8
\$\begingroup\$

enter image description here

The typical approach to interrupt the flow through a circuit is interrupt it at a single location as in the left image.

Do any power loads require both power lines disconnected by the switch as in the right image?

Does AC/DC matter? And in the case of DC, does +/0 vs +/- power matter?

If so, why?

\$\endgroup\$
2
  • 1
    \$\begingroup\$ Require in what way? Physically require, or require by statutory regulations, or require in some other way? \$\endgroup\$
    – Hearth
    Commented Sep 29, 2023 at 3:53
  • \$\begingroup\$ @Hearth functionally required \$\endgroup\$
    – dtech
    Commented Sep 29, 2023 at 3:56

6 Answers 6

15
\$\begingroup\$

The American, residential, split-phase power distribution system has 240 V across two live lines and 120 V between each line and neutral.

enter image description here

In this system , a single pole switch is sufficient for 120 V loads with only one line being live and the other neutral.

enter image description here

However, a two-pole switch is required for 240 V loads with both both the lines being live.

enter image description here

The following three-phase system example also shows the requirement of a two-pole switch for 208 V loads connected between two live lines.

enter image description here

Incidentally, a three-pole switch is a must to switch 3-phase loads with all three lines being live.

\$\endgroup\$
6
  • 2
    \$\begingroup\$ residential American split-phase power... A three pole switch is required for industrial loads connected between all three legs of a three-phase system. - However many poles there are, all of them need to be shut off. (Define "both power lines", because you'd better not be talking about the neutral.) \$\endgroup\$
    – Mazura
    Commented Sep 29, 2023 at 18:57
  • \$\begingroup\$ @Mazura - Hi, I thank you for your feedback. I have edited my answer accordingly. \$\endgroup\$
    – vu2nan
    Commented Sep 30, 2023 at 4:44
  • 4
    \$\begingroup\$ This doesn't actually answer the question. a) functionally (as is asked) these devices will be off also with a single switch. b) even safety wise using a single switch on single phase stuff is dangerous if Land N are changed somewhere along the wiring (e.g. with the usual unpolarized connectors) \$\endgroup\$
    – tobalt
    Commented Sep 30, 2023 at 5:19
  • \$\begingroup\$ What is the consequence, if any, of the fact that multi-pole switches won't all (dis)connect at exactly the same instant? I'm guessing since this is an analog circuit and things like bouncing can be ignored for just a single pole it won't matter? \$\endgroup\$
    – Michael
    Commented Oct 2, 2023 at 8:11
  • \$\begingroup\$ @tobalt don't some appliances use 240V for motors and heaters, but 120V for controls? In that case you could expect misbehaviour switching only one leg \$\endgroup\$
    – Chris H
    Commented Oct 2, 2023 at 8:29
12
\$\begingroup\$

While there may be regulatory or safety reasons you would want to switch both power rails, there is no functional or physical reason it would ever be required. The electrons need to go through a complete circuit, and breaking that complete circuit in even just one place is enough to stop them from moving.

Note that there may be more than one complete circuit involved in a device and in general all of those circuits must be opened, e.g. a three-phase device with three input rails needs at least two of the lines switched off to fully remove power.

More insidiously, sometimes a complex IC can be accidentally powered through its inputs, via the protection diodes intended to stop ESD damage. This is one reason it's usually recommended to switch off the positive supply rather than the ground for digital chips--or to use a built-in disable function rather than switching the power supply at all.

\$\endgroup\$
2
  • 1
    \$\begingroup\$ If there are no devices that use line/neutral 120V downstream of the disconnect point, switching one rail may be functionally adequate, but if e.g. a 220V oven has a 120V light bulb between one hot and neutral, disconnecting just that hot upstream of the bulb may leave a current path from the other hot, through the heater, and through the bulb. \$\endgroup\$
    – supercat
    Commented Sep 29, 2023 at 21:20
  • 1
    \$\begingroup\$ @supercat This is the situation I'm referring to in the second paragraph. \$\endgroup\$
    – Hearth
    Commented Sep 29, 2023 at 23:07
7
\$\begingroup\$

You have to consider the purpose of the switch (including any reasonably foreseeable misuse/misunderstandings) and the characteristics of the supply.

In general there are two reasons to disconnect all of the lines feeding a device rather than disconnecting all but one* of the lines.

  1. To make the electrical conductors supplying the device safe to touch so the device can be worked on. In some situations, the neutral may be considered safe to touch and therefore not need to be isolated. In other situations there either is no neutral or the neutral is not considered safe to touch.
  2. To ensure that the device is turned off with a very high degree of reliability. Consider for example a railway signal that, if it incorrectly showed a green, could cause two trains to crash. Railway signalling wires run long distances in large groups under sub-optimal conditions so the risk of two wires shorting cannot be discounted. Switching both legs of the feed to the signal practically eliminates the risk of it lighting up uncommanded.

* Some devices have more than two lines feeding them. For example a 3 phase motor.

\$\endgroup\$
4
\$\begingroup\$

