1
\$\begingroup\$

I have a heavy duty, 62V, 600A continuous battery which I want to connect to an electric vehicle controller with a 5,2mF capacitor in it.

To prevent a massive spark when connecting it, I am thinking about precharging the controller / capacitor with a laboratory power supply up to 62V, then quickly disconnecting the power supply and connecting the battery to the controller, instead of the power supply.

What do you think? Is this a good idea or will this burn my house down?

I have already tried out an anti-spark circuit from a friend of mine with a MOSFET, but the N-channel, selflocking MOSFET is broken, I think. The supply and drain are shorted, even though no voltage is connected and even though all connectors have been shorted so there should be no voltage on it anymore.

\$\endgroup\$
7
  • \$\begingroup\$ all mosfets will always conduct from source to drain, there's a parasitic diode in the structure. They can only block conduction from drain to source. \$\endgroup\$
    – Hearth
    Feb 5, 2022 at 1:52
  • \$\begingroup\$ Also, what is a "selflocking mosfet"? \$\endgroup\$
    – Hearth
    Feb 5, 2022 at 1:54
  • \$\begingroup\$ @Hearth are you sure about that? I mean the purpose of the mosfet is to only conduct, when I apply a voltage from gate to source, isnt it?` Selflocking = bad german translation for "only conducts, when there is a voltage between gate and source". \$\endgroup\$ Feb 5, 2022 at 3:04
  • 1
    \$\begingroup\$ MOSFETs are one-way switches, turning a FET off blocks conduction from drain to source but does not block conduction in the reverse direction from source to drain. If you need to block in both directions, you'll need to use two MOSFETs, or some other setup. Also, the correct translation for what you call "selflocking" would be "enhancement-mode", if I'm understanding you correctly. \$\endgroup\$
    – Hearth
    Feb 5, 2022 at 3:08
  • 1
    \$\begingroup\$ I worked on a scooter that supported battery swapping. The way we handled it was that we pre-charged the capacitor when the battery was initially plugged in (discharge FET's were off), and then enabled the battery discharge FET's only after the cap was charged up. The battery had a low-current VBAT pin that we used for pre-charge. Once we debugged it all it worked OK. But this required both the battery and the scooter to support the feature. It is more difficult if you are not free to add features to the battery (such as a pre-charge contact). \$\endgroup\$
    – user57037
    Feb 5, 2022 at 3:50

1 Answer 1

2
\$\begingroup\$

You can precharge the big capacitor via a resistor so the initial currenta are low enough to not cause problems. In your case, a 62 ohm resistor would limit peak charging current to 1 ampere. Remember that the capacitor initial volts could be zero and the capacitir will have very low ESR. When the capacitor is deemed to be charged, then you could automaticaly bypass the resistor with a heavy duty relay or MOSFET. The capacitor AC currents are significant when the motor controler is running so rate your switch accordingly.

\$\endgroup\$
2
  • 1
    \$\begingroup\$ Peak dissipation is also 62 Watts, although it falls off rapidly. The time constant would be 62 * 0.005 = 40 ms. So it might require around 200-300 ms to charge the cap sufficiently. \$\endgroup\$
    – user57037
    Feb 5, 2022 at 4:10
  • \$\begingroup\$ @Autistic instead of bypassing the resistor with a Mosfet, can I just bypass it by connecting the wires of the resistor? Then I would not need the Mosfet anymore. \$\endgroup\$ Feb 5, 2022 at 12:04

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.