Alright, so a bit of history on the car here. I have a 1987 Iroc-Z, which had a sluggish 305 in it. I overrevved it and spun a bearing so I decided to put a built 383 in it. Flat top pistons, AFR 195 heads, Comp XFI cam (their mid-range one, I can find the stats if it's relevant), dual exhaust with no h or x pipe. I just got the engine in not long ago, and have started tuning it using an Ostrich 2.0 and an ALDL datalogging cable.

I'm learning from scratch and the forums here, which is helping greatly (there are some awesome posts here, but they're sometimes hard to find). I'm still WAY in over my head though so you might expect to see many more questions from me.

First question: I have a Wideband O2 sensor about 3ft down from the shorty header collectors on the passenger side. I have a stock narrow band o2 sensor running into the ecu from the drivers side header pipe, about 8" from the head itself. My question stems from an odd condition I have noticed at idle, where my wideband reads dead rich, and the ECU thinks it's lean (high BLM/INT). The odd thing, is that when the wideband is reading 12-13:1, the narrow band is flipping back and forth like it normally should. IF I'm reading horribly rich, shouldn't the narrow band read solid rich too due to it's accuracy only being within a couple of tents of stoichiometric?

Could this be caused by a high overlap cam, and intake air being pushed into the exhaust while both valves are open, or is it expected for the narrow band to flip back and forth, even with the real AFR so rich and the ecu thinking it's lean?

 

yes.... I recently tuned a 383 HSR AFR with a big cam in it and had to change the O2 calibration tables alot to have it idle good in closed loop with AFR around 14-15......

 

But which should I believe, the ECU, or the wideband since one says dead rich and one says dead lean?

 

What cam is it?

How does the plugs look like? and how does the exhaust smell?

On the 383 I tuned the exhaust smelled very rich at idle!

my first guess is trust the WB

 

I chased a similar symptom for over a year. Found that I missed a ground pig-tail when I installed my motor. Once hooked up everything worked fine.

 

Is it possible that an excessively rich condition can be read by NB02 sensor as lean as having incomplete combustion with high levels of unburned 02 present?

Maybe unrelated. I remember that with EGT probes monitoring combustion temp there is a "bell curve" to WOT ext temps where excessively lean can show as lower temp rather that hotter as you may expect. I think that was due to fresh air cooling down ext manifold.

 

I'd trust the Wideband.

Careful on trusting either the WB or the factory O2 at idle with a cam with lots of overlap.

I have run my car on its 383 all year in open lop as I haven't taken time to figure out the closed loop control with the 02. Car's running great in open loop...but I want to get back to closed loop to help fuel economy.

Anywho, good luck.

 

I tried OL for a couple months but due to varying air temps here in WI my mileage was poor(-3 mi/gal) and tune obviously was therefore off.

I believe the WB is Ok with overlap at WOT. right? Only at lower RPMs is the reversion an issue.

I used to run EGT on by 2 stokes. Unit allowed 3 probes. Does anyone know of anyone to have tried EGT? Maybe just one probe in header at head just to see what temps look like. System not all that costly.

 

WB02 is fine at WOT...

 

You are going to have issues with overlap at idle with the stock O2 tables. You will have to offset them some to match 14.7 or so air fuel ratio and BLM of 128, else your ecm will report different activity than whats actually going on.

Trust the spark plugs over everything else. At idle let it run and then check the plugs to see whats going on. They will verify which sensors are right.

I ran closed loop for awhile on my 383 and ignored the idle BLM and just went by wideband and plugs and I have more overlap than any of the XFI lineup. During normal operation as rpms increased, there was less overlap effect, and BLMs worked out just fine, around 128 +-3 or so, and air fuel in the mid 14's most of the time.

I eventually went open loop and kept the laptop in the car at all times to adjust summer vs winter tune.

There are o2 tables to correct for this and I have played with them alittle but never did get it all worked out before going open loop. they are based on g/s MAF air flow, so figure out what your idle air flow is which should be around 13-15 grams a second i believe, and adjust the table milivolts. I assume the lower end of the table around idle has higher milivolts than other high flow regions because of AIR systems on these cars for emissions. With fresh air being injected the ecm needs to account for that and the 02 mv swing point was increased to the rich side to bring idle back in line. My guess is that increasing that some more may help with an overlap situation, but not 100% sure. I'm tuning a car now with a 280xfi cam on a 360 L98 based motor and at idle its showing dead lean but at normal cruise BLM's and air fuel is correct. I will attempt to play with o2 stuff again.

Narrow band flips back and forth when its operating correctly in closed loop at the right temperature. )2 sensor has to get hot enough to work right. Only at WOT will it report one value and not flip. If you are rich at WOT, it will be a higher value. Generally 12-13 to 1 air fuel will probably be around 880-910mv. Most richer cars in the 11 to 1 range or richer seem to show 930 and higher. Very lean will show much lower mv at WOT. Its a good tuning tool to verify what the wideband is saying as well.

