long time ago that Ralph had mentioned the decel popping and such was that the PCM pulses the injectors to cool the catalytic converters. Off road PCM eliminates the decel pulsing of the injectors because the majority of vehicles using it have no cats.

[Steve M]

Slightly off topic, I thought I read something a long time ago that Ralph had mentioned the decel popping and such was that the PCM pulses the injectors to cool the catalytic converters. Off road PCM eliminates the decel pulsing of the injectors because the majority of vehicles using it have no cats.

I would not be surprised at all...DFCO is a fuel saving feature, but the Viper PCM also seems overly protective of the catalytic converters, and opts to douse them with additional fuel when it deems appropriate. I do enjoy listening to it grumble on its way to a stop.

[Cowger]

Slightly off topic, I thought I read something a long time ago that Ralph had mentioned the decel popping and such was that the PCM pulses the injectors to cool the catalytic converters. Off road PCM eliminates the decel pulsing of the injectors because the majority of vehicles using it have no cats.

Seems pretty on-topic, actually, and that's good to know. I had always assumed the "burble" / decel popping was programming just because it sounded cool, like what a big-lobed, cammed-out, 4 bbl engine sounds like on deceleration. Knowing that it's more than that helps in this problem solving riddle.

I checked out what it takes to replace a camshaft sensor, and it looks like it's no big deal. Just need to make sure that the engine is rotated correctly before installing it. I'll give that a shot, next, and report back...

[Alex Mills]

Thanks, guys. It doesn't seem like there are any solid solutions out there for a wireless OBDII plug-in and iDevice app, so I think I'll just go down to the auto parts store today and borrow theirs or just buy one. I'll let you know what I find...

I use a PLX Kiwi to make a WiFi hotspot that I connect my iPhone to: http://www.plxdevices.com/product_cat.php?id=GSST

The iOS app I use is DashCommand: http://www.palmerperformance.com/pro...hone/index.php

[Cowger]

I use a PLX Kiwi to make a WiFi hotspot that I connect my iPhone to: http://www.plxdevices.com/product_cat.php?id=GSST

The iOS app I use is DashCommand: http://www.palmerperformance.com/pro...hone/index.php

Thanks, Alex. I looked at DashCommand and was intrigued by all the functionality, but the reviews I saw weren't very good. How has it worked for you?

[Alex Mills]

Thanks, Alex. I looked at DashCommand and was intrigued by all the functionality, but the reviews I saw weren't very good. How has it worked for you?

It's worked fine for me, but I have mostly used it for error code reading and clearing. I don't connect it to an iPad for live stats while driving or anything for example.