[-qaz-]

So lets say one wanted to adjust the air to fuel ratio without a chip or a tunner, for example because of a new injector size or a blower etc etc... Based on the info i collected this is what i came up with.

Granted the MAF and the O2 sensors determine the air:fuel mix.

The MAF sends an analog DC voltage to the computer of 0.5 - 5V depending on air flow. The O2 sensors send a voltage of 0-1.1V where <.4V is lean and >.6V is rich.

Now if lets say i give the computer 5v dc as the MAF input i think that will give 100% duty cycle for the injectors and ofcourse make it extremely rich (on a prettty much stock setup). I dont know what will happen when the O2 sensor notices that its rich, if it does have higher priority then the MAF input then the computer will componsate.

Now what is to stop me from giving the computer a 1V for O2 sensor input and a 5V for the MAF input? This should therefore make the car rich constantly. Yes terrible milalge will result and carbon deposits become a problem. But if one wanted to control the fuel to air mixture of the car manualy then it can be accomplished this way, by adjusting the voltage value of the imputs to be in a moreor less reasonable range a average setting can be obtained. Using some relays one can set up switching network between the stock setup and the manual setup.

Thoughts?

Comments?

 

[americanmotorsport.com]

This is not necessarily correct. See my reply in the other thread.

Also, when chaging injectors, you DO NOT readjust the mixture but rather, the correct way is o change the injector coding in the file.

 

[-qaz-]

QUOTE

Yes and no.

If you were ever seeing 5v, first off all, you would be in open loop, which means the 02's aren't even used. The EEC goes solely off the fuel base table Second, you will probably have pegged the MAF, which means a lot of that air would not be measured and the car will actually go lean.

AND there is one thing not considered: most EEC's have a cap at 980 AD counts, whcih is liek 4.79 volts...so after this point the EEC would set a failed MAF code and the above condition would occur. Now I can make this go away and set the limigt at 5v...

Anyway, the stock O2's are not wideband O2's but rather HEGO or EGO, Heated Exhaust Gas Oxygen sensor. This is a sensor in the exhaust system, before the catalytic converters that measures the content of the oxygen in the exhaust. The output of this sensor is from zero to five volts. The sensor only has the ability to tell if the air-fuel mixture is either rich or lean of stoichiometric. It cannot give feedback as to what actual air-fuel ratio you are running is. Regarless, in open loop, the EEC has NO IDEA how rich or lean you are.

This also has NOTHING to do with injector duty cycle...


Also, the point of teh resistor mod, which yes it does have the sideffect of adding SOME fuel in, is to add timing by tricking the EEC into thinking it is a colder/higher density day...but how then do you compensate for warmer days and what not? The system is set up the way it is for a reason...you may at times end up with too much timing and do some damage, since now you are taking control from the control system...

This post has been edited by americanmotorsport.com on Jan 27 2004, 11:24 PM


--------------------

Alberto
www.americanmotorsport.com[/b]

Thank you for your response, I apriceatte the info.

Well you are the expert here not I. However according to my Ford 1995 service manual cd this is a quote from the O2 sensor section:
QUOTE

The Heated Oxygen Sensor (H02S) (Figure 9) detects the presence of oxygen in the exhaust and produces a variable voltage according to the amount of oxygen detected. A high concentration of oxygen (lean air/fuel ratio) in the exhaust produces a voltage signal less than 0.4 volts. A low concentration of oxygen (rich air/fuel ratio) produces a voltage signal greater than 0.6 volts. The H02S provides feedback to the Powertrain Control Module (PCM) indicating air/fuel ratio in order to achieve a stoichiometric air/fuel ratio of 14.7:1 during closed loop engine operation. The H02S generates a voltage between 0 and 1.1 volts.[/b]

also to quote the MAF section:

QUOTE

The Mass Air Flow (MAF) Sensor (Figure 15) measures the air flowing into the engine. The sensor output is a DC (Analog) signal ranging from 0.5 to 5.0 volts that is used to calculate the injector pulse width required for stoichiometry. The sensing "hot wire" element is maintained at a temperature higher than ambient as measured by a constant "cold wire" element.

