non-invasive test for injectors

Now all engines have injectors for fuel, they are managed electronically by a control unit, in which decides the injection time or the time for which the injector must remain open and therefore the amount of fuel to enter, depending on several parameters detected in real time by the same ECU. A fundamental part is the needle injector, that surging by means of a solenoid, He uncovers the holes by which the fuel is sprayed. An electric impulse given by the solenoid enough to lift the pin and open the supply port.

In this circuit we will check that the solenoid is not interrupted.

Principle of operation

First a bit of theory on the essential heart of this comparator circuit.

The figure 1 shows a conventional design for a comparator. This arrangement works using a voltage divider (Rx e Ry) to establish the minimum threshold voltage.

The comparator should evaluate and compare the input signal or the voltage (Wine) with the set threshold voltage (Vth).

The voltage at the comparator input to be compared is connected to the inverting input, accordingly the output has an inverted polarity.

Whenever Vin> Vth is supposed that the output approaches the negative supply (GND or low logic level to the flowchart shown) e, when Vln <Vth then the output is close to the positive supply (Vcc = 5V or high logic level in this example).

While the voltage of the input signal arrives at the set limit (via the voltage divider network) (Vth = 2,5 V), rule above and below the minimum threshold.

Consequently, l’output oscilla anche in base all’input. In current circuits, this unstable output could easily cause problems.

By way of example, think that the input signal is a parameter of temperature and that the output response is crucial application based on temperature, interpreted by a microcontroller.

The response to the fluctuating output signal may not provide information to the microcontroller and produce faithful results “confusionari” for the microcontroller that operates with precise and constant levels.

Furthermore, imagines that the comparator output is used to drive a motor or a valve. This inconsistent during the switching threshold limits could force the ignition / the valve is turned off or the engine several times in the course of crucial threshold situations.

The graph more clear than words this situation, in the case of a signal with a noise threshold superimposed on the main signal.

In the chart below a portion of the upper graph is expanded to explain the concept.

This solution is used when there is the certainty that any on / off output fluctuations do not pose any problem, or when the changes are so slow that the noise figure can be considered negligible. It lets you decide if a real signal, eg, the temperature is higher than a predetermined value or as I realized these days for a friend of a sensor operation of the injectors of a car with a hall sensor.

The sensor, a AH49E, You can also find similar acronyms produced by other manufacturers such as SS49E is easily affordable network at an acceptable price.

It has three pins, two power (5V) and a third which produces an analog output proportional to the magnetic flux which invests. The only flaw is that the magnetic flux can have positive or negative polarity, to overcome this drawback, manufacturers fix the output of the sensor to approximately half of the power supply in the absence of magnetic flux.

I used this scheme

The operational amplifier is powered by the battery voltage to eliminate the problem of the drop-out typical of economic opamp not provided with rail to rail output.

The 78L05 feeds the sensor and the voltage divider formed by two fixed resistors 33K and 5K from the trimmer, the trimmer function is to fix the threshold voltage of the comparator, and also even has the effect “collateral” sensitivity adjustment.

Calibration is simple, in the absence of magnetic fields adjusts the trimmer until the led off. At this point, each magnetic field is also small entities picked up by the sensor will cause the ignition of the LED.

As previously mentioned this sensor settles to about 2,53V in the absence of magnetic field to vary linearly its output in more or less than this value as a function of the magnetic flux and its polarity. I exploit the fact that any injector is crossed by a discrete current and it induces a small magnetic field detectable. If the injector is healthy as I note from its magnetic field. With this small, non-invasive tool I must not remove anything to control this aspect of the engine, the power take it directly from the battery, always present in the bonnet, the diode in series it guarantees that nothing will break in case of power supply polarity inversion, simply does not work, the circuit draws less than 10mA with LED access.

Here is a short video of the test the first prototype did on the field.