Take a look at the schematic in Figure 3-87, which is built around our old friend the 2N2222 NPN transistor. When the switch is closed, it connects the base of the transistor to the negative side of the power supply through a 1K resistor. At the same time, the base is connected with the positive side of the power supply through a 10K resistor. Because of the difference in resistances and the relatively high turn-on voltage for the LED, the base is forced below its turn-on threshold, and as a result, the transistor will not pass much current. The LED will glow dimly at best.
Figure 3-87. In this demonstration circuit, when the switch is opened, it interrupts negative voltage to the base of the transistor, causing the transistor to lower its resistance, allowing current to reach the LED. Thus, when the switch is turned off, it turns on the LED.
Now what happens when the switch is opened? The base of the transistor loses its negative power supply and has only its positive power supply. It becomes much more positive, above the turn-on threshold for the transistor, which tells the transistor to lower its resistance and pass more current. The LED now glows brightly. Thus, when the switch is turned off and breaks the connection, the LED is turned on.
This seems to be what we want. Imagine a whole series of switches instead of just one switch, as shown in Figure 3-88. The circuit will still work the same way, even if the switches are scattered all over your home, because the resistance in the wires connecting the switches will be trivial compared with the resistance of the 1K resistor.
Figure 3-88. A network of switches, wired in series, can be substituted for the single switch in Figure 3-87. Now any one switch will break continuity and trigger the transistor.
I have shown the switches open, because that’s the way the schematic for a switch is drawn, but imagine them all closed. The base of the transistor will now be supplied through the long piece of wire connecting all the closed switches, and the LED will stay dark. Now if just one switch is opened, or if anyone tampers with the wire linking them, the base of the transistor loses its connection to negative power, at which point the transistor conducts power and the LED lights up.
While all the switches remain closed, the circuit is drawing very little current—probably about 1.1 mA. So you could run it from a typical 12-volt alarm battery.
Now suppose we swap out the LED and put a relay in there instead, as shown in Figure 3-89. I don’t mind using a relay in this location, because the relay will not be “always on.” It will normally be off, and will draw power only when the alarm is triggered.
Figure 3-89. If the LED and 680Ω resistor shown in are removed, and a relay takes their place, the relay will be activated when any switch in the sensor network is opened.
Try one of the 12-volt relays that you used previously. You should find that when you open the switch, the relay is energized. When you close the switch, the relay goes back to sleep. Note that I eliminated the 680Ω resistor from the circuit, because the relay doesn’t need any protection from the 12-volt power supply.
Self-Locking Relay
There’s only one remaining problem: we want the alarm to continue making noise even after someone who has opened a door or window closes it again quickly. In other words, when the relay is activated, it must lock itself on.
One way to do this would be by using a latching relay. The only problem is that we would then need another piece of circuitry to unlatch it. I prefer to show you how you can make any relay keep itself switched on after it has received just one jolt of power. This idea will be useful to you later in the book as well.
The secret is to supply power to the relay coil through the two contacts inside the relay that are normally open. (Note that this is exactly opposite to the relay oscillator, which supplied power to its coil through the contacts that were normally closed. That setup caused the relay to switch itself off almost as soon as it switched itself on. This setup causes the relay to keep itself switched on, as soon as it has been activated.)
In Figure 3-90, the four schematics illustrate this. You can imagine them as being like frames in a movie, photographed microseconds apart. In the first picture, the switch is open, the relay is not energized, and nothing is happening. In the second, the switch has been closed to energize the coil. In the third, the coil has pulled the contact inside the relay, so that power now reaches the coil via two paths. In the fourth, the switch has been opened, but the relay is still powering its own coil through its contacts. It will remain locked in this state until the power is disconnected.
Figure 3-90. This sequence of schematics shows the events that occur when a relay is energized. Initially, the switch is open. Then the switch is closed, activating the relay. The relay then powers itself through its own internal contacts. The relay remains energized even after the switch is opened again. Power switched by the relay can be taken from the circuit at point A.
All we need to do, to make use of this idea, is to substitute the transistor for the on/off switch, and tap into the circuit at point A, running a wire from there to the noisemaking module.
Figure 3-91 shows how that would work. When the transistor is activated by any of the network of sensor switches, as previously explained, the transistor conducts power to the relay. The relay locks itself on, and the transistor becomes irrelevant.
Figure 3-91. The self-locking relay depicted in Figure 3-90 has been incorporated in the alarm circuit, so that if any switch in the network is opened, the relay will continue to power the noise maker even if the switch is closed again.
Because I’ve been adding pieces to the original alarm noisemaker circuit, I’ve updated the block diagram from Figure 2-112 to show that we can still break this down into modules with simple functions. The revised diagram is shown in Figure 3-92.
