A Bipolar Junction Transistor (BJT) is essentially two diodes sharing a common terminal (the base). Because of this internal physics, you do not need an expensive semiconductor curve tracer to find out if it is dead. You can definitively test a transistor using the diode-test function on a standard digital multimeter (DMM).
The 60-Second Verdict: A healthy silicon NPN transistor will read between 0.500V and 0.800V when forward-biased from Base to Emitter and Base to Collector. It will read OL (Over Limit) in reverse bias and across the Collector-Emitter junction. If your meter reads 0.000V (a dead short) or OL in both directions on a junction, the transistor is destroyed and must be replaced.
Multimeter Setup and Safety Categories
Before touching any probes to silicon, configure your meter correctly and verify your safety environment. Testing a loose component on a bench is vastly different from probing a soldered joint on a live or recently powered board.
If you are testing a transistor in-circuit on a mains-powered device (like an offline Switch Mode Power Supply or a variable frequency drive), the primary bulk capacitors can retain lethal voltages (up to 400V DC) long after the device is unplugged. You must de-energize the circuit, lock out the power, and safely bleed the filter capacitors using a high-wattage discharge resistor before testing. Never probe a live SMPS primary-side transistor.
Meter Setup Block:
- Dial Position: Set to Diode Test mode (symbol: →| ). Do not use the resistance (Ω) mode, as the test voltage is often too low to forward-bias a silicon junction.
- Red Lead: Insert into the VΩ (Volts/Ohms) jack.
- Black Lead: Insert into the COM (Common) jack.
- Range: Auto-ranging is standard for diode test. If your meter is manual, set it to the 2V DC range.
- Safety Category (CAT Rating): A CAT II meter is sufficient for isolated bench work. If probing mains-connected control boards (e.g., HVAC contactors or appliance motor drives), you must use a CAT III or CAT IV rated meter with fused test leads to protect against transient spikes.
Step-by-Step Probe Placement for NPN and PNP BJTs
To test a transistor accurately, you must first identify the Base (B), Collector (C), and Emitter (E) pins. Always consult the manufacturer's datasheet for your specific part number (e.g., BC547, 2N2222, TIP31) because pinouts vary wildly even within the same package style (like TO-92). Once identified, follow this sequence for an NPN transistor. (For a PNP, simply reverse the red and black probe roles).
- Test 1 (Base-Emitter Forward): Place the Red probe on the Base and the Black probe on the Emitter. You are forward-biasing the base-emitter diode.
- Test 2 (Base-Collector Forward): Keep the Red probe on the Base and move the Black probe to the Collector.
- Test 3 (Base-Emitter Reverse): Swap probes. Black on Base, Red on Emitter.
- Test 4 (Base-Collector Reverse): Keep Black on Base, move Red to Collector.
- Test 5 (Collector-Emitter Check 1): Place Red on Collector, Black on Emitter.
- Test 6 (Collector-Emitter Check 2): Swap probes. Black on Collector, Red on Emitter.
Expected Readings: Good vs. Bad Junction Values
A silicon BJT junction drops approximately 0.6V when forward-biased. Germanium transistors (rare today, mostly found in vintage audio gear) will drop around 0.2V to 0.3V. The table below outlines exactly what your DMM display should show for a healthy modern silicon NPN transistor.
| Test Point (Red → Black) | Junction State | Expected 'Good' Reading | Reading Indicates 'Bad' (Failed) |
|---|---|---|---|
| Base → Emitter | Forward Biased | 0.500V – 0.800V | 0.000V (Short) or OL (Open) |
| Base → Collector | Forward Biased | 0.500V – 0.800V | 0.000V (Short) or OL (Open) |
| Emitter → Base | Reverse Biased | OL (Over Limit) | Any numeric voltage (Leaky/Short) |
| Collector → Base | Reverse Biased | OL (Over Limit) | Any numeric voltage (Leaky/Short) |
| Collector → Emitter | Blocking | OL (Over Limit) | Any numeric voltage (Punch-through) |
| Emitter → Collector | Blocking | OL (Over Limit) | Any numeric voltage (Punch-through) |
For a comprehensive breakdown of BJT internal physics and junction behaviors, refer to the bipolar transistor tutorials at Electronics-Tutorials.ws. Note that the Base-Collector junction will sometimes read slightly lower (e.g., 0.580V) than the Base-Emitter junction (e.g., 0.620V) due to differences in doping concentrations during manufacturing. This is normal.
