I'm building a device that has a mains power component, which I need to prototype before building to make sure everything works. I know I need to make sure wiring—and insulation!—is thick enough to cope with the higher current, and that I need to make sure there's enough space between exposed components so there's no arcing. I'll also be stashing whatever I prototype in a nice, insulated box before applying power, and I'll be include a small-value (250mA) fuse just in case. I'll be considering everything double-insulated rather than earthed (plastic enclosure, and there's nothing I'll need to touch while it's operating).

My plan is to solder (through-hole) components directly to each other, then insulate everything (with heatshrink tubing or electrical tape) as much as possible, then use screw-type cable joiners to connect the power cable.

I know I can't use traditional prototyping tools such as breadboards and stripboard for mains voltages, but what can I use to make prototyping at mains voltages easier?

(To be clear, 'mains voltage' in my area means 230V AC +10/-6% at 50Hz; GPOs are generally rated at 10A.)

  • \$\begingroup\$ What circuit are you prototyping? What else is it connected to besides the mains? \$\endgroup\$ Commented Jun 2, 2014 at 3:19
  • \$\begingroup\$ I was hoping for a generalised answer, but in this particular case I'm attempting to power an LED from mains power; there's nothing else connected, it's just the LED with a resistor, (mains-rated) capacitor, and diode. \$\endgroup\$
    – Calrion
    Commented Jun 2, 2014 at 3:29
  • 1
    \$\begingroup\$ I ask because infinitely more things can go wrong if you have a mains-connected circuit connected to test equipment, development systems, computers, etc. \$\endgroup\$ Commented Jun 2, 2014 at 3:42
  • \$\begingroup\$ Of course; I'd not thought of that (though I don't really need to in this case). What I'm after is something like a breadboard or stripboard, but that can be used for mains-power stuff. For four components, what I'm doing seems ok, but what about more complex circuits? \$\endgroup\$
    – Calrion
    Commented Jun 2, 2014 at 3:50

1 Answer 1


There are several potential safety issues with mains-connected circuits. I don't know about Australian legal issues, so I have no comment on that.

  • shock, electrical burn or electrocution hazard
  • fire hazard
  • energy related injuries from arcing etc. (damage to vision, burns)
  • damage to other equipment caused by a circuit being completed through connected devices
  • indirect safety hazards (for example, an oscilloscope probe or case becoming electrically 'hot' due to connection to a mains circuit, harm from falling off a ladder or whatever due to a shock, etc.)

Your simple circuit is unlikely to present an energy related hazard with a 250mA fuse (assuming you don't do something pathological like using a 5x20mm fuse on an industrial circuit). In tests I was able to make glass fuses literally explode, projecting a cloud of glass shards and molten metal for a fair radius with no more than 240VAC (they were rated 250VAC), but a large interrupting current (well beyond their rating).

Shock hazard on your simple circuit- it might be an issue. The LED is not safety agency approved for mains connection- and it will stick through the case, I'd imagine. Probably you'll see an additional bit of transparent plastic in front of any mains-connected LED. They're worried about possible damage to the LED, maybe a molding short-shot that could expose wires, what a kid could do poking at the LED with a fork or whatever.

You mention the capacitor is mains-rated. Do you mean it's safety agency listed, or just rated for what you think is sufficient voltage? An "X-rated" (or Y-rated) capacitor should be okay.

As far as the mechanics of prototyping mains circuits, usually what I do is to hack the power stuff together on a perf board or a proto PCB with everything very, very neat and clear and visible isolation gaps between high voltage connections etc. If there is complex control stuff that will not be isolated eventually from the line, I'll add adequate galvanic isolation for development. Most people will recommend using an isolation transformer, which means you have to do two things wrong rather than one to be electrocuted in most (but not all) cases, and makes test equipment damage less likely. An RCD (GFCI)-protected circuit should be a bit safer. Everything gets buttoned down, secured and mounted on a non-conductive base before power is applied.

Stuff that gets connected- if the inputs are earthed (most oscilloscopes, computers etc.) then even a momentary accidental mains connection can blow the heck out of stuff, so galvanic isolation and keeping everything clean and neat is vital. A stray cut lead or a wire that comes off and goes a-sparking can be big trouble, as can a probe or screwdriver that slips with the power on. I've seen every single chip in a computer blown by a stray ground strap that touched the 'hot'. If it's not grounded, the attached device itself can become a shock, burn or fire hazard.

An appropriately-rated fuse is an excellent idea- sometimes a series incandescent lamp works even better (it's effectively a PTC resistor).

Arcing can cause eye damage- if everything is not adequately enclosed for whatever could happen, safety glasses at least, and perhaps a full face shield are good. Really high power stuff requires precautions against arc flash. Even low voltages can be hazardous if potential fault currents are high- spraying molten metal in your face, for example, or heating a ring on your finger up to red heat in a fraction of a second.

One frequently stated bit of advice is to work with (at least, LOL) one hand in your pocket, which avoids the possibility of current flowing from one hand to the other (through the heart), but still leaves the possibility of shock and burn injuries.

You should treat working with mains voltage like any other potentially hazardous activity. Educate yourself on safety precautions, consult with local experts if there is any doubt, work slowly and deliberately, don't do it when you're tired or otherwise off your game, and if you find yourself making small mistakes, take a break before you make a big one.

  • \$\begingroup\$ The perfboard I have has about 0.5 mm (19 mil) separation between the tracks; my thought was that's not enough for mains voltage, am I wrong or are you talking about a different product? The capacitor has a "Volt Rating: 275/250VAC". I've never seen a voltage rating on an LED, are 'mains-rated' LEDs something I can get? (You're right, the LED dome will protrude beyond the case inside a plastic bezel.) \$\endgroup\$
    – Calrion
    Commented Jun 2, 2014 at 4:47
  • 4
    \$\begingroup\$ The perf board I use for this purpose has no tracks or copper. It's literally just perforated FR-4. You can get housed LEDs with mains power. apem.com/files/apem/brochures/UK/… \$\endgroup\$ Commented Jun 2, 2014 at 5:03
  • 3
    \$\begingroup\$ P.S. if you're using a dropping capacitor circuit, note my comment about indirect hazards, and put a bleeder resistor across the capacitor or input. You don't want to pull the plug out, get startled by a shock from the prongs and plummet 30 stories out the window or something (okay, more likely just a tingle). \$\endgroup\$ Commented Jun 2, 2014 at 5:09
  • \$\begingroup\$ Use a ceramic not a glass fuse if you can - they don't tend to explode in the same way! As well as keeping one hand in your pocket a thick rubber-ish mat to stand on might be helpful as well & given that your metal workbench will be properly earthed a 2nd one for your project wouldn't go amiss - and try and ensure the project PWB is securely fixed in place - it will help to prevent probes slipping as things move, though you really want to have probes fixed in place before energising the supply... - finally have someone else around who knows what you are doing and how to switch it all off...! \$\endgroup\$
    – SlySven
    Commented Oct 9, 2016 at 20:16
  • \$\begingroup\$ Where can I get myself some of those X-rated capacitors? ( ͡° ͜ʖ ͡°) \$\endgroup\$
    – Sparkette
    Commented Jan 15, 2018 at 19:19

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.