0
\$\begingroup\$

Here's the most basic capacitative dropper circuit I could find with ten seconds on Google. Allegedly the voltage across the load will be about 5V. (I say "allegedly" because I couldn't make CircuitLab actually simulate this, it kept complaining about invalid starting conditions, with or without a fake ground node. Anyway, that's not what the question is about.)

schematic

simulate this circuit – Schematic created using CircuitLab

We all know that these are unsafe for a variety of reasons. What I'm wondering is, does it make it any less unsafe if you put another Y-rated capacitor and another input resistor on the bottom AC rail (what's currently a direct connection from Vin(-) to BR1)? Adjusting the values of everything as necessary, of course.

It seems to me that this would eliminate the safety issues that arise if you plug one of these in backward, but that the risk of (at least one of) the Y caps failing short remains higher than the risk of an isolation transformer failing short. I don't know how to assess any of the other risks involved with a design like this.

\$\endgroup\$
2
  • 1
    \$\begingroup\$ Please define "backward"? Depending on which kind of mains sockets and plugs you have in your or any country, or which kind of mains inlet your device has (C8 for example) there may be two equal ways of plugging something in, and so neither is backwards. \$\endgroup\$
    – Justme
    Commented Jun 26 at 18:22
  • \$\begingroup\$ It would still have the potential to kill you if you touch it so no, not really any safer. \$\endgroup\$
    – Finbarr
    Commented Jun 26 at 18:31

2 Answers 2

1
\$\begingroup\$

I'd say it makes even more unsafe for variety of reasons.

First of all, let's assume you live in a country where mains sockets and plugs are polarized and always wired correctly and there is never any wiring errors.

With only a single capacitor on Live and Neutral being directly connected, and each part being undamaged so no fauls in the components, the diode bridge negative terminal will always be within 1V of the 0V Neutral input voltage. Relatively safe and touching the output should not be deadly, but still, never touch it, and also such a device must not have any external connections because they are not isolated.

Of course if something is wired wrong and Live goes to diode bridge directly, the output is live and hazardous, as the 5V differential output rides on top of full 120VAC mains.

If you now put a capacitor on both Live and Neutral wires, assuming the capacitors are perfectly balanced and identical, with 120 VAC input, the output is still 5V differetial but riding on half mains AC with 60VAC common mode voltage compared to 0V Neutral. May not be a deadly voltage either and again if every component is working perfectly, it should be fine, 60 VAC is approximately the limit of safety low voltage depending on your jurisdiction, but still, unisolated, and it should not be touched.

The second resistor being irrelevant to the subject and can be omitted, one resistor is enough for limiting inrush current but it could be split into two for symmetry.

But two capacitors in series needs to be double in capacitance to equal the capacitance of a single capacitor. If for some reason the Neutral side capacitor blows open, the Live side capacitor gives double the current than if it were a single capacitor design. Even the 1uF capacitor alone has impedance that passes 45mA at 60Hz, so it is also deadly alone, if for some reason the neutral is switched.

If either capacitor of the two short-circuits, it equals a single cap system that now provides double the current, which stresses the zener and it may overheat, damage, stop working and allow the capacitor that normally sees 5V to charge to much higher voltage, to about 180VDC.

Adding the capacitor makes it no more safe, it seems to make it only less safe.

It also does not depend on having polarized plugs or unpolarized plugs, as for safety reasons, these kind of capacitive dropper supplies must not have connections to outside world or otherwise exposed for users, these must be contained in a box with no user access to touch the circuitry.

It sounds safer that any fault will either open the circuit or cause large enough short so that the fusible resistor acts as a fuse, or a separate fuse blows, or in the event of neither, the circuit breaker blows, instead of just some amount of extra current that may ignite a fire.

\$\endgroup\$
0
\$\begingroup\$

If you use a circuit like this, then every node, before and after BR1, must be isolated from the possibility of touching it. Once that is done, it doesn't matter whether the capacitor is in the live or neutral line, the circuit will be just as safe.

Using a dropper capacitor in both lines guarrantees that the circuit is at mid line voltage, rather than being a 50% chance of full line voltage or near to neutral. With its peak voltage of 170 V, even nominally 120 V supplies are still not touch-safe (less than 50 V) when halved by using two capacitors.

So as it's still safe if you use insulation, and still unsafe if it's not insulated, the number or distribution of dropper capacitors is irrelevant to saftey.

\$\endgroup\$
1
  • \$\begingroup\$ It may be irrelevant if only looking at a working circuit, don't think it's irrelevant as a whole, for example one capacitor shorting out (which should not happen with a properly rated capacitor with proper safety class rating). \$\endgroup\$
    – Justme
    Commented Jun 26 at 20:00

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.