This is a long read that I hope will offer some insight as to what occurrs when your car gets tuned. Others will likely chime in and may even argue some areas. What I have discovered is that only the manufacturers designers know for SURE how they work but MANY people have theories on how it works.
Since so many people want to know how these EFI cars work, I am going to explain to you the principles on how Electronically Fuel Injected cars work and operate based on my experiences. When I wanted to know how to do this I was shocked at the lack of information out there. I learned through the old school of trial and error and through watching someone do it that I paid just so I could see what they do. The best reason I could find for the lack of knowledge available is the good ol' mighty dollar. Many tuners don't want you to know how it works cause it cuts into their $$. Being the type of person that likes to squash this, I'm going to give you the basics and you can do with it what you want. My experiences are actually in GM land but the principles are the same. Maybe someone in Ford land can elaborate on how these theories apply in Ford land.
With EFI, there are many components required to for the engine to function. Your basics are still there. It takes three things to make an engine run. Fuel. Spark and Air.
Lets talk about fuel first. In an EFI car, an electric fuel pump pumps the fuel out of the tank, into the fuel line/rail and through the injectors. The injectors open and close very fast for a very specific amount of time. To visualize, if you turn on the garden hose and hold a kink in it to keep the fuel from coming out, you are simulating the injector in a closed state. When you un-kink it for a second or two, a certain amount of water will come out while you are unkinked and when you kink it again, the water stops. You are simulating a fuel injector opening and closing. It just opens and closes quicker, as in a certain amount of miliseconds. If you hear the term Pulse Width, that term is defining how long the injector will stay open when it opens.
If you are at my house, I have low water pressure. If I open the kink for one second, a certain amount of water will be allowed out. If you have more water pressure at your house, and you open the kink for one second, more water will come out because you have more water pressure.
Same goes for Fuel Pressure. If you have 43 pounds of fuel pressure and you open the injector for 10 ms, x amount of fuel will come out. If you have 50 lbs of fuel pressure and open the injector for 10 ms, Y amount of fuel is going to come out. So there are two ways to get the injector to give you more fuel. One is to increase the amount of time the injector stays open (called increasing the pulse width) and second is to up the fuel pressure so that more fuel is injected while the injector is open. Injectors are rated by how much fuel they can push. For example, 24 lb/hr. The size of the nozzles dictate how much fuel is allowed to come out during each open pulse along with how long the injector is held open (pulse width). So if you have a large fuel requirement, your injectors may not be able to provide enough fuel thus the need for larger injectors. Injectors like to be run within a certain duty cyle threshold. Typically no more than 80 % of it's maximum duty cycle (wide open). The reason you don't typically want to exceed this 80% mark is that as long as you don't, you still have 20% of what the injector is physically able to do left to play with. Not to mention that an injector forced to run 100% would die early. This means you want to be able to have the correct amount of fuel provided with the injector being open no more than 80% of the time. There is a lot of other areas that could be talked about here, but remember these are the basics.
Now lets talk about AIR. It has been determined that the correct a AFR (Air Fuel Ratio) is apx 14.7:1. This means the optimum medium between power and economy is 14.7 parts Air to every one part Fuel. This AFR is called Stoichiometric or just Stoich for short. This is a very relevant number as you will see shortly. Your car operates in two modes called Open Loop (OL) and Closed Loop (CL). While in OL, the computer is using a very specifc set of programmed parameters to allow the car to run. While in CL, the car uses the O2 sensors to determine what the current AFR is. The sensors will read how much oxygen is in the exhaust gases and based on what it reads, it will put out a certain voltage. The computer reads this voltage and can decide if you are running too rich or too lean. If you are running too lean, it will increase the pulse width on the injectors to add more fuel or decrease if you are running too rich. Your O2 sensors have to be warmed up to work right so there is typically a delay between OL and CL to allow the O2 sensors to warm up to be able for them to function properly.
Now, if you know that you want the fuel ratio to be 14.7:1, to get there you need to know how much air is coming in and once you know that, you do the math and add the right amount of fuel by setting the proper pulse width for the injector. You know that if you have 43 lbs of fuel pressure behind the fuel injectors that opening them for a specific amount of time will cause a specific amount of fuel to be injected. The way you know how much air is coming in is via the Mass Air Flow sensor. This sensor is essentially a resistor that changes as air is brought accross it. As more or less air is run accross it (via being sucked into the engine) it causes the resistor to change the amount of voltage reported to the computer. The sensor is calibrated to say that if the voltage is X then there must be Y amount of air coming in. Now that you know how much air is coming in, the computer can calculate how much fuel is required get a stoich AFR.
