One of the most difficult aspects of modifying an EFI or engine managed car on a small budget is knowing what air/fuel ratios are occurring at different combinations of loads and revs. The ignition timing maps won't usually need much tweaking for a budget car to run well, but more power needs more fuel, and knowing how much extra fuel to add can be quite tricky. However, there is an easy and cheap way of getting a rough idea of the air/fuel ratios being used. It is certainly not a technique accurate enough to guide you when mapping programmable management, or to use when modifying a mega-dollar engine, but it is far better than tuning by "feel" alone. It is also much cheaper than having professional exhaust gas analysis carried out on a chassis dyno every time you make a tuning change.
So how do you do it?
All unleaded petrol cars use an exhaust gas oxygen sensor. As the name suggests, this sensor is mounted in the exhaust flow - usually in the exhaust manifold - and sniffs the composition of the exhaust gas. Specifically, it measures the ratio of oxygen in the exhaust gas with that in the atmosphere. It does this is to determine whether the air/fuel ratio is rich, stoichiometric, or lean. The ECU uses this information as part of its self-learning technique, and in some cars a dash warning light is also illuminated if the mixture stays rich or lean for too long a period.
The most commonly used sensor generates its own voltage output, which varies between 0-1 volt. In round terms, if the sensor output is about 0.2 volts or less the mixture is lean, and if the output voltage is over 0.8 volts it is rich. However, the precise value of the output voltage is less important than its relative value - whether it is "richer" or "leaner" than the mid-point voltage.
Monitoring the sensor output can be done with a digital multimeter, but in practice the response time of many multimeters is too slow to keep up with mixture fluctuations. This is because when the ECU is in closed loop mode, the air/fuel ratio fluctuates in a rapid rich-lean-rich-lean sequence. This occurs because if the oxygen sensor output indicates that the mixture is rich, the ECU leans it out a little. When the mixture is a little lean, the ECU richens it. This constant process causes a cycling of the mixtures around the stoichiometric point. Depending on the particular engine management system (and the health of the EGO sensor!) this can occur up to about 10 times a second. That's what makes it hard to read on a digital multimeter! (Incidentally, don't use an analog meter - it loads the circuit down too much.)
The speed of change of the EGO sensor output means that it is easiest read on a bar-graph display, with a LED bar graph fast in response and cheap. Ten coloured LEDs can be used, with two red LEDs indicating lean mixtures (remember red for danger!), 6 green LEDs showing mixtures in a 'normal' range, and 2 yellow LEDs showing rich mixtures. If you wish, you can reduce the number of green LEDs and substitute more red and yellow ones. It is important that coloured LEDs be used (as opposed to an all-red bar-graph display, for example), because it is far easier to see at a glance the mixture strength by simply looking at the LED colour, rather than its position in the display. Jaycar Electronics produces a kit allowing the easy construction of a Mixture Meter - it is available in the AutoSpeed shop here.. Later in this feature we show you how to construct the meter, step by step - so even if your electronics knowledge is minimal, you will still be able to build it.
The meter is powered by an ignition-switched 12 volt source, and so you will need to access such a supply at either the fuse box or another switched device (like the radio). Connect positive 12 volts to the '+12V' pin on the meter, and connect the 'GND' pin to earth (the chassis or negative lead of the battery). Make sure that these wires are connected the right way around - the meter is not protected against reverse power supply polarity! The final meter connection is to the signal output of the oxygen sensor.
Oxygen sensors are commonly available in single or 3-wire configurations. If your car is fitted with a single wire sensor, simply connect the signal lead from the meter to this wire. Don't disconnect the oxygen sensor output from the vehicle ECU; instead wire the meter in parallel. The Mixture Meter (unlike some commercially available air/fuel gauges) does not draw enough current to affect the reading that the ECU is making of the sensor. The easiest way of making the connection is to access the EGO sensor wiring near to the sensor itself. Bare a short length of the wire by using a razor blade to carefully slice away the insulation covering, and then solder the mixture meter signal lead to it. Wrap the joint with good quality insulation tape or seal the connection with heat-shrink tubing. Alternatively, you can access the oxygen sensor signal wire at the ECU.
If your car's sensor is the 3-wire (or 4-wire) type, you will need to locate the signal output wire. The extra wires found in this type of sensor are to power an internal heater, which brings the sensor up to temperature more quickly than occurs solely by heat transfer from the exhaust gas. With the car running and up to operating temp, one wire will have +12 volts on it, another 0 volts (the earth), and the final wire 0.4-0.6 volts. It is the latter that is the EGO sensor output, which must be connected to the '02INPUT' on the Mixture Meter.
With the car running, the connected meter should light some of its LEDs. If the EGO sensor is still cold, the 'lean' red LED may be the only one to light, but as the sensor comes up to temp, other LEDs will also glow. With the sensor up to temperature, a blip on the throttle should cause the lit LED to run up and down the scale. If all the LEDs light up at once - and there is a burning smell coming from the meter - switch off the car immediately and check the orientation of the IC. If no LEDs light, check the polarity of the power supply wiring, and if you find that you have connected the meter wrongly, buy another IC and try again! If your meter works perfectly but dies after a few days in the car, buy a new IC and then make sure that none of the wires leading to the meter are near to ignition leads.
