Switch-Mode Power Supplies for Beginners: An Efficiency Primer Part 1

May 13, 2008 10:04 AM
By Daniel Wagner, Associate Member of the Technical Staff, Applications, and Roger M. Kenyon, Director, Customer Applications Engineering, Maxim Integrated Products, Chandler, Ariz.

Like the MOSFET, the diode also exhibits switching loss. However, this loss depends to a large extent on the reverse-recovery time (t_RR) of the diode used. Diode switching loss occurs during the transition of the diode from a forward- to reverse-biased condition.

Charge present in the diode due to forward current must be swept out of the junction as reverse voltage is applied to it, resulting in a current spike (I_RRpeak) opposite of forward current. This action results in a V × I power loss, since reverse voltage is applied across the diode during this reverse-recovery event. Fig. 5 presents the simplified plot of a pn diode reverse-recovery period.

When the reverse-recovery characteristics of the diode are known, the following equation is used to estimate the switching power loss (P_SWdiode) of the diode:

P_SWdiode 0.5 x V_REVERSE x I_RRpeak x t_RR2 x f_s

where V_REVERSE is the reverse-bias voltage across the MOSFET, I_RRpeak is the peak reverse-recovery current, t_RR2 is that portion of the reverse-recovery time after I_RR peaks. For the stepdown converter, V_IN reverse-biases the diode after the MOSFET turns on.

To demonstrate the diode equations, Fig. 6 displays the voltage and current waveforms observed for the pn switching diode in a typical stepdown converter. V_IN = 10 V, V_OUT = 3.3 V, measured I_RRpeak = 250 mA, I_OUT = 500 mA, f_S = 1 MHz, t_RR2 = 28 ns and V_F = 0.9 V. Using these values:

P_TOTALdiode = P_SWdiode + P_CONDdiode

(1 - V_OUT / V_IN ) x I_OUT x V_F + 0.5 x V_IN x I_RRpeak x t_RR2 x f_S

= (1 - 0.33) x 0.5 x 0.9 + 0.5 x 10 x 0.25 x 28 x 10^-9 x 1 x 10⁶

= 301.5 mW + 35 mW

= 336.5 mW

This result coincides with the average power loss of 358.7 mW indicated in the lower plot in Fig. 6. Due to the large value of V_F and the lengthy diode conduction interval, and since t_RR is relatively fast, conduction losses (P_SWdiode) dominate the diode.

Given the previous discussion, what can be done to mitigate the losses presented by the switching components of the power supply? The simple answer — choose MOSFETs with low R_DSon and fast switching transients, and diodes with low V_F and fast recovery periods.

Several phenomena directly affect the MOSFET on-state resistance. Naturally, R_DSon increases with die dimensions and drain-source breakdown voltage (V_BRdss), due to an increased amount of semiconductor material in the device. So, oversizing a MOSFET may introduce efficiency penalties that a smaller, properly chosen device might not have.

Further, due to the positive temperature coefficient of the MOSFET, R_DSon increases as die temperature increases. Thus, proper thermal management practices must be followed to keep junction temperatures cool and ensure R_DSon does not grow excessively.

The on-resistance also varies inversely with gate-source bias, up to a point. Therefore, a maximum gate-drive voltage is recommended to achieve lowest R_DSon, with consideration given to the increased gate-drive loss introduced by doing so. However, gate-drive voltage is often not adjustable in SMPS ICs. That is, unless an option allows the user to do so, such as bootstrapping the IC supply, or when an external gate driver is used for an SMPS design.

MOSFET switching losses depend on the capacitances found in the device. Larger capacitances are slower to charge, causing switching transitions to last longer and to dissipate more power. Miller capacitance, commonly termed reverse transfer capacitance (C_RSS) or gate-drain capacitance (C_GD) in MOSFET data sheets, is a major contributor to transition times during switching.

Want to use this article? Click here for options!
© 2007 Penton Media, Inc.