0
\$\begingroup\$

In DC circuits, if I have a device that gets switched on and has a large inrush current, I can just put a large charged capacitor near the power switch (parallel to the device) to source the surge. This protects the DC power supply from the inrush current and the DC source can take its sweet time recharging the cap.

In the same circuit, if the power source was AC, then the voltage in the capacitor would be constantly changing. If the device's power switch was switched on at exactly when the voltage in the capacitor was zero, there'd be no energy stored to source the rush.

Is there an AC circuit equivalent to my DC capacitor/in-rush scenario?

\$\endgroup\$
1
  • \$\begingroup\$ Fly wheels can used for AC energy storage, but that may be a bit larger than what you're asking for. \$\endgroup\$
    – uhours
    Aug 15 '17 at 22:58
1
\$\begingroup\$

In DC circuits, if I have a device that gets switched on and has a large inrush current, I can just put a large charged capacitor near the power switch (parallel to the device) to source the surge.

You could also use a slower switching time, or ramp-up the power to the thingy which has huge inrush current. DC-DC converters usually have a slow-start feature, in order to startup with the load has lots of decoupling caps.

Is there an AC circuit equivalent to my DC capacitor/in-rush scenario?

Well, I've seen it in audiophile amplifiers, because these guys like to put enormous values of capacitance after the rectifiers, thus when you turn the amp on, the fuse blows, which is kind of a bummer.

There are two ways to do this: the brutal way and the smart way.

The über-brutal way is a NTC thermistor in series with mains. It has lots of resistance when cold, and low resistance when hot. Thus it limits inrush current, but only when it's cold when the power switch is pressed.

The brutal way puts some resistors in series with the transformer primary, which limits current when charging the caps. A few seconds after power is switched on, a relay closes and shorts the resistors.

If someone cycles the power button ten times in a row on the two previous circuits, something may catch fire. YMMV.

The smart way uses a triac dimmer, just like a halogen dimmer. It slowly ramps up the voltage on the transformer primary, by advancing the phase angle. So the caps charge slowly. But you need to design the triac dimmer properly for the transformer load.

Modern power supplies (like PC ATX)

\$\endgroup\$
1
\$\begingroup\$

There is no benefit to adding a capacitor across the AC supply (other than for power-factor correction - but that's another story).

You can control the switch-on using a zero-cross switch - generally a triac-based solid-state relay (SSR). While this will not reduce the inrush current of, say, a motor or cold filament lamp, it will prevent switch-on at peak voltage and eliminate EMF (electro-magnetic interference) and solves the problem of contact arc (by eliminating the arc).

enter image description here

Figure 1. An SSR circuit. Source All About Circuits.

enter image description here

Figure 2. Resultant waveform. Source: CCI Power.

enter image description here

Figure 3. A pre-packaged SSR.

The other option which may or may not be applicable depending on the load is to gradually increase the voltage using a non-zero-cross SSR.

enter image description here

Figure 4. By controlling the switch-on point in the mains half-cycle the RMS voltage can be adjusted from zero to almost 100%. Source: @Transistor.

In the case of Figure 4 the DC control signal would have to ramp up at a suitable rate.

\$\endgroup\$
2
  • \$\begingroup\$ And the zero cross switch suppresses the inrush current? \$\endgroup\$ Aug 15 '17 at 21:23
  • \$\begingroup\$ See the updates. \$\endgroup\$
    – Transistor
    Aug 15 '17 at 21:31

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.