Component Signature Analyser

The Huntron^® Tracker^® is an amazingly useful test instrument, but sadly it is far too expensive for the hobbyist.  This project is a greatly simplified version, and uses your oscilloscope as the display.  Although it is not recommended for sensitive logic devices, the unit described is perfect for fault finding in audio equipment and most other applications where everything can tolerate a couple of milliamps of input current.

The idea of component 'signature analysis' is not new, but as far as I know, it was commercialised by Huntron with the original Tracker unit many years ago.  This signature analyser is the same as one that I built nearly 30 years ago (and still use), and although it doesn't get a great deal of use these days, that's mainly because I do very little service work.

Each class of component produces a display that identifies it, and a faulty part will not produce the pattern you expect.  Signature analysis is especially useful if you have one working and one non-working item.  For example, if a stereo amplifier is faulty, it is rare that both channels will fail.  The working amplifier gives you a pattern for each test location, and when the faulty component(s) are probed in the other amp the pattern will change.  Like most test instruments, it takes some time to get used to the tester and what to expect, but once you have used it a few times you will quickly get a feel for what it is telling you.

You will need an oscilloscope (CRO - cathode ray oscilloscope or a digital scope) that is capable of operating in X-Y mode.  Normally, the timebase provides the X (horizontal) axis signal, but in X-Y mode both axes are external.

The general (and very basic) principle of operation is shown in Figure 1.  The test signal is simply derived from the mains, and is a sinewave at 50 or 60Hz.  In most locations, the sinewave will be distorted, but this barely matters.  With nothing connected to the DUT (Device Under Test) terminals, the oscilloscope simply displays a horizontal line.  This represents voltage, and is applied to the X axis of the oscilloscope.  If the DUT terminals are shorted, the display changes to a vertical line (Y axis).

Figure 1 - Operating Principle

This behaviour is easily explained by looking at Figure 1.  With the terminals open, there is no voltage across R1, so nothing is applied to the Y axis of the oscilloscope.  The full voltage is applied to the X axis, and produces a horizontal line.  R1 is selected to limit the current through the DUT to a safe value.

Now, if we short the test leads, the signal to the X axis is shorted out, and the voltage from the transformer is now across R1, and is directed to the Y axis resulting in a vertical line.  When any component that is not open or short-circuited is probed, there will be a mixture of X and Y axis signals applied to the scope, and a distinctive pattern is produced.  Diodes, inductors, capacitors, transistor junctions and resistors all provide a unique signature, and any mixture of components will produce a result that is easily recognised.

Figure 2 shows some typical patterns (signatures) for a variety of components.  While not comprehensive, it shows what to expect.  You need to change ranges to suit the part being tested - for example a large electrolytic capacitor will register as a short circuit on a low current range.  If big enough, such a cap will show as a short on any range, because its impedance is so low.

Figure 2 - Typical Component Signatures

Transistor junctions are usually a very good test, since the signature of a component that is still working may show that the junction is degraded.  This could be from excess temperature or other component damage.  The junctions of a good transistor will show a sharp transition, but a degraded junction may show a slow (curved) transition and/ or evidence of leakage (the breakdown region is sloped instead of vertical).

The tester is easy to build, and no PCB is needed.  Although construction is somewhat fiddly because of the switching, it is very straightforward.  The switching is easy to get wrong though, so refer to the circuit diagram and make sure that everything is where it should be.  All switch positions are shown on the diagram ... both voltage and current switches are in the 'Low' position.

The transformer can be any size you have available or can find - the current drain is no more than 20mA.  As shown, a 30V multi-tapped transformer is ideal.  Because the first tap on those commonly available is 9V, the transformer is used the other way around - the 30V tap is the common, the 24V tap now gives a more sensible 6V output, and the 0V tap gives the full 30V.  Remember that these are all RMS voltages, so the peak voltage is 8.5V (approx.) in the low position, and 42V for high.

