Tuning...some beginner and basic questions

[starscream2k1]

So was looking into tuning and trying to wrap my head around some things heres what i understand so far...

1) 14.7 AFR is optimum for a N/A engine, for a FI engine you would drop the AFR as you go into boost (more boost, more AFR drop).

2) When your tuning timing, wtf are you actually changing? The timing of what exactly (spark? fuel? what?)

3) What are big things to watchout for when tuning? AFR's and what else?

4) 3rd Gear or 4th Gear pulls? which one works best?

[terbow]

I will answer these as basically as i can cause Tuning is really complex. Basic tuning is easy, however its complex because there are a million things that can change something.

1) 14.7 is for anytime your not in boost (aka cruising). Once you hit boost its really up to you, the richer you go at 1 psi the more hesitation your going to see. 1 psi you can do 13.0 2 psi 12.5 3 psi 12.0 4-Xpsi is where u decide your conservativeness. 10.5-11.5 is pretty safe 12.0-12.3 is optimum power but less room for errors. It all depends on the car and setup.

2) Your changing your spark timing (ignition). Usually you are retarding timing during boost to elevate the level of safety and remove knock (in turbo cars).

3) You want it essentially to be a straight line once your at boost. taper down to your target boost afr and keep it there all the way through the wideband. Its better to start rich than lean. Shoot for 11.0 then decide from there . This is where its best left to a professional.

4) 4th is preferred because its usually a 1:1 ratio on the gears. Youll notice your speedo and tach match when driving. This gives you the most acurate reading.

again, this is all in a nutshell. its really best left to a professional or someone whose done it.

[TurfBurn]

well... I have some documents that you can read if you want that I've written on this topic... but in response to your questions:

1) Optimum AFR varies wildly depending on all sorts of engine factors... but short version is that max power for both NA and turbo when fuel is the only consideration is typically in the range of 12.4-13.1.

2) When you tune timing you are changing how many degrees before top dead center the spark occurs. The whole reason behind it is you are trying to time your flame front and the pressure rise within the cylinder to create your highest pressures at about 14 degrees after top dead center. Due to changes in fuel mixture, pressures, temperatures etc, the number of degrees BTDC that you have to do your spark is different in every occasion... in some cases you can not get your spark early enough to hit that goal without running into pre-ignition or detonation (different things).

3) Biggest thing to watch out for is detonation. AFR and timing are what you are adjusting for your power and output... but detonation or pre-ignition are your two enemies that you have to fight. That said, when you tune there are a lot of things you do for gas mileage, emissions, driveability, smoothness of power delivery etc that are not all compatible with one another... so the tune is focused on whatever your goals are.

4) Doesn't matter what gear you do a dyno pull in. the drivetrain losses are near identical regardless etc. Generally I prefer to pull in 3rd gear due to the lighter loads on the engine... but it is harder to tune on a 3rd gear pull because things happen too quickly.

[starscream2k1]

Thanks so far guys...Turf think you can sent me or pm me that info? That would be greatly appreciated!

[TurfBurn]

Here is the current revision of the document. I try to add to it every now and then and fill in more... but this should be of some decent use for you for now.

Thanks!

Steve

[starscream2k1]

thanks a fucken ton! :D

[ItsSlow]

Id suggest you get something to watch knock as well..

[1moreMPH]

Id suggest you get something to watch knock as well..

^Drive on a freeway in a lane close to one of those big cement walls. you can hear knock quite well off of those. just a little tip if you don't have a sensor.

[ItsSlow]

You cant hear 6-7 counts of knock off a wall with a car with a loud exhaust..

[Keno1542]

subs

[TurfBurn]

If you can hear the knock... it's already causing damaged. Plain and simple. Even with a microphone on the block and earphones you are missing a lot of low level detonation that can be causing problems. That's why you need to run things like a J&S if you are pushing your timing and AFR's and boost levels to regions where detonation is a possibility.

[starscream2k1]

Well i think tuning for 8psi to 10psi on a stock block shouldnt cause detonation or pinging. I mean i am not venturing too far off stock i mean with a FMIC, Intake and Exhaust i should be @ 7. So...

But yeah the J&S is expensive...is there any other way to detect knock other then hearing it and a J&S.

Like i am thinking tune for fuel first and then work on timing slowly. Since i am a novice. But then again sometimes i think of taking it to a shop, but if i do that.. how the hell do i learn? you get what i mean? hence why maybe start @ 8psi (something relatively safe) and start slowly till i feel comfortable and then go on to bigger things (9psi then 10psi). Then once i 'think' i have it good. take it to a tuner and see what adjustments he makes. Since i'll have a good base for understanding and hands on experience i'll have more of a clue of what he does.

[ItsSlow]

Stand alone would possibly be the only other way to detect knock other than the j&s and thats only if the stand alone has that feature..

What you need to know is that when you adjust your afr's your timing count rises/lowers in relation to the afr.. If it rises too much and you're running a 12:1 afr then more than likely you're going to cause some damage that you couldnt see happening on a screen.. Either predetonation or it will start knocking.. If the timing lowers itself, well, you'll just be down on power..

Alot of timing + a 12.1 or higher afr will not leave any room for mistakes..

[bstdtimsp]

Take your car to fourstar motorsports there are around toronto talk to frank he is a good guy and a good tuner i used to work with him very closely, they do mostly subaru's but he can probably fix you up and if you have the money pick up a micro tech system and switch from maf to map you should like that more

[1moreMPH]

12.1 afr is kinda dangerous neways...

[ItsSlow]

12.1 afr is kinda dangerous neways...

Yes and no.. Its just doesnt react well upon mistakes.. But it doesnt make that much more power than a little richer afr..

[solo.]

