exhaust leak cause artifical AFR's?
 

ok so my wideband is reading anywhere from 16.1 to 13.3 under WOT full boost, i have no knock at all, my wideband sensor is located just before where cat would be (DC Sports downpipe) its on the correct angle and the weld is great, i replaced the sensor and borrowed a friends gauge and i replaced the donut gasket nothing different happened....

a few people i know who have tuned suggested that i might have an exhaust leak from either the donut of where the o2 housing meets the hotside or a crack in the o2 housing itself, sound correct? any suggestions on sealing the gasket to the hotside better?

An upstream leak generally appears leaner than you really are so thats probably it. The areas that can leak are the exhaust manifold gasket, turbo inlet, turbo outlet and the downpipe donut gasket. As you can see its real easy to have a leak. Donut gasket is most common, followed by the turbo gaskets and then the exhaust manifold gasket. Has the exhaust manifold ever been removed?

the manifold has never been removed but the car does have 36k on it of daily driving, including winter snow and rain

Well, a few things I find helpful to tell if there is a gasket leak is to look around the gasket area for carbon where air is being forced out. Also, after a car has sat through a cold night and you first start it up, sometimes condensation is pushed out of the leaks so look for any water being forced out when you initially start it for the first time in the morning.

Donut gaskets are definitely the most common leak but I suggest getting one from the dealer because the closest I've found at an autoparts store was for a 2003 Honda CR-V but it might need some sanding on the inside with a dremel to fit correctly. If you go that route (I don't suggest it) make sure you can fit the donut gasket on the O2 housing by HAND, if a hammer is required it will crack. Thats all the advice I have, maybe some other guys will chime in ;)

well i already replaced the donut gasket and the old one seemed to be in good shape, so idk, i have a new setup going in soon, and im building it with alot of care and im going to tighten everything very well, and go with torque specs and copper seal on all the gaskes, this may also help...but this condition only happens under boost, at idle and under low load no boost conditions there isnt a leak the AFRs appear correct, then as i make boost they go all outta wack

I have the same problem going on. I am leaking exhaust between the turbo and the exhaust manifold. Took my Afr's from mid 11's to mid 13's, scared the crap out of me at the track but no knock so it is most likely artificial. I still don't understand how an exhaust leak would cause the wideband to read lean. There is less exhaust but wouldn't a small sampling of exhaust coming out of the engine have the same a/f ratio as the whole thing. If I put a/f sensors on each cylinder wouldn't they read the same as one sensor downstream? I don't get it. Or is it that the engine is sucking in exhaust thereby screwing with the a/f ratio but wouldn't that make it read richer? Any engineers in the audience?

IIRC, it is because the exhaust actually will create a decent amount of vacuum and "suck" in the fresh air from the leaking area.

Ok, but how? The exhaust is pressurized and leaking out, how does it suck air in? I don't get it. If my garden hose was leaking water it would not suck in air, I would just have less pressure at the nozzle. Same for the exhaust no? Does it actually cause air to come in from the exhaust outlet, I still find that hard to believe but slightly more plausible. Anyway. I guess all that really matters is that it seems to be happening.

The exhaust fumes follow the path of least resistance and your exhaust piping is just that. There is a GREAT deal of exhaust trying to escape at once and it will try to "follow the pack", for lack of a better term,lol. Once it creates a flow, it will indeed suck air IN from the leaks and introduce air into the stream like zl1560 mentioned. This suction is known as "scavenging" and is why sometimes its better to use an exhaust system rather than go with open headers on some setups because the 'scavenging' effect literally aids in the exhaust being pulled from the combustion chamber, helping to lower the back pressure even more than an open header could.

+1 I went for the short answer.

The same theory applies in an exhaust scavenging crankcase ventilation system .

Those are a bit tricky and I really want to do that to my evo. The trick is the exposed tip that protrudes into the exhaust has to be just right or you actually create too much vacuum and pull oil thru.

Go put your hand over the exhaust at the end, feel the "pulsing"?

in between the pulses is when air can get into the exhaust..

im aware of it it just sucks cause i dont want to bother correcting my tune untill i put my new setup in, and all thats stopping me from doing that is 2 bolts and time lol bothers me cause i need to get it done

I have the Moroso one from Summit and it works very well. Comes w/ instructions and all. You won't pull oil through if you attach the hose to the PCV valve at the top of the valve cover. BTW, I hollowed out my PCV too, to reduce resistance to the vacuum.

http://store.summitracing.com/instru...%2D25900%2Epdf

I agree that the vacuum idea is prevailant if the leak is past the turbine, but if it is before the turbine the gasses are so pressurized and flow is so turbulent that I think the only direction that gas is moving is out. I blew my turbo gasket last weekend at the race, I'll check my logs when I get time.

Please do, very interested to see as I also blew out my turbo gasket, the bolts loosened up on me, lock washers on them now. I was at Watkins Glen when it happened. What track were you at?