The circuit shown below may not look anything like that in Figure 1, but they are (more-or-less) the same.  It looks 'odd' because of the transformer wiring and all the switching, but the principle is identical.  The 30V tap on the transformer is the common, and current is sensed by resistors R1-R4.  There are two sets of current sense resistors, with the appropriate set selected by Sw2B to account for the two voltage ranges.  Without the extra switching and resistors, the current would change widely (by a factor of 6:1) when the voltage is changed.  The extra set of current-sense resistors prevents this from happening.

Figure 3 - Circuit Diagram of Complete Tester

The LED is optional, but recommended - it tells you that the power is on.  The diode in parallel is to prevent excess reverse voltage which can damage the LED junction.  The 1k resistor (R9) limits the current to about 6mA peak.  Sw2C and Sw2D are used to maintain the same voltage at the oscilloscope terminals when the voltage is changed.  This saves you from having to change ranges whenever the test voltage is changed.  The higher voltage is divided by the ratio of 1M to 330k (in parallel with the oscilloscope's input resistance - usually 1MΩ), so the oscilloscope voltage is reduced by a factor of five when the high range (30V) is selected, giving about 6V.  You may choose to use trimpots so that the high and low test voltages give identical displays on the scope if you want to do so.

If you can't find a suitable transformer, you can use two.  The primaries will normally be in parallel and the secondaries in series.  As noted above, current drain is very low, so even the smallest of transformers will work fine.  The voltages are not especially critical, but the networks of R5-R6 and R7-R8 are voltage dividers, designed to maintain the same voltage(s) to the oscilloscope inputs regardless of the selected voltage range.  Additional voltage and current ranges can be added if you wish, but you will have to work out the dividers yourself.

You could (if willing to experiment enough) use an audio oscillator and small power amplifier to provide the test signal.  You will still need a transformer, because the signal must be fully floating for the circuit to work.  Variable frequency is very useful for caps and inductors but makes no difference with resistors or semiconductors.  You do get a nice clean sinewave, which you most certainly will not get from the mains.  The disadvantage (of course) is that the unit becomes much more complex, so much so that if you think you want to go that far, I'd consider buying a real Huntron Tracker.  Not cheap, but they have different frequencies, and a wide range of voltage and current ranges (as well as an in-built colour display), plus many other features.  See the Huntron website for more information.

The current ranges are selected by SW3.  The resistors are selected to give the same current at either voltage, and as shown they will provide an RMS current of either 2mA or 20mA (2.8 and 28mA peak current respectively).  While these currents are fairly safe with most small signal and power transistors, be very careful if you test the gates of MOSFETs - the voltage must be on low range, and even this may be too high for some devices.  If the gate voltage exceeds the rated maximum, the MOSFET will be destroyed!

Note:  Most tests should be conducted at low voltage and low current.  Even though the higher voltage and current are relatively safe, some components may be distressed or damaged.  Only use the high ranges if you are absolutely certain that the device(s) under test can withstand the peak values that the tester can supply.

As you can see from Figure 2, each component type has a signature, but it will take time to learn what to expect.  The most common way to use this type of tester is comparison - comparing the waveform of a good board to that of a faulty board.  As you get closer to the fault, expect to see the 'good' and 'bad' waveforms differ more and more.  The beauty of this method of testing is that it only ever requires a working board to allow analysis, so construction errors can often be found just by the signature.

Many parts, such as low value resistors, small inductors or large capacitors will show as a short circuit.  In some cases increasing the current will work, but if the impedance is low enough it will still show as a short.  While this is a limitation, in reality it is not as bad as it may seem.  Most real faults will make themselves apparent fairly easily, and the more you use it, the more proficient you'll become at interpreting the results.

Don't expect to become an expert immediately.  Despite the apparent simplicity of the circuit and its principle of operation, it will take you quite a while to become proficient in its use.  Because it's primarily intended for 'in-circuit' testing (no removal of parts should normally be required), make sure that the circuit is powered off before you start to probe anything.  If you probe with power applied, there's every chance that you could damage the circuit, the tester, or both.

Main Index

Projects Index