Just hooked up my wideband a few days ago and it's been reporting my A/F ratio of 13.4 at WOT from 4000rpm all the way up to 6000rpm. I have SMIC, hardpipes, 2.5" midpipe and DSM AFC running at 7.5psi.

Have I been really lucky for the past few months? :)

[ItsSlow]

Have I been really lucky for the past few months? :)

Yes, yes you have.. You should be throwing some fuel at it like crazy right now..

[1moreMPH]

yeap, lucky lucky. My guess is you havn't had a big problem because you're still at stock boost, and you're in Canada...cold = safer. I would defnitely add fuel to your map, probably a good amount if you're going to increase the boost.

[solo.]

Just retuned my car and from 4000rpm at WOT to 6500rpm, it stays at a steady 10.3 - 10.5 ratio.

Should I lean it out a little to 11? Or is that getting too brave?

[ItsSlow]

You can lean it out to a low-mid 11 and you should be safe..

[JDM Sam]

Lean it out to the 11s. Turbo XS knock lite is a cheaper knock sensor detector you can use. If you have a good reliable tune you don't need a knock retard system.

[ItsSlow]

Lean it out to the 11s. Turbo XS knock lite is a cheaper knock sensor detector you can use. If you have a good reliable tune you don't need a knock retard system.

You got a link to one of those?

[JDM Sam]

http://turboxs.com/more_info.php?ID=212
http://*************.com/mazdaspeed5.htm
I have one sitting here ready to go in my car. Just need to figure out where to mount it.

[solo.]

http://turboxs.com/more_info.php?ID=212
http://*************.com/mazdaspeed5.htm
I have one sitting here ready to go in my car. Just need to figure out where to mount it.

Interesting, where does it go? In the block?

[SleepyMSP]

wow glad I found this thread...subz...

[69RMSP]

Goatse it....



Interesting, where does it go? In the block?

[MazdaManiac]

Well, seeing as how there is a whole bunch of bad info in this thread, I figured I'd cut-n-paste the epic I wrote for the folks over at the RX-8Club. Some of this is rotary-specific, but the science/art is the same for all internal combustion engines running on gasoline.
If this stuff doesn't make sense to you, DO NOT attempt to tune your engine management.

HOW PROGRAMMABLE DIGITAL FUEL AND TIMING SYSTEMS WORK


Programmable engine management systems require a tuner to calibrate the basic raw fuel and ignition curves for an engine by defining tables of numbers that specify spark timing advance and base fuel delivery for pulse width-modulated (PWM) electronic fuel injection as a function of engine speed and airflow. Such tables are typically represented to the tuner in graphical or tabular format via user interface/editing software running on a laptop.

This PC-based engine management software simultaneously controls communications-driver software that has the ability to interact, via serial cable (or USB), with the onboard electronic control unit (ECU) that actually manages the spark and fuel requirements for an engine. The software allows the user to move data back and forth between ECU memory and the laptop for tuning purposes.

Typically, programmable engine management systems provide both the ability to interactively modify many operating parameters and table cells in the ECU while the engine is running, as well as the ability to upload the entire set of all engine management tables into a Windows PC for subsequent offline calibration and eventual download back to the ECU.

A programmable engine management system is, therefore, comprised of the onboard ECU, firmware, and calibration data, plus a Windows-based laptop computer (or, in a few cases, an alternate dedicated interface module), laptop engine management software, and data-link hardware. For full functionality of the system, all components must be present and operational. Once the ECU is calibrated, it is fully self-contained and will run the engine without additional components. However, without the laptop present and connected, the ECU is in no sense programmable.

Although user-interface software of various engine management systems may present internal engine management tables to the tuner with its own unique graphical appearance and although the underlying resolution or granularity of the tables may, in fact, actually vary, behind the scenes the basic engine management operating concepts are exactly the same:
When it is approaching time for the digital microprocessor managing an engine to schedule the next fuel injection or spark event, the processor first checks engine sensor data to get a snapshot of how fast the engine is turning and how much air it is ingesting.
The ECU then performs a series of look-ups to the appropriate entry for the current speed and airflow in the various fuel and timing tables to determine when to open and close the injectors to spray a pulse of fuel into one or more combustion chambers and how many degrees before top dead center to trigger an external igniter or coil to fire the next spark plug. If engine speed or loading fall somewhere in between two specified points in an engine management table, the ECU will provide base fuel or spark or corrections by applying a weighed average of the closest several points on either side of the actual value.

Regardless of the exact means used, the process of calibrating an engine management system on a new engine has the ultimate goal of optimizing torque, fuel economy, or emissions at each and every breakpoint of engine speed and loading defined in the fuel and ignition tables (there are usually between 24 and 800 such points).

In theory, this can be a laborious process, but it is one that can frequently be expedited with the use of automated map-builder computer modeling algorithms, pre-existing calibrations from similar engines, intelligent extrapolation and interpolation between rpm ranges and load points, and even self-learning routines that make use of a wideband 02 sensor and the dyno. Nevertheless, it is worth keeping in mind that OEM car manufacturers frequently spend months or even years calibrating the engine management systems for a particular engine in a particular vehicle.

The magnitude of this task is ameliorated by the sensitivity of gasoline in combustion - the spread between the air/fuel ratio producing lean best torque (LBT) and rich best torque (RBT) That is the range of air/fuel ratios over which the engine produces best torque. (Best torque is not a single point but rather a range.) In a normally aspirated power plant that does not require surplus fuel for combustion cooling and knock control, the spread between RBT and LBT at wide-open throttle is typically 11.5 to 13.3, with mean best torque in the 12.0 to 12.5 range. This spread narrows at high engine speeds as highest possible flame speed becomes increasingly critical to achieving complete combustion in the time available.

