Why does my car backfire when compression braking?

[markjs]

When I go downhill and the forward motion of the car turns the engine it backfires quite a bit. It seems to run good on all four, never backfires or misfires on acceleration, so why does it do it when compression braking? I have noticed this on many other manual transmission cars and motorcycles. It can't be normal because I have also noticed that new cars do not seem to do it.

[bgd73]

An exhaust system just right does away with it. check immediately after muffler (tail pipe) for any holes. This noise is harder to do away with when there is no sound resignator(this is in midpipe), but can be done.

[PoorManzImpreza]

It has to do with your idle air fuel mixture, how well your exhaust flows and size of the economizer in the carb (I'm assuming carburetion)..you can usually illiminate it by running a Cat LOL seriousely adjusting idle air fuel a little bit rich sometimes helps..This is classic with a bigger exhaust..I'll quote a site that says it better than I could:

 

"In general, backfiring on deceleration (as opposed to acceleration) is generally caused by a lean condition in the pilot circuit. What happens is that the mixture leans out enough to where is fails to ignite consistently. This, in turn allows some un-burnt fuel to get into the exhaust pipes. Then when the engine does fire, these un-burnt gasses are ignited in the exhaust pipe, causing the backfire. Newer carbs have an 'enricher circuit' which cuts in on deceleration to help this problem. Earlier carbs do not have this. I would say check the pilot circuits, and set them a bit richer. It is hard to do this right without some equipment to test where you are. Aftermarket pipes often increase backfiring, probably because they make the bike run leaner."

[bgd73]

excellent explanation! I'm hell bent on keeping my 38 mpg tho!

[NorthWet]

And now for my contrary 2 cents:

 

Backfiring/afterburning is usually caused by RICH mixtures entering the exhaust system, combining with extra air somewheres along the way, and igniting due to residual high temperatures.

 

Think about the conditions on overrun: You close the throttle to its idle position while the engine is still trying to suck air for a much higher speed. Manifold vacuum goes to its highest value (5-10 inHg greater than when idling). This vacuum is trying to suck mixture from the barely cracked throttle plate and the carburetor's idle circuit, which normally supplies a richer mixture than the main circuit. But overrun isn't "normal", and you suck even more fuel from the idle circuit producing an even-richer-than-idle mixture. This mixture may or may not ignite in the cylinder (low effective compression ratio combined with rich mixture), and unburnt fuel enters the exhaust system.

 

If this unburnt fuel doesn't get combined with extra air prior to surrounding temperatures dropping below ignition point, it goes quietly into the night. If however, extra air is added (exhaust system leak/hole) AND the temps are still high, it ignites in pulses (forward and backward of the extra air) burning out any burnable mixture.

 

Car/carburetor designers deal with this by having a dashpot that holds the throttle open for a little bit, an auxillary air valve ("gulp valve"), and/or an idle-circuit "cut" solenoid (like the anti-diesel valve).

 

If a charge fails to ignite in the combustion chamber due to it being a lean mixture, it is unlikely to ignite in the exhaust system.

 

A nonleaky exhaust system will go a long way towards preventing backfires.

 

Like I said, my .02

[PoorManzImpreza]

why does richening idle mixture make it go away?

[NorthWet]

why does richening idle mixture make it go away?

How are you richening the mixture? (Its not that I don't know how, but sometimes the procedure we think is correct isn't. )

[Adam N.D.J.]

Also, new cars don't do this because they are fuel injection. The computer systems that monitors everything takes into account engine RPM, airflow and vacume, it will notice that the car is in hard decel and adjust for this. Also a lot of cars actually kill the fuel injectors on hard decel so that there is no fuel in the cylinders during ignition, and there for no fuel will enter the exhaust. They do this to save cat-converts more than anything.

[555Ron]

The air suction valve can also be a cause. If you look at you two cylinder heads, where the exhaust joins there will be a small 1/2" pipe. This will be connected to the ASV. This joins to a black plastic silencer then to the air intake. This injects air into the burnt mixture to get air into the cat converter to help it burn hydrocarbons. If it is too rooted it will put too much air in and let air flow in both direction, causing a backfire. You can buy a genuine one from Subaru of America for $14, a replacement costs >$300 if you replace it with the original, but the cheap one will bolt up easily and work and last better.

 

Only carbed, non feedback systems have this feature. Check it out first and PM me if you want the part number.

 

Also, one of those things i forget the name of that hold the carb open after hard deceleration are meant to help also. The long manifold means the fuel can take a while to reach the cylinders and there is no air because of the closed carb so you have a very rich mix. Not sure how on the money this theory is, as others say it shouldn't backfire if the exhaust and engine are all working correctly

[NorthWet]

The air suction valve can also be a cause.

...

 

Only carbed, non feedback systems have this feature. Check it out first and PM me if you want the part number. ...

In the USA, we have plenty of EA81 (and I believe EA82) with feedback carbs and ASVs. On these models, the ASVs are only activated during the first 2 minutes or so after startup, before the cats and EGO sensor come up to temp.

[555Ron]

oh, ok. My bad, sorry.

[NorthWet]

oh, ok. My bad, sorry.

If this is in response to my post, no worries. Our various emissions laws in North America (Canada has some slightly different equipment) causes us some confusion. EA81s here have none, one or two ASVs, and the y-pipe is different between, at least, the one and two ASV models. I had to pull a replacement engine out after I realized that it had the wrong number of ASVs for the Brat that it was going into.

 

If this was NOT in response to my post, then MY bad.

[KStretch55]

Perhaps a bit of clarification will help get everyone on the same page, too. What the original poster described is not "backfire". Backfiring is usually due to incorrect valve timing resulting in combustion/exhaust cycle gases coming "back" through the intake manifold/system. If it happens to you, you'll know exactly why it's called "backfire". The best way to put it out is to keep cranking the engine so it will suck the flame back into the engine.

 

What the poster described is afterburning, results from unburned fuel getting into the exhaust system and being ignited by following exhaust gases or hot components. It can be caused by incorrect timing, faulty ignition parts, rich mixture, etc. But, as someone stated, the usual culprit is the dashpot or a similar component. The dashpot keeps the throttle from slamming closed and creating a rich fuel condition. If you watch auto racing on TV and see flame coming out of the exhaust as they go into the corners, you're seeing the same thing. It's prevalent during racing because the exhaust systems are usually set up to have little or no back pressure (headers, big pipes, no muffler or CAT). A very slight amount of back pressure is engineered into modern engines and exhaust systems. If you remove it all to "help your engine breathe", sometimes you're not helping.