Figure 3-92. This block diagram previously shown in Figure 2-112 on page 90 has been updated to include the magnetic-switch network and locking-relay control system.
Blocking Bad Voltage
One little problem remains: in the new version of the circuit, if the transistor goes off while the relay is still on, current from the relay can flow back up the wire to the emitter of the transistor, where it will try to flow backward through the transistor to the base, which is “more negative,” as it is linked through all the magnetic switches and the 1K resistor to the negative side of the power supply.
Applying power backward through a transistor is not a nice thing to do. Therefore the final schematic in this series shows one more component, which you have not seen before: a diode, labeled D1. See Figure 3-93. The diode looks like the heart of an LED, and indeed, that’s pretty much what it is, although some diodes are much more robust. It allows electricity to flow in only one direction, from positive to negative, as shown by its arrow symbol. If current tries to flow in the opposite direction, the diode blocks it. The only price you pay for this service is that the diode imposes a small voltage drop on electricity flowing in the “OK” direction.
So now, positive flow can pass from the transistor, through the diode, to the relay coil, to get things started. The relay then supplies itself with power, but the diode prevents the positive voltage from getting back into the transistor the wrong way.
Perhaps a more elegant solution to the problem is to connect the NO (normally open) leg of the relay via a 10k resistor to the base connection. When the relay is not energized, the NO leg is inert and simply behaves as a parasitic capacitance on the node. When the relay becomes energized, the NO leg shunts +12V through the common terminal via a 10k resistor into the base of the transistor. In this circuit configuration, the transistor is never exposed to a potentially harmful voltage and you are not depending on leakage currents of non-ideal elements to protect devices.
However, I needed an opportunity to introduce you to the concept of diodes. You can check the following section “Essentials: All about diodes” to learn more.
Figure 3
-93. Diode D1 has been added to protect the emitter of Q1 from positive voltage when the relay is energized.
Essentials
All about diodes
A diode is a very early type of semiconductor. It allows electricity to flow in one direction, but blocks it in the opposite direction. (A light-emitting diode is a much more recent invention.) Like an LED, a diode can be damaged by reversing the voltage and applying excessive power, but most diodes generally have a much greater tolerance for this than LEDs. The end of the diode that blocks positive voltage is always marked, usually with a circular band, while the other end remains unmarked. Diodes are especially useful in logic circuits, and can also convert alternating current (AC) into direct current (DC).
A Zener diode is a special type that we won’t be using in this book. It blocks current completely in one direction, and also blocks it in the other direction until a threshold voltage is reached—much like a PUT.
Signal diodes are available for various different voltages and wattages. The 1N4001 diode that I recommend for the alarm activation circuit is capable of handling a much greater load at a much higher voltage, but I used it because it has a low internal resistance. I wanted the diode to impose a minimal voltage drop, so that the relay would receive as much voltage as possible.
It’s good practice to use diodes at less than their rated capacity. Like any semiconductor, they can overheat and burn out if they are subjected to mistreatment.
The schematic symbol for a diode has only one significant variant: sometimes the triangle is outlined instead of filled solid black (see Figure 3-94).
Figure 3-94. Either of these schematic symbols may be used to represent a diode, but the one on the right is more common than the one on the left.
Completing the Breadboard Alarm Circuit
It’s time now to breadboard the control circuit for your alarm noisemaker. Figure 3-95 shows how this can be done. I am assuming that you still have the noisemaker, which functions as before. I’m assuming that you still have its relevant components mounted on the top half of the breadboard. To save space, I’m just going to show the additional components mounted on the bottom half of the same breadboard.
It’s important to remember that you are not supplying power directly to the left and right “rails” on the breadboard anymore; you are supplying power to the relay-transistor section, and when the relay closes its contacts, the relay supplies power to the rails. These then feed the power up to the top half of the breadboard. So disconnect your power supply from the breadboard rails and reconnect it as shown in Figure 3-95.
Figure 3-95. The schematic that was developed in the previous pages can be emulated with components on a breadboard, as shown here. S1 is a DPDT relay. Wires to the sensor switch network and to the power supply must be added where shown.
Because it’s a double-pole relay, I am using it to switch negative as well as positive. This means that when the relay contacts are open, the noisemaking section of the circuit is completely isolated from the rest of the world.
The breadboarded relay circuit is exactly the same as the schematic in Figure 3-93. The components have just been rearranged and squeezed together so that they will fit alongside the relay. Two wires at the lower-left corner go to the network of magnetic sensor switches that will trip the alarm; for testing purposes, you can just hold the stripped ends of these two wires together to simulate all the switches being closed, and separate the wires to simulate a switch opening.
