MicroNote #108 By Mel Clark and Kent Walters Determining Clamping Voltage Levels for a Range of Pulse Currents Clamping voltage (VC) is specified only at the maximum limit on most silicon transient voltage suppressor (TVS) data sheets. Often the designer needs to determine the VC at some intermediate level between breakdown voltage (V(BR)) and maximum VC. The curve based on surge test data has a more-shallow slope than the curve interpolated through calculation. This indicates that the devices are conservatively rated and the formula given is adequate for interpolating intermediate values of VC for a fractional part of IPP. The value can be calculated with the data sheet parameters using the formula: The linear relationship between IP and VC can be applied in determining greater IPP ratings for applications requiring lower than normal values of VC. In the equation above, insert the desired value for VC and solve for the higher IPP value. This often requires upgrading to a higher peak pulse power (PPP) rated device. VC = (IP/IPP)(VC max – V(BR) max) + V(BR) max Where: IP = actual test pulse current IPP = maximum rated peak pulse current VC = clamping voltage at IP VC max = maximum specified clamping voltage V(BR) = upper limit of breakdown voltage 24 Clamping Voltage - Volts Based on previous data, a linear increase in VC can be assumed between V(BR) and VC max for this formula. The VC versus IP relationship for the SMCJ15A for a 1.5 kW TVS between V(BR) and VC as calculated by this method is shown in Figure 1. Results are as expected. This calculation assumes the TVS to be at the V(BR) upper limit, hence it would be conservative for most of the distribution. Note that when IP equals IPP, VC equals VC max. SMCJ 15A Clamping Voltage vs. % Peak Current 22 20 d late lcu Ca l Typica ured Meas 18 16 If only V(BR) min is listed on the data sheet, V(BR) max can be approximated. For “A” suffix parts, multiply V(BR) min by 1.20 and for non-suffix parts, multiply by 1.25 to obtain V(BR) max. Figure 1 An example of a calculated curve compared to one derived from test measurements (Figure 1) illustrates the feasibility and conservative aspects of this method. Surge tests were performed on a 20 piece sample at 25 degrees C with a 10/1000 µs waveform. Corporate Applications Engineering Department, 480-941-6300, Ext 433 or 524 www. .com 1 mA 25% 50% Percent of IPP 75% 100% 1