This month, I want to talk about why circuits are ground switched. For years, most circuits on a car were switched in the live side of the circuit, it was considered safer because the circuit was only live when switched on. So why the change?
This change came about when electronics started to take over the role of switching circuits. So why would that make a difference? Well, it is down to the way in which transistors work and how the control circuits are powered. Let’s start by looking at transistors and how they are used as switches in automotive circuits. The two most basic types of transistors are known as NPN and PNP. The names are to do with the way they are constructed. They are made of sandwiches of material and are either N-type or P-type.
NPN is by far the most common. Figure 1 shows a typical control circuit using a NPN transistor. There are three connections on the transistor; Base, Emitter and Collector. The transistor works a bit like a relay, it uses a small current to switch a larger current. Passing a small current through the Base connection to the Emitter allows a much larger current to flow through from the Collector to the Emitter. The voltage at the Base must go approximately 0.7 volts above the voltage at the Emitter for the transistor to start conducting current from Collector to Base, and below 0.7 volts it switches off again. The flow of current through the transistor from collector to emitter is actually proportional to the current flowing through the Base to Emitter. On a basic type of transistor there will be around 20-100 times more current flowing through the Collector/Emitter than through the Base/Emitter. For the transistor to act like a switch, we need to supply enough current into the Base to ensure the transistor is fully switched on, but not too much or there is a risk that the transistor will release the magic smoke. When the transistor is fully on, known as saturated state, the voltage drop across the Collector/Emitter is close to 0 volts, meaning the full supply voltage is across the device being switched.
Most devices on a car are now controlled by a microcontroller, a small single chip computer. Many years ago, the operating voltage for these devices was standardised at 5 volts, this being the lowest voltage that was not prone to electrical noise from other parts of the circuit switching, which could cause errors in operation. Minimum voltage also means minimum heat dissipation, something that is important if you are going to put hundreds or thousands of transistors onto a single chip and gives the ability to run the circuit from small batteries.
If you look at FIGURE 1, you’ll notice the transistor is placed on the negative/ground side of the circuit. This means the emitter is connected to the ground and the microcontroller’s 5 volt output is well above the 0.7 volts needed to switch the transistor on. In fact, we use a resistor in the circuit to limit the current and stop the smoke escaping. The micro controller can easily switch the transistor on and off straight from a digital output.
Now look at FIGURE 2. Here, I have drawn the circuit with the NPN transistor on the positive side. For the transistor to be switched on, the voltage at the base must be 0.7 volts above the emitter, so if we supply 12 volts to the Base there will be an initial flow of current though the lamp, but as the transistor reaches saturation point and the voltage drop across the Collector/Emitter becomes close to zero, the Base and Emitter voltages will be the same. We no longer have the 0.7 volt difference needed and the transistor will switch itself off. In addition, the microcontroller can’t provide the 12 volts to initially switch it on from a 5-volt output without added components and expense.
FIGURE 3 shows a circuit using a PNP transistor, these are better suited to switching the positive side. A PNP transistor is switched on by taking the Base voltage 0.7 volts lower than the Emitter (note that the terminals are the opposite way around to the NPN transistor). Our microcontroller can again manage this fine, with a current limiting resistor to stop it being damaged, however the base now needs to be taken up to the same voltage as the emitter to switch it off. If we are switching a 12V circuit, then the base needs to reach 12V to turn off. The 5V supply from the microcontroller is never going to be able to achieve this, so the microcontroller will never be able to switch the circuit off.
Hopefully, you can see the only easily achievable way we can use a microcontroller to switch 12-volt devices is by using an NPN transistor and switching the ground side, and this is the reason for the change to ground side switching.