In addition, in GM land we use a MAP sensor. It tells the computer what the load on the engine is. I honestly don't know if this is used in Ford engines but I suspect not for a variety of reasons. Point being, you need some way to tell what the load is on the engine and Fords will have some method of determining the load. This is kinda like a vacuum gauge.
Sooo...all that said, there are tables as shown below in the computer that say "If the engine is at X amount of load, and it has Y amount of air coming into it, I need Z amount of fuel. Below you will find a shot from the main fuel tables from my 96 corvette to give you an idea of what you are looking at.
http://www.cerinosolutions.com/tuning/vetable.jpg
What you are looking at here is the Volumetric Efficiency table. There is MUCH debate on this table but i will give you my opinion on it. An engine is an air pump. It pumps air in and pumps it back out. If it were to pump all the air out that it pumped in, then it would be 100% efficient which we know doesn't happen. So, this table tells the computer how much air is actually being utilyzed by the engine. The idea is that if it is pulling in 5 lbs of air (just throwing that number out there) as indicated by the MAF sensor, and you are under 60 KPA load as indicated by the MAP sensor and running at 2000 RPM, only 78.9% of that air is actually being used and you would want to know that for determining how much fuel to add to get the AFR of 14.7:1 (if in closed loop). More on this later.
Spark:
Spark is pretty much as it has always been. Certain conditions like certain amounts of timing advance. One of the best things about an EFI setup is that you can precisely tune the timing to just about any condition. This vs traditional systems where you have to use springs and weights and that is what you have to work with. In EFI, you get to say exactly what timing you want and where.
Below you will find one of the timing tables. The one shown is the main spark table for my 96 Corvette and it says what ignition advance you want vs the RPM vs the load on the engine (KPA)
http://www.cerinosolutions.com/tuning/mainspark.jpg
Here are a bunch of things to consider in tuning:
The enemy to an engine is Spark Knock (detonation). Many things can cause Spark Knock. Things like timing being too far advanced for your current load and rpm and a lean fuel condition which can be a kiss of death.
One of the main things that a tuner would do to improve your power is to readjust your timing map (table). The mustang is a really good example. The car is made to be able to run with 87 octane. Because of this, I can only imaging that the timing table is set very conservative. Well, what if you have 93 octane available at all times? This is a perfect candidate to allow some more timing throughout the RPM range thus increase the amount of power for you.
In the case of my 96 Corvette, the person that I bought it from just had a bad combination of engine parts and the result was too high of a dynamic compression ratio which caused pinging or detonation. Fortunately, GM (and Ford) have something built in to combat this. They are called knock sensors. These sensors listen for the noise that pinging makes. If it hears it, it commands the computer to back the timing off. There are tables within the computer that you can set to determine how much timing to take out and for how long when pinging is detected. Thanks to this, my vette even with too much timing was spared un untimely death. However, had it gone for a prolonged period of time, it certainly would not have been good. The sensors don't hear knock until it occurs so the fact that it was pulling timing means that knock was occurring (generally speaking, there is such a thing as false knock detection). I scanned my mustang to see what it shows and the Fords are a little different. In Ford land I was able to scan a voltage being registered by the knock sensors. This tells me that the computer reads that voltage and based on what it is, it pulls timing out with respect to what it sees that voltage at. In GM land, the scan showed me exactly how much timing was being pulled.
Lets talk about scanning for a minute. In order to know what changes you need to make to the computer, you need to know what the car is currently doing. In the case of my 96 Corvette, I needed to use one tool for scanning (Auto Tap) and one tool for modifying the tables and pushing the new tune to the car (DST). In addition, I won't tune without a wide band O2 sensor so I know that I am not running lean. Autotap would allow me to plug into the car and record. So I would get in the car, and drive it under many different conditions recording the whole time. Then when I was done I would look for things like spark knock occurring. In my case, I had quite a bit at the cruising RPMS when I would just start putting my foot in it. Cruising down the road at 55 was about 1600 RPM with a KPA load of about 40. Start to put your foot into it and the load would go up to about 80KPA. In my case, 80KpA and 1600 RPM had knock. Since I now knew the location that had the problem, I knew where I needed to make the change in timing. What you see below first is my original timing map and then the last one I loaded to the car. Note that there is a significant difference in timing around the 1600 RPM range which is where I had the most problems. The whole map needed changed but those areas were the worst:
Before:
http://www.cerinosolutions.com/tuning/mainspark.jpg
After:
http://www.cerinosolutions.com/tunin...sparkafter.jpg
Now modifications aren't just made to the timing tables. Remember that stoich is 14.7:1. However, experience has told many that when tuning for power, 12.5 to one is a good number to have. From the factory, the car manufacturers like your engine to run rich under WOT conditions. This protects you from detonation blowing up your engine with the cost being not having as much power. For example, in my 96 the factory tune had the AFR at about 11:1 under WOT (wide open throttle).
