Power Electronics
Driving Automotive Power Supplies to Higher Frequencies

Driving Automotive Power Supplies to Higher Frequencies

If protected against load dumps and other transients, switching regulators with relatively low voltage ratings can be configured to operate efficiently at high switching frequencies.

The increasing sophistication of electronic systems in automobiles presents unique challenges and opportunities for power system designers. Most automotive modules need low voltages like 3.3 V and 5 V. The voltage conversion from battery to such lower voltages using linear regulators means a significant power dissipation. High power dissipation makes the thermal management difficult and expensive. The higher power requirement of faster processors and ASICs have steered the power-conversion method from simple, low-cost, inefficient linear regulators to the more complex but higher-efficiency switching converters.

The size of a switching converter depends on the switching frequency. The passive components like power inductors and capacitors become smaller with higher switching frequency. These high-efficiency converters reduce power dissipation by eliminating bulky and expensive heatsinks. The entire power supply can shrink significantly when using switching converters. These advantages make the switching converter an obvious choice for power management of body electronics, infotainment and engine-control modules.

Selecting the proper switching frequency is critical because a switching converter poses its own set of problems. For example, electromagnetic radiation caused by the switching converter can interfere with other electronics. The AM radio receiver is sensitive to the interference between 530 kHz and 1700 kHz. The switching converter's fundamental switching frequency and third harmonics can contribute significant interference to other electronics around the power supply. The even harmonics cancel each other out, and the odd harmonics higher than fifth order typically have little energy, which is usually easy to filter out. Selecting a switching frequency greater than 1800 kHz eliminates fundamental and other harmonic interference from the AM frequency band.

However, a high switching frequency increases the power loss, partly offsetting the advantage of using a switching converter. The switching losses increase with higher input voltages, as they are proportional to the square of the operating voltage. The automotive environment typically demands high-voltage processes (40 V or higher) for power controller ICs to withstand overvoltage transients such as load dump. High-voltage processes use relatively larger geometries and higher gate thicknesses. The resulting longer channel lengths lead to longer propagation delays. Obviously, high-voltage processes are inherently slow and could be very inefficient as the transition losses increase due to longer rise and fall times of the switch.

Certain fabrication processes at Maxim are suited for extremely high-speed converters at moderate voltage levels. A recent example is the MAX5073, a dual-output 2.2-MHz buck or boost converter that can tolerate up to a 23-V input. An effective switching frequency of 4.4 MHz is achieved by using ripple phase operation. The switching converters are supposed to be immune to the interference present on the power source. As far as automotive applications are concerned, high-voltage controllers are not an absolute necessity when designing these switching converters. This article describes the most common automotive power disturbances and ways to protect the low-voltage electronics from them.

Power Line Overvoltage Stress Conditions

Overvoltage (OV) protection devices are used to isolate and protect electronic circuitry from excessive voltage conditions that are conducted via electrical connections to the automotive electrical system, in particular those connections to the main voltage supply. The ability to withstand conducted disturbances is generally known as conducted immunity (CI).

Automotive manufacturers and standards organizations specify various test methods to evaluate the CI of electronic components and systems. While each automotive OEM tends to have specific requirements, the ISO7637 standard provides the basis for many of these. The following is not meant to be a comprehensive description of all CI requirements, but rather a brief summary of the typical OV conditions relevant to automotive applications.

Steady-State OV Conditions

Certain OV conditions are of long enough duration to be considered steady-state from an electronic circuit standpoint. For example, any OV condition that persists comparatively longer than the thermal time constant of an electronic device can be considered steady-state. In these situations, the continuous power dissipation and resulting temperature rise are of primary concern. Conditions that can be considered steady-state include a failed alternator regulator, a double-battery jump-start and reverse battery connections. The following is a brief description of these conditions.

The output of the alternator is regulated with respect to speed, load and temperature by regulating the magnitude of the current in the field winding. This function is typically provided by an electronic circuit (that is, voltage regulator) that pulse width modulates the field winding to achieve a constant, regulated output voltage from the alternator. The output set point of the voltage regulator is typically about 13.5 V. It is possible for the alternator voltage regulator to fail in such a way as to provide full field current irrespective of load or output-voltage conditions. When this happens, voltage in excess of the typical 13.5 V may be applied to the entire system, with the actual voltage level being dependent on vehicle speed, loading and other conditions. The typical OEM failed-regulator test requirement is on the order of 18 V for 1 hour. Most systems are required to operate under these conditions, although certain comfort or convenience functions are allowed to deviate.

Another OV condition that is effectively steady-state is the double-battery jump-start. This typically occurs when a tow truck or service station uses a 24-V system to jump-start a disabled vehicle or charge a dead battery. The typical OEM double-battery test requirement is on the order of 24 V for 2 minutes. Certain engine management and safety-related systems are required to operate under these conditions.

It is possible for a steady-state reverse potential to be applied to the vehicle electrical system during manufacturing or service. In general, most systems are required to survive but not operate under this condition. The typical requirement is -14 V for 1 minute. This can be a challenging requirement for systems with high current or low-voltage drop input requirements.

