4

For an RC (mostly LiPo) battery charging setup I've got a 48V Meanwell AC->DC PSU supplying power to the charger. I'm considering simple ways to power the charger either by this PSU or directly by a 24VDC deep cycle battery bank. The charger is smart enough to have profiles that have limits so I can be sure I won't draw too much from my chosen power source.

Off the top of my head, one simple way would be to just tap into the lines that run between the PSU and the charger, and connect the 24V battery there. This is obviously a really bad idea if the PSU was on, but given my use case the whole setup will either be in a location where AC is available or it'll be in a location where only the deep-cycle batteries are available.

My question is: Will it harm the PSU or be dangerous if 24VDC is seen on the PSU's output terminals when the PSU has no AC on the input side?

kbyrd
  • 239
  • 2
  • 9
  • 3
    Depends on the power supply. (a) what does its spec say? or (b) do ya feel lucky? –  Apr 02 '15 at 17:41
  • I don't know what to look for in the specs. How would this be shown? To keep things simple, I'm hoping to avoid a "protection circuit" of some sort, but if I can't I'll end up doing that. – kbyrd Apr 02 '15 at 17:51

2 Answers2

9

Some PSU's have physically disconnectable outputs (i.e. a relay on the output) and/or some other kind of over voltage detection/protection when on. OVP is usually a marketed feature, but doesn't necessarily mean it will work when the PSU has no power at all. Physically disconnectable outputs is not a feature I've typically found advertised.

Regardless of whether it's safe or not, you can add 2 diodes to protect both the PSU and the battery:

schematic

simulate this circuit – Schematic created using CircuitLab

This will select the output with the higher voltage and use that to power the charger while protecting the other source from harmful reverse currents/voltages.

You should pick diodes with relatively low forward voltage drops (i.e. Schottky diodes) and capable of handling the currents the charger will draw.

helloworld922
  • 16,600
  • 10
  • 54
  • 87
  • Just heard back from MeanWell: "...I would recommend adding a blocking diode on the output..." – kbyrd Apr 02 '15 at 20:03
  • So if I built things this way, and then accidentally connected both sources, then I have both sources connected in parallel, right? So the "to Charger" load would see roughly 48V+24V? – kbyrd Apr 03 '15 at 01:19
  • No, the charger would see 48V; only the largest source supplies any power to the charger. – helloworld922 Apr 03 '15 at 01:57
  • Followup question: The charger is current limited at 40A, so will need to pas a max of 40A. Will a regular or Schottky diode handle this? Should I be using a different diode-like device instead? – kbyrd Apr 04 '15 at 03:04
  • At 40 amps, you really need to do a proper power analysis on an actual parts with heat dissipation system to determine which parts will work. There are solutions based on MOSFET's (not diode-like), but they will be even more complicated, and you will still have to worry about heat dissipation issues – helloworld922 Apr 04 '15 at 04:15
0

As it is a switching supply there is almost certainly a blocking diode in the output already as part of the circuit (usually the output circuitry will be a transformer winding feeding through a diode or diode bridge to a smoothing capacitor.

So I would say it is very unlikely to be a problem, but if you want to be quite sure, use an external blocking diode, preferably a Schottky type to minimize the voltage drop.

THe diode will of course need to have sufficient current rating to handle the output current of the power supply when it is active. That is likely to mean a 48A or 50A diode on a decent heatsink (at 40A the diode will be dissipating around 10W).

Alex Gray
  • 1