Sound source locator uses PreChamp

This circuit could be regarded as a simple alternative to the Sooper Snooper device featured elsewhere in this issue. It is based on the PreChamp preamplifier featured in the July 1994 issue of SILICON CHIP. It can be used to pinpoint or locate sounds coming from inconspicuous sources such as a noisy bearing in a VCR containing many bearings.

Other examples are an audible gas leak or a puncture in a slowly deflating tyre, provided the background noise level is considerably less than the source noise.

The device uses an electret microphone. This is attached to the end of round Texta casing of the same diameter, then covered with spaghetti sleeving and overlapping the electret by 5mm to make it more directional.

The electret was then wired to the two-transistor Pre-Champ preamplifier (available from Dick Smith Electronics, Jaycar & Altronics as a kit). The 2.2kΩ feedback resistor in the preamplifier is replaced with switch S1 and four resistors – 2.2kΩ, 4.7kΩ, 10kΩ and 22kΩ – to give gains of 23, 48, 101 and 221, respectively.

After passing through the modified PreChamp, the signal goes to the base of Q1 which is biased on the verge of turning on by the 100kΩ and 6.8kΩ resistors. When the electret microphone picks up a sound, the amplified signal is fed to the base of Q1 turning it on and drawing current via diode D1 to reduce the stored charge in C1, the 220μF electrolytic capacitor. This results in a dip in the reading of analog meter M1.

Meanwhile C1 is constantly being charged via the 10kΩ resistor and the 50kΩ trimpot VR1. VR1 is set to provide full scale deflection on meter M1 when no signal is present. If a 50uA meter movement is used it will required a suitable shunt resistor to suit the circuit.

Summing up, as the unit is used to home in on a sound source, louder sounds cause the meter reading to drop. Frequencies below 20Hz will cause the meter pointer to flutter.

Switch S2 and the associated 1kΩ resistor are provided to quickly discharge C1 to enable repeated measurements.

P. Hetrelezis, Noble Park, Vic. ($30)

Building a synchronous clock

The quartz clocks which have dominated time-keeping for the past 20 years or so have one problem: their errors, although slight, are cumulative. After running for several months the errors can be significant. Sometimes you can correct these if you can slightly tweak the crystal frequency but otherwise you are forced to reset the clock at regular intervals.

By contrast, mains-powered synchronous clocks are kept accurate by the 50Hz mains distribution system and they are very reliable, except of course, when a blackout occurs. This circuit converts a quartz clock to synchronous mains operation, so that you can have at least one clock in your home which shows the time.

First, you need to obtain a quartz clock movement and disassemble it down to the PC board. For instructions on how to do this, see the article on a "Fast Clock For Railway Modellers" in the December 1996 issue of SILICON CHIP. Then isolate the two wires to the clock coil and solder two light duty insulated hookup wires to them (eg, two strands of rainbow cable). Drill a small hole in the clock case and pass the wires through them. Then reassemble the clock case.

To test the movement, touch the wires to the terminals of an AA cell, then reverse the wires and touch the cell terminals again. The clock second hand should advance on each connection.

The circuit is driven by a low voltage AC plugpack. Its AC output is fed to two bridge rectifiers: BR1 provides the DC supply while BR2 provides positive-going pulses at 100Hz to IC1a, a 4093 NAND Schmitt trigger. IC1a squares up the 100Hz pulses and feeds them to the clock input of the cascaded 4017 decade counters. The output at pin 12 of IC3 is 1Hz.

This is fed to IC4, a 4013 D-type flipflop, which is connected so that its two outputs at pins 12 & 13 each go positive for one second at a time. As these pulses are too long to drive the clock movement directly, the outputs are each fed to 4093 NAND gates IC1b & IC1c where they are gated with the pin 3 signal to IC4.

This results in short pulses from pins 3 & 10 of IC1 which drives the clock via limiting resistor R1. The value of R1 should be selected on test, allowing just enough current to reliably drive the clock movement.

A. J. Lowe, Bardon, Qld.

Blown fuse indicator

This blown fuse indicator will work with a wide range of DC supply voltages from 5V to 50V. It illuminates LED1 when the fuse blows. With the fuse intact, Q1 is held off and there is no bias current available for the base of Q2. So the LED is off.

When the fuse blows, a small current flows via the base-emitter junctions of Darlington transistor Q1, through its base resistor R1 and then via the load. Typically this current will be around 20μA and this turns on Q1 which provides base current to Q2 which then turns on to illuminate the LED.

The emitter current of Q2 is limited by Q3 which turns when the current reaches about 10mA, to shunt base current away from Q2.

The three resistor values not given in the circuit are dependent on the supply voltage and can be calculated from the following simple equations:

R1(kΩ) = V(DC)/0.02 = 560kΩ for 12V DC

R2(kΩ) = V(DC)/2 = 5.6kΩ for 12V DC

R3(Ω) = V(DC)/0.02 = 560Ω for 12V DC

R3 should be included for voltages above about 20V otherwise the heat dissipation in Q2 will be too great. At lower voltages it can be omitted.

Any general purpose NPN transistors can be used for Q2 and Q3, provided they will handle the DC supply voltage. The PNP Darlington, Q1, could be an MPSA65, available from Dick Smith Electronics (Cat Z-2088).

Keith Gooley, via email. ($40)