[MazdaManiac]

OVERVIEW OF THE COMBUSTION PROCESS


Combustion is the rapid release of energy from a fuel - in this case gasoline.
A finite quantity of gasoline contains an equally finite amount of energy, which can be released by combining it with a specific quantity of oxygen at an absolute ratio.
1 kilogram of gasoline contains 43 megajoules of energy that can be released via oxygen at a ratio of 1:15.179 (a ratio referred to as "stoichiometric"). That is about 41,700 BTUs. Plenty.
The trick with an internal combustion engine (ICE), whether it be rotary or other wise, is to control that combustion in space and time.
We do this by causing the combustion process to occur in the combustion chamber at a precise time and OVER a precise quantity of time to convert that heat into torque.
To effect this level of control, we must take a fixed quantity of space (about 650 ml in the case of the 13b-MSP), fill it with a quantity of gas and air as proscribed by the above ratio, compress it to a precise degree and ignite it at precisely the right time as to cause the maximum pressure increase resultant from that combustion to occur at the time of maximum delta for the combustion chamber's swept cyclic volume (that is to say at the precise moment that the combustion chamber is starting to get bigger again after it just got done getting smaller to compress the charge).
The beauty of this process is that it can occur completely independent of any change in factors in the outside world - temperature, pressure, altitude, pollution, humidity, whatever - as long as we can be assured that these conditions inside the combustion chamber are constant.

The problem is, we can't.

Because the process of getting a fixed volume of O2 into the combustion chamber at a proscribed density (meaning temperature via Avagadro) is complicated by the fact that this air is supplied by the available atmosphere, we are straddled with the effects of varying density on the combustion charge.
What that means is we must compensate for the volatility of gasoline as it responds to the varying charge densities. At differing charge densities, the amount of energy necessary to start the combustion process and the time it takes to complete the combustion process changes in a not so linear fashion.
So, what we do is vary the amount of fuel we add and ignite the process on a adjustable schedule based on what information we can obtain about the conditions of the air going into our ICE.
What we measure in the case of the RX-8 to know these conditions are these:

Air Flow
Intake Air Temperature
Barometric Pressure
Coolant Temperature
Throttle Position
Eccentric Shaft Position

By computing all of these measurements together, the engine control unit (ECU - sometimes called PCM for powertrain control module) can determine APPROXIMATELY the density of the air charge in the combustion chamber at any given time. It is a shame, really, that there is no way to measure the density directly or we could forgo all of this.
Two factors that can't be measured by the above methods are important to the whole equation as well.
First is volumetric efficiency (VE) or the amount of air, as a percentage of maximum, that the engine actually ingests as a result of the physics of mass and inertia. This number is fixed to some degree and changes at different RPM.
The other is the latent temperature of the actual combustion chamber as a result of the combustion cycles that proceed the cycle under scrutiny at that moment. This changes as a result of RPM as well, but it is also tied to 'load' or the increase of RPM over time as a proportion to charge density.
What that leaves us with is a very crude measurement of the total charge density.
How do we compensate for that?
By conservative 'hedging' on the bet that is ignition timing through advancement and retardation of the onset of the spark and by introducing elements into the gasoline that seek to stabilize its volatility. That is what octane is for and how much it affects the combustion process is measured by various methods including, but not limited to, Research Octane (RON), Motor Octane (MON) and the Anti Knock Index (AKI - and average of the RON and MON numbers). Unfortunately for the average motorist, many other ingredients are added to the fuel we use to affect its environmental impact that are not directly computed into the AKI. Ingredients are added to lower the boiling point and vapor point, reduce the hydrophilic nature of gas and reduce the amount of oxides of nitrogen after the combustion process. Many, if not most, of these ingredients change the combustion process in ways that may not be consistent from sample to sample. They also alter the total energy content of the fuel itself.

Having the ignition process starting at the wrong time (especially too soon) is a bad thing and can (especially in the case of the rotary ICE) quickly destroy a motor. So what is done, more often than not, is to err on the side of safety and bracket the combustion process with extra fuel and start the ignition with a slightly delayed spark. What this does is lower the temperature of the intake charge and insure that the combustion process is slower and later than optimal and never faster or sooner. Raising the AKI of the fuel used will accomplish the same thing but since the manufacturer of the vehicle can't insure that the fuel used will always have the proper AKI to achieve this or won't contain additives that adversely affect the ignition onset, they don't depend on their octane recommendation alone.
What is done by "tuners" then, is to take into account this margin of error and dial some of it out for more power which can be achieved by charge composition that is closer to optimal. To achieve this, they depend on the operator to use fuel that takes up the slack in AKI and remove some of this extra fuel and spark retardation.
Really, that is all there is to it. How good a tuner can be is dependent on his or her ability and knowledge with respect to the events within the combustion chamber in question.

[MazdaManiac]

VOLUMETRIC EFFICIENCY AND ENGINE FUEL REQUIREMENTS


An internal combustion engine is, in a sense, a self-propelled air pump that breathes in a quantity of air through the throttle related to the displacement of the engine, and exhales nitrogen and combustion products through the exhaust. But even at wide-open throttle (WOT), the amount of air that enters a spark-ignition internal combustion engine varies considerably depending on, well, stuff. There are many factors that affect how much air actually enters the engine's combustion chambers under various operating conditions (almost always less than the static displacement of the engine except with highly optimized performance engines utilizing resonation-effects tuned intake and exhaust systems, or on turbo or supercharged engines, which may easily operate at 100 percent or more of static displacement, also known as the volumetric efficiency or VE).

Torque is the instantaneous twisting force that an internal combustion engine is able to impart to a crankshaft/eccentric shaft, averaged over the combustion cycle by a flywheel. Over time, torque can be used to do work such as accelerating a car. Peak torque occurs at the engine speed and loading at which an engine (with power-adders, if present) is most efficient at ingesting air into the cylinders. Therefore, peak torque is also peak volumetric efficiency, or VE. Peak torque requires the largest amount of fuel per combustion cycle, hence, the longest injector open-time or pulse width. Even though engine breathing is less efficient at speeds above peak torque and therefore each putt is less powerful, more combustion events are occurring per time, and the engine will make its peak power at faster speeds than peak torque. Due to the definition of power as a function of torque/5250, these two measurements are always the same at 5250 rpm. A properly tuned engine will always use the most fuel per time at peak power.

