Understand the thermal derating characteristics of PWM ICs to obtain optimal system performance

introduction

The final operating environment of the chip is usually not determined during the IC evaluation stage, and the operating environment of the chip in different applications is also different. Since the operating conditions of the final system often depend on the ventilation and cooling conditions on site, it is very important to understand the temperature derating characteristics of portable and non-portable products. It is with this in mind that the temperature derating characteristic curve is given in the data sheet of pulse width modulation (PWM) controllers with built-in MOSFET. The derating curve shows the power that the chip can dissipate at a specified airflow intensity and ambient temperature.

In order to ensure that the heat in the application environment does not cause the PWM controller to overload, the temperature derating characteristic curve can be used to select the appropriate PWM IC. The temperature derating test box provides an effective way to evaluate the temperature derating performance of PWM chips.

This article explains how to build and calibrate the temperature derating test box, and the test results of two PWM controllers are very close to the actual operating conditions.

Standard thermal properties test

For a given module size (including heat sink) and airflow intensity, the maximum power that the chip can dissipate is related to physical characteristics, so the chip's thermal characteristics can be estimated. However, without a standardized controlled IC test environment, it is difficult to keep the temperature derating characteristics provided by different manufacturers for different packages consistent.

One way to generate standard device derating data is to actually measure the module's temperature characteristics under different airflow intensity and ambient temperature conditions. The temperature derating test box will hold an evaluation (EV) board with a PWM IC installed, and the fan can be adjusted to obtain uniform airflow. The test results will be published in the data sheet and used as a basis for the correct module selection for the specific application.

Test Equipment and Calibration

Build a standard thermal derating test box (Figure 1) that is adjustable to fit the largest EV kit and keep the box clean. The minimum plan size of the box is 1 foot x 1 foot and the maximum height is 3 inches to allow for the installation of fans. Two or more fans can be installed to provide uniform airflow. The box should be made of low thermal conductivity material, such as polycarbonate, polypropylene, or glass epoxy, and should be at least 3.2 mm thick to provide adequate rigidity.


More image (PDF, 5.50MB)
Figure 1. The thermal derating test box surrounds the EV kit and maintains a preset airflow level to ensure repeatability.

Place the box horizontally in the chamber and calibrate the box by measuring the airflow on the left, center, and right sides of the box opening using an average airflow meter (Figure 2) to confirm uniform airflow inside the box. When the module is placed inside the box, ensure that it is at least 2 inches away from the fans. Use a voltage-controlled variable speed fan to adjust the airflow level by increasing or decreasing the voltage.


More image (PDF, 4.57MB)
Figure 2. The airflow meter at the test box opening is used to measure airflow and calibrate the thermal characterization parameters.

Start with a standard evaluation board that can be mounted on a larger board for easy access and testing. The larger board should be at least 1 inch from the bottom of the test box to ensure uniform airflow underneath the board.

Once the test box is inside the oven, verify that the oven airflow does not affect the derating characteristics of the test box inside. If the oven airflow affects the airflow of the test box, place a larger box around the test box to maintain uniform airflow inside the test box. To minimize interference, consider mounting the derating test box with the fan airflow at an angle to the oven airflow.

The test box can be secured with ordinary set screws and adhesives, just make sure they can withstand the temperature inside the oven. Mount the thermocouple in the center of the IC geometry. Be careful when mounting the thermocouple on the IC to ensure that the thermocouple does not act as a heat sink for the chip. It is best to use a small amount of thermal epoxy to mount the thermocouple to the IC.

The output of the thermocouple used to measure the temperature should be electrically floating so that the temperature reading is not affected by any voltage applied to the IC. Thermocouple lead wires can be 30 AWG or smaller and should be wired in a way that is as insulated from airflow as possible. The same considerations apply to the wiring of the power supply and other test boards.

Before actual measurements, the thermal derating test box must be calibrated (Figure 2). Table 1 lists the calibration data when the box is empty; Table 2 shows the calibration data when the evaluation board is installed in the box. Figures 3a and 3b show the test curves, both of which maintain a nearly uniform airflow inside the box.

Once the box is installed, the evaluation board is mounted on the test board and the entire setup is placed in the temperature chamber. The temperature chamber is turned on for a period of time to allow the temperature to reach a specified equilibrium point. The evaluation board is then loaded and run to achieve a stable temperature reading. If the difference between the two temperature readings is no greater than 0.2°C over a 5-minute interval, thermal equilibrium is assumed.

It is important to note that it is difficult to obtain thermal derating parameters without airflow because the natural flow of air is erratic, especially with high-temperature, high-power ICs.

Table 1. Calibration of airflow intensity when the test chamber is empty

	Air Flow (LFM)
Fan Supply (V)	Left	Center	Right
3.3	102	98	91
4	215	207	187
6	339	337	323
8	449	447	445
10	565	585	581
12	685	709	673

Table 2. Airflow calibration when the evaluation board is installed in the test chamber

	Air Flow (LFM)
Fan Supply (V)	Left	Center	Right
3.3	91	90	89
4	189	160	205
6	333	321	325
8	455	445	443
10	561	581	565
12	673	689	671

Figure 3a. The test box was empty and the results of the three airflow intensity levels were measured.


Figure 3b. The evaluation board was placed in the test box and the airflow intensity measurement was repeated to determine the difference with the airflow test results when the box was empty.

Test Results

To verify the performance of the test box, the temperature characteristics of two Maxim PWM ICs ( MAX15035 and MAX8686 ) were tested. The test results are shown in Figure 4a and Figure 4b respectively. The MAX15035 is a 15A step-down regulator with an internal switch, and the MAX8686 is a single-phase/multiphase, step-down DC-DC converter that can provide up to 25A per phase. The MAX15035 evaluation board is a four-layer board with a size of 2.4 inches x 2.4 inches and 2 ounces of copper; the MAX8686 evaluation board is a six-layer board with a size of 3.5 inches x 3.0 inches and 2 ounces of copper. Figure 4a. The relationship between the maximum output current of the MAX15035 and the ambient temperature, obtained by testing the MAX15035 evaluation board in the test box, and the temperature derating curves correspond to three different airflow intensity levels. Figure 4b. The relationship between the maximum output current of the MAX8686 and the ambient temperature. The temperature derating curve of the MAX8686 shows that the controller can achieve the rated current of 25A at an ambient temperature of +50°C and an airflow of 100 LFM. Using actual measurement data, designers can more accurately determine the temperature derating characteristics required for specific applications. Based on these temperature derating characteristic curves, end users can choose the right device for their application. For a given ambient temperature and airflow intensity, the characteristic curve can be used to obtain the maximum power that the PWM IC can provide without exceeding the safe operating range of the chip. If the PWM IC cannot meet its safe operating conditions in the actual application, the user can only meet the design requirements by increasing air circulation for cooling or improving heat dissipation.