Fullscreen HTML5Med Res (??MB)HTML4

Chapter 18 Peak-Hold Injector Adaptor This simple adaptor board allows the Digital Pulse Adjuster, Independent Electronic Boost Controller and Digital Duty Cycle Meter to work with cars using peak hold injectors. Which Cars? So how do you find out if your car has peak-hold or conventional (they’re called “saturated”) injectors? In short, the only definitive way is to use an oscilloscope. However, if the injector resistance is low (eg, 3Ω) and if the duty cycle measurement of the Digital Pulse Adjuster, Independent Electronic Boost Controller or Digital Duty Cycle Meter is erratic, it’s likely the car is using peak hold injectors. 108 PERFORMANCE ELECTRONICS FOR CARS M OST FUEL INJECTORS are operated with a pulse waveform – power is applied to switch them on, they stay open for a short time, and then the power is switched off and they close. However, there is one injector type that doesn’t work this way. These are known as peakhold injectors – they look completely standard but the way they operate makes measuring their duty cycle much more difficult. Since three of the major projects in this book measure injector duty cycle, that could create some problems for us. However (and sound the trumpet), after quite some work, we have developed a simple standalone module that allows these projects to be used with peak-hold injectors. As a bonus, it also allows a normal duty cycle measuring multimeter to read injector duty cycle on peak-hold cars, something which normally can’t be done. But what’s peak-hold all about, anyway? Peak Hold Peak-then-hold injectors are switched on with full power but once they are open, the power is reduced. This reduction is carried out by quickly switching siliconchip.com.au Parts List 1 PC board coded 05car151, 79 x 50mm 1 PC board coded 05car152, 53 x 15mm 3 2-way PC-mount screw terminals with 5.08mm spacing 4 6.3mm male PC-mount spade connectors with 5mm pitch 1 UB5 plastic box (optional; not in kit) Semiconductors 1 LM358 dual op amp (IC1) 3 16V 1W zener diodes (ZD1-ZD3) 1 4.7V 1W zener diode (ZD4) 1 1N4004 1A diode (D1) Capacitors 1 100µF 16V PC electrolytic 1 10µF 16V PC electrolytic 2 100nF MKT polyester (code 104 or 100n) 1 10nF MKT polyester (code 103 or 10n) 2 100pF ceramic (code 100 or 100p) Fig.1: this diagram shows the parts layout for the PC board and the details for connecting the monitoring resistor. The 12V feed to an injector is broken and the 0.1Ω 5W resistor is inserted in series with it. Signal wires from either side of the resistor run back to the Peak Hold Adaptor which is mounted in the cabin. The signal output from the adaptor connects to the input of the Digital Pulse Adjuster, Independent Electronic Boost Control or Digital Duty Cycle Meter. Resistors (0.25W, 1%) 2 1MΩ 4 1kΩ 1 470kΩ 1 470Ω 1 10kΩ 1 150Ω 2 4.7kΩ 1 10Ω 1 2.2kΩ 1 0.1Ω 5W wirewound (R1) the voltage to the injector on and off. This is done so fast that the injector doesn’t shut – it just sees a lower average voltage while this process is occurring. At the end of the injector opening time, the power is switched off and the injector closes. Measuring the duty cycle involves detecting when the injector opens and closes – in other words, the “edges” of the waveform. However, in peak-hold injector waveforms, it’s very hard to detect the edges and filtering has to be used to ensure that the system ignores the very quick switching that occurs during the “hold” portion of the injector opening period. This filtering also takes care of the sharp voltage spike that occurs part way through the opening period, when the injector changes to “hold” mode. Fig.3(a) shows the complex waveform of a peak-hold injector. siliconchip.com.au The Peak Hold Adaptor is constructed on two small PC boards. The 0.1Ω 5W resistor is mounted close to the injector, while the main PC board can be housed in a box inside the cabin. The best way to sense the injector duty cycle in a peak-hold system is to monitor the injector current instead of the voltage. That way, we can be sure when the injector is switched on and off. This is because when the injector is switched on, there is current flow and when the injector is off, there is no current through it. This current is detected using a small value series resistor. The re- sistor is small enough that it does not affect the injector operation. By monitoring and amplifying the voltage across this current sensing resistor, we can use a comparator to switch its output level when there is current flow detected. Fitting The series resistor – a 0.1Ω 5W unit – is mounted in the engine bay. PERFORMANCE ELECTRONICS FOR CARS 109 How It Works The circuit is based on dual op amp IC and just a few other components. As shown, resistor R1 is placed in series between the +12V supply and the injector. Op amp IC1a is connected as a differential amplifier and monitors the voltage across this resistor. When no current flows through R1, no voltage is developed across it. Conversely, when the injector is powered, there is current flow and so there is a small voltage drop across R1. In practice, the top of resistor R1 has +12V applied to it. This is reduced to +6V at the junction of the two 1kΩ divider resistors. ZD2 clamps any high voltages to protect IC1a while the 100nF capacitor filters the signal to reduce high-frequency noise. The following 4.7kΩ and 1MΩ resistors form a voltage divider to reduce the signal by a factor of 0.995. The gain applied to the signal at IC1a’s non-inverting input (pin 3) is set by the ratio of the feedback resistors connected to pin 2 – ie, to 1 + 1M/4.7k, or +213.77. The voltage at the injector side of R1 when it is switched off is also 12V. Therefore, the voltage at the junction of the 1kΩ divider resistors for the invert- ing input, pin 2, is also +6V. The gain for this signal is -1M/4.7k or -212.77. Therefore, the gain for the signal fed to the non-inverting input is slightly higher than for the inverting