Why In-Circuit Testing Lies to You (And How to Fix It)
The single most common mistake makers and technicians make is attempting to test a transistor while it is still soldered into the PCB. In-circuit testing is inherently unreliable for BJTs.
When a transistor is soldered to a board, the Base, Collector, and Emitter are connected to surrounding passive components. Bleed resistors, snubber networks, and biasing voltage dividers create parallel electrical paths. If you place your DMM probes across the Base and Emitter in-circuit, you are not just measuring the semiconductor junction; you are measuring the junction in parallel with a 4.7kΩ biasing resistor.
According to Fluke's official testing guidelines, parallel resistance will drastically lower the forward voltage reading. A perfectly good transistor might read 0.250V in-circuit, leading you to falsely conclude it has a leaky junction. Furthermore, if the circuit contains large electrolytic capacitors, the DMM's diode test current will charge the capacitor, causing the voltage reading to continuously climb or fluctuate, rendering the measurement useless.
The Fix: You must desolder at least two of the three transistor legs (usually the Base and Collector) to lift them off the PCB pads before testing. If you are troubleshooting a blown power supply, the main switching transistor is almost certainly dead if the board fuse is blown; just desolder it and test it on the bench.
The Final Decision Path: Replace or Keep?
Use the decision matrix below to determine your next step based on your DMM readings. Do not attempt to 'rehabilitate' a failing transistor by heating it or re-tinning the legs; semiconductor lattice damage from thermal runaway or overvoltage is permanent.
| Observed Symptom / DMM Reading | Diagnosis | Action Required | Concrete Replacement Pick (2026 Standard) |
|---|---|---|---|
| BE & BC read 0.5V-0.8V; all others read OL. | Healthy Silicon BJT | Keep. Reinstall or return to inventory. | N/A |
| CE reads 0.00V to 0.05V (with audible continuity beep). | Collector-Emitter Punch-Through (Dead Short) | Discard. Check surrounding driver resistors for collateral damage. | ON Semi 2N3904G (Small Signal) or ST TIP31C (Power) |
| All junctions read OL in both directions. | Internal Bond Wire Severed (Open Circuit) | Discard. Usually caused by severe overcurrent. | ON Semi 2N3906G (PNP equivalent) |
| BE or BC reads between 0.100V and 0.350V (Silicon). | Leaky Junction (Thermal Damage) | Discard. Will cause severe thermal runaway if powered. | Diodes Inc 2N2222A (Metal Can) or BC547B |
| Readings fluctuate wildly or read ~0.2V in-circuit. | Parallel PCB Path Interference | Desolder 2 legs and re-test on the bench. | N/A (Re-test before replacing) |
Final Procurement Rule: If your diagnostic path terminates in a failed component, do not waste time hunting for the exact OEM part number if it is an obsolete or proprietary house-brand transistor. For 95% of general-purpose small-signal NPN applications (under 40V and 200mA), standardize your bench inventory on the ON Semiconductor 2N3904G (TO-92 package). For PNP circuits, stock the 2N3906. If you are repairing audio amplifiers, motor controllers, or linear power supplies requiring TO-220 power packages, default to the STMicroelectronics TIP31C (NPN, 100V/3A) and TIP32C (PNP). These four part numbers cover the vast majority of hobbyist and bench repair scenarios, ensuring you always have a verified, high-quality replacement ready to solder.