Two more wires bring power to the breadboard on either side of the relay. This is where you should connect your power supply during testing. The output from the relay, through its top pair of contacts, is connected with the rails of the breadboard by a little jumper wire at top left, and another at top right. Don’t forget to include them! One more little wire at the lower-left corner (easily overlooked) connects the lefthand side rail to the lefthand coil terminal of the relay, so that when the relay is powering the noisemaker circuit, it powers itself as well.
When you mount the diode, remember that the end of it that is marked with a band around it is the end that blocks positive current. In this circuit, that’s the lower end of the diode.
Try it to make sure that it works. Short the sensor wires together and then apply power. The alarm should remain silent. You can use your meter to check that no voltage exists between the side rails. Now separate the sensor wires, and the relay should click, supplying power to the side rails, which activates the noisemaker. Even if you bring the sensor wires back together, the relay should remain locked on. The only way to unlock it is to disconnect the power supply.
When the circuit is active, the transistor followed by the diode drops the voltage slightly, but the 12-volt relay should still work.
In my test circuit, trying three different relays, they drew between 27 and 40 milliamps at 9.6 volts. Some current still leaked through the transistor when it was in its “off” mode, but only a couple of milliamps at 0.5 volts. This low voltage was far below the threshold required to trip the relay.
Ready for Perfboarding
If the circuit works, the next step is to immortalize it on perforated board. Use the type of board that has a breadboard contact pattern etched on it in copper, as shown in Figure 3-72 on page 116. Check the following section,
“Essentials: Perfboard procedure,” for guidance on the best way to make this particular kind of solder joint—and the subsequent section for the most common problems.
Essentials
Perfboard soldering procedure
Carefully note the position of a component on your breadboard, and then move it to the same relative position on the perfboard, poking its wires through the little holes.
Turn the perfboard upside down, make sure that it’s stable, and examine the hole where the wire is poking through, as shown in Figure 3-96. A copper trace surrounds this hole and links it with others. Your task is to melt solder so that it sticks to the copper and also to the wire, forming a solid, reliable connection between the two of them.
Take your pencil-style soldering iron in one hand and some solder in your other hand. Hold the tip of the iron against the wire and the copper, and feed some thin solder to their intersection. After two to four seconds, the solder should start flowing.
Figure 3-96.
Figure 3-97.
Figure 3-98. To establish a connection between a section of wire and a copper trace on perforated board, the wire is pushed through the hole, and solder (shown in pure white for illustrative purposes) completes the connection. The wire can then be snipped short.
Allow enough solder to form a rounded bump sealing the wire and the copper, as shown in Figure 3-97. Wait for the solder to harden thoroughly, and then grab the wire with pointed-nosed pliers and wiggle it to make sure you have a strong connection. If all is well, snip the protruding wire with your cutters. See Figure 3-98.
Because solder joints are difficult to photograph, I’m using drawings to show the wire before and after making a reasonably good joint, which is shown in pure white, outlined with a black line.
Actual soldered perfboard is shown in the photographs in Figures 3-99 and 3-100.
Figure 3-99. This photograph was taken during the process of transferring components from breadboard to perforated board. Two or three components at a time are inserted from the other side of the board, and their leads are bent over to prevent them from falling out.
Figure 3-100. After soldering, the leads are snipped short and the joints are inspected under a magnifying glass. Another two or three components can now be inserted, and the process can be repeated.
Tools
Four most common perfboarding errors
1. Too much solder
Before you know it, solder creeps across the board, touches the next
copper trace, and sticks to it, as depicted in Figure 3-101. When this happens, you have to wait for it to cool, and then cut it with a utility knife. You can also try to remove it with a rubber bulb and solder wick, but some of it will tend to remain.
Even a microscopic trace of solder is enough to create a short circuit. Check the wiring under a magnifying glass while turning the perfboard so that the light strikes it from different angles (or use your solder wick to suck it away).
Figure 3-101. If too much solder is used, it makes a mess and can create an unwanted connection with another conductor.
2. Not enough solder
If the joint is thin, the wire can break free from the solder as it cools. Even a microscopic crack is sufficient to stop the circuit from working. In extreme cases, the solder sticks to the wire, and sticks to the copper trace around the wire, yet doesn’t make a solid bridge connecting the two, leaving the wire encircled by solder yet untouched by it, as shown in Figure 3-102. You may find this undetectable unless you observe it with magnification.
You can add more solder to any joint that may have insufficient solder, but be sure to reheat the joint thoroughly.
Figure 3-102. Too little solder (or insufficient heat) can allow a soldered wire to remain separate from the soldered copper on the perforated board. Even a hair-thin gap is sufficient to prevent an electrical connection.
3. Components incorrectly placed
It’s very easy to put a component one hole away from the position where it should be. It’s also easy to forget to make a connection.
I suggest that you print a copy of the schematic, and each time you make a connection on the perforated board, you eliminate that wire on your hardcopy, using a highlighter.
Make: Electronics Page 18
