Powering car electronics with no switching noise and 99.9% efficiency: Page 5 of 7

August 20, 2020 //By David Megaw, ADI
Powering car electronics with no switching noise and 99.9% efficiency
Powering automotive electronics systems can be challenging due to requirements of high reliability while contending with a relatively unstable battery voltage. This article describes a solution that keeps voltage stable – at lowest power dissipation and without the need for fuses.

The switching operation of a buck-boost regulator depends on the relationship between the input and output voltages. If the input is significantly higher than the output, the boost top switch turns on continuously while the buck power stage steps down the input. Similarly, when the input is significantly lower than the output, the buck top switch turns on continuously while the boost power stage steps up the output. When the input and output are roughly equal (within 10% to 25%), the buck and boost power stages switch simultaneously in an interleaved fashion. In this way, efficiency is maximized in the various switching regions (buck, buck-boost, boost) by limiting switching to only the MOSFETs required for regulation for input voltages above, roughly equal to, or below the output, respectively.

ISO 16750-2 Solution Summary

Figure 3 summarizes the various solutions to addressing the load dump, reverse input voltage, superimposed alternating voltage, and engine starting profile tests along with the pros and cons of each approach. Several key takeaways begin to emerge:

  • A series N-channel MOSFET with its drain facing the input is extremely desirable because it can be used to limit current and disconnect the output whether it’s used as a switch (for example, in a buck power stage) or linearly controlled (for example, in a surge stopper).
  • In the case of reverse input protection and superimposed alternating voltage, using an N-channel MOSFET as the rectifying element (source facing the input) significantly reduces power loss and voltage drop compared to a Schottky diode.
  • A switch-mode power supply is preferable to a linear regulator because it alleviates the reliability concerns and output current restrictions that come with operating within the SOA of the power device. It can regulate at input voltage extremes indefinitely, whereas linear regulator and passive solutions have inherent time limitations that complicate design.
  • A boost regulator may or may not be necessary depending on the severity classification of the starting profile and the details of the ECU (what’s the highest voltage it must provide).

If boost regulation is called for, then a 4-switch buck-boost regulator combines the above desirable traits into a single part. It can efficiently regulate through severe undervoltage and overvoltage transients at high current levels for extended periods of time. This makes it the most robust and straightforward approach, from an application’s standpoint, despite the increased design complexity. Nevertheless, a typical 4-switch buck-boost regulator has some drawbacks. For one, reverse battery protection is not naturally provided and must be addressed with additional circuitry.

The primary shortcoming of the 4-switch buck-boost regulator is that it spends much of its operational life in the lower efficiency, noisier buck-boost switching region. When the input voltage is nearly equal to the output (VIN ~ VOUT) all four N-channel MOSFETs are actively switching to maintain regulation. Efficiency drops as a result of increased switching losses and application of maximum gate drive current. Radiated and conducted EMI performance suffer in this region as both the buck and boost power-stage hot loops are active and the regulator input and output currents are discontinuous.

Design category: 

Vous êtes certain ?

Si vous désactivez les cookies, vous ne pouvez plus naviguer sur le site.

Vous allez être rediriger vers Google.