Let me also throw this out there. This is going to pi$$ off many of the tuners and cause much debate but there is no way in he11 I would do a tune without seeing the effects of that tune with a Wide Band O2 sensor. This means that I would have a real hard time accepting a tune from anybody that wasn't near by me that didn't tune with a Wide Band. For example, altitude changes the way your engine behaves because you have thinner air. If a tuner was in Denver and created my tune, there is no way I would use it because that tune is good under the conditions that are local to them. It is going to be a whole different tune in my car, at my location. Someone like Diablo (or competitors) do their tunes most likely on a Dyno for a given car (for example a Mustang GT) and then I am sure they check the tunes by monitoring on the road. I can only hope that they do a conservative tune that leaves some margin for error. Each car is going to run different. A car with 1000 miles using Tune A is going to run different than a car with 50000 miles using that same tune. That's just the nature of cars. I will admit that I have a problem with mail order tuning just because the tune they created may be ok in a car that they made it with but that doesn't mean it is going to be good in mine. If you don't have a Wide Band O2 sensor, you don't reallly know what the effects of that tune are. The reason this is a problem is that you have to remember that lean conditions are a really bad thing. How do you know if it is lean if you don't have a wide band? You don't. Ya, when the tuner made the tune the car they were using may have been fine but that doesn't guarantee that yours is fine. Lets hope so, but for me, I need to know for sure so I have a wide band. Now, this is just my personal opinion and MANY people are going to have a hard time with that but it is still something to consider.
I use a LM1 Wide Band O2 sensor. I don't have a dyno. I log everything that the car does and then make changes based on what was reported. It is fine to do this. The advantage that the dyno has is that you don't have to drive all over the place logging. You can do it on the dyno and it will report the power that you are gaining from your modifications.
In addition to everything that we have looked at, there are other things to consider. One of them is what's called the Long Term Fuel Trends (LTFT or BLM's). Here's how this works. When your car goes into CL (Closed Loop) the O2 sensors are monitoring your AFR getting you as close to Stoich as it can. They use a table of conditions such as 2000 RPM at a KPA of 45. (typical of just cruising down the road) Visualize it like you see the timing table from above. If under those conditions you are running lean the car will add more fuel to get you to the proper AFR. What's unique here is that the computer will remember that it needed to do that at that RPM and KPA so next time you are driving at 2000 RPM and 45 KPA it will allready know that it needs to make that adjustment. If circumstances change that throw off the Stoich, it will just adjust that LTFT again. To visualize this, look at the main VE table shown above. This table gives the car a basis of how much fuel it needs as it runs through certain conditions. Well the air could be cold that day or you could be on a trip at different altitudes so those values are kinda estimates. You need some way to adjust for different conditions. The LTFT's allow this to occur by storing how far off you were from Stoich (14.7:1 AFR). If you have heard how cars can "learn" new mods, that is what they are talking about. All of this happens very fast many times a second. If you are at 2000 RPM and 45KPA and are running lean, the O2's will detect that, make the change, and put into the LTFT for that condition that you were off and how much it had to adjust so next time you are under these conditions, it remembers this. The amount of change that can be made via LTFT's are limited. In the case of my 96, 15.6% is the most. If your AFR is more than this in either direction for an extended period of time, this is when the good ol' Check Engine Light comes on indicating a lean condition or a rich condition.
Ok, so what kind of things can we tune? It is important to note that there is MUCh controversy on how things work and you can expect some argument from this area. My 96 Corvette is a great example. The VE table that you see above, well some claim it is never even used unless the MAF sensor or something goes bad. Others will say it is definately used for everything. I will tell you in my experience, there were no changes to the car using the VE table however others may swear that there were changes. I don't know who was right and who was wrong. All I can say is how it worked for me. I can tell you that changing the calibration of the MAF sensor did make a difference. The guy that shows me tuning told me that MAF calibration is how Fords are tuned. Take that with a grain of salt. I know it works in GM land but I have never tuned a Ford yet to know for sure. Maybe others will chime in and I suspect there maybe some argument of this statement.
Lets talk about my 96 Corvette (ya, I know it's a GM but the facts are similar regardless of brand). A larger cam was put in the car and it idled like crap. Furthermore, the AFR was very rich under WOT conditions. This not to mention the timing issues.
