Finding Faults on Transistors
Testing BJTs and JFETs
Once you have identified the transistor pins, or at least the base, if the fault is not already obvious you can use the method in Fig. 3.1.27b to identify a fault on any bipolar transistor that is not connected in a circuit.
Use a multi meter switched to a range suitable for testing about transistor junctions, as discussed on the Meters for transistor testing page. Follow the numbered sequence of tests below to find out if the transistor is good or faulty.
Before you begin these tests, make sure you know which of your meter leads is positive and which is negative. If you are not sure, read part 2 of this section "Meters for transistor testing"
1. Test the resistance between collector and emitter.
2.Then reverse the positive and negative meter connections and test again.
If the meter reads zero or a few ohms in tests 1 and 2, there is a short circuit between collector and emitter and the transistor is faulty. If both readings are infinity, continue with test 3.
3. Now connect the positive meter lead to the base and test the resistance of both junctions by connecting the negative meter probe to one of the other two pins. It doesn't really matter whether this is the collector or the emitter, in our test we are simply testing a junction.
4. Now leave the positive lead on the base and move the negative lead to the other untested (collector or emitter) pin and measure the resistance of this junction.
For tests 3 and 4 you should get a typical forward resistance reading of less than 1k in both cases.
5. Now connect the negative lead of your meter to the base and the positive lead to another pin as shown at 5 in the diagram above.
6. Lastly connect the positive probe to the other untested pin as shown at 6 in the diagram above.
In tests 5 and 6 both junctions should read infinity. If all of these six tests are ok you have a good transistor. If one or more of the tests has failed, so has the transistor!
Fault identification on FETs
The results of resistance tests on FETs are generally not as easy to interpret as in bipolar transistors. Because of the high impedances involved the results will be more variable and practice is needed to gain confidence in the results obtained. In addition, the handling requirements for IGFETS with regard to electrostatic voltages mean that testing these devices out of the circuit is very likely to cause more damage than good! The only effective test for IGFETs is by substituting a known good device, making sure that the handling precautions mentioned earlier are observed. JFETs however can, with care, be tested with a multi-meter in much the same way as bipolar transistors.
Fig 3.1.28 JFET Junction Model (A single PN junction and a Resistive Channel)
Fig 3.1.28 shows a junction model for testing a JFET. With this device we can consider that we are testing a single PN junction attached to a channel that is basically a resistor. The resistance of the channel between source and drain will be very high (several Megohms) but may vary considerably if we have the positive meter lead connected to the drain and then even touch the very high impedance gate with our fingers. This can put enough voltage on the gate to operate the transistor! The actual results you get we vary depending on such things as the type of meter used, the resistance of your skin and even the humidity of the room.
We can normally test the PN junction by connecting the meter between gate and source, first one way and then reversing the polarity. The result should be a low reading of about 1k ohms in the forward bias direction (positive to gate in the case of a N channel device) and infinity (open circuit) in the reverse bias direction (negative to gate).
Testing transistors in circuit
Although you can sometimes use the above methods for testing transistors still in a SWITCHED OFF circuit, provided that any resistors in the circuit around the transistor have high values and don´t have much effect on the actual transistor resistances you are measuring, the above methods assume you are going to test the transistor, having first unsoldered it and removed it from the circuit. Of course this is only one way to test a transistor, and usually used to confirm earlier tests done with the circuit in "working" (though faulty) condition. These tests involve measuring the voltages on the suspect transistor with the circuit switched on and are part of a full fault finding process. There are however some simple voltage indications that can indicate if an already suspect transistor is faulty.
1. More than 0.7V difference between base and emitter voltages indicates an open circuit b-e junction.
2. The same voltage on two or more terminals MAY indicate one or more short circuit junctions.
3. A LOWER than expected collector voltage generally means that the transistor is conducing heavily (turned on).
4. A HIGHER than expected collector voltage generally means that the transistor is not conducting (turned off).
Note: Indications 2,3 and 4 can also be caused by other circuit conditions.
Warning: You should never work on "live" circuits unless you know AND USE safe working practices. Many circuits that derive power from the mains (line) supply (and some that don´t) contain LETHAL voltages as well as other hazards. Live circuits must only be worked on by fully trained personnel. Before attempting any work on live circuits using any information provided on this web site, please read the important DISCLAIMER.