The MAF sensor is a calibrated emission component. The screen, metal body and electronics are trimmed as a complete unit during the manufacturing process. The electronics module and/or screen cannot be interchanged into different metal bodies. If replacement is required, the entire MAF sensor should be replaced.[/b]

"used to calculate the injector pulse width required for stoichiometry" is the pulse width not another term for duty cycle?

I have not done any testing to actually verify these numbers I am just theorizing based on the numbers and info i've found so far. The numbers themselvs are not important the theory is, if i can supply 1V i can supply 5V. Well as you can see from the quotes it does not state that the sensor can give an accurate meassurement of the mixture I assumed that, which may very well be wrong but it still does tell the computer if the mixture is rich, lean or stoichiometric. So if lets say the MAF input has 4V to prevent the faild MAF condition and the O2 sensor outputs stoichiometric condition (due to adjustment to the inputs of the computer) then it will keep the fuel mixture i set by giving the 4V at the MAF input.

Ofcourse you are correct, this is a step backwards and the control becomes more like that of a carburated engine and ofcourse the computer will no longer be able to adjust for the changes of the air imput since it no longer know anything but what i have told it (in this case 4v) so the mixture will always be rich unless the engine actually gets more air then the 4v setting in which case it will go lean. However if the injectors are large enough and then that condition can be avoided since from my understanding the voltage here contolls the pulse width of the injectors.

As you can see this is by no means a substitude for a chip but mearly a way to buy time or test some modifications without getting the chip or tunner untill the chip or tunner can be obtained. For example, lets say i put a charger on my car and want to see if it runs, i dont want to go lean since that can be pretty bad so this meathod can be used to see the performeance of the charger without getting the chip.

The side effects of timing I am not aware of and would appreciatte any input you can give me since you probably know the computers code much more then I do.

Thank you for the info, I look forward to hear your response and comments.

 

[americanmotorsport.com]

Injectors are PWM or pulsewidth modulated. In the newer returnless cars, the fuel pump is also, but that is anotehr subject.

All that section says is that wile in closed loop (and this is key and probably where your confusion lies) feedback from the 02's are used to determine whether the car is running a stoichiometric a/f, or 14.64 (if this is what is commanded, which it is for a stock file and is usually no changed for an aftermarket file).

SO bascially the 02's ONLY tell the EEC isf the mixture is rich or lean from stoich, and only at closed loop. At WOT (which commands open loop) the EEC has no idea what the a/f is..and thecommanded a/f is much richer than 14.64...

Now if you artifically raise the volatge on the MAF, thenyou are doing two things: you are makignteh car run rich, as the EEC thinks it is seing more air than actual, and two is you are arificially fudging load calculations which will in turn result in load being calculated toohigh and timing being reduced in turn.

You must also take note on the fact that MAF input even affects how an automatic transmission feels while shifting.

Let me add some comments:

The air meter actually measures mass of air that is flowing into the engine. The EEC also knows the mass flow rate of the injectors. If you are flowing, for example, 20#/min of air into the motor, and want to run 14.64:1 air/fuel ratio (which happens to be the chemically correct a/f raito for heptane, basically gasoline). So with 20#/min of air and a desired a/f ratio of 14.64, you'd need 20/14.64, or 1.366#/min of fuel. Let's multiply this by 60 to get #/hr of fuel, so we need 1.366 times 60 or 81.97#/hr of fuel through all the injectors. If you had say 8 injectors, each injector would have to inject 81.97/6 or 13.66#/hr of fuel. With a 24# injector, this is 13.66/24 or 56.9% of full injector opening.
This part is straight forward. If you change air meter and don't tell the EEC, you will deliver the wrong amount of fuel, period.

But something else happens:
The EEC calculates what's called load, which is essentially volumetric efficiency or VE. VE is the efficiency of the engine to pump air. For a 3.0L engines, in two engine revolutions, you can move in and out 3.0L of air. If the engine only moves 1.5L of air, then it's VE is 1.5/3.0 or 50%. Since the EEC knows mass of air, air temp and barometric pressure, it knows the volume of air that is moving through the engine at all times, so it can calculate VE.

