By Kent Walters and Mel Clark
Zener diodes are affected by temperature changes associated with their voltage temperature coefficient (delta VZ). This characteristic delta VZ is usually included in the applicable zener specification data sheet where it is often stated in a percent change in zener voltage per degree centigrade (%/°C), or occasionally in mV/ °C . The delta VZ can be as low as -0.09 %/°C for low voltage zeners, or as high as +0.110 %/°C for high voltage zeners. This is further illustrated in Figure 1.
(INSERT) FIGURE #1: TEMPERATURE COEFFICIENT VS ZENER VOLTAGE
The temperature coefficient is used in predicting voltage-temperature behavior. The zener junction temperature (TJ) will be affected by both the ambient environment and self heating effects from applied zener power (P = VZ x IZ). Self heating is generated by zener package thermal resistance from junction to case (R(theta)JC) or lead (R(theta)JL). This MicroNote will focus primarily on outside ambient temperature (TA) and its effects on zener voltage VZ. Internal heating and heat sink mounting considerations will be further described in MicroNote 204.
Figure 1 shows that a negative to positive temperature coefficient "transition" will occur in the vicinity of five volts for most zener product designs. When well above five volts, the positive delta VZ for avalanche breakdown is generally independent of operating current (IZ). When in the five volt zener device region or below where transition to field emission or "tunneling breakdown" occurs, negative values of delta VZ are observed that are also notably affected by various operating current values. This phenomenon and its typical characteristics are illustrated in Figure 2 for 500 mW zeners. At very low zener voltages where field emission predominates, the delta VZ is again unaffected by operating current. Although the aVZ is provided at test current (IZT), zeners in the "transition" region will usually specify a sufficiently wide tolerance of temperature coefficient for general consideration with typical operating current variations.
(INSERT) FIGURE #2: TRANSITION TEMPERATURE COEFFICIENT VS ZENER VOLTAGE NEAR FIVE VOLTS
Zeners are specified for voltage (VZ) at ambient TA of 25 °C. The voltage change (delta VZ) may be calculated for PN junction temperature change delta TJ from an initial TA of 25 °C using the formula:
delta VZ = delta VZ x VZ x delta TJ / 100 .
For low power levels or pulse test methods, the TJ temperature will approximate TA. However, most zener JEDEC "1N" diode registrations are specified for VZ at dc thermal equilibrium conditions at their specified test current IZT. These dc power test conditions require a period of time in self heating before thermal equilibrium is achieved with internal zener junction temperature TJ above ambient TA. A difference will then exist in zener voltage pulse testing compared to specified thermal equilibrium that often require 20 seconds or longer to achieve. The described voltage shift is easily observed for axial lead package configurations tested at 25% of full rated power, despite heatsinking at typical specified lead lengths of 0.375 inch (10 mm) from the body.
When zener thermal equilibrium conditions have already been included for VZ reference and an external ambient temperature change of delta TA causes further changes in voltage delta VZ, a similar calculation applies. This can simply be expressed as:
delta VZ = delta VZ x VZ x delta TA / 100 .
Additional important considerations to package mounting and thermal resistance on PN junction temperatures and zener voltage will be further described in MICRO NOTE 204.
Copyright Microsemi Corporation 1996
203.HTM