Building a New Airflow Meter, Part 2

Last week in Building a New Airflow Meter, Part 1 we showed you how to build a larger than standard airflow meter that still takes the factory sensing element. It’s good for reducing intake flow restriction and increasing the headroom before the meter reaches maximum output. (And it also allows you to mount in-line a meter that was previously integrated into the airbox.) However, an upsized airflow meter will give an altered signal voltage output, resulting in a change in the mixtures.

In this story, we show you how to correct the air/fuel ratios, and as a bonus, tune the open-loop mixtures as well.

Changed Output

As the body of an airflow meter is increased in size, the voltage output from the sensor will be reduced. This is because the air speed through the meter is lower and so the sensing element sees less air going past. So if you upsize the airflow meter but don’t make any electronic changes to its output, the air/fuel ratio will be lean as the ECU will think that there’s less air going into the engine than there really is.

However, there’s a cheap and elegant solution in the shape of the Silicon Chip Digital Fuel Adjuster. This electronic kit (it’s also available fully built and tested) allows the output of the airflow meter to be adjusted up or down at up to 128 points across its load range. The design interpolates between the points (ie smooths the curve of the adjustments) and gives factory driveability. The programming of the DFA is via a hand-controller that uses a backlit LCD. For more on the DFA, see The Digital Fuel Adjuster, Part 1. In kit form the DFA costs just AUD$80 and the hand controller AUD$60.

Calibration

• No Air/Fuel Ratio Meter

It is possible to upsize the airflow meter and then correct the output voltage of the new design with the DFA without ever measuring air/fuel ratios. Taking this approach has the major advantage that an accurate air/fuel ratio meter is not needed. So how’s it done?

The first step is to use a multimeter to carefully measure the standard airflow meter’s output voltage at a variety of engine loads during road testing. (Obviously, use an assistant to read the meter or drive the car!) For example, the output voltage can be measured at idle, full power (ie redline at full throttle), and at perhaps eight points between these extremes. These ‘in between’ load points can be set using a manifold vacuum gauge (or vacuum/boost in a forced aspirated car) and the tacho – so for example, one of these points might be 1 psi of boost at 4000 rpm, 3^rd gear, another might be at 7 psi of boost at the same revs and gear, and so on.

Then, when the new airflow meter is installed and the DFA wired into place, the output signal voltage of the DFA is measured and the hand controller used to adjust the output voltage until it matches the original airflow meter at all the previously measured load points. When the output voltage is correct at these load points the rest of the curve can be set up using these points as the reference (ie basically join the dots). If doing it in this way, gradually work your way up to higher loads, setting the DFA correction ahead of your current load point so the car never runs lean under power.

Once you’ve done all of that, the air/fuel ratios should be fine.

• With an Air/Fuel Ratio Meter

If you have access to a professional air/fuel ratio meter, just use the DFA to set the mixtures at all load points. Note that when the car is operating in closed loop, you should bring the mixtures back to standard - outside of closed loop you can set them to whatever you want.

Alternatively, any tuning shop should be able to dyno-tune the Digital Fuel Adjuster to suit the revised airflow meter – just take along the Hand Controller instruction sheet that shows the function of each button.

Or finally, if you’re really careful with how you go about it, you may well be able to get away with tuning via the output of a conventional oxygen sensor – see Real World Air/Fuel Ratio Tuning

In any case, don’t forget to operate the ‘lock’ button on the DFA when the tuning is done.

If the airflow meter upsize isn’t great, the DFA should be set to fine mode (and bench-calibrated in that mode) and this will allow ultra-fine tuning of the mixtures. If the airflow meter upsize is great, or absolute accuracy in mixtures isn’t needed, the normal coarse mode is fine. So what air/fuel ratios do you want at different loads? See Tuning Air/Fuel Ratios for more on this.

Doing It

In the case of the NHW10 Toyota Prius on which the upsized airflow meter was fitted, a MoTeC air/fuel ratio meter was used to monitor the outcome of tuning changes being made with the DFA. However, even before the car would start and happily idle, the airflow meter’s output level needed to be boosted.

With the DFA set to coarse mode and calibrated to work over the 0-5V range that suits this airflow meter, a +10 correction was put in at all load sites around idle. In other words, the voltage output of the new airflow meter was increased by about 0.4V. The +10 value was arrived at by trial and error – at this value the car idled happily and the ECU worked in closed loop, as indicated by an air/fuel ratio of about 14.7:1.

But why do you need to set any correction at all if the car is in closed loop? Won’t it just learn its way to the new mixtures? The answer in most cases is ‘no’. This is because the ECU doesn’t have unlimited ability to change the mixtures in response to the oxygen sensor feedback. In the case of the Toyota being covered here, the decreased voltage output of the larger airflow meter resulted in mixtures too lean for the ECU to learn around – a 0.4V boost at idle was needed so that the car would run in closed loop. This is an important but oft-overlooked aspect of the DFA – returning signal levels to close enough to standard that the ECU self-learning will work.

With the +10 value working well at idle, this value was put in at all higher load sites – right up to #128. This car runs in closed loop all the time and adding this value across the whole load range made it driveable at all loads. However, some fine tuning of the DFA was then needed as the car could be seen to be still learning – that is, when a certain load site was first reached, the air/fuel ratio was leaner or richer than standard before it quickly learnt its way back to 14.7:1. By making further minor corrections to the DFA map, it was possible to reduce the amount of learning the ECU needed to do and so mixtures stayed closer to 14.7:1 all the way through the load range.

Conclusion

If your car uses a detachable airflow sensing element, making a new larger airflow meter to house it is easy. Almost as straightforward is the adjustment of its output with the Digital Fuel Adjuster.

Digital Fuel Adjuster