seegongsik
Saved words
F2 · power electronics

The Switching Device: A Real Switch Heats Up Two Ways

Earlier we treated the switch as ideal. A power MOSFET is that switch, but a real switch loses power through a small on-resistance while on, and through the overlap of voltage and current in the instant of turning on and off. Sweep the switching frequency yourself and see how a conduction loss independent of frequency and a switching loss proportional to it set the ceiling on efficiency.

Change the switching frequency and see where the two losses part

The waveform is the voltage Vds across the device and the current Id through it over one switching period. While on, the voltage is almost zero so only a small conduction loss (amber) flows; in the instant of the turn-on and turn-off transitions the voltage and current rise together and a red loss spike appears. Sweep the switching frequency. Raising it shortens each period, so the transition spikes take up a larger share and the switching loss overtakes the conduction loss.

Switching frequency fswfsw = 200 kHz
Loss and efficiency
P_cond = 0.48 W P_sw = 0.96 W
P_tot = 1.44 W · η = 98.2%
The switching frequency is in the middle. The conduction loss and switching loss are comparable. This is the crossing where the two losses part and decides whether to push the frequency higher.
The two losses are comparable

The on cost: on-resistance makes the conduction loss

An ideal switch has zero voltage across it when on, but a real power MOSFET keeps a small on-resistance Ron even when on. So when a current Io flows, a voltage Io·Ron appears across the device and a conduction loss P_cond = Io²·Ron turns to heat. The fraction of each period the switch is on is the duty ratio D, so on average P_cond = Io²·Ron·D. This loss does not depend on how often you switch, only on the current and the on-resistance. So a device with a lower Ron carries the same current with less heat.

The flip cost: overlap makes the switching loss

A transition does not finish in an instant. Turning the device on and off takes a short time tsw, during which the voltage has not yet fully fallen while the current is already rising, so the two overlap. In that instant where voltage and current are both large, P = Vds·Id surges, and each transition loses an energy E_sw = ½·Vds·Io·tsw as heat. Since this loss occurs every time you switch on and off, switching fsw times per second makes the switching loss P_sw = E_sw·fsw, directly proportional to the frequency. Unlike the conduction loss, the faster you switch, the more often it leaks away.

The trade, and why faster and lower-resistance is chased

Why bother raising the frequency at all? As we saw in F1, the switching output is set by the duty ratio, and smoothing the slices is the job of the inductor and capacitor. Raising the frequency achieves the same smoothing with a smaller inductor and capacitor, so the supply becomes small and light. But the price is switching loss. So design is always a trade: raise the frequency to shrink it and the switching loss grows; lower the frequency to cut the loss and the circuit grows. The way to win both at once is a better device. A lower on-resistance cuts the conduction loss, and a faster transition (a shorter tsw) cuts the switching loss. Wide-bandgap devices like gallium nitride (GaN) and silicon carbide (SiC) lower both losses at once, pushing power electronics smaller and more efficient.

ObservePcond = I² Ron D
The conduction loss is the current squared times the on-resistance and the duty ratio.
ChooseEsw = ½ Vds Io ?
The loss of one transition is proportional to the time voltage and current overlap.
Fill inPsw = Esw ?
The switching loss is the energy of one transition times the switching frequency.
On your ownPtot = Pcond ?
The total loss is the sum of conduction and switching loss and sets the ceiling on efficiency.

Back to the first screen

When the switching frequency was low, the loss was almost all the amber conduction loss; raising it made the red transition spikes appear often and the switching loss overtook the conduction loss. The on cost was P_cond = Io²·Ron·D from the on-resistance, and the flip cost was P_sw = E_sw·fsw from the overlap of voltage and current. This is exactly what the switch we assumed ideal in F1 really is. The way to lower both losses is a lower on-resistance and a faster transition, that is, a better device, and the basis of that device was the energy bands we saw on the first screen of this track. The single thread that ran from bands to carriers, to junctions and diodes, to the amplification and switching of the transistor, closes here.

A power MOSFET is F1’s switch but not ideal. While on, a small on-resistance Ron makes a conduction loss Pcond = Io²·Ron·D, which is independent of frequency. In the instant of turning on and off, voltage and current overlap and each transition leaks Esw = ½·Vds·Io·tsw, so the switching loss Psw = Esw·fsw is proportional to frequency. Raising the frequency shrinks the inductor and capacitor but grows the switching loss, so the total loss Pcond + Psw sets the ceiling on efficiency and the heat. The way to lower both is a lower Ron and a faster transition, that is, a better device like GaN and SiC.

Closing the track

The electronics track ends here. We set out from the energy-band question of what lets a material conduct, then doping set the carriers, drift and diffusion moved them, and the pn junction became a one-way valve. With that valve we rectified and shaped waveforms, and the thin-base BJT and insulated-gate MOSFET held a large flow with a small input to give amplification. With the differential pair and the op-amp, its limits of bandwidth and slew rate, and the filter, we handled signals, and at last we came to using the same transistor as a switch to handle power itself efficiently. A single thread, how to handle the electron, ran through all twenty-four units from beginning to end.