input and this is why the pin 3 signal is reduced slightly (ie, by 0.995). Thus, when the injector is off, both input signals on either side of R1 are at +12V and so the same +6V is produced by both sets of 1kΩ divider resistors. The subsequent signal path gains in each case are effectively the same; however, the signal on the injector side of R1 is inverted compared to the +12V side of R1. Consequently, the output of IC1a will be at 0V. In other words, this +6V “common mode” signal is rejected while any difference signal (ie, the voltage drop across R1) is amplified and appears at pin 1 of IC1a. Let’s now see what happens when the injector is driven. In this case, there will be a voltage drop across R1 and so IC1a’s output voltage will rise accordingly. This typically increases to about +2V when the injector is in its hold mode and to +12V during the peak current drive. This voltage change is filtered using a 100pF capacitor across the 1MΩ feedback resistor for IC1a. Further filtering is provided by the 2.2kΩ resistor and 10nF capacitor at IC1a’s output. This filtering removes any sudden voltage changes that may cause false detection of the injector on/off current. The filtered signal from pin 1 of IC1a is then fed to op amp IC1b which is connected as a Schmitt trigger. Pin 5, the non-inverting input, is connected to the wiper of trimpot VR1. Zener diode ZD4 provides a stable +4.7V reference voltage for VR1. It is fed via a 470Ω resistor from the +12V supply and its output filtered using a 10µF capacitor. VR1 is the threshold control for IC1b. The 470kΩ and 10kΩ resistors at pin 5 of IC1b are there to provide a small amount of hysteresis for the Schmitt trigger. This means that the voltage at pin 6 needs to go about 200mV higher than the voltage at VR1’s wiper before the output of IC1b switches to 0V. Similarly, pin 6 needs to go about 100mV below VR1’s wiper before the output switches high again to 12V. This hysteresis prevents IC1b’s output from oscillating when the voltage on pin 6 is close to the switching threshold. It can be soldered to the small sub-PC board provided in the kit and the assembly mounted in a small metal box (making sure that the connections are insulated from the box). Alternatively, the resistor can be connected directly in-line in the injector wire. It’s important to note that this resistor is not placed on the switched side of the injector but instead in the +12V feed to the injector. The easiest way to find this wire is by unplugging the injector and probing the plug with a multimeter. One side of the plug should have +12V on it – that’s the wire into which the resistor is inserted. Two signal feed wires are used to connect each side of the resistor to the module, which should be mounted in the cabin. These connections are shown in Fig.1. The signal “out” from the Peak Hold Adaptor connects to the “input” of the device that you’re working with – eg, the input of the Digital Pulse Adjuster. Initially, leave the lid off the box so that you can access the trimpot (VR1). At this point, set it to about the middle of its travel. Start the car and see if the device that’s monitoring injector duty cycle works – eg, the load site number on the Hand Controller of the Digital Pulse Adjuster varies up and down with load. If there are problems, try adjusting the input pot on the DPA (or the RESISTOR COLOUR CODES 110 Value 4-Band Code (1%) 5-Band Code (1%) 1MΩ 470kΩ 10kΩ 4.7kΩ 2.2kΩ 1kΩ 470Ω 150Ω 10Ω 0.1Ω brown black green brown yellow violet yellow brown brown black orange brown yellow violet red brown red red red brown brown black red brown yellow violet brown brown brown green brown brown brown black black brown not applicable brown black black yellow brown yellow violet black orange brown brown black black red brown yellow violet black brown brown red red black brown brown brown black black brown brown yellow violet black black brown brown green black black brown brown black black gold brown not applicable PERFORMANCE ELECTRONICS FOR CARS siliconchip.com.au Fig.2: the circuit is based on a dual op amp IC (IC1). IC1a operates as a differential amplifier while IC1b is wired as a Schmitt trigger. Note that IC1b’s output follows the injector voltage so when the injector is off, pin 7 is high (+12V) and when the injector is powered, pin 7 is low (0V). Power is obtained from the switched +12V ignition supply of the vehicle. D1 provides reverse polarity protection, while zener diode ZD1 clamps spike voltages above 16V. The 10Ω resistor limits the current through ZD1 when there is a voltage transient and the 100µF capacitor filters the supply.      ) b) Fig.3(a) is the scope view of a peak hold injector waveform. The sequence of events is as follows: (1) the voltage drops to zero when the fuel injector is switched on; (2) an inductive spike occurs as the drive switches from peak to hold; (3) the hold voltage is controlled by rapidly “turning” (or switching) the injector on and off but at a rate that’s too rapid for the injector to actually open and close; (4) there is another, larger Independent Electronic Boost Control, if that’s what you’re working with), or adjusting the pot of the Peak Hold Adaptor. If there is still no joy, try siliconchip.com.au inductive spike as the injector is switched off; (5) the signal voltage returns to the battery voltage (5). Sensing when the injector is open and when it is shut is very difficult but our adaptor overcomes that problem. Fig.3(b) is the scope view of the Peak Hold Adaptor output. As you can see, it’s nothing very exciting – just a square wave. But that’s exactly what we want – a waveform that’s easily monitored for duty cycle. swapping the signal leads from the resistor – you may have these the wrong way around. Finally, if it still won’t work cor- rectly, try the resistor in the other arm of the injector feed – in some cars, it can be very hard to work out which  wire is which. PERFORMANCE ELECTRONICS FOR CARS 111