The spark and fuel tables are tables for the amount of spark to deliver and the a/f ratio to command if you are NOT open loop. Open loop is when you are richer than 14.64:1 a/f raito and no longer using the O2 sensors for feedback. This could also be called power enrichment, a throw back to the carb days. The spark and fuel tables have RPM on the X axis and load on the Y axis. The output is then spark or a/f ratio. If your air meter transfer function is off, this is the input voltage from the meter to the EEC and what the flow is at that voltage, you will calculate VE wrong and hence grab the wrong spark and fuel values from the table.
This is why the air meter transfer function is so critical to making the engine work right. VE gets an input into almost everything, even what the trans feels like when it is shifting as I said above.

Now, someone will say that you can get an air meter calibrated for a given size injector and not change the EEC. This is right and wrong. Let's say you have 19# injectors and some air meter. You want to put in 38# injectors. You send your air meter off what they do is TRY to make it so at every voltage point your reading out from the air meter is original injector size divided by new injector size, in this example, 19/38 or 1/2 of the original value. Now, the EEC thinks there is very little air flow into the engine. Let's say before at 2 volts it was 12#/min, now with the "new" air meter, to get that same 12#/min you are at 1 volt. So, you now running at 1 volt with 12#/min of air and the EEC thinks you are running much lower flow, thus giving you a lower fuel pulsewidth, in fact, if the air meter was set up right, you will have 1/2 the pulsewidth and the amount of fuel going into the engine will be correct.

This is the part that's right. Now the wrong part. Load is calculated incorrectly, in this example, it's 1/2 of what it should be. So, you grab the wrong values out of the spark and fuel tables. This typically ends up in knock on some cars, since you are leaner with more spark.

I hope that shines some light on this...

 

[-qaz-]

Ah I see. So the moment i make it run rich it will go Open loop. Now im a bit confused, when would the computer go in open loop if no mods are made to any sensors? Its in open loop until o2 sensors warm up, and at WOT. When eslse is it in open loop? Ones it goes open loop what sensors does the EEC use if any?

Hmm this is interesting, since it seems that the MAF and the injectors are linked together by the EEC's firmware it would be reasonable to assume that if i were to transplant the injectors along with the MAF from lets say a v6 mustang everything should still match up and work alright or would this be the same scenerio as the recalibrated air meter you mentioned?

Right, injectors are PWM aka PDM so doesn't this directcly change the duty cycle (longer for higher, value shorter for lower)?

So with my original plan I see what you are saying. I'm curios to see what you think of the MAF and injectors transplant idea, but if i may comment on the side-effects of doing this as i understand them.

What kind of shifting issues would the tranny have? Wouldnt the VE value be a relatively small part of the equation? I would expect that the main components of that equation would be Tachometer and VSS. Besides if the EEC thinks it has high load wouldnt it shift faster and/or firmer?

Right the fuel amount will always be wrong since there is no more real MAF reading but it will always be wrong heavily on the rich side.

With the timing, this is new to me, but wouldnt i be able to compansate for that by advancing the timing at the distributor? The the EEC will retard it but it was already advaced so it will be withing normal conditions.

Again thanks for this interesting discussion.

 

[americanmotorsport.com]

No, the computer goes open loop based on throttle position. At open loop is simply does not usethe o2's.

And no, you cannot just swap a MAF and injectors together and have everything be alright. The MAF transfer function and injector coding must still be modified in the EEC.

VE and load are key to shifting and how the tranny feels. MHP and what not is only used for the shiftpoints.

The fuel CAN be compansated for by swapping out the MAF, to an extentent, like I said above, but timing would be off...I have explained it all ind etail in my previous post.

No you cannot just re-adjust timing manually and have it be optimal. First of all, not all cars have distributios anymore. Even if you did, the distributor only detrmines base spark. In older cars, spark is modified using barometric pressure, ECT, and ACT. In the new spark system, spark values are calculated by LOAD, and modified by ECT, ACT, a/f, etc.

You need to do things the right way.

