Moving a wideband affects tune... how?

[white92tsi]

I just installed a 3" turbo back exhaust (no real muffler to speak of). The wideband was moved from right on the down pipe to a location about 36" from the down pipe. I see that my tune was affected. My AFR's are a little wilder in closed loop (using narrowband simulation on ecmlink), and for some reason, I'm getting a bit of knock on the top of my pulls. I'm keeping my foot out of it for now, and will be tuning it out soon.

I'm just wondering if it's the movement of the wideband, or the 3" turbo back that is affecting the changes to the tune more. Anyone have any input on this?

Thanks

 

[1G Eclipse94928]

any reason why you cant keep it up closer? I thought there was a upstream and downstream sensor, but maybe not?

 

[donniekak]

The 3" turbo back will affect your tune, especially if you are running speed density. The further away from the turbo the o2 sensor is, the longer the delay between actual afr, and what the sensor sees. Not really a big deal, unless you are using the sensor for feedback in closed loop with the ecu. The stock o2 sensor location provides the best fuel trimming in closed loop.

 

[c2ypt1c]

unless you are using the sensor for feedback in closed loop with the ecu.

In which case there is a delay in fuel adjustment from the ecu because of the time it takes for the exhaust gas to exit the cylinders and travel to the sensor. But this really only applies to closed loop operation.

Having the sensor further downstream increases the potential for exhaust leaks because there might be additional flanges, connections, etc, involved, which would certainly affect the open loop tune.

 

[donniekak]

In which case there is a delay in fuel adjustment from the ecu because of the time it takes for the exhaust gas to exit the cylinders and travel to the sensor. But this really only applies to closed loop operation.

Having the sensor further downstream increases the potential for exhaust leaks because there might be additional flanges, connections, etc, involved, which would certainly affect the open loop tune.

He is using the wideband in link for a narrowband signal for fuel trimming.

 

[c2ypt1c]

He is using the wideband in link for a narrowband signal for fuel trimming.

Yeah I know I was just agreeing with you and clarifying a bit more. Also expanding on how it can affect open loop

 

[white92tsi]

I assumed there would be a small delay in the closed loop... hence the wider swings back and forth. I'm not very concerned about that so much as the knock that was introduced at the top of my pulls. The reason it concerns me is because my timing was pretty conservative and I had 0 knock before making the change. I'm hoping I'm not going to have to turn the timing down much to get rid of it... or at all... I might recalibrate my wideband (LC1) and do a bit of tuning with fuel before I touch the timing. It's holding at about the same AFR as before though.

I can't put the wideband back in the stock location because the turbo back is a straight downpipe setup that was made for a car that had an FP manifold on it. It will not fit in that location with the stock exhaust manifold. I wish I could though. I had 0 issues with it there.

There are actually fewer possibilities for exhaust leaks because the pipe has no joints from the downpipe connection to the turbo until after the wideband bung. It's actually a pretty nice setup... except for the extra pain in the ass getting it installed (without removing the turbo).

How does the 3" exhaust change the tune in a way that would add knock where none existed before?

Thanks for the discussion.

 

[c2ypt1c]

Now that I think about it, placing the the sensor further downstream can also affect open loop because of the delay. In open loop, the ecu tries to target an AFR based on rpm vs load factor. When we dial in our airflow, using mafcomp or the vetable, we adjust it in order to match AFRest with wideband AFR (In other words, zero WBFactor). Note that AFRest is an estimation at a particular rpm and loadfactor.

When you move the sensor downstream, AFRest could be different by the time the wideband sees the corresponding exhaust pulse, because rpm and load factor are changing within that time period. So this might affect the tune because the airflow is not dialed in correctly.

However, this shouldn't change anything in your case because you already dialed the airflow in when the sensor was closer...unless you made changes after moving the sensor.

 

[white92tsi]

I haven't touched anything on the tune yet. I'm just keeping my foot out of it for now....

I will try to tune out the knock and just deal with the wider open loop swings for now. When I can, I'll pick up a FP exhaust manifold and move the wideband back up to the front.

The open loop AFRs look the same. Almost spot on at 11.0:1 10.9 in a few spots. I hadn't quite gotten to the point of leaning it out when I put the exhaust on and started seeing knock again... it's a never ending affair.

 

[c2ypt1c]

The open loop AFRs look the same. Almost spot on at 11.0:1 10.9 in a few spots

Yeah, what I said is pretty much the theory behind it, which might not really be practical (imagine placing the wideband extremely far down, near the muffler for example). My guess is that the exhaust gas is traveling fast enough, in open loop, to not make a noticable difference.

I got an idea! Do you have before and after logs? Post em!

 

[white92tsi]

No logs from before. I dont tend to save them past tuning. I might have something recent though.

 

[Mello]

I don't think your O2 sensor really has anything to do with your tuning change. I think it's from the 3". You now have less restriction and your motor VE is different. My guess is your peak torque is happening sooner in the RPM range and the timing at those load/RPM point is to advanced for your given fuel. Another guess is it's happen between 2500-3000rpm at about 0-5psi with pump gas(?)

 

[white92tsi]

It's actually happening a little higher in the RPM range than that, but yes, it does seem to be happening a bit earlier than any knock I have gotten before... somewhere between 4500-5000, so that very well could be. I'm going to be putting my Evo III 16g in tomorrow, so I'm holding off on refining the tune until then. I don't care to waste the gas with how expensive it is.

I'm not getting any knock on part throttle (about 70-80%), but she seems to have quite a bit more pep after the exhaust install.

I was thinking something along the same lines... that the opening of the exhaust from the down pipe has changed the power curve. She's basically hitting a higher load (peak boost) slightly faster than before which would change the route the ecu takes through the VE and timing tables. Seems to me like a complete re-tune is in order, and definitely will be after I put the big 16g on.