[Rich Z]

Well, OK. I've learned a lot since I started poking around with this tuning stuff, but it seems like people are missing something important when talking about a closed loop system. Heck, even an open loop system has issues that I have yet seen anyone address.

First off let's describe how a typical closed loop MAF based system, like we normally see on our later model Corvettes operates. You have a MAF sensor that determines the air charge based on a frequency generated by the air flow passing through the sensor. This tells the PCM what charge of fuel to spit into whatever cylinder that will be firing when that air charge hits the cylinder. Then afterwards, the oxygen sensor in the exhaust detects the proportion of oxygen in the exhaust from that cylinder firing and reports back to the PCM about whether or not that fuel charge was correct to produce the desired air/fuel ratio. Pretty over simplified, I think, as there is a devil in the details.

Let's start with the MAF. There is a conversion table used by the PCM to convert the frequency detected to an actual measurement the PCM can use for airflow. This table can be modified via aftermarket tuning, and obviously in a stock tune it is an average of all the vehicles that GM did the development of the tuning with. Fine tuning, in an aftermarket sense, can tailor this table to be more realistically relevant to the engine in YOUR car at THAT time. Maybe a more finely tuned MAF table will better enable the PCM to more accurately determine what that aircharge is going to be in the engine's cylinders. But then again, maybe your engine is just average, and tuning will not improve this at all.

So here's part of a problem. When the MAF detects the airflow THAT air charge is still in the MAF, and not in the engine. So there is a delay that has to have been programmed into the PCM (and as far as I know, there is no way for aftermarket tuning packages to change that) that tells the PCM, "hey, get ready, the aircharge that is going to be in the cylinders is XX.XX sometime in the near future." And, of course, the VELOCITY of that air will change with engine speed, along with volume. And this is a complication because velocity and volume can have the same effect as far as the frequency of the signal detected. The MAF works via a heated element that is cooled by the passage of air over that element and generates a proportional frequency. A heated element can have the temperature changed by both volume and speed of the airflow. But I'm sure this has all been calculated out by GM during the tuning development using a STOCK air intake system. So here's the rub. Change ANYTHING in your air intake that changes either volume, air speed, or both, and you throw those calculations right out the window. Change the DISTANCE between the MAF and the engine cylinders and you've really thrown a wrinkle into things. Now that charge of air that the MAF has told the PCM is heading towards the engine cylinders might get there before or after the PCM thinks it SHOULD be there. So the fuel charge might be completely off because the calculated fuel charge doesn't apply to the air charge that is actually in the engine cylinders at that time.

So let's muddy the waters even further. In a CL-MAF (closed loop) system, there are the O2 sensors on each of both of the exhaust pipes that are reporting back to the PCM what the oxygen content of the exhaust is so the PCM can determine how close it's calculation of the proper fuel charge to apply to the air charge detected by the MAF was. So if the calculation was incorrect because of the before mentioned CHANGES in the airflow the MAF is detecting, then the PCM is going to be jumping around trying to give an AFR calculation that satisfies the oxygen sensors. Which, in reality, it might NEVER be able to do, because it doesn't know that it is basing it's calculations of a fuel charge on the wrong air charge. Granted, at an engine speed that doesn't change, this would not be an issue, since the air charge should be stable, and correspondingly, the fuel charge should be as well. But during rapid changes of engine speed, well, things could be changing faster than the the system can realistically work with. How fast is TOO fast? Beats me. Maybe it's really not that much of a problem because the changes can't be drastic enough to be very significant.

So back to the oxygen sensors. They are also trying to detect a particular charge of air and fuel based on the amount of TIME it takes for that charge to get from the cylinder(s) producing the exhaust to the sensor itself. So then you go an put headers on, that will generally move the position of the oxygen sensors further away from those firing cylinders. So what does this mean? Are those sensors now detecting a charge of air and fuel that is NOT the charge or air and fuel that the PCM thinks it is getting feedback from? So, again, the PCM could be thrown akilter trying to compensate for either a too lean or too rich mixture based on this feedback that is not accurately produced by what the PCM calculated at all, because the detected exhaust is NOT of that charge the PCM thought it was.

So yeah, if you go and change your intake to a cold air intake, put in a larger throttle body, put on a bigger MAF, and put on headers, you just threw a monkey wrench in all those signals that the PCM is counting on in order to create the proper calculation for the air/fuel ratio. So heck, in my case, I go and put a 427 cid engine right in the middle of this, which is going to have different air movement needs altogether from the LS6 engine, and you've got a real mess on your hands.

There are tables in the PCM that are able to be edited to tell the PCM some things about such changes, but so far I haven't seen anything that can touch the TIMING between the events needed in a Closed Loop MAF based system. So it seems to me that using such a tuning system in a moderately to heavily modified car will be difficult, at best, to get a perfect tune for any longer.

The reality of it, from my naive and novice knowledge so far, is that an Open Loop Speed Density tune is the ONLY way to realistically go. This sort of tune relies solely on the VE (volumetric efficiency) table for the air/fuel ratio calculations the PCM does. Don't use the MAF at all. The air charge is a calculated value from the VE table based on the vacuum or boost of the engine and the speed of the engine AT THE TIME that this information is needed by the PCM. Being Open Loop, the PCM doesn't care what the oxygen sensors are telling it. It is presuming that the values in the VE table are correct and that is all it needs.

Of course, there is still a wrinkle in this. When we are trying to tune the VE table, we are still dependent on feedback from a wideband O2 sensor. And the same issues apply with trying to determine if the air charge that we are getting feedback from at the wideband is actually applicable to the VE table entry that the PCM is using. This can be compensated for pretty handily, though, by spending a lot of time doing steady state tuning, meaning, gathering sampling data from the engine at consistent engine speeds over the entire range of combinations of engine vacuum/boost and engine speeds. If you have those values as close to perfect as you can get, without any oddball steep transitions from one cell to the next, that will be as close to a perfect tune as you can expect to get. At least as far as the basic air/fuel ratio the PCM calculates using the VE table. Of course, there are other types of modifiers that the PCM uses (coolant temp, incoming air temp, etc.) but they can be managed much more easily if the basic values used by the PCM are accurate.

Anyway, that sort of stuff has been bothering me about this tuning stuff. It may not be really all that significant, but I read so many accounts of people doing tunes on cars where they put the car on a dyno, make a few pulls, make some changes to the tune under wide open throttle, and call it a day. Seriously, if you do ANY changes that changes the airflow characteristics of your engine, it is my opinion that you are getting short changed. Especially if you are still using a basically stock tuning methodology using the MAF and oxygen sensors in a Closed Loop configuration. Quite likely, when you are not driving in WOT, your PCM is working overtime trying to compensate for values that are making it swing widely from producing lean and rich fuel ratios. A good way to check this is to disconnect your battery for a while, and then after reconnecting it, take the car for a drive. Does it run rough for a short time and then eventually smooth out? Well, your tune is screwed up, and is getting smoothed over by the short and long term trims. THEY are compensating for a tune that is not even close to being optimal for your car.

IMHO, of course.

I'll proof read this some other time. It was tough enough just typing it in...