 

[-qaz-]

Hello again Alberto,

I have carefully pondered the info you have given me and I see now that giving the EEC a false MAF reading is not a good approach since so many calculation would be offset. So i took a step back and though of what my goal is:

Use larger fuel injectors, while maintaing the overall adeptiveness of the EEC.

My original plan ofcourse totaly ignored the 2nd part of the goal. I have come up with a couple different approaches that I would like to get your input on.

My first new approach was a variation of the old one, but instead of using the input of the MAF i would simply modifythe O2 sensor input to give a stoich reading and ignore the output to the injectors all together and supply them with my own generated signal. This ofcourse ran into a seriouse problem quickly, timing totaly wrong, the engine probably wouldnt work at all or run terribly. So that idea is scrapped.

Next idea was to modify the signal from the EEC to the injectors. This i should explain in more detail. My car is a 1994 Taurus with a 3.0 Vulcan engine, these come stock with 14# injectors which are good for about 157 HP i believe; which is a short way from the stock 140hp. So lets say the 14# injectors are replaced by 19# injectors 14/19=.73 or 73% so for me to have the same amount of fuel i need to shave 27% of the signal going to the injectors since the new 19# ones would need that much less of a pulse. So The PWM needs to be demodualted, adjusted, remodulated and send to the injector. Now im sure i can make a electronic circuits needed for this but i run into a problem... delay, i dont think i can accomplish all those tasks fast enough. while pondering this it hit me. Why do this? If i simply plug in 19# injectors isntead of 14# when in open loop the car will run 27% richer then it would otherwise, however while in closed loop would the EEC not compensate for the extra fuel being sensed by the O2 sensors by reducing the signal to the injectors, advancing timing or changing the spark? I dont care how it componsates as long as the car is at stoich what does it metter? So i guess then my question is this:

Would the only side effect of changing the injector size and not modifying the EEC code be the richness of the a/f ratio at open loop?

Thank you.

 

[americanmotorsport.com]

You need to stop trying to trick the EEC.

It is not as simple as that. First fo all, the EEC, from the factory, can only compensate 25% each way. Second, from the factory, it applies corrections at one point to a lot of other points not learned, even WOT.

Just do it the right way and get a chip or reflash. If you could do stuff like that, everyone would do it...

 

[-qaz-]

Well its not so much me tricking the EEC, there are always different ways to acheive a goal. Modifying the software is one way and its a good way to go since by modifying the software it is not needed to redesign the whole system just because one thing changed.

For me however I dont think its worth it to spend over $300 for a chip or a reflash, mainly because I am not done with modification and am still tinkering with things. A chip should ideally be the last step, a tunner would be great for me but I dont see the need for it right now either, I am only looking to change injectors to get more options as far as hp upgrades. After i got a setup I want to keep, then I'll get a chip but during the testing and design stage it seem premature to do so.

Since the EEC can compansate +/- 25% then as long as the new injector isnt more then 25% larger then the old one it should work. Ofcourse with open loop issues but those i dont think will be too bad since 90% of the time its in closed loop anyway.

The other downside would be it wont be able to compensate for the injector wear and tear since it will already be almost maxed out. However i think this will nevertheless be quite good.

I guess I would need to go ahead and do this to determine if the theory is corrent or not.

Thanks for the info, I'll let you know how it goes.

On a sidenote, Can you reprogram the chip to use a diferent tranny? Like a mazda 6 ATx on a taurus? Also would it be doable to install an obd 2 controlled tranny on an obd 1 controlled car?

I'm sure the tranny will crap out sooner or later (probably sooner) with any hp mods more then a cone filter.

 

[americanmotorsport.com]

Just because an injector is 25% larger does not mean it will work. It is not as straightforward as thst. There are injector slope and breakpoint values in the EEC. It is a lot more complicated than siply takign out "X" percent fuel inthe EEC whenchanging injectors. A lot more than you think will be affected. I've tried to come on here and explain to you why doing stuff like that is not a good idea. What you do, it's up to you. It's your car.

I can program for a different tranny if the electronics are sorted out between models..