Download this article in .PDF format
This file type includes high resolution graphics and schematics when applicapable.

Initial studies by James Janky, at ITT Semiconductors Applications Engineering Department in September, 1966, show that the rf breakdown voltage of a transistor is higher than its dc value (in a CEO or CEX mode) and can approach its BVCBO value at high frequencies. Attempts to measure this value may be accomplished by applying rf directly to the collector and measuring the result. The result is very dependent on the circuit in the base and the capacitive current through COB. A circuit, which leaves the base open at rf, tends to approximate the typical condition of an amplifier circuit but will give much lower breakdown values than if the Cob current is shorted at the base (BVCES) which will measure closer to BVCBO at rf.

Fig. 6. Rf breakdown test circuit.

A simple, straightforward rf breakdown test circuit is shown in Fig. 6. The transistor is normally in the OFF condition; a reverse bias is applied to the collector; and rf peak voltages are applied to the unit until breakdown occurs.

The BVCEO of the 2N3375 with 40 V bias is shown in Fig. 7. Only a 10 percent increase is apparent here. BVCES will be considerably higher.

Fig. 7. Dc vs rf breakdown voltage for transistor 2N3375 operating at 130 Mc with 40-V bias.

Transistor epitaxial thickness and basewidth are important factors because these affect the avalanche multiplication factor. Table 1 gives the range for other device types.

Table 1: Dc and rf breakdown-voltage comparison for typical microwave transistors

Transistor type

DC BVCEO (Volts)

RF BVCEO @100 Mc (Volts)
















Much work must yet be done to determine how the actual operating-circuit breakdown varies. This is because the foregoing fails to account for an rf base drive and the fact that the transistor is going through a transition—from a charge-controlled (ON) to an (OFF) state where the charge-storage time may be a considerable portion of an rf cycle.