On modern normally aspirated street engines, engineers strive to design engines to have as high a VE (torque) as possible across the range of engine operating speeds. One hundred percent VE is the amount of air that would be in the cylinder of an engine at bottom dead center at rest, minus the combustion chamber clearance volume (the actual displacement of the engine). It is extremely difficult or impossible to achieve 100 percent VE when the engine is running, because it takes a certain amount of time for air to rush into the combustion chambers and there is a limited amount of time (eight or nine thousandths of a second at 7,000 rpm), and there are restrictions in the way. Most street internal combustion engines cannot achieve anything like 100 percent VE, settling instead for maybe 70 to 90 percent, but engineers work to build a high, flat torque/VE curve across the rpm range.

A number of factors conspire to prevent a full 100 VE in normally aspirated street engines. There is normally at least a slight pressure drop through a throttle body, even at wide-open throttle. All fuel injection throttle bodies restrict airflow - producing a pressure drop - when the throttle blade is partially closed (and sometimes even when it's wide open). Intake ports and valves offer at least some restriction at some engine speeds. The exhaust stroke does not expel all burned gases because some exhaust is trapped in the clearance volume. The exhaust ports, valves and exhaust pipes offer some restriction as well. The camshaft profile of an engine (when so equipped) or port timing has a huge effect on the VE of an engine at various speeds and loading.

If a normally aspirated engine had a perfectly flat torque curve across the rpm range, injection pulse width would not vary at all with engine speed, but would be a function only of engine loading. If VE is entirely independent of engine speed, then the amount of air entering a combustion chamber on the intake stroke will depend entirely on the amount of air pressure in the intake manifold, hence, injection pulse width is a linear function varying precisely with the voltage from a manifold absolute pressure sensor, with peak VE and torque at the highest achievable manifold pressure. Since many modern street engines do have a pretty flat torque curve, a calibration in which pulse width varies in a linear fashion from zero at zero manifold pressure to maximum power pulse at full achievable manifold pressure is a good approximation of an ideal steady-state air/fuel table optimized for peak power.

Given the varying VE of an engine, the amount of air in the cylinder at any given breakpoint is not perfectly predictable. Again, you can assume that the maximum VE occurs at the point of peak torque. This will be the point of maximum injection pulse width, with pulse width falling off both above and below this. Maximum power will occur at a higher rpm than peak torque where the engine is making more power strokes per time increment, although the power strokes are less efficient at speeds above the torque peak and will therefore require less fuel to be injected per power stroke.

In some cases, engine torque and VE vary considerably with RPM. Normally aspirated engines optimized to run efficiently at very high speed with resonation-effects intake-runner and exhaust tuning at the expense of very weak low-rpm torque capabilities are one example (think S-DAIS). Some such engines have achieved as much as 10 psi positive pressure at the intake valve from inertial and resonation effects. Another example of varying VE is turbocharged engines optimized for high-speed operation that cannot make boost at lower rpm. However, even if the torque curve is not flat, given the relatively broad range of air/fuel mixtures at which an engine will still produce peak torque or near-peak torque, a straight-line approximation of injection pulse may still produce relatively good performance even with no VE corrections.

As an engine wears out, its VE decreases - but this may not occur evenly across all speeds and loading ranges or between combustion chambers. The newest original-equipment factory engines vary less than 1.5 percent in power, but historically, factory-built engines have varied greatly in VE due to significant variations in cam lobe profiles or port timing and other machining, with perhaps 5 percent very fast (high VE), 15 percent very slow, and the rest somewhere in the middle.

If the load-measuring device for an engine management system is a mass airflow (MAF) sensor (output voltage or frequency highly dependent on engine speed), then the basic fuel calculation is based on the MAF voltage divided by engine rpm. Note that MAF-based fuel calculations require no temperature correction for air density, since the MAF reading already reflects air density effects on mass airflow into the engine.

[MazdaManiac]

AIR/FUEL RATIOS


Air/fuel ratio has a major impact on engine octane number requirement (ONR), increasing octane requirements by +2 per one increase in ratio (say from 9:1 to 10:1). Ideally, air/fuel ratio should vary not only according to loading but also according to the amount of air present in a particular combustion chamber at a particular time (combustion chamber VE). Richer air/fuel ratios combat knock to some extent by the intercooling effect of the vaporization of liquid fuels and a set of related factors. The volatility of fuels affects not only octane number requirement but drivability in general.

The chemically ideal air/fuel mixture (by weight), at which all air and gasoline are consumed in combustion occurs with 14.68 parts air and 1 part fuel, which is usually rounded to 14.7. This ratio is referred to as stoichiometric. Stoichiometric mixtures vary according to fuel, from a low of nitromethane at 1.7:1, to methanol's 6.45: 1, ethanol's 9:1, up to gasoline at 14.7:1 and beyond to natural gas and propane, which are in the range of 15.5-16.5:1. Mixtures with a greater percentage of air than stoichiometric are called lean mixtures and occur as higher numbers. Richer mixtures, in which there is an excess of fuel, are represented by smaller numbers. Mixtures, by the way, are often expressed as a percentage of stoichiometric, usually referred to as lambda. Therefore, the stoichiometric APR is 1.0 lambda; the best-power ratio for gasoline (12.2:1) is 0.83 lambda.

