# Why is my battery showing a huge negative inrush current in my designed charger model?

I have a three-phase charger model on simulink as shown below, with a three-phase active front end converter followed by a phase-shifted full bridge DC/DC converter.

The circuit is working pretty well, but the battery shows a huge negative inrush current at the beginning of the order of 30kA, where as my battery current is just 125A in steady state. I have added resistor as inrush limiter which I switch out with a relay (using 'ideal switch' on simulink), but that doesn't seem to help at all, adjusting the value of resistor shows absolutely no impact on the negative inrush. I am finding it very unusual that this current is negative, because what I know about inrush is that its the initial current flown due to the capacitor, but that should have been positive. Is this some other phenomena that I am seeing here? How to mitigate this?

Zoomed in view of initial inrush

Zoomed out view of the current: (limiting resistor is switched at 0.1s)

Edit: This is the zoomed in image of the circuit as someone suggested:

This is the active rectifier:

This is the phase shifted full bridge DC/DC converter with battery:

• A bit difficult to see details in the schematic. Should be split into 4 parts. Have you considered the "initial conditions" ? Commented Mar 20, 2022 at 20:58
• does the simulator use ideal components? Commented Mar 20, 2022 at 21:06
• By 'negative', do you mean the current is coming out of the battery? What's the ESR of the 200uF capacitor? Commented Mar 21, 2022 at 2:33
• @Bruce Abbott yep that's what negative current would be. The ESR is 0 in the model. Commented Mar 22, 2022 at 6:10
• @jsotola yes all components are ideal. Commented Mar 22, 2022 at 6:10

When the switches are turned off your circuit simplifies to this:-

simulate this circuit – Schematic created using CircuitLab

The battery is putting out (say) 30 V. The capacitor has 0 V across it. In this ideal circuit the capacitor has zero ohms internal resistance, and the wiring has zero resistance and inductance. Therefore the only thing limiting current is the battery's internal resistance. At 1 mΩ the initial current is 30/0.001 = 30,000 amps. The RC time constant is 1 / (0.0002 F * 0.001 Ω) = 200 ns, so the capacitor will only take a few microseconds to fully charge.

But why is the current negative? It's coming out of the battery and going into the capacitor. Due to the way the ammeter is connected it will read negative current, which means the battery is discharging into the charger when initially connected.

This simulation is unrealistic because it uses 'ideal' components. However the same effect will occur in a real circuit, just with lower current. If the battery and capacitor both have low internal resistance the current surge could be very large, causing arcing at the connector when the battery is plugged into the charger. If the battery is a flooded Lead-acid type (which releases hydrogen and oxygen gas during charging) there is an explosion risk. This too is not just theoretical - I have seen it happen. If there is any possibility of your charger being used in an environment where explosive gasses may be present then it should be designed to avoid surge currents that cause arcing.

Your battery is being charged at 125 A so it must have very low internal resistance and act like a huge capacitor itself. Therefore the 200 μF capacitor hardly seems necessary.

• Thank you for your answer. I removed the cap and the issue seems to be solved, however, I have always seen caps in battery chargers after DC/DC converters. Isn't this a more widespread problem then? Why is the cap used so commonly then? Commented Mar 22, 2022 at 14:31
• Chargers used in 'float' mode are actually operating as a power supply, which may be required to provide smooth DC voltage even if the battery is not connected. The capacitor also helps to reduce ripple, EMI, and possible inductive effect of long battery leads. Portable SCR based lead-acid battery chargers and motor vehicle alternators generally don't have a smoothing capacitor on the output because it isn't needed. Commented Mar 22, 2022 at 21:45