Disc Thyristor Testing: Clamping Is Essential, Not Optional

With the correct clamp installed, the disc thyristor gives repeatable junction readings and can be tested under a controlled load. Without compression, the meter may show an open circuit, unstable resistance, or a reading that changes every time you touch the package. That is not automatically a failed semiconductor.

Start here: identify the package construction and obtain its specified mounting force. A low-current multimeter check cannot prove forward conduction or rated blocking performance, and a bench vise cannot apply the calibrated, uniform pressure the package requires.

Recognize a false failure indication

A pressure-contact disc, often called a puck package, contains a semiconductor wafer between heavy conductive faces. Some designs use a bonded connection on one side and a pressure contact on the other. Other internal arrangements also exist, so classify the actual device before interpreting an open-circuit reading.

Symptom	Likely cause to check first
Main terminals read open with no clamp	The internal pressure contact is not established.
Resistance changes when hand pressure is applied	Contact pressure or contact uniformity is controlling the reading.
Readings vary between repeated setups	The device is being loaded unevenly, the faces are contaminated, or the meter is operating near an unstable contact condition.
Gate-to-cathode resistance is measurable, but the main path will not conduct	The meter has checked only the gate junction; it has not proved triggering or main-die conduction.
The device triggers at low current but fails in service	The test did not reproduce the required gate drive, load current, thermal path, or blocking stress.
A previously functional puck fails after uncontrolled clamping	Excess or nonuniform force may have distorted the package or fractured the die.

Compare devices only after controlling clamp force, orientation, temperature, test current, and connection sequence. Otherwise, differences between readings may belong to the fixture rather than the thyristors.

Confirm that the package requires compression

Inspect the datasheet or mounting instructions for a specified clamping force and a prescribed clamp or heatsink arrangement. A published force confirms that compression is part of the electrical and thermal interface. Read the value for the exact device; package diameter alone does not establish it.

The construction explains the symptom. Pressure brings the conductive faces into uniform contact with the wafer and establishes the intended current path. It also creates the thermal path from the wafer through the pole faces to the heatsinks. With too little pressure, localized or incomplete contact raises resistance and produces erratic readings. With excessive or tilted pressure, the assembly can bend and crack the wafer.

Examples of large disc devices have used forces from about 2,000 lb to 15,000 lb. Those figures show the scale of the mechanical load; they are not selectable settings for an unidentified device. One dismantled Eupec T1189, described as a 1,200 A SCR, had a 48 mm face and was 26 mm thick. Those dimensions and that rating do not define its required clamp force.

Decide which property you need to test

Separate the test into three questions. Each needs different equipment and proves a different part of the device.

1. Gate junction: Measure between gate and cathode with the device isolated. Field measurements on these devices may range from a few ohms to about 100 ohms, but resistance mode, polarity, temperature, internal gate structure, and device design affect the result. Use the exact datasheet limits or compare against a known device of the same type under identical conditions.
2. Main-path triggering: Apply a current-limited DC source, a controlled load, and a gate pulse that meets the device requirement. A multimeter supplies too little main current to demonstrate full conduction or latching.
3. Forward and reverse blocking: Use controlled insulation or leakage-test equipment within the device rating. A resistance measurement at meter voltage does not prove blocking capability at working voltage.

If the only requirement is sorting obviously damaged units, begin with visual inspection and repeatable gate-junction comparisons. If the requirement is release for service, test blocking, triggering, on-state behavior, insulation, and the mechanical interface according to the manufacturer’s procedure.

Obtain the specified clamp force

Do not estimate force from handle torque or how hard the clamp feels. Friction in screw threads and contact surfaces makes hand torque a poor indication of axial force. A vise also concentrates load through its jaws and can introduce tilt across the pole faces.

Retrieve four items before loading the device:

1. The minimum, nominal, and maximum mounting force, if all three are published.
2. The required contact or pole-face area and permitted surface condition.
3. The clamp geometry, including any spring stack, pressure plate, centering feature, or indicator.
4. The tightening sequence and the permitted parallelism or flatness limits.