 

A few notes and observations:

It turns out that either the o2 sensor itself was bad, or it is too close to the head. Turns out the bung I put my o2 sensor in was supposed to be for an EGT sensor, it's about 3-4" closer to the head than the stock manifold has it. I ran my LC-1's narrow band simulated output into the ECU (Which is 12" past the collector), and immediately I was seeing mid 14's on the wideband, the ECU adjusting like it's supposed to in my mind.

While in closed loop if I hit the gas, it will go rich (drop to 12-13:1) and then adjusts, showing I need to tune it. That's not the interesting thing though, the narrow band goes hard rich (pops up near 1v) and stays there until the wideband gets back near the 14's, which is what I would assume that it SHOULD do. This is in contrast as to what was happening before and can only lead me to believe that having an o2 sensor that close to the head with a high duration/lift cam can cause issues with fresh air being in the exhaust. I suppose I should adjust the o2 sensor calibration to bring the idle BLM's closer to 128 now?

Another question for you guys: I have three coolant temp sensors; One in the head for the dash gauge, one in the intake for the ECU CTS, and I assume the other one is for a dummy light? The only reason I'm concerned, is that the datastream I have has a place for a intake temp sensor, but I don't have one. All three of these sensors were in the coolant path on the old engine, so I put them back on this engine I just want to make sure that I'm not doing something wrong here.

Overall you guys have all been super helpful, I've got a book of information from posts on here as I learn, and it's constantly growing

 

The third sensor could just be a fan switch....Its located in the Pass side head on a std TPI engine.

As a note... On my buddies car I will now swap to MAP system instead, much easier to tune with a big cam.......

 

I'd put the narrow band in the header collector to get the average of all 4 cylinders on one bank. One cylinder could show leaner or richer than the others depending on air distribution in the intake manifold, so if your nb was reading one cylinder or just 2 cylinders, it could be getting different results than the wideband if the wideband is reading all 4 cylinders.

WHen you say hit the gas, are you going WOT or just half throttle acceleration or something? If somewhat quick hit on the gas but not necessarily going WOT, it could be AE kicking in which will momentarily show rich conditions and eventually lean back out to stoich as it decays out. I think the INT/BLM values drop out when AE kicks in so it may jump back to 128 momentarily. I know it does this in PE mode. When this happens the o2 sensor may jump to a steady milivolt value, which near 1 volt would be extremely rich. This happens in PE mode but not sure if it happens in AE.

I'd really work on getting most of the part throttle stuff dialed in to see if your BLMs/INTs hover around 128 when the wideband is in the mid 14's to 1 area. If thats the case, the o2 sensor parameters are fine for the higher MAF flows, but at idle you may need to adjust the parameters for low grams/sec flows.

Give it what it wants at idle to make motor happy and stable, and watch what the BLM/INT's do. YOu could ignore them at idle due to possible cam overlap effects skewing the sensor readings, and just read the plugs to be sure the mixture is ok, or you can go ahead and try to adjust the swing points/o2 settings at those idle grams/sec flows. You also may find that blm of 128 at idle may be much leaner than 14.7 to 1. Not all engines want to idle well at that air fuel ratio anyway, so just work it til its happy.

 

I am currently using the wideband's simulated narrow band output and it is working fine, do you see any issues to continuing like that? Is it worth pulling the exhaust apart to weld in another bung for a real narrow band sensor?

Yeah it's just a part throttle blip (like 1/2 throttle enough to bring the rpm's up some. I haven't actually driven the car enough to get any real acceleration values. That will be my next step once I get idle and stopped-in-drive load values more acceptable.

I was thinking, adjusting the o2 parameters would change what A/F you're actually running at when BLMs/INTs are at 128, correct? So I should tune the MAF to 128 while driving around, and then adjust the o2 sensor offsets if it dosen't quite match the A/F I'm looking for, or am I misunderstanding what these values do?

To be honest, it seems fairly happy slightly rich, I'll play around and see if once I bring the BLM/INT down to 128 if it still seems as happy as it does now. The plugs seem to match what I'm seeing on the wideband, so I'm fairly confident with the wideband measurement.

 

Well it looks like tuning is on hold for a while. I'm runnning into the while '165 communication issues that everyone seems to have. The ecu refuses to talk to me all of a sudden after sitting for a week. The car still starts and runs fine, but nogo on the communication. Hopefully if I leave it for another week it will start talking again :/. It has always had intermittent communication but this is the first time I haven't been able to connect for more than an hour.

Edit: On a side note, 160baud communications do not work via WinALDL either.

 

Yeah I do believe you will change the air fuel ratio with swing point modifcations up to a certain gram/sec air flow. Some motors like idling lean and some can handle slightly rich. 14.7 to 1 at idle isnt necessarily optimal and I believe at the lower MAF flows for idle I think a BLM of 128 actually gives something abit leaner on the wideband. The factory offsets are higher for those values to compensate for AIR injection I believe so a 128 BLM idle could be 15-17 air fuel.

I've never used the narrow band output signal but some have used it just fine. I have also heard some guys having problems with it being lazy, not switching back and forth causing erratic fueling in closed loop.

 

I've been using the narrow band output on my Innovate LC2 to feed my '165 ECM for 3 years now. Car runs good and plugs look great. Free air calibrate once per year...