At high loading and wide-open throttle, richer mixtures give better power by making sure that all air molecules in the combustion chamber have fuel present to burn. At wide-open throttle, where the objective is maximum power, all four-cycle gasoline engines require mixtures that fall between lean and rich best torque, in the 11.5 to 13.3 gasoline range. Since this best torque mixture spread narrows at higher speeds, a good goal for naturally aspirated engines is 12.0 to 12.5, perhaps richer if fuel is being used for combustion cooling in a turbo/supercharged engine.

Typical mixtures giving best drivability are in the range of 13.0 to 14.5 gasoline-air mixtures, depending on speed and loading.

OCTANE REQUIREMENTS


A fuel's octane rating represents its ability to resist detonation and pre-ignition. Factors influencing an engine's octane requirements are listed below:

• Effective compression ratio
• Atmospheric pressure
• Absolute humidity
• Air temperature
• Fuel characteristics
• Air/fuel ratio
• Variations in mixture distribution among an engine's combustion chambers
• Oil characteristics
• Spark timing
• Spark timing advance curve
• Variations in optimal timing between individual combustion chambers
• Intake manifold temperature
• Combustion chamber coolant temperature
• Condition of coolant and additives
• Type of transmission
• Combustion chamber hot spots.

When an engine knocks or detonates, combustion begins normally with the flame front burning smoothly through the air/fuel mixture. But under some circumstances, as pressure and temperatures rise as combustion proceeds, at a certain point, remaining end gases explode violently all at once rather than burning evenly. This is detonation, also referred to by mechanics as knock or spark knock. Detonation produces high-pressure shock waves in the combustion chamber that can accelerate wear of an engine or actually cause catastrophic failure. Pre-ignition is another form of abnormal combustion in which the air/fuel mixture is ignited by something other than the spark plug, including glowing combustion chamber deposits, sharp edges or burrs on the surfaces of the combustion chamber, or an overheated spark-plug electrode. Heavy, prolonged knock can generate hot spots that cause surface ignition, which is the most damaging side-effect of knock. Surface ignition that occurs prior to the plug firing is called pre-ignition, and surface ignition occurring after the plug fires is called post ignition. Pre-ignition causes ignition timing to be lost, and the upward movement of the piston or forward movement of the rotor on compression stroke is opposed by the too-early high combustion pressures, resulting in power loss, engine roughness, and severe heating of the piston crown or rotor face. It can lead to knock or vice versa.

The single most important internal engine characteristic that requires specific fuel characteristics is compression ratio, which generally increases the ONR +3 to +5 per one ratio increase (in the 8-11: 1 compression ratio range). High compression ratios squash the inlet air/fuel mixture into a more compact, dense mass, resulting in a faster burn rate, more heating, less heat loss into the combustion chamber surfaces, and consequent higher combustion chamber pressure. Turbochargers and superchargers produce effective compression ratios far above the nominal compression ratio by pumping additional mixture into the combustion chamber under pressure. Either way, the result is increased density of air and fuel molecules that burn faster and produce more pressure against the piston or rotor. Another result is an increased tendency for the remaining gases to spontaneously explode or knock as heat and pressure rise.

Keep in mind, high peak combustion chamber pressures and temperatures resulting from high compression can also produce more NOx pollutants. Lower compression ratios raise the fuel requirements at idle because there is more clearance volume in the combustion chamber that dilutes the intake charge. And because fuel is still burning longer as the piston descends or the rotor progresses, lower compression ratios raise the exhaust temperature and increase stress on the cooling system.

Until 1970, high-performance cars often had compression ratios of up to 11 or 12 to one, easily handled with vintage high octane gasoline's readily available in the 98-99 ((R+M)/2) range. By 1972, engines were running compression ratios of 8-8.5:1. In the 1980's and 1990's, compression ratios in computer controlled fuel injected vehicles were again showing up in the 9.0-11:1 area based on fuel injection's ability to support higher compression ratios without detonation, coupled with the precise air/fuel control and catalysts required to keep emissions low. Race car engines typically run even higher compression ratios. In air unlimited engines, maximum compression ratios with gasoline run in the 14-17:1 range.

Ratios above 14:1 demand not only extremely high-octane fuel (which might or might not be gasoline), but experienced racers use tricks like uniform coolant temperature around all combustion chambers, low coolant temperature, and reverse-flow cooling. Extremely high compression ratios also require excellent fuel distribution to all combustion chambers, retarded timing under maximum power, very rich mixtures, and probably individual combustion chamber optimization of spark timing, air/fuel ratio, and volumetric efficiency.

[MazdaManiac]

SPARK ADVANCE


Spark advance, which is optimally timed to achieve best torque by producing peak combustion chamber pressure at about 15 degrees ATDC, increases octane requirements by a half to three-quarters of an octane number per degree of advance. Spark advance increases combustion chamber pressure and allows more time for detonation to occur.

Engine speed range and fuel burn characteristics affect ignition timing requirements. As an engine turns faster, the spark plug must fire at an earlier crank position to allow time for a given air/fuel mixture to ignite and achieve a high burn rate and maximum combustion chamber pressure by the time the piston or rotor is positioned to produce best torque, which is dependent not only on engine speed but on mixture flame speed, which, in turn, is dependent not only on the type of fuel but on operating conditions that change dynamically, such as air/fuel mixture.

Therefore, an independent variable affecting the need for spark advance as RPM increases is the need to modify ignition timing corresponding to engine loading and consequent volumetric efficiency variations, which demand varying mixtures. Throttle position, for example, affects combustion chamber filling, resulting in corresponding variations in optimal air/fuel mixture requirements.

