Make Electronics

Getting Somewhat More Serious 139

Experiment 15: Intrusion Alarm Revisited

Transferring components from the breadboard to the perforated board should
be fairly simple, as long as you don’t try to move too many at once. Follow the
suggestions described in previous section “Essentials: Perfboard procedure,”
and pause frequently to check your connections. Impatience is almost always
the cause of errors in this kind of work.

Figure 3-103 shows the noisemaker section of the circuit on perfboard, with
the components positioned to minimize wasted space. Figure 3-104 shows
the perfboard with the relay and its associated components added. The two
black wires will go to the loudspeaker, the black-and-red pair of wires will
bring power to the board, and the green wires will go to the magnetic sensor
switches. Each wire penetrates the board, and its stripped end is soldered to
the copper beneath.

Test it now, in the same way that you tested the same circuit on the bread-
board. If it doesn’t work, check the following section, “Essentials: Real-world
fault tracing.” If it does work, you’re ready to trim the board and mount it in a
project box.

essentIAls

Real-world  fault   tracing

Here’s a real-life description of the procedure for tracing a fault.

After I assembled the perfboard version of the combined noisemaker and relay

circuit, I checked my work, applied power—and although the relay clicked, no

sound came out of the loudspeaker. Of course, everything had worked fine on

the breadboard.

First I looked at component placement, because this is the easiest thing to

verify. I found no errors. Then I flexed the board gently while applying power—

and the loudspeaker made a brief “beep.” Any time this happens, you can be

virtually certain that a solder joint has a tiny crack in it.

The next step was to anchor the black lead of my meter to the negative side

of the power supply, and then switch on the power and go through the circuit

point by point, from top to bottom, checking the voltage at each point with the

red lead of the meter. In a simple circuit like this, every part should show at least

some voltage.

But when I got to the second 2N2222 transistor, which powers the loudspeaker,

its output was completely dead. Either I had melted the transistor while solder-

ing it (unlikely), or there was a bad joint. I checked the perfboard beneath the

transistor with the magnifying glass, and found that solder had flowed around

one of the leads of the transistor without actually sticking to it. The gap must

have been less than one-thousandth of an inch, but still, that was enough.

Probably, the problem had been caused by dirt or grease.

This is the kind of patient inquiry you need to follow when a circuit doesn’t

work. Check whether your components are placed correctly, check your power

supply, check the power on the board, check the voltage at each stage, and if

you are persistent, you’ll find the fault.

Figure 3-103. The noisemaker circuit has

been transplanted from breadboard to

perforated board, with no additions or

changes.

Figure 3-104. The relay-transistor control

circuit has been added. Wires to external

devices have been stripped and poked

into the perforated board, where they are

soldered in place. The green wires connect

with the sensor network, the black wires

go to the loudspeaker, and the red-and-

black wires supply power.