Transient OV Conditions

The majority of transient OV conditions in an automobile are due to switching inductive loads. Examples of such loads include the starter motor, fuel pump, window motors, relay coils, solenoids, ignition components and distributed circuit inductances. Whenever current is interrupted in these inductive loads, an OV pulse will typically be produced. Due to the amplitudes and durations involved, filters, metal oxide varistors (MOVs) or transient voltage suppressors are required for suppressing these types of OV transients. A description of pulses based on the ISO 7637 standard, shown in the table, is as follows:

Pulse 1 is a negative-going repetitive pulse ranging from -80 V to -150 V in amplitude with a duration of 1 ms to 140 ms. The source impedance is typically on the order of 5 Ω to 25 Ω. Pulse 2 is a positive-going repetitive pulse ranging from 75 V to 150 V with a typical duration of 50 s. The source impedance is typically 2 Ω to 10 Ω. Pulse 3a is a series of negative pulses that are on the order of -150 V and 100 ns. Pulse 3b is a series of positive pulses on the order of 100 V and 100 ns. The impedance of the signal source is typically 50 Ω.

Pulse 5, also known as a load dump, is a condition that occurs when an alternator is supplying high current to a discharged battery and the battery is suddenly disconnected. Since the alternator is a magnetic device, the sudden reduction in stator current induces a high voltage at the alternator output to maintain the energy of the system. The duration of this transient is based on the electrical time constant of the alternator field circuit and regulator response time.

Due to conditions described earlier, the battery voltage cannot be fed directly to the low-voltage, high-performance switching converters. Transient voltage suppressors like MOVs and bypass capacitors, followed by traditional input voltage limiters are used. These circuits are simple and built around the p-channel MOSFET (Fig. 1). The p-channel MOSFET must be rated at 50 V or 100 V, depending on the voltage transients expected at the input voltage (VBAT). The 12-V Zener diode Z1 limits the gate-to-source voltage of the MOSFET below the VGSMAX. The MOSFET operates in saturation when VBAT is below the breakdown voltage of the Zener Z2. During the input voltage transient, the MOSFET blocks the voltages higher than the Z2 breakdown voltages. The disadvantage is an expensive p-channel MOSFET and too many components around it.

Another approach is to use the npn transistor with collector connected to the “plus” terminal of the battery and the emitter to the downstream electronics. A clamping device (VZ) then clamps the npn base voltage, which regulates the emitter voltage at VBE below the VZ. It's a lower-cost but inefficient (PLOSS = IIN × VBE) solution. The drop also increases minimum operating battery voltage, which is especially critical through cold crank. The third possible solution is using an n-channel MOSFET. N-channel MOSFETs are widely available, cheaper and may be used as a blocking element. However, the gate drive is more complicated because it needs to be higher than the source voltage. The MAX6398 includes an internal charge pump to drive an external n-channel MOSFET (Fig. 2).

Fig. 2 shows the implementation of an n-channel MOSFET switch as a blocking device. The MOSFET can be completely turned off as soon as VBAT increases above the set limit during the load dump. The MOSFET remains off as long as the VBAT remains above the set voltage. The MAX6398 controls the n-channel MOSFET to protect the high-performance power supply from the automotive OV events, such as double-battery jump-starting and load dumping. The MAX5073, a 2-MHz, two-output compact buck converter is connected downstream.

As depicted in Fig. 3, the MAX6398 effectively blocks automotive load-dump pulses and regulates the voltage seen by low-voltage, high-performance electronics. The strategy of using a combination of protector and low-voltage, high-frequency power electronics saves space and cost compared to the high-voltage solutions operating at significantly lower frequencies.

Table. Examples of typical OEM conducted immunity requirements.
Pulse type ISO 7637-1 OEM #1 OEM #2 OEM #3 OEM #4 OEM #5 OEM #6 OEM #7
Pulse 1 VP -75 V to -100 V -100 V -100 V -100 V -150 V -100 V -100 V -80 V
TD 2 ms 2 ms 2 ms 2 ms 2 ms 5 ms 50 µs 140 ms
RS 10 Ω 10 Ω 10 Ω 10 Ω 10 Ω 25 Ω 10 Ω 5 Ω
Pulse 2a VP 50.5 V to 63.5 V 163.5 V +50 V +100 V +75 V
+200 V 110 V
TD 50 µs 50 µs 50 µs 50 µs 50 µs
2 ms 5.7 µs
RS 2 Ω 4 Ω 4 Ω 10 Ω 2 Ω
10 Ω 0.24 Ω
Pulse 3a VP -98.5 V to 136.5 V -300 V -150 V -150 V -112 V
-150 V -260 V
TD 100 ns 50 µs 100 ns 100 ns 100 ns
-150 V -260 V
RS 50 Ω 4 Ω 50 Ω 50 Ω 50 Ω
50 Ω 34 Ω
Pulse 3b VP 88.5 V to 113.5 V +100 V +100 V +100 V +75 V
+100 V
TD 100 ns 100 ns 100 ns 100 ns 100 ns
100 ns
RS 50 Ω 50 Ω 50 Ω 50 Ω 50 Ω
50 Ω
Pulse 5a VP 78.5 V to 100.5 V 73.5 V 32 V 113.5 V 82.5 V 80 V

TD 40 ms to 400 ms 150 ms 400 ms 400 ms 250 ms 120 ms

RS 0.5 Ω to 4 Ω 0.5 Ω 0.5 Ω 0.5 Ω 0.5 Ω 2.5 Ω

Hide comments


  • Allowed HTML tags: <em> <strong> <blockquote> <br> <p>

Plain text

  • No HTML tags allowed.
  • Web page addresses and e-mail addresses turn into links automatically.
  • Lines and paragraphs break automatically.