An important factor that affects VE - and potentially flame speed - is valve/port timing. Remember, a denser mixture burns more quickly, and a leaner mixture requires more time to burn. Valve/port timing has a great effect on the speeds at which an engine develops its best power and torque. Adding more lift or earlier opening and intake/ exhaust valve/port opening overlap allows the engine to breathe more efficiently at high speeds. However, the engine may be hard to start, idle badly, bog on off-idle acceleration and produce bad low speed torque. This occurs for several reasons. Increased valve overlap allows some exhaust gases still in the combustion chamber at higher than atmospheric pressure to rush into the intake manifold (or be trapped inside the combustion chamber) exactly like EGR, diluting the inlet charge - which continues to occur until rpm increases to the point where the overlap interval is so short that reverse pulsing is insignificant. But large ovelap results in gross exhaust gas dilution of the air/fuel mixture at idle, which consequently burns slowly and requires a lot of spark advance and a mixture as rich as 11.5 to 1 to counteract the lumpy uneven idle resulting from partial burning and misfires on some cycles. Valve/port overlap also hurts idle and low-speed performance by lowering effective manifold vacuum. Since the lower atmospheric pressure of high vacuum tends to keep fuel vaporized better, large overlap and port timing resulting in low vacuum may have distribution problems and a wandering air/fuel mixture at idle, which may require an overall richer mixture in order to keep the motor from stalling, particularly with wet manifolds.

Coming off idle, a large port overlap engine may require mixtures nearly as rich as at idle to eliminate surging, starting at 12.5-13.0 gasoline air/fuel mixtures and leaning out with speed or loading. Mild porting and overlap will permit 14:1 - 15:1 gasoline mixtures in off idle and slow cruise.

With medium speeds and loading, the bad effects of big ports and overlap diminish, resulting in less charge dilution, allowing the engine to happily burn gasoline mixtures of 14 to 15:1 or higher. At the leaner end, additional spark advance is required to counteract slow burning of lean mixtures.

With large ports and overlap, the spark advance at full throttle can be aggressive and quick; low VE at low rpm results in slow combustion and exhaust dilution, lowering combustion temperatures and reducing the tendency to knock. Part-throttle advance on large port overlap engines can also be aggressive due to these same flame speed reductions resulting from exhaust dilution of the inlet charge due to valve/port overlap.

In the 1970s, automotive engineers began to de-tune engines to meet emissions standards that were increasingly tough. They began to retard the ignition timing at idle, for example, sometimes locking out vacuum advance in lower gears or during normal operating temperature, allowing more advance if the engine was cold or overheating. Since oxides of nitrogen are formed when free nitrogen combines with oxygen at high temperature and pressure, retarded spark reduces NOx emissions by lowering peak combustion temperature and pressure. This strategy also reduces hydrocarbon emissions. However, retarded spark combustion is less efficient, causing poorer fuel economy, reduced power, and higher heating of the engine block as heat energy escapes through the combustion chamber walls into the coolant. The cooling system is stressed as it struggles to remove the greater waste heat during retarded spark conditions, and fuel economy is hurt since some of the fuel is still burning as it blows out the exhaust ports, necessitating richer idle and main jetting to get decent off-idle performance. If the mixture becomes too lean, higher combustion temperatures will defeat the purpose of ignition retard, producing more NOx. I
By removing pollutants from exhaust gas, three-way catalysts tend to allow more ignition advance at idle and part throttle. Undesirable products of combustion include formaldehyde, NOx, CO2, and fragments of hydrocarbons.

In any case, various high-performance fuels vary in burn characteristics, particularly flammability, flame speed, emissions, and so on, and all affect spark timing requirements. Gasoline engines converted to run on propane, natural gas, or alcohol require a different timing curve due to variations in combustion flame speed of the air/fuel mixture.

[MazdaManiac]

COMPUTING TARGET AIR/FUEL RATIOS, PULSE WIDTH, AND TIMING



It is relatively easy to calculate or estimate the volume of a combustion chamber and therefore the weight of air that enters the combustion chamber under ideal operating conditions. It is also fairly easy to make a good guess at how long an injector should stay open to spray in an amount of fuel that is a particular fraction of the weight of the air (such as the 14.7: 1 stoichiometric or chemically correct air/fuel ratio).

Of course, a complicating factor in designing a fuel curve is that even if you know how much air is entering the combustion chambers at a given time, it is not necessarily clear what air/fuel mixture is optimal. Most tuners want the maximum power possible at wide-open throttle, using the least possible fuel to accomplish this (lean best torque). But under some circumstances, it may be desirable to run at rich best torque (or even richer) in order to design in a safety factor to help prevent detonation with bad gas.

What about throttle response versus efficiency at part throttle? Are you willing to sacrifice some idle quality for fuel economy or reduced emissions? Does the overlap of your port timing the idle mixture so you need a particularly rich mixture for an acceptable idle? These kinds of questions complicate tuning even further.

Additionally, there's the real-world performance of the circuitry that activates the injectors and the physical response of the injectors themselves, which can vary from injector to injector.

Given the complexity of designing a theoretically correct fuel map, most tuners don't even try. If possible, they use a preexisting map from a similar vehicle or engine. Otherwise, they build a startup map designed purely to be rich enough to get the vehicle running and warmed up. This is made easier by the fact that warmed-up fuel-injected engines will run with mixtures as rich as 6.0: 1 on up to lean mixtures near 22.0: 1, and even cold engines are fairly flexible about air/fuel mixtures that will run the motor.

With the engine running (ideally on a dyno), tuners adjust the engine at each breakpoint combination of speed and load and set ignition timing and injection pulse width to achieve low emissions, best torque (lean or rich), or a specific air/fuel mixture or some combination of all three. Then they road test the vehicle and fine tune it under actual driving conditions, also fine tuning the enrichment maps.

If your EMS has closed-loop capability and is running in a speed-throttle position range in which closed-loop operation is activated, your EMS may be able to tell you what the mixture would have been if you were running open-loop based on what amount of mixture correction was required to achieve a stoichiometric air/fuel mixture. You can then make corrections to the raw fuel map based on this information.