The computer has some basic tables in it. In my case one of those tables wanted to know the size of the engine. It was programmed for a 350 yet I put a 383 in it so right off the bat, it was off. I changed it to a 383. Under scanning while idling, my LTFTs were showing lean. The way that I was able to cure that was to get the MAF sensor to report more air at idle. For example, Take a look at the stock MAF table shown below:
http://www.cerinosolutions.com/tuning/mafbefore.jpg
Now after I tuned it:
http://www.cerinosolutions.com/tuning/mafafter.jpg
Take a look at the range from 2640 HZ to 3408 HZ. The scan showed me that at Idle, these were the ranges of air coming in. The LTFT's showed me that I was lean by X percent. Something you need to understand about tuning is that the actual numbers in this case don't mean squat. The adjustments are done in percentages. Since the LTFT's showed me that I was off by X percent at the idle area, that told me that I need the computer to add X percent of fuel in these areas. So how can I get the computer to add more fuel? Well since it wants a certian AFR, and I need to add more fuel, just trick the computer into thinking it has more air and it will bump the fuel to keep the desired AFR. So I multiplied the AFR calibration in the idle ranges by what I was short at according to the LTFT's. This tells the computer that if you see a reading of X HZ, you are actually taking in THIS amount of air where as you used to think it was THAT amount of air. So now, the computer adds Y amount of fuel under these conditions and my lean condition went away.
Mind you, this is an example. Making certian mods cause certain things to happen. Adding headers may improve the volumetric efficiency of the engine. If the VE at 2000 RPM and 60 KPA is 78.9% volumetric efficiency, adding headers may have improved that efficiency. Since LTFT's will allow you some slop, the check engine light may never come on. However, it it improves it so much that the LTFT's are not able to compensate for the change, that is when you may get the Check Engine Light because the LTFT's were run at their boundries (which you can set by the way) for a period of time.
OK, finally let's talk about the 12.5 AFR we like under WOT conditions.
The computer has a table called the PE table. (at least in my 96 Vette). This stands for Power Enrichment. Another table defines when the car goes into PE mode. In plain terms it adds fuel under heavy load such as WOT. A table defines the conditions that say when the car uses the PE values and you can set this.
While logging the wide band, as mentioned above my car was running at about 11.1 or so AFR. We know from experience that you will get the most power running about 12.5 to 1 AFR. If you ever look at someones dyno report, somewhere on there you will see a line that runs pretty flat across the board. That is typically the AFR and if it was done right, that will usually be around 12.5 across the board.
In my case, I was about 12% away from my goal of 12.5. (11.1 is 88% of 12.5). There are two tables in my Corvette that handle WOT conditions. One of them is how much fuel to add in PE mode based on current operating temperature. The other is how much change for PE based on RPM. Here is what those tables look like:
http://www.cerinosolutions.com/tuning/pe1.jpg http://www.cerinosolutions.com/tuning/pe2.jpg
Corvettes run at around 92 degrees C so this area and the ones above and below it are where I made my changes in the first table. The table shows how much fuel to add. Since I was 12% too rich, I changed those numbers to 12% less fuel to be added (values in table were changed from 15.6 to 13.7). I flashed the computer and went out and ran it. I was at the desired 12.5 up until I got in the higher RPMs where I still went a little rich. I used the second table and played with the values above 4000 RPM and flashed in between until I got 12.5 across the board. It took a few times but after tweaking, I was finally there. I the dyno, I gained 40 HP on m stroked 383 so I was pretty happy.
Ok, that is the crash course and should be enough to get you thinking. You can rest assured that there may be much argument born from this thread. What you see here is what I learned. With these cars, every new generation almost guarantees new revisions on the computer system. This means that although I have these tables for my 96, the tables for an LS1 car are different and I don't know what they are called in Ford land. The theory is still the same. You just need to find which tables do what.
In addition, each tuning software is much different too. My software calls any given table one thing, another piece will call it something else. In addition, for my 96 Corvette, it was a ba$tard child in the tuning world. It was the only year that there doesn't exist a piece of software to do both the scanning and the tuning. I don't know what is available to you in Ford world but I will say you do not have the same flexibility as we did in GM world. Seems they like it for you to have to go to their "certified tuners" a LOT.
I suggest to all who wish to have their car tuned to have it done on a local dyno with wide band O2 capability. That way you know for sure what YOUR car does with whatever changes that were done for the tune. I can see the handhelds being ok for a car that is completely stock and that is me assuming that they go conservative on the tunes to accomodate a margin of error for different physical conditions.
I hope this helps. I am by no means a total expert at this which is why I just list the basics. Each car is different and each manufacturer handles thier computer systems differently. GM will have some tables that Ford won't and vice versa.
I am sure others will chime in. Feel free to ask any questions and if I have experience in that area I will do my best to answer what I can.
I will likely make changes to this as I think of other things.
Cheers!
Want to learn about tuning? Read me. 
- Her Toy. 
