ECN Forum Forums General Topics of the Trade Non-US Electrical Systems & Trades Questioning the electrical norms

Correct. At least the one in my home (or more correctly, my mother's home), a Heller TAH2S (stainless steel, 2 slices in one long slot, nominal power of 800W at 230V which appears accurate from a measured 66 Ohms across the plug with the lever held down), shows no conductive path between either the line (usually referred to in Australia as "active") or neutral plug pins and the element with the switch off. Of course, anyone who sticks a knife in with the toaster on deserves what they get (although having RCDs increases the chances of survival).

Hilariously, this is even true of the North American toasters nowadays! (If you go here, you can find repair guides for a few of them, and the pictures show the detail quite clearly. So obviously, the only reason why they still bother with the polarised plugs is tradition, and the public perception of the safety "advantage". But anyway, I'm quite sure that they are not double-insulated but are merely exempt from the normal rules. At least, I fail to see where the supplementary insulation is in those examples, and you can even find a designated earthing point on the chassis of the PS77401. Of course, it stands to reason that insulation failure is rare enough anyway that the UL and CSA can get away with it, knowing that the public will buy their argument despite it having zero basis in reality - that's what I meant when I said before that the UL is corrupt; well, they're a private corporation established by a group of insurance companies instead of a national safety agency, which I guess explains it.)

As for the Edison screw fittings, at least the European ones have the side contact relatively deep into the lamp socket (the socket thread itself being plastic or ceramic) - unlike the North American ones where the entire threaded section makes the connection (nasty). So with a DP switch, I suppose they aren't that much worse than the bayonet fittings predominantly used in Australia and the UK.

Even if you were able to verify 100% the polarity, a broken neutral upstream on any circuit shared between multiple loads still presents a risk (even a relatively small electric heater, for example, might as well be a straight wire as far as a resultant shock is concerned) if you rely on only SP switches.


It's a matter of feasibility; the simplest way of meeting the double insulation standard, of course, is with a plastic casing serving as supplementary (or, in some cases, reinforced) insulation. With a metal chassis or cover, it gets trickier, although there are still some possibilities; for example, you could:

• Insulate the inner surface in the vicinity of live circuitry
• Insulate the entire outer surface; this way, the chassis can still become live but the user is still protected, as long as they don't damage the outer insulation.

Another method that I've read was used by some vintage fan heaters was to have a dual-layered casing, with insulating spacers between the inner and outer layers (modern cheap fan heaters, of course, just have plastic cases).

(Components such as motors that are themselves only single-insulated may be used in a Class II appliance if mounted in such a way that they are hidden from user contact, and isolated from any externally accessible metal. Dissipating the waste heat may then become harder, though - I've seen inside the base of a badly-designed fluorescent desk lamp whose (inductive) ballast ran so hot that its 105°C rated lead-out wires deteriorated quite quickly, although it hadn't stopped functioning (at the time - but I subsequently took it to pieces). Suffice to say that the result of the ballast leads shorting together wouldn't be a pretty sight...)

Double-insulated hook-up wire could also be used, although I don't know if any is provided with insulation other than boring PVC. Or add-on tubing (available in various materials, which may or may not be heat-shrinkable) can be used for supplementary insulation.

Suitably guarding the terminations may be a challenge, of course. When you're simply connecting wires to other wires, it wouldn't exactly be insurmountable; enclosing the wire joiners in a robust plastic (or other suitable insulating material) box with a suitable clamping arrangement for the insulation should do. With connections to other hardware (controls, motors, heating elements, etc.) things may be trickier. Still, some types of terminals are quite strong; in particular, the crimp terminals that hold onto the insulation as well as the conductor are nearly indestructible - provided that they are matched to the wire size, crimped solidly, and no attempt is made to join more than two wires to a single terminal (and that 2-wire combinations are examined on a case-by-case basis - just jamming the wires in there without proper testing won't cut it with me). Wire terminations to printed circuit boards can be reinforced with Amp-In(tm) or similar connectors (photos from DigiKey):



From left to right, a Mini Amp-In for 26-22AWG wires, another version for 22-18AWG wires, the larger Amp-In for 18-14AWG wires and finally the biggest Amp-In, for 12AWG or 10AWG wires. (They can be found in this category by selecting "Amp-In" and "Mini Amp-In" for the series, and "Through Board" and "Through Board, Locking" for the terminal type - except for the first one shown, which was accidentally placed in the category for "Rectangular Connectors - Contacts".)

Basically, the wire is stripped, inserted into the connector and crimped, then the connector is inserted into an appropriate-sized through-hole in the board, whereupon its tab(s) will lock it into place, and it is then soldered as normal.

For pluggable connectors, we have, among others, the ever-popular Faston(tm) line (and equivalents) - and the "Positive Lock" variations, which add a mechanical latch (enabling stronger retention force, but also easier disconnection at will), look particularly appealing for the purpose I have in mind here:



Even with that out of the way, though, there are other things, such as the sealed heating elements. I don't know if there's a way to make those so that they can never short or leak to the outer casing, although I presume that engineers have tried but without success.

But enough of the technical details. I take that, in the end, the choice between double-insulation or earthing comes down primarily to which approach costs less to build. As long as the earth connection holds solid, it wouldn't exactly change the world were someone to work out how to build a double-insulated fridge tomorrow.

Of course, there is one category of devices that are typically built as Class II, but really shouldn't be - those switching PSUs. Most of them require small "Y"-class capacitors (rated for use across safety insulation - Y1 class across double/reinforced, Y2 class across basic or supplementary) connected between the mains lines and the output return in order to pass EMI tests; of course, these pass a current (however small) through which can be felt at times. Probably worse, though, is if they are accidentally discharged into audio/video circuits; this is much nastier than "traditional" static discharges (already a serious danger to many electronic parts), being capable of one-shotting even a discrete bipolar transistor (as explained in detail in this article). This highlights a significant flaw in the regulations - they still seem to treat each "appliance" as a stand-alone entity, even though much audio/visual equipment is pretty much useless without interconnections. Interconnecting multiple devices with Y1 class capacitors will also add together their leakage currents, so if you hook up enough gear, you could get quite a significant shock touching it even with no component failure.


If I recall right, they may have either a normal resettable GFCI (or RCD as known to the rest of the world), or a single-use ALCI (Appliance-Leakage Circuit Interrupter, or as the Sci.Electronics.Repair FAQ dubbed it, the "Ground-Fault Circuit Killer"). The rest-of-the-world models don't have additional protection (that I know of).

As for the old-style mains transformers, while they are heavy, can't do auto-ranging input and their idle draw is higher than the PC environmentalists are happy with, they are at least more reliable and it's much easier to guard against catastrophic failures in them (one thermal fuse in the primary and you're covered, basically; there's much more that can go wrong in a switcher), plus with almost no EMI (which can matter a lot for audio equipment).