TARGET AIR/FUEL RATIOS


Gasoline, being a stew of hydrocarbons of varying structure, is not a single homogenous molecule that burns with oxygen in a precisely predictable fashion. Nonetheless, given unlimited time for combustion and a perfectly mixed batch of fuel and air, it takes between 14.6 and 14.7 pounds of air to burn a pound of gasoline in a reaction that produces 100 percent water and carbon dioxide (in this ideal reaction, the nitrogen in air is purely along for the ride). So an air/fuel ratio of 14.7:1 is the chemically correct or stoichiometric ratio, sometimes abbreviated by the Greek letter lambda, as in lambda of 1.0. For example, lambda of 0.9 translates as 0.9 x 14.7, or 13.23 air/fuel ratio.

In reality, in a high-speed internal combustion engine, there is limited time for combustion, the air and fuel are not perfectly mixed, and tiny amounts of nitrogen will be burned by hot combustion temperatures and pressures. And inevitably some gasoline molecules will not be burned and some oxygen will not find any fuel to burn. Best power always occurs when there is at least a small fuel surplus above stoichiometric because air (oxygen) is the scarce commodity in spark-ignition internal combustion engines. The best power strategy is therefore always to make sure there is enough fuel available so that virtually every molecule of oxygen finding its way into the engine will be able to burn fuel. The best economy strategy is to make sure there is a small oxygen surplus so that every bit of fuel will get burned. The best emissions strategy is complicated by the fact that lowest hydrocarbon (Hc) emissions occur, as you'd expect, with an air surplus, but such lean mixtures produce hotter combustion, which burns more nitrogen and increases oxides of nitrogen, known as NOx. The best overall emissions strategy, therefore, turns out to be to target 1.0 lambda, the stoichiometric ratio, which does not minimize either HC or NOx, but minimizes the sum of both.

Let's first consider target air/fuel ratios for steady-state operation, when an engine is operating continuously at a particular speed and power output and the various engine fuel, air, and cooling systems have reached equilibrium.

Assuming the engine is not knock-limited and therefore does not require additional surplus fuel to cool combustion via heat vaporization effects, for best torque with optimal fuel economy at wide-open throttle, aim for a 12.5:1 air/fuel ratio midway between lean and mean best torque (12.8-12.2 AFR). If you're willing to sacrifice a little torque, mixtures as lean as 0.92 lambda (13.5: 1) on mild street engines that are not turbocharged will produce excellent peak-power fuel economy at the cost of about 4 percent power.

On turbocharged or other boosted power plants that are knock limited, at a very minimum you are aiming for mean best torque of 0.82 lambda (12.2 AFR), if not rich best torque of 0.8 lambda (11.76 AFR). This has the side benefit at very high flame speed (which increases all the way to 0.75 lambda, at which point a healthy flame front will move through this air/fuel mixture at 80 feet per second).

Under very high vacuum (deceleration conditions), run 15.5:1 or 16.0:1 AFRs (1.05-1.09 lambda) or even fuel cut on deceleration to prevent lean exhaust system backfiring.

Mild-cammed or ported engines with tuned runners and little valve timing overlap operating at full operating temperatures will idle at 14.7:1, but such a 1.0 lambda mixture is fairly close to producing misfires, and most engines produce a better, smoother idle and less stalling with richer air/fuel ratios. It is common practice to set idle lambda at 0.9 (within the correction range of the oxygen sensor) and allow the closed loop system to pull mixtures leaner to stoichiometric at idle and light cruise under 02 sensor control. If, for any reason, the 02 sensor fails, the engine will only idle better. Big-cammed or highly ported engines will not idle well at stoichiometric mixtures, needing at least 13.0-13.5: 1 mixtures (0.88-0.92 lambda) to idle as smoothly as possible, even at full operating temperatures.

Subject to engine VE distortions, for best drivability, air/fuel mixtures should smoothly increase fuel in a linear fashion from light cruise at low rpm toward heavier loadings and peak-torque speeds. Build in smooth increases in mixture as the engine increases in speed toward best torque and as loading increases.

[1moreMPH]

^decent read. i would hope that most people already know that though. Care to point out the bad information presented in here?

[ForceFed]

^decent read. i would hope that most people already know that though. Care to point out the bad information presented in here?
There was a ton..I LOL'd here for a few days..and I was gonna post myself..Bt every time I tell someone how they are wrong..I end up getting into a fight with an Idiot..And we all kno what happens when you do that.

[1moreMPH]

There was a ton..I LOL'd here for a few days..and I was gonna post myself..Bt every time I tell someone how they are wrong..I end up getting into a fight with an Idiot..And we all kno what happens when you do that.

the idiot gets pwnt? haha. im still missing the bad info though, except maybe my ghetto-knock test on the freeway, which i actually can't take credit for; that's a JDM-Sam special.(naughty)

[MazdaManiac]

^decent read. i would hope that most people already know that though.


I'm willing to bet that not a single person that reads that knew more than half of it.
It might seem simple because you are familiar with the essentials, but there is a lot of devil in those details.

Care to point out the bad information presented in here?

Almost all of the A/F recommendations are wrong, and not only because the proper value falls in a range that is dictated by VE.

[1moreMPH]

I'm willing to bet that not a single person that reads that knew more than half of it.
It might seem simple because you are familiar with the essentials, but there is a lot of devil in those details.



Almost all of the A/F recommendations are wrong, and not only because the proper value falls in a range that is dictated by VE.

i mean... it may be "extremely technically and specifically" wrong if you are building an engine from scratch to your own specs. but with the FS-DE as is, on pump gas, and without a standalone, i'd say the recommended afr's are right on. They weren't just thrown out there -- these cars get the most power in the mid/low 12's. Much above and they seem to pop, and lower than that is usually a good idea for a decent safety factor and not much power suffer.

I agree on your writing, but for basic tuning i would say that the information given is ok. for the normal driver there's not much need to dwell into the gritty stuff to figure out exact degrees for ignition, what fuel octane is perfect for what afr, etc. do you not agree?

edit: do you even have a msp?

[MazdaManiac]

do you even have a msp?

Nope.
But I've tuned several FS powered cars to within and inch of their life.
And some KLs.
And BPs.
And B6s.
And 13bs.
Etc...

[1moreMPH]

then how can you say that the afr's are wrong? As far as i can see, all the big tuners among this site agree that ideal power is right around the low 12's, and reasonably safe would be low 11:1 with the stock CR, not overly-aggressive timing, and 93octane.

[MazdaManiac]

Because saying "low 12" or "low 11:1" is meaningless without knowing a bunch of other things like Ve, EGT at peak Ve (torque peak) and net ignition advance.

[TurfBurn]

MazdaManiac's writeup is correct. A couple minor tweaks here and there that my clarify or get rid of some people reading in some confusion. But I agree/second/verify what he typed.

[TurfBurn]

I'm willing to bet that not a single person that reads that knew more than half of it.
It might seem simple because you are familiar with the essentials, but there is a lot of devil in those details.



Almost all of the A/F recommendations are wrong, and not only because the proper value falls in a range that is dictated by VE.


well for what it's worth I knew everything you had in that writeup, and even did research on it in college. It's also the basis of all the theory I use in working up a tune when I do a customer's car. But major credit to you to having the patience to type it all up and get all the correct numbers etc for it. Not an easy task. I've typically gone the generalized/incomplete route (NOT incorrect mind you) as I lack the time and patience to put together what you did.

The stuff I posted was generalized information based on the work we've done with the Protege's and MSP's and the FS-DE engine's I've built. It is also factoring in the general knowledge and technology available for tuning to most people with these cars/on these boards. So it does not cover every possible combination, and it also isn't aiming/going for that last 4-5% of output that can be yielded with a fully technical tune.

[TurfBurn]

Because saying "low 12" or "low 11:1" is meaningless without knowing a bunch of other things like Ve, EGT at peak Ve (torque peak) and net ignition advance.

I'd disagree a bit here as even your writeup points to where those generalized statements come from. I, as an engineer, fully appreciate and understand your discomfort with the generalized ranges given because it is NOT "technically correct" but for the majority of people who can't grasp the entirety of what is being put in front of them the generalized ranges work as an okay "rule of thumb"

Here are the points from your writeup:


he spread between RBT and LBT at wide-open throttle is typically 11.5 to 13.3, with mean best torque in the 12.0 to 12.5 range. This spread narrows at high engine speeds as highest possible flame speed becomes increasingly critical to achieving complete combustion in the time available.

Which is exactly an example of where someone says high 11's to low 12's being okay... you are "generally" hitting MBT in that range, but exactly as you point out as you get to higher speeds, or detonation becomes a limiting factor that "range" will narrow down to a more specific AFR.

And again:

On turbocharged or other boosted power plants that are knock limited, at a very minimum you are aiming for mean best torque of 0.82 lambda (12.2 AFR), if not rich best torque of 0.8 lambda (11.76 AFR).


Which I did notice one "inconsistency" in your writeup around the above quote. One problem that you can run into with a street car is that given fixed items like VE, and the knock index of the motor and the quality of fuel in a given area you peak actual output for torque from the motor is not necessarily achieved in the MBT/RBT range as detonation/pre-ignition can become a factor that may require you to actually operate outside of that ideal range. You follow/agree or do you not see what I'm saying here?

[MakeMeGoFast]

A book that I recently picked up. Is by Jeff Hartman titled How to Tune and Modify Engine Management system. Has a Wealth of info , much of which has been discussed already.

[MazdaManiac]

You follow/agree or do you not see what I'm saying here?

I agree, though the two "contradictory" statements you picked were for an NA motor and then a turbo motor.

[crashkelly]

subbing for quality knowledge

[Witchdoktor]

subbing for quality knowledge

no doubt

MazdaManiac - great reads. much appreciated!

[1moreMPH]

A book that I recently picked up. Is by Jeff Hartman titled How to Tune and Modify Engine Management system. Has a Wealth of info , much of which has been discussed already.

+1. That's right up there with CB's 'Maximum Boost'. God we engineers are so nerdy...(flash)

[Velocifero]

+1. That's right up there with CB's 'Maximum Boost'. God we engineers are so nerdy...(flash)

hey i love that book and I am no where near an engineer. i work in the biology field.

[TurfBurn]

I agree, though the two "contradictory" statements you picked were for an NA motor and then a turbo motor.


I don't think anything I quoted I said was contradictory. I don't see a contradiction in there at least looking at it now. I was only illustrating the point that it IS reasonable to give a rough range for AFR as a recommendation.

and when I said there was an inconsistency I was pointing out that some motors are SOOOO knock sensitive that you can't even reach the AFR's given and you'll actually make your "best" torque at an even lower AFR etc... in other words even if you retard timing all the way to 0 (not a good idea for other reasons as well) you would still not be able to go "lean" enough to get to the AFR's for MBT/LBT

[MazdaManiac]

Which is where peak EGT comes into play.

Timing is a funny thing. On rotary motors, it isn't uncommon to fire ATDC at peak torque.

[TurfBurn]

I know jack abuot Rotaries :) So I won't even question there.

I've never found EGT to be super useful (at least not standard gauge based like most consumers have... not to be confused/compared to lab type measurement) as it just is too slow and often biased based on probe location. I think it provides some good information, but it needs to be understood properly relative to the application and setup to be truly of any use or to be reliable.

[MazdaManiac]

EGT is only useful after you know your peak torque.
On a dyno, you shoot for the A/F that produces peak torque, look at the EGT and use that as a reference for tuning the rest of the RPM range and the ignition timing.