Use a purpose-made heatsink and clamp assembly that centers the puck and spreads force over its specified faces. Spring elements matter because they maintain force as the assembly heats, cools, and settles. Calibrated fixtures may indicate force through spring deflection or another defined measurement.

Too little force causes inadequate electrical and thermal contact. Too much force can distort the package and fracture the die. A cracked wafer may still pass a light bench check and then fail without warning under current or thermal cycling. Replacing the meter or increasing test voltage will not correct either mechanical condition.

Reject uncontrolled fixtures before electrical testing

Inspect both pole faces and the fixture. Remove foreign material using the handling method permitted for the device. Check that the puck sits squarely and that conductors, gate leads, and insulating parts cannot apply side load.

Do not place the device between narrow vise jaws, even with insulating covers. Electrical insulation does not correct force concentration, unknown load, or misalignment. Hand pressure may make a continuity reading appear, but it is useful only as a clue that the package uses pressure contact. It is not a qualification test.

Confirm that the electrical circuit is isolated and discharged before connecting test equipment. Use current limiting appropriate to the intended test, and contain the device and connections against fault energy. The higher the test voltage and available current, the less useful an improvised fixture becomes.

Run the tests in increasing order of stress

Use this sequence so a simple wiring or contact problem does not become a destructive high-energy test.

1. Identify the terminals. Confirm anode, cathode, and gate from the device documentation. Do not infer polarity from an unfamiliar puck’s physical appearance.
2. Inspect the package. Stop for cracked ceramic, damaged seals, burned faces, displaced gate hardware, or evidence of arcing.
3. Install the calibrated clamp. Center the device, load it through the specified area, and bring the assembly to the published force using the prescribed sequence.
4. Check gate to cathode. Record the meter mode, polarity, reading, temperature, and clamp condition. Compare only with datasheet criteria or an identical known-good unit tested the same way.
5. Check unintended terminal shorts. A near-short across the main terminals without gate excitation indicates a different failure from an open, unclamped pressure contact.
6. Perform the controlled conduction test. One successful field arrangement used a 110 V DC source with halogen lamps or a resistor load bank to draw about 20 A. Use those values only when they remain within the exact device, fixture, load, and laboratory ratings. Some large devices may require gate current approaching 1 A; obtain the actual trigger requirement from the datasheet rather than raising gate current by trial.
7. Perform blocking or leakage testing. Connect the insulation tester before energizing it. Keep voltage, polarity, duration, and leakage limits within the device documentation.
8. Discharge the circuit. Verify zero stored voltage before touching or rearranging connections.

Connection order matters during blocking tests. Some test instruments contain an internal capacitor. Connecting a pre-energized tester can dump stored charge into the device, unintentionally trigger it through the structure, and create a current rise beyond what the thyristor can tolerate. The same mechanism makes triggering into an already energized, low-impedance circuit hazardous unless the circuit limits current rise.

Verify the resolving branch

A corrected clamp should change the result from unstable to repeatable. Prove that mechanically before accepting the electrical result.

1. Record the specified force and the fixture indication used to reach it.
2. Take the same low-energy reading at least twice without changing meter mode or lead polarity.
3. Remove and reinstall the device, restore the same force, and repeat the measurement.
4. Compare the result with an identical known-good device under the same clamp, temperature, and connections.
5. Run the controlled conduction test and verify that gate removal, load-current interruption, and reapplication behave as expected for an SCR circuit.
6. Run the specified forward and reverse blocking checks, recording leakage rather than only a pass/fail indication.
7. After testing, inspect the package and fixture again for marks, tilt, ceramic damage, or abnormal heating.

If compression restores only continuity but the functional or blocking test still fails, the clamp was one problem, not the whole diagnosis. Keep the unit out of service until its trigger, leakage, and on-state results meet the limits for the exact part.

FAQ