Balance current to achieve the best performance of automotive-grade smart drives-EEWORLD

Collect

Author: by Giusy Gambino, Marcello Vecchio, Filippo Scrimizzi STMicroelectronics, Catania, Italy

When doing distributed intelligence design in automotive power management systems, for intelligent power switches, it is crucial to ensure whether the protection mechanism is truly intelligent, especially in scenarios involving multi-channel drivers, because even a slight current imbalance or accident Load short circuit will affect the protection effect.

Smart drivers play a key role in managing and distributing automotive battery packs to various components (ECUs, motors, lights, sensors, etc.). These multi-channel drivers simultaneously control different electrical loads, e.g. resistive actuators, inductive actuators and capacitive actuators. Balanced current flow across all channels is critical for the drive to function properly and ensure the vehicle operates properly and efficiently. In circuit layout, any unforeseen circumstances such as minor current imbalances that cause current to concentrate through specific metal paths, damaged or failed loads, and improper wiring can cause current concentration effects in local circuits. Current imbalance will cause the chip to overheat and accumulate hot spots, eventually damaging or burning components.

Although thermal simulation experiments and preventive measures have been taken, the implementation of the intelligent protection mechanism still needs to be checked and verified, which can help identify potential problems that may affect the timeliness of intervention.

Thermal detection in smart switches

High-side switches need to handle high currents in compact packages with very little space, and current balancing is an important factor in efficiently managing heat. Smart power switches are often installed in enclosed areas with poor ventilation and heat dissipation, making thermal management even more important.

Therefore, the intelligent performance of the protection mechanism depends on embedded thermal diagnostic functions. These diagnostic functions based on thermal detection and protection mechanisms are used to monitor the temperature of the drive and perform protective actions when the temperature exceeds a preset threshold. Accuracy is a difficult problem faced by temperature measurement technology, because the current balance of multi-channel drivers has a great impact on temperature measurement accuracy.

A sudden increase in local current density or short-circuit situation is a problem that designers are very concerned about. These two phenomena will produce scattered hot spots, causing sudden heat accumulation effects and causing the temperature to rise suddenly. These conditions can lead to overheating and component failure, which can be costly to repair.

To prevent thermal shock from damaging components, protection circuitry is designed to limit current and keep the power MOSFET within the safe operating area (SOA) until the thermal shutdown function is triggered, shutting down the driver. However, this type of protection may create physical stress on the power device surface. To meet power surge requirements and process tolerances, the current limit value needs to be set higher. However, when driving a short-circuit load, a higher current limit value will cause the temperature of the chip surface to rise rapidly. Sudden temperature changes will produce huge thermal gradients on the chip surface, resulting in thermomechanical stress and affecting the reliability of the device.

The solution of VIPower M0-9 is to integrate a temperature sensor in the low temperature area and high temperature area of the high-side driver (as shown in Figure 1).

Figure 1: Schematic of a smart switch with different temperature sensors

The temperature sensor uses polysilicon diode manufacturing technology because the temperature coefficient of polysilicon diodes remains very linear over the entire operating temperature range. The low temperature sensor is placed in the low temperature area inside the driver close to the controller side, while the high temperature sensor is located in the power stage area, which is the area with the highest temperature inside the driver.

This dual-sensor technology can limit the temperature rise of the driver because when the temperature reaches the over-temperature threshold, or the dynamic temperature difference between the two sensors reaches the threshold, the thermal protection is triggered. Once the overheating fault disappears, the smart switch reactivates when the temperature drops to the recovery value.

This approach helps reduce thermal fatigue caused by thermomechanical stress on the switch. Thermomechanical stress can increase over time, resulting in reduced switching performance and reliability.

Thermal map

In addition to thermal simulation experiments and prevention methods, infrared (IR) thermal imaging technology is also an effective technology to obtain driver thermal maps, which can provide designers with a comprehensive understanding of heat distribution within integrated circuits and reveal all potential risk factors.

