To check a transistor using a multimeter, you must use the Diode Test mode, not the Ohms (resistance) setting. A good silicon Bipolar Junction Transistor (BJT) will show a forward voltage drop of 0.500V to 0.700V across its base-emitter and base-collector junctions. A good power MOSFET will show 0.400V to 0.600V across its internal body diode. Any reading of 0.000V (short) or OL (open) in forward bias means the semiconductor junction is destroyed.

Below is the exact bench procedure for verifying NPN/PNP BJTs and N-Channel/P-Channel MOSFETs, including the specific numeric thresholds that separate a good part from a dead one.

Meter Setup and Safety: The Right Mode and CAT Rating

Before touching any probes to silicon, configure your digital multimeter (DMM) correctly. Using the wrong mode or ignoring safety categories is how meters get bricked and technicians get shocked.

Meter Configuration

  • Dial Position: Set to Diode Test (symbol: an arrow pointing into a perpendicular line, often shared with the continuity buzzer).
  • Lead Jacks: Black lead in COM. Red lead in V/Ω (Voltage/Ohms).
  • Range: Auto-ranging is standard. Open-circuit test voltage should be between 2.0V and 3.0V (verify by touching probes together; it should read 0.000V, then pull them apart to see the meter's open-circuit voltage, usually around 2.8V on a Fluke 87V).
⚠️ Safety Category (CAT) Requirements:
If you are testing a transistor on a loose bench, CAT II probes are sufficient. However, if you are probing a transistor inside a hardwired Switch Mode Power Supply (SMPS), a solar inverter, or any device directly connected to mains branch circuits, you must use CAT III 1000V or CAT IV 600V rated probes. Mains-connected circuits can harbor high-energy transient spikes that will arc across standard probe gaps. Always de-energize the circuit, lock out the breaker, and bleed large filter capacitors (using a 10kΩ 5W resistor) before testing in-circuit.

Testing Bipolar Junction Transistors (NPN & PNP)

A BJT is essentially two diodes sharing a common anode (PNP) or cathode (NPN). We test it by forward-biasing and reverse-biasing these internal diodes. For this example, we will assume a standard silicon transistor like the ubiquitous 2N3904 (NPN) or 2N3906 (PNP). Germanium transistors are rare today but will show lower forward drops (~0.2V to 0.3V).

Step-by-Step NPN Test (e.g., 2N3904, 2N2222, TIP31C)

  1. Base to Emitter (Forward): Red probe on Base, Black probe on Emitter. Expect 0.500V – 0.700V.
  2. Base to Collector (Forward): Red probe on Base, Black probe on Collector. Expect 0.500V – 0.700V (often slightly lower than B-E by ~0.020V).
  3. Base to Emitter (Reverse): Black probe on Base, Red probe on Emitter. Expect OL (Overload/Open).
  4. Base to Collector (Reverse): Black probe on Base, Red probe on Collector. Expect OL.
  5. Collector to Emitter (Both ways): Probe C-E, then swap. Expect OL both ways.

Step-by-Step PNP Test (e.g., 2N3906, 2N2907, TIP32C)

  1. Base to Emitter (Forward): Black probe on Base, Red probe on Emitter. Expect 0.500V – 0.700V.
  2. Base to Collector (Forward): Black probe on Base, Red probe on Collector. Expect 0.500V – 0.700V.
  3. Reverse & C-E Tests: Same as NPN. Reverse bias should read OL, and Collector-to-Emitter should read OL both ways.
Test Point NPN Expected (Red on Base) PNP Expected (Black on Base) Failure Indication
Base-Emitter 0.500V – 0.700V 0.500V – 0.700V 0.000V (Short) or OL (Open)
Base-Collector 0.450V – 0.680V 0.450V – 0.680V 0.000V (Short) or OL (Open)
Collector-Emitter OL (Both Directions) OL (Both Directions) Any numeric voltage drop

Testing Power MOSFETs (N-Channel & P-Channel)

MOSFETs do not have PN junctions between the Gate and the Drain/Source; the Gate is insulated by a thin layer of silicon dioxide. Therefore, the Gate should read completely open (OL) to all other pins. The primary test for a MOSFET relies on its intrinsic body diode, which exists between the Drain and Source. For an N-Channel MOSFET (like the IRFZ44N or IRF540N), the diode's anode is the Source and the cathode is the Drain.

