0
\$\begingroup\$

I've heard that if you can have negative current in PSU, you can use that to discharge batteries. Is this true?

If yes, how to design/modify a linear bench PSU (which have transformer with several secondaries) in a way that be able to discharge batteries?

What's the working principle of it, I have no idea where to start.

\$\endgroup\$
5
  • \$\begingroup\$ I imagine you would connect the + temrinaly of the supply to the - terminal of the battery and vice versa. You need current limiting, for sure or bad things will happen. However, nothing stops you from discharging the battery to 0V if you do this, which would destroy the battery. Perhaps they are referring to connecting + to + and - to - and having the supply sink current from the battery by setting the PSU to the minimum discharge voltage. You still need current limiting and the PSU has to be able to sink current, which linear supplies cannot do. \$\endgroup\$
    – DKNguyen
    Commented Feb 13, 2020 at 17:10
  • \$\begingroup\$ @DKNguyen The bench PSUs have current limit feature, but it can only control positive current. do you mean an extra current limiter for discharging the batteries? so PSUs can't be modified and should add a whole new circuitry just for discharging using mosfet etc to sink current? \$\endgroup\$ Commented Feb 13, 2020 at 17:11
  • 1
    \$\begingroup\$ It has to have an output stage similar like AB class amplifier. You won't find them for cheap, these are extremely expensive PSU as it can cross conduct if not made correctly. \$\endgroup\$ Commented Feb 13, 2020 at 17:16
  • 1
    \$\begingroup\$ Does it have to be a PSU? Does it need to do power conversion (e.g. feed back power to the grid)? What's wrong with just dissipating the battery's energy in a power resistor or something? \$\endgroup\$
    – anrieff
    Commented Feb 13, 2020 at 18:37
  • \$\begingroup\$ @anrieff IDK just read it some where and was wondering how it can be done. \$\endgroup\$ Commented Feb 13, 2020 at 19:43

1 Answer 1

2
\$\begingroup\$

In general this will not work.

schematic

simulate this circuit – Schematic created using CircuitLab

Figure 1. Really simplified schematics of (a) what you've got and (b) what you want.

Most bench power supplies are of the form of Figure 1a. They are designed to provide current, not sink it. The output stage will have beefy transistors from the positive power rail and these regulate the amount of current through them to maintain the required voltage at the output. If the output voltage is higher than the setpoint then Q1 will turn off. No current will flow back into the PSU unless you blow the output somehow.

How Figure 1a works:

  • V1 provides the raw DC. R1 provides a reference voltage between 0 and 30 V to the non-inverting amplifier input. The op-amp output will increase, turning on Q1 until the feedback to the inverting input equals the setpoint. There is no high-current discharge path to ground.

To do what you want would require something like Figure 1b.

  • This circuit will control the discharge current from the batteries. R3 is a current shunt and provides negative feedback in proportion to the battery discharge current. R2 provides the setpoint as before but since the feedback voltages will be low the reference voltage setpoint will need to be restricted by R4.
\$\endgroup\$
2
  • \$\begingroup\$ Thanks for the explanation, in a bench PSU we have all the required components, except the regulator is a transistor usually. do you think it's a good idea to use a mechanical relay to switch the op amp output between transistor base and a mosfet gate? this way we can have both, with two separate output of course... \$\endgroup\$ Commented Feb 13, 2020 at 19:48
  • 1
    \$\begingroup\$ No. What I have drawn is a highly simplified schematic. Why would you use a power supply at all when you have no power requirement other than a few tens of mA to control a current sink? \$\endgroup\$
    – Transistor
    Commented Feb 13, 2020 at 19:55

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.