In order to evaluate the protective effect of intelligent protection circuits in harsh automotive environments, the heat distribution within the driver must be analyzed under two different application scenarios and harsh short-circuit conditions:

· Terminal short circuit (TSC)

· Load short circuit (LSC)

A terminal short is a condition when there is a low-resistance connection between the terminals of a component or device, as shown in Figure 2.

Figure 2: Temperature measurement test circuit under TSC conditions

On the other hand, a short-circuit condition in the load occurs when there is an inductive path between the load and the power supply, causing a sudden surge in current (Figure 3).

Figure 3: Temperature measurement test circuit under LSC conditions

The test conditions are as follows:

· Tamb = 25 °C

· Vbat = 14V

· When thermal imaging, Ton = 1 ms

· Ton = 300 ms when capturing temperature of thermal sensor and hot spot

· TSC conditions: RSUPPLY = 10 mΩ, RSHORT = 10 mΩ

· LSC conditions: RSUPPLY = 10 mΩ, LSHORT = 5 µH, RSHORT = 100 mΩ

in,

Tamb is the ambient temperature

Vbat DC battery voltage

Ton is the short circuit duration

RSUPPLY is the internal resistance of the battery

RSHORT is the short circuit resistance

LSHORT is the short circuit inductance

To generate the thermal map, we used an infrared camera to capture the infrared light emitted by each location and then converted it into a temperature value. Calibration is an essential and important process in order to ensure that a specific color is converted to the correct temperature value. The process is to compare different colors captured by the sensor with known temperature values, analyzing specific thermal parameters and their trends as temperature increases. By analyzing these parameters, the calibration process ensures that the heat map accurately reflects the temperature distribution of the area being scanned.

To calibrate the infrared camera sensor, the forward voltage (VF) of the MOSFET body-drain diode is chosen because it has a linear relationship with temperature. However, the diode needs to be pre-calibrated to accurately determine its temperature coefficient. The temperature coefficient of a diode can be determined by measuring the voltage VF at a constant forward current (IF) while varying the temperature from 25°C to 100°C. To prevent temperature rise due to current and its associated power dissipation, the value of IF should be in the range of 10mA to 20mA.

Use the VF values collected under different temperature conditions to perform linear interpolation and mathematical fitting calculations to obtain the temperature coefficient of the diode, as shown in Figure 4.

Figure 4: Pre-calibration of MOSFET body-drain diode

Calculate (1) using the following formula:

in:

Dt is the temperature change;

DVF is the forward voltage change;

K is the temperature coefficient of the diode.

To create a heat map, each temperature point is first photographed at 1ms intervals using an infrared imaging sensor. Once all spots on the chip have been photographed (which takes about 3,000 seconds), specialized software generates a heat map depicting the temperature of each spot based on the minimum spatial resolution of the infrared sensor. By placing the heat map on top of the chip map, the hottest hot spots in the work area can be identified. When current flows through the device, the coordinates of these hot spots can be determined.

Figure 5 shows the heat map of the VND9012AJ dual-channel smart switch under TSC conditions.

Figure 5: Heatmap of VND9012AJ channel under TSC conditions

Thermal mapping method uses different colors to describe the temperature distribution of each channel of the driver in the temperature range of 25°C to 150°C. This is an important method to detect any overheating areas and ensure that the driver operates within a safe temperature. By providing a heat map of each channel under different operating conditions, the heat map test method can verify the operating reliability of the drive without increasing the temperature to a maximum threshold.

In order to find hot spots and monitor the temperature changes of the high-temperature sensor and low-temperature sensor, and verify the effect of the thermal shutdown mechanism, extending the short-circuit duration to 300ms must be considered in the experiment.

Figure 6 shows the temperature change observed for the VND9012AJ at TSC.

Figure 6: Temperature variation of two sensors under TSC conditions

The image above shows that the high temperature sensor detects hot spots in both channels of the VND9012AJ with maximum temperatures in the range of 150 °C.

Figure 7 shows the heat map of the VND9012AJ under LSC conditions.