# Whats your rule of thumb method of testing a transistor

Status
Not open for further replies.

#### tytower

##### Banned
In your parts boxes you pick up a transistor -any type -no markings . How do you test it and decide if its PNP or NPN ,Polar or FET, find its gain etc.?

I am interested in the on the job approach , what do you use when the tranny is out of circuit and maybe any tricks you use while it is still in the circuit.

#### Hero999

##### Banned
The only fail safe way is to buy a component tester.

Second to that, you can easily find the base of a BJT using your DVM's diode tester.

If it's NPN, the base will appear as an anode and the collector/emitter will be cathodes. If it's PNP, the base will appear as a cathode and the collector/emitter will be anodes.

Measuring the gain is easy. Connect the transistor configured as an emitter follower (with a known emitter load) to a 3V to 5V (no more than 5V because it will be damaged if the collector and emitter are reversed) supply.

Measure the impedance of the base. How do you do that? Simple measure the current using a sense resistor.

Measure the emitter current using Ohm's law.

Work our the gain.

Here's a worked example:

$I_B = \frac{2.33 \times 10^{-3}}{100} = 23.3 \times 10^{-6}A\\ I_E = \frac{2.35}{1000} = 2.35 \times 10^{-3}A\\ Hfe = \frac{I_E-I_B}{I_B}=\frac{2.35 \times 10^{-3}-23.3 \times 10^{-6}}{23.3 \times 10^{-6}}=99.86$

Normally, $Hfe = \frac{I_E-I_B}{I_B}$ can be simplfied to $Hfe = \frac{I_E}{I_B}$ as Ib is normally so small it doesn't matter.

For PNP transistors, just reverse the power supply.

If you don't know which pin is the collector and which is the emitter, try the above circuit both ways, the gain should be higher when connected the right way round.

For power MOSFETs, there's always a diode connected between the drain and source (negative to drain for N-type and positive to drain for P-type) and there should be no electrical connection between the gate and the other terminals.

#### Attachments

• 2.3 KB Views: 584
Last edited:

#### Mr RB

##### Well-Known Member
I don't keep transistors with no markings.

The ones that are marked, I refer to comprehensive transistor substitution books that tell the specs, or sometimes google a datasheet (which only takes about 1 minute).

Really, why would you work blind with a transistor with no markings?? Even the poorest person can pull tons of well marked transistors from equipment and get the datasteets from google.

#### Hero999

##### Banned
I don't know why I suggested using a 100R resistor for sensing 23.3µA.

Use a 10k resistor, 212mV is much easier to measure with a DVM.

As you can see the results will be slightly different but that's to be expected. A high voltage drop across the base resistor will reduce the base and therefore the emitter voltage.

EDIT:
Another good idea is to make the base resistor equal to the emitter resistor, that way you can calculate the gain by dividing the voltage across the emitter resistor by the voltage across the base resistor.

Here's a component tester:
Test Equipment - multimeters, oscilloscopes, function generators, thermometers & electrical testers

#### Attachments

• 2.2 KB Views: 468
Last edited:

#### kpatz

##### New Member
I don't keep transistors with no markings.

The ones that are marked, I refer to comprehensive transistor substitution books that tell the specs, or sometimes google a datasheet (which only takes about 1 minute).

Really, why would you work blind with a transistor with no markings?? Even the poorest person can pull tons of well marked transistors from equipment and get the datasteets from google.
X2... plus if it's unmarked and has 3 leads, it may end up not being a transistor at all. Could be a SCR, triac, 3-pin IC...

#### Hero999

##### Banned
Here's the schematic to go with my edit.

Just divide the emitter resistor voltage by the base emitter voltage to get the approximate Hfe.

In this case
$Hfe = \frac{2.33}{23.1 \times 10^{-3}}=100.9$

Note the gain will be slightly different because it depends on the collector current.

The disadvante is you need a DVM that can accurately measure 23.1mV.

#### Attachments

• 2.2 KB Views: 548
Last edited:

#### RODALCO

##### Well-Known Member
I chuck unmarked transistors out as well.

Marked TRS, I test with a multimeter on diode schale. You should get about a 0.6 volts drop across both junctions.
When testing in circuit generally that voltage drops to about 0.4 volts because of the parrallel paths.

Status
Not open for further replies.