2 Design Methodology

The circuit design in Figure 1-1 was used to have an output of 1 V over 24.9 Ω (40.16 mA). High-precision resistors and capacitors were used to reduce sources of error for the circuit, as well as reliable supply voltages. Four test cases were identified in order to quantify the difference in the junction-to-ambient thermal resistance (Θja) of the HVSSOP and emulated SOIC packages:

In each of these test cases, the ambient temperature was swept from 20 °C to 80 °C and the supply current was recorded. Since the relationship between supply current and internal junction temperature was calculated through simulation results, the measured supply current could be immediately converted to a respective junction temperature. Next, the calculated temperatures were used to find the estimated Θja of both package types.

To demonstrate how to use the thermal resistance metrics, an estimation using the verified values in the OPA2828 data sheet can be made. The power dissipation of the circuit needed to be calculated, in order to convert thermal resistance to temperature change. The following equation shows the power calculation of the power dissipation on the amplifier for the cases listed in Table 2-1, with Vs being the respective supply the output is using.

For both ±5 V and ±15 V supplies, the estimated amplifier power dissipation was calculated to be 0.275 W and 0.9 W respectively, when using a high temperature quiescent current of 11 mA. Using these power values, the estimated heating for the HVSSOP package can be determined by multiplying the Θja (49.9 °C/W) to the power values.

This can be directly compared to a package that doesn't include a thermal pad. The SOIC package has a Θja value in the datasheet for the single OPA828 (121.5 °C/W) which can be multiplied to the same power calculations.

This estimation shows the benefit of using a thermal pad, as a SOIC package would be 19.69 °C and 64.44 °C hotter than the HVSSOP package for the ±5 V and ±15 V supply voltage cases respectively.