I am designing a power circuit based off an older circuit for a similar machine, and have had a few issues in LTSpice simulator, where massive currents are drawn for the capacitors charging.

The image above shows the capacitors spike at the frequency of the AC input, quickly charging, but drawing a huge amount of current per cycle.

The instantaneous current draw (approx 120Hz) is 18A, and I am wondering whether this value is harmful to the operation of the circuit (currently under a 3.15A fuse), as the normal DC load is under the fuse normally.

Circuit voltage vs total current draw Graphs showing current vs Voltage

Circuit outline (Simplified) Circuit design

(I've got ways to manage inrush current, but not the constant spikes)

  • 1
    \$\begingroup\$ The simulator can calculate you the RMS if you worry about your fuse, but you should also worry about the diodes and the cap(s) \$\endgroup\$
    – PlasmaHH
    Commented Jul 16, 2018 at 8:47
  • \$\begingroup\$ Is that a 3 ohm resistor across C1? \$\endgroup\$
    – Andy aka
    Commented Jul 16, 2018 at 9:30
  • \$\begingroup\$ In the original circuit the fuse is right at the start, but the capacitors themselves aren't wired any different from the circuit above, so I'm not sure how sudden spikes of current are handled due to the capacitor charging. The I1 and I2 are loads, the 24V regulator has a 0.125A load, and along the main wire there is a 3A load \$\endgroup\$ Commented Jul 16, 2018 at 9:31

1 Answer 1


This is perfectly normal and is something of a downside of the capacitor input power supply, but they all do it.

Generally (unless you model the parasitics correctly) spice models of this sort of thing are a bit worse then the charging pulses are in reality, because real transformers have leakage inductance and resistance and real grids have both inductance and resistance.

The old school solution is a choke input supply, but it adds quite a lot of weight and needs a minimum current load to avoid the voltage shooting up, this was popular in the vacuum tube era where lots of volts at low current was the usual case. The modern way if a PFC circuit, but it adds a lot of complexity.

How much ripple can you tolerate? Reducing the size of the cap increases ripple, but raises the conduction angle, so less cap is better in that respect.

What almost everyone does is just to massively over size the diodes and size the transformer for the actual VA requirement (which will be quite a lot bigger then the average power), and carefully consider the cap ripple current.

  • \$\begingroup\$ Thanks for the answer, this covers everything and I'll go do some more research. I can tolerate about 3-4V ripple, before I start getting major regulator dips, and the transformer currently being used is what the original circuit used (I'm also using a similar capacitance), so I'll have to do some digging to find the information for it. \$\endgroup\$ Commented Jul 16, 2018 at 9:48
  • \$\begingroup\$ @GeoffrySkionfinschii there is a simple trick using an extra diode and capacitor that overcomes the main ripple being seen on the regulator input. \$\endgroup\$
    – Andy aka
    Commented Jul 16, 2018 at 10:30

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

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