There are two ways of calibrating the meter - on the road or on a chassis dyno. Easier is on the road, although note that this won't be appropriate in a car which has already been highly modified.
With an assistant installed in the passenger seat and with the engine up to operating temp, drive at a constant speed - say 60 km/h - and with a steady throttle opening. The lit LED should start oscillating up and down the display, as the ECU makes the mixtures alternately rich and lean in closed loop operation. Adjust the trim-pot so that the oscillations in either direction are symmetrical around the middle LED. Now, use full throttle and watch what happens to the mixture read-out. The display should instantly show a rich mixture (either of the two yellow LEDs lit), and this mixture should be constantly held. Lift the throttle abruptly and the display should blank, as the injectors reduce their flow on the over-run - and so the mixture goes full lean. At idle, the meter should again show the closed loop oscillations.
If you are installing the meter on a highly modified engine, in-car calibration can still be done - but with the proviso that the mixtures may be all wrong to start with! Safest with this type of car is to use a chassis dyno and an exhaust gas analyser so that the mixture meter can be calibrated according to the gas analyser's read-out.
It is important that you realise that are a couple of substantial limitations to the Mixture Meter. Firstly it is incapable of separating air/fuel ratios of (say) 11:1 and 12:1. The meter will just show full rich! Secondly, mixtures will always appear lean when the sensor is not up to temperature. This means that you should be consistent in your testing procedure, so that when you make adjustments to the air/fuel ratio, changes that appear on the meter are changes that you have caused -not just the meter warming up!
If you have a leaded car that lacks an exhaust gas oxygen sensor, you can source a sensor from a wrecker and install it in the exhaust yourself. However, running leaded petrol will soon poison the sensor, and so this approach should be used only for tuning purposes, with the sensor removed for everyday use.
Once you have installed a LED Mixture Meter on your dashboard, it is very hard to go back to having no mixture indication at all. You simply feel blind without it.
Building the LED Mixture Meter
Even if you have had no electronics experience, you will still be able to assemble the LED meter. You'll need a fine-tipped soldering iron, diagonal cutters and the kit. It will be useful to also have a multimeter, some needle nosed pliers, and some more solder.
Make sure that you know what each of the components looks like. From top then clockwise: printed circuit board (PCB), hook-up wire, LEDs, potentiometer (pot), resistor, pins, capacitor, solder, and (in the middle) the integrated circuit (IC).
Each of these components needs to be soldered to the PCB with their wires going into exactly the right holes. The first component to be placed on the board is the resistor. This can be inserted either way around.
Using the needle nosed pliers, bend the resistor wires at 90 degrees and then insert the wires into the holes shown here. When working on the PCB, you should always orientate the board so that the multiple row of holes at the base of the board is closest to you, as shown here.
Turn the PCB over and you will see the long resistor wires protruding through the holes in the silver PCB tracks. You need to now solder these wires to the track.
Touch the tip of the soldering iron to the wire and PCB track, applying solder at the same time. When the junction is heated sufficiently, the solder will flow around the join, forming a shiny solder blob and electrically joining the two. This should be a quick operation - don't leave the soldering iron applied for more than a few seconds. If the solder doesn't "take" properly, apply the iron again.
After you have soldered the resistor into place, the board should look like this. Note that huge blobs of solder are not wanted - there should be only enough to surround the wire and cover the track on the PCB. Make quite sure that you haven't formed a bridge to another PCB track - if you have, apply the iron and remove this surplus solder.
Solder the other resistor wire and, once you are happy that the soldered joins are good, use the diagonal cutters to nip off the wires close to the PCB. Don't chop into the solder join itself - cut the wires off just proud of the solder. The soldering technique used for the resistor is used to attach each electronic component to the board.
Next, insert the pot. This uses three wires and so fits onto the PCB only one way around. The wires are already short, so they don't need to be cut off after you have soldering this component into place.
The next step is to solder the LEDs into place. These must be orientated the right way around, or they will not work. Place one of the yellow LEDs next to the board as shown here. Make sure that the longer lead (the anode in technical speak!) is at the left hand side, nearest to the 'A' marked on the back of the PCB.
Bend the wires at 90 degrees and insert the yellow LED, pushing its body snug up against the edge of the board. Doing this means that all the LEDs will line up neatly. Solder the two wires into place and nip off the surplus wire.
Insert the rest of the LEDs, soldering their leads into place as you go. Make sure that you get all of the LEDs orientated the right way around. One way to check this is to hold the board up to the light and check that the internal shape of the LEDs are all orientated the same.
Next the capacitor can be soldered into place. The capacitor must go onto the board the one way around. If you look at the capacitor, you will see that one lead is marked with a (-) on the body. This is the negative wire. The other wire is therefore the positive (although it's not marked) and the positive lead goes closest to the LEDs.
Nearly done! The IC socket is soldered on next. The IC socket has a cut-out at one end, and the socket is soldered into place so that (with the board orientated as shown here) the cut-out goes to the right. The wires from the socket are quite short and so won't need to be cut-off after they have been soldered. Be very careful that solder bridges are not formed between the pins - they are spaced together tightly.
Insert the gold coloured pins into place (they just make it easier to solder wires to the board) and then solder a different coloured length of hook-up wire to each pin. Carefully insert the IC into its socket - making sure that all of the pins actually go in without being bent - and then the Mixture Meter is finished!