Supercircuit Header - works great but MIL on, how to fix?

[foama]

foama - I do not believe that is correct. Not at all. And, you're making BIG assumptions. And I don't believe your assumptions are correct. And besides, I don't think the 1st O2 sensor is the issue with the Check Engine light that dikulus is getting. I don't remember if it was this forum that mentions this. But (as I understand, granted I might be wrong), the 1st sensor is reporting AFR (oxygen) data to the ECU. The 2nd sensor is reporting probably oxygen data to the ECU ... but that signal is used in order to confirm catalyst performance. Therefore it should be the 2nd sensor that is the issue at hand. I'm probably making assumptions too. For one thing, I'm only referring to Mirages that run 2 O2 sensors. dikulus and Basics are those.

Don't misunderstand me. I don't think you're dumb. Quite the opposite. I just don't believe you're correct in this case. There are a lot of reasons why I don't think your assumption is correct. I actually run an AFR gauge. It is probably as high tech, possibly more so, as what is used by OEMs. It is attached to one collector on my V8 engine. Both sides have a bung and I can switch sides. It is MORE THAN capable on just the one side.

However, why I think your assumption is wrong is because, outside of actual design work, the 2nd leading amount of time spent as a design engineer is VA/VE or cost out. Every OEM will push their engineers to make everything on the car as inexpensive as possible. In turn, design teams will push tier 1's (suppliers, like NGK, Denso, etc) to do the same thing. Tier 1's will have maybe as much as quarterly, but at minimal once a year reviews with OEM engineers to review every possible idea and variation of the products they supply to said OEM, to reduce cost. These reviews are detailed down to dissecting a gnat's anus. I disliked these efforts. Too time consuming, for small reductions. But across thousands of parts, the reductions can be quite substantial. So it is always a big effort. The point being, if slowing down the sampling rate in some way reduces cost but continues to provide enough sampling to run well, the sampling rate will be slowed to minimal. I'm not saying that's the case, I don't actually know, it's an example.

I'm not saying an OEM O2 sensor cannot "sniff" every puff, it could likely take multiple samples per exhaust stroke. And heckfire, if you place it in a manifold where all cylinders' exhaust accumulate, how would it then distinguish between the 3 individual "puffs?" They blend. What I am saying is that it is not necessary to do so in order to gain the data needed in order to run properly. Perhaps it's sampling more than every individual exhaust stroke, perhaps it's sampling less. Its sampling at whatever rate the sensor is rated to sample, based on meeting lifecycle (warranty) requirements and cost standards. Automotive engineering is VERY complex. Decisions about how to design and solve problems are sometimes based on the most UNobvious reasons. Been through MANY decision meetings at Toyota and those meetings were a REAL eye opener. Much respect to the Toyota process of making design decisions. Hyundai simply makes decisions on how much things are like Toyota and, hey it's 1 cent cheaper, do it.

Basic - I sent dikulus a PM asking exactly what error codes he sees. I didn't send you one because, I guess I've missed where you mentioned your set up. As I recall you purchased ... drawing a blank, major contributor here / previous autocrosser, his blue Mirage. Now your blue Mirage (3 pedals, no cruise control).

However, if you don't mind telling me your setup, and what codes you see (dikulus too), we can go from there and see what we can learn. There is a solution. Will we find it? Dunno. It may be too complex and/or too expensive.

I agree the check engine light is probably due to the readings of the second sensor. It expects to be behind a cat.

The over-rich fuel/air ratio could be coming from the non plausible readings of the first sensor. Since the fuel/air ratio of these cars is calculated in realtime for each cylinder by using the data of the first sensor, the ECU increases the fuel amount as a fail-safe measure.

[7milesout]

The over-rich fuel/air ratio could be coming from the non plausible readings of the first sensor. Since the fuel/air ratio of these cars is calculated in realtime for each cylinder by using the data of the first sensor, the ECU increases the fuel amount as a fail-safe measure.

I don't want to sidetrack the thread too much, so I'm going to have to *try* to not be such a PITA. And maybe I missed something, I always do.

But, I don't know from what source you have that indicates an "over-rich fuel/air ratio."

What substantiates what you mention is "non plausible readings" of the first O2 sensor?

How can the AFR be calculated for each cylinder if the O2 sensor is located in a manifold that allows for mixing of all 3 cylinder's exhaust?

The ECU might richen the mixture under certain circumstances, but you're implying it is doing it simply by running this header. I don't think that has actually been established.

Like I mentioned, I believe you're making too many assumptions.

I'd like to know the error codes for Basic and dikulus, compare them to the 1 or 2 codes I got from my son's RX8. The primary code I got on his car was P2270, O2 Sensor Signal Biased/stuck Lean - Bank 1 Sensor 2.

This is what I think (don't know) goes on with the rear O2 sensor under normal circumstances. A catalyst reaction consumes the oxygen in the exhaust. With little to no oxygen remaining in the exhaust after the catalyst, an O2 sensor that is made to detect oxygen would detect none or very little. Thus the "normal" signal from a post cat O2 sensor would be a "rich" signal.

Why? A rich combustion event consumes all the oxygen. A lean combustion event leaves oxygen behind. So after a nominally functioning catalyst, an O2 sensor's signal would provide a "rich" signal to the ECU. So the ECU would be programmed such that the rear O2 sensor reading rich means that the catalyst is working correctly.

In the case of the RX8, with no cat, there would definitely be more oxygen in the exhaust. Hence the P2270 code (stuck lean) making sense. Does that mean the engine is running lean? No. The engine AFR would not have changed, it is controlled by the first O2 sensor. And from what I've read, the ECU uses ONLY the first O2 sensor for AFR adjustments, the 2nd O2 sensor signal is used exclusively for cat status.

The translation of the P2270 code means the catalyst has failed. Theoretically a minicat should fix that problem. And in the case of the minicat I'm running on my son's RX8, it *seems* to be working. There's no CEL any longer. I should go check the codes anyway, there may be one pending. Dunno.

[foama]

@ 7milesout: Maybe I missed or mixed up something too.
The AFR can indeed be sensed by the linear AFR sensor. This kind of sensor gives practically instantaneous response. It sits in the manifold where each exhaust "puff" must pass. The info coming from it goes to the ECU, and the ECU also controls timing and injection amongst other things. Therefore the sensor sends info about the AFR of each individual cylinder to the ECU. This is one of many things or "tricks" done to reduce fuel consumption. If the injectors put out differing amounts of fuel, it would at least hurt economy if not corrected.

If the aftermarket header has the linear AFR sensor in one of three branches before they unite, it will only sniff the gas of one cylinder.
If the length of the branches are too long before the uniting point, with the linear AFR sensor placed within one branch, that linear AFR sensor can not distinguish the gasses (AFR) of individual cylinders any more.
If the sensor is at the uniting point and the branches too long, the reading also becomes non plausible because of delays after correction.

If there is no second cat any more, then of course the downstream AFR sensor will see a lot more oxygen than with a second cat in place.

There are plenty of seemingly little unimportant details in our cars that are often not seen or comprehended. Before modifying anything, we should thoroughly think of what we are doing and the consequences thereof.