N-Channel MOSFET Procedure

  1. Discharge the Gate: Briefly touch the Black probe to the Drain and the Red probe to the Gate. This bleeds off any residual gate charge.
  2. Body Diode (Reverse Bias): Red probe on Drain, Black probe on Source. Expect OL.
  3. Body Diode (Forward Bias): Red probe on Source, Black probe on Drain. Expect 0.400V – 0.600V.
  4. Gate Isolation: Measure Gate to Drain, and Gate to Source, in both directions. All four combinations must read OL.
Pro-Tip: The Gate Charge Test
To prove the gate oxide is intact and the channel can actually form, leave the Black probe on the Source and the Red probe on the Drain (reading OL). Without lifting the Black probe from the Source, use your finger or a jumper wire to briefly touch the Gate pin to the Drain pin. This applies the meter's internal test voltage to the Gate. Move the Red probe back to the Drain; the reading should now drop to near 0.000V, indicating the MOSFET has turned on. Touch the Gate to the Source to turn it back off (returns to OL).

Common Mistakes That Give False Readings

If your readings don't match the tables above, the transistor might not be dead—your testing environment might be lying to you. Watch out for these three bench traps.

  • Testing In-Circuit (Parallel Paths): If you test a transistor while it is still soldered to a PCB, surrounding components (pull-down resistors, snubber diodes, transformer windings) will create parallel current paths. A good transistor might read 0.200V or show continuity across Collector-Emitter. Fix: You must desolder at least two pins (preferably all three) to isolate the part for a definitive test.
  • Finger Resistance: If you hold the metal tips of both probes with your bare fingers while testing a reverse-bias junction, your body's resistance (typically 1MΩ to 5MΩ) will allow a tiny current to flow. The meter might display 1.200V to 1.800V instead of the expected OL. Fix: Hold only the plastic insulation, or use alligator clip leads.
  • Using Ohms Mode: The Ohms (Ω) setting applies a variable test current and calculates resistance. Semiconductor junctions are non-linear; their resistance changes depending on the meter's test current. This yields useless, fluctuating numbers. Always use Diode Test mode, which measures the actual forward voltage drop at a standardized test current (usually 1mA to 2mA).

The Decision Tree: Pass, Fail, or Replace?

Use this decision matrix to determine your next step. Do not reinstall a part that shows borderline behavior; silicon degradation is usually catastrophic under load.

Observed Symptom Diagnosis Action & Concrete Replacement Pick
Forward bias reads 0.000V or buzzes continuously. Shorted Junction. The silicon has melted internally, usually from thermal runaway or overcurrent. Discard. Replace NPN with ON Semi 2N3904G or STMicro TIP31C. Replace N-Ch MOSFET with Infineon IRFZ44NPBF.
Forward bias reads OL (Open) instead of 0.5V-0.7V. Open Junction. The internal bond wire has snapped or the die has cracked from mechanical stress. Discard. Replace PNP with ON Semi 2N3906G. Replace P-Ch MOSFET with Infineon IRF9540NPBF.
C-E or D-S reads a voltage drop in both directions. Punch-Through Failure. The depletion region has collapsed. Common in MOSFETs subjected to voltage spikes exceeding Vds_max. Discard. Check your snubber circuits before installing the replacement.
MOSFET Gate to Drain/Source reads any value other than OL. Gate Oxide Rupture. Caused by ESD (Electrostatic Discharge) or exceeding Vgs_max (usually ±20V). Discard. Use a 12V Zener diode between Gate and Source on the replacement to prevent recurrence.
All readings match the spec tables perfectly. Pass. The basic semiconductor junctions are intact. Keep. Reinstall with fresh thermal compound (e.g., Arctic MX-4) if it is a TO-220/TO-247 power package.

For deeper theory on semiconductor junction behavior and advanced curve tracing, refer to the All About Circuits semiconductor textbook. For foundational multimeter safety and diode testing mechanics, review the SparkFun Multimeter Tutorial. Always ensure your test equipment meets the IEC 61010 measurement category standards for your specific work environment.