The ever-increasing number of “do-it-yourself” horsepower enthusiasts helped to explode the market for standalone EFI systems. These self-contained systems are designed to work independently from any other vehicle systems. This means they work equally well in a classic restoration project, dedicated racecar, or retrofit to most modern road-going vehicles. Almost all stand-alone aftermarket systems are speed density to reduce the complexity of the MAF and its necessary plumbing. These systems can be installed under the same inlet system as a common four-barrel carburetor or use almost any OEM intake manifold.
This Tech Tip is From the Full Book, ENGINE MANAGEMENT: ADVANCED TUNING. For a comprehensive guide on this entire subject you can visit this link:
SHARE THIS ARTICLE: Please feel free to share this article on Facebook, in Forums, or with any Clubs you participate in. You can copy and paste this link to share: http://www.musclecardiy.com/performance/introduction-to-standalone-efi-systems/
Most of these systems allow the calibrator to view data in real time. This means that actual lambda, spark advance, engine speed, load, etc. can all be viewed while making adjustments to the tables on the fly. This reduces the down time associated with taking measurements, changing the file, and reflashing the module before returning to take more measurements. Tuning in real time allows the calibrator to instantly see whether a change was beneficial or not. For steady state calibration, the ability to “arrow up/arrow down” quickly makes airflow modeling and finding MBT incredibly easier.
The complexity of these systems, while intimidating to most beginners, pales in comparison to modern OEM systems. Their calibration is surprisingly simple. Many programs have a tool to estimate a starting VE map or come preloaded with one that is relatively close. If the system is installed and wired properly, one should be able to quickly move on to the business of fine-tuning the base VE map after confirming proper fuel system setup and injector modeling.
Starting with medium load and engine speeds before attempting to idle helps the calibrator identify trends in the engine’s VE characteristics that can be extrapolated downward. Taking advantage of the stable nature of the engine at elevated speeds, the VE map is adjusted until delivered lambda equals target lambda at every cell in question. As long as cells exist in enough of the right locations to build an accurate airflow model, fueling should be stable regardless of speed and load. Most engines require more cells in the low engine speed range, since this is where the most drastic changes in actual VE occur.
These necessary shifts can save a lot of time chasing an unstable idle fuel mix that is experiencing misfires from being too far out of range. Likewise, break points necessary to accurately model the airflow can be chosen to improve accuracy. Modeling VE at idle, slightly above, and below gives the PCM the ability to maintain a constant lambda as VE changes, resulting in a more stable idle condition.
Only after steady state, part load, and idle are sufficiently calibrated should WOT tuning be performed. Just like with idle, the trends seen in the base VE map should be extrapolated upward for WOT as well. Again, the VE table is adjusted until actual measured lambda equals the commanded values in the tables.
After all steady state values of the base VE table have been calibrated, we can move on to transient fueling. Acceleration enrichment tables can now be adjusted to provide subjectively good throttle response. Resist the urge to adjust the VE table at this point, since the changes in measured lambda are a result of the wall wetting phenomena under changing conditions.
After the engine has had a chance to cool to ambient conditions, it should be started again to verify calibration of the temperature-compensation curves. In a speed density system, these curves have significant authority on delivered lambda. Many tuners mistakenly recalibrate the base VE table when encountering poor lambda control the day after tuning. If there is a legitimate temperature or weather change, and the tuner is confident in the previous day’s work, the only change should be to the compensation curves. Ideally, the temperature compensation should be checked near the limits of the predicted temperature range the vehicle will see to ensure proper operation under all conditions.
Many of these aftermarket PCMs have an “Auto Tune” feature that corrects fuel delivery based on feedback from an onboard wideband oxygen sensor. For the hobbyist looking to just get something running on a car, these help reduce the tuning effort and time. For those looking to perform a proper calibration, I recommend that these be turned off. There is rarely any substitute for an accurate wideband and an experienced calibrator making the proper mathematical adjustments. I always get better results by holding the engine in steady state on a dyno and making the necessary corrections versus allowing the included strategy to “learn” the proper tune. Remember that the engine is almost never steady state while driving on the road, so the learning function never really gets a good look at any particular trim cell for very long. If the steady state calibrations are done correctly, there will be little work for the closed loop correction later anyway. This leaves less room for error or drivability problems a week later.
The Accel DFI GenVII stand-alone EFI system is one of the most widely used and flexible aftermarket engine controllers. Scaling of the base maps is infinitely adjustable with respect to engine speed and load. Load can be configured as in-Hg, kPa, or TPS (true alpha-N) and can be used with any standard GM MAP sensor to accommodate up to 30 psi of boost. TPS range adjustment can be conveniently done from the driver’s seat by simply pressing the pedal in the calibration mode.
All base fuel calibrations are referenced from predicted air mass. Even though the actual mass flow number is not directly shown to the tuner, it is being calculated based on base VE, MAP, and modeled air temperature at the port. The result is an OEM-like calibration ability for lambda control. A properly calibrated GenVII system should drive just like a stock vehicle regardless of weather or load changes.
Accel DFI GenVII is the first widely available and affordable standalone to employ true airflow modeling, including Tau (wall film) prediction. Normal acceleration enrichment tables are present for those accustomed to using them. Accel goes one step further by adding tables to represent actual wall film fueling contribution during both acceleration and deceleration with respect to speed and temperature.
The integral datalogger for the Accel GenVII is plenty capable for most WOT or drivability recording needs. Although limited to six channels, acquisition rate is fast enough to provide sufficient feedback to the calibrator. The ability to overlay a recorded run with the base VE or spark maps makes it simple to see which cells were being used for the fuel and spark calculations during the run.
Another one of the most popular aftermarket standalone EFI systems is the Fuel Air Spark Technology (F.A.S.T.) XFI system. This PCM is a true speed density system much like the Accel DFI, where an actual air mass is being calculated under all conditions. Base maps are completely user-configurable for speed and load break points at an almost infinite number of break points.
The load axis for the base VE table can be configured for MAP in inches of mercury or PSI as well as TPS for alpha-N configuration if desired. Acceleration enrichment is handled in the traditional manner with a group of tables featuring modifiers based on both MAP and TPS values for predicted fueling. The datalogger for the FAST system is equally comprehensive, with the ability to log and graph any PCM input including wideband air/fuel error.
AEM (Advanced Engine Management) offers a standalone PCM called the “Programmable Engine Management System.” This system is very popular primarily with the import performance crowd. The real benefit is that AEM often packages the system together with an adaptor harness as a “plug and play” solution. This allows the user to simply replace the OEM PCM with the AEM unit and, using the preloaded tables for a stock vehicle, immediately start and run the car.
Because the AEM system is geared toward a wide variety of possible OEM combinations (and their vast array of sensors), it can be configured with unique individual sensor transfer functions. This means that the same PCM box can work with the engine sensors on a Toyota Supra, Honda Accord, or Mitsubishi Lancer Evolution. Calibration of the AEM system is fairly straightforward once the proper sensor configuration has been established. However, if an incorrect transfer is selected, you are in for a long and frustrating tuning session.
The system reacts like a typical speed density PCM, and the majority of the calibration work is done in the main VE table. There are separate tables for acceleration enrichment and a whole aggregate for boost compensation to fueling. One caveat is that for serious boost levels, proper fuel tuning requires the use of their specific compensation logic. If you intend to use this system on a high-pressure application, read up on this particular part of the instruction manual closely to avoid headaches later.
Another interesting feature is the variable wastegate duty cycle control. This allows for OEM style variable control where spool up time can be reduced and boost overshoots can be controlled. The ability to tie this logic to a vehicle speed sensor effectively gives traction control to the turbocharger crowd.
A relative newcomer to the stand-alone PCM market, the MegaSquirt system is a low cost alternative to its competitors. The system is available either complete or as a do-it-yourself electronics project requiring some assembly. The MegaSquirt is a traditional speed density system much like the Accel DFI and F.A.S.T. systems.
Like the other systems, it can be configured for either alpha-N or MAP based load calculation. MegaSquirt also employs Tau-based wall film compensation for those looking for ideal tip-in fueling response. The tuning process for these PCMs is identical to other stock or aftermarket units, so experienced calibrators should not have much trouble.
Actual calibration of the unit is best handled using their MegaTune software, but can also be accomplished with a number of other PC based editors. Although some of the controls appear a bit crude compared to the more polished and established aftermarket competitors, all the right knobs are there to turn for a proper tune. For some, the difference in cost more than outweighs this minor issue.
The Electromotive system is widely popular with the import performance crowd, and to an arguable extent with the domestic enthusiast as well. The system really shines with its extreme flexibility in controlling various coil pack and coil on plug ignition systems. It is a slightly more complex speed density system that packs a large amount of flexibility. However, its calibration and operation are significantly different from the traditional speed density systems previously mentioned.
The basic theory behind the operation of the Electromotive systems is their self-described “linear thermodynamics” calculation of pulsewidth. This should not be confused with a traditional thermodynamic calculation for engine performance. Rather, it is essentially their marketing terminology used to describe their pulsewidth calculation. The calculation of pulsewidth begins with a reference maximum referred to as “Gamma.” The time for one Gamma (TOG) is then multiplied by a series of factors that determine the final injection time. This essentially boils down to a linearization of the injection time versus manifold pressure (MAP) with a few other factors thrown in for good measure. The net result is that the engine can be started with less initial work, but fine-tuning the system is more involved and slightly less intuitive.
It has been my experience that these systems work best when the TOG is calculated as the maximum possible mechanical injection time at redline. Since some of these systems don’t allow for an injection time in excess of TOG, this ensures the full range of the injector is usable where it is needed most, at WOT near redline.
In the users’ manual, VE correction is not emphasized strongly enough, as they depend heavily upon the “liner thermodynamics” to produce the proper fuel delivery. This essentially skips the step of airflow modeling, going straight from MAP sensor to injection time. As we have already seen, plenty of other engine parameters come into play in the airflow calculation, so we must still make adjustments for changes in actual VE.
Once the basic engine and sensor parameters have been properly input to the Electromotive system, I prefer to finish the calibration by adjusting the VE correction table. This table falls under the “few other factors” description mentioned above, but is critical to proper operation. The values in this table do not reflect actual engine VE, but still have a mathematical effect on fuel delivery in the same manner. Mercifully, a 5% change to this table should still yield a 5% change to delivered lambda when the injector is properly modeled for dead time, voltage offset, and flow rate. Even with this approach, an actual MAF number is never really achieved, but the overall delivered fuel ratio can remain under reasonable control.
Another very keen feature is the ability to blend TPS and MAP readings by a factor to maintain better control near idle on vehicles with large cams. Since the large camshaft can have relatively poor vacuum at and near idle, the Electromotive system allows for the partial (or complete) replacement of the MAP in the fuel equation with TPS at certain speeds. Effectively, the system switches from MAP based speed density to alpha-N as needed. This results in a more stable signal, and less variation in the fuel delivery calculation in a region that can be critical to good driving behavior.
Written by Greg Banish and Posted with Permission of CarTechBooks