1
\$\begingroup\$

I need to charge two 1500uF-450V capacitors off of a 3-phase 220Vrms-60Hz line using a very robust bridge rectifier (datasheet) as well as both capacitors datasheet (datasheet). capacitor info highlighted in yellow

Here's the simulation I performed taking into account ESR of approx \$63\,m\Omega\$ for each cap.

simulation

WITHOUT the inrush limiting resistor \$R_2\$, the inrush current peaks \$4.2\,kA\$ in one of the diodes (the max forward surge for each diode is \$7500 A\$ for \$10\,ms\$) and \$1.5\,kA\$ at each capacitor. The duration of both surges is approx 1ms.

The problem is I don't think this kind of surge is good for long-term in the circuit since it might cause tripping of breakers, and it's a very large current so I'm taking the safe path and trying to limit it with an inrush resistor.

The crux of the problem is how to properly design/choose both bypass switches S1 and S2. S1 bypasses the inrush resistor while S2 connects the load R3 to the DC voltage after reaching a reasonably okayish value (200 - 250 V).

I've looked up some solid state relays but most are rated for 40VDC max so it'd not help. Using an IGBT for both switches I think would suffice in my POV however I don't have the time to come up with a PCB or any other additional circuitry at the moment so an off-the-shelf solution would be best here.

Any thoughts on how could I approach that with available solutions?

\$\endgroup\$
7
  • \$\begingroup\$ On the left side of your drawing, is that Neutral? If yes, it is tied to ground somewhere. Your circuit is also referenced to ground. D2/D4/D6 will be directly across the line, big boom. \$\endgroup\$
    – Mattman944
    Oct 26, 2023 at 17:20
  • \$\begingroup\$ @Mattman944 Circuit ground is arbitrary, and this is a fine reference point for simulation purposes. \$\endgroup\$ Oct 26, 2023 at 17:35
  • \$\begingroup\$ It sounds like you need an inrush current limiter element, or circuit, or capacitor precharge. This is an interesting design exercise, but it sounds like you might not have much design time available; and keep in mind asking for off-the-shelf solutions is discouraged here. \$\endgroup\$ Oct 26, 2023 at 17:38
  • \$\begingroup\$ What are the specs for "charging"? \$\endgroup\$
    – Antonio51
    Oct 26, 2023 at 17:40
  • \$\begingroup\$ @Mattman944 No, it's not neutral. That GND is just for ease of simulation so I could measure voltages with respect to load ground. In the real-world application I won't have a neutral. \$\endgroup\$ Oct 27, 2023 at 5:23

2 Answers 2

2
\$\begingroup\$

The classic lossless alternative to resistors is to use an inductor to limit the inrush current. You see these used extensively in switchable power factor correction capacitor banks, and you can even get PFC caps where the inductors are built into the three phase capacitor case. This may or may not give you the desired behaviour in your rectifier circuit.

You could also consider using NTC thermistors to limit the initial inrush current. Whether this works will depend on how quickly the load can be switched off and on again (so whether they get to cool down).

\$\endgroup\$
1
  • \$\begingroup\$ I came across the NTC thermistor idea and might do well. I have an 70A rated inductor which might help in reducing the inrush but its inductance might not suit the circuit so that'd lead to me spending money which I'm kinda out of at the moment for the project. For future ideas though, that's clearly a nice way to go. \$\endgroup\$ Oct 27, 2023 at 5:30
2
\$\begingroup\$

Classics is to use inductors for limiting inrush fire currents.

Here is a simulation with 1 mH inductors. RR = 1 Ohm ...

Crest currents should be lower than 250 A for 10 ms.
Note the high "power" for resistors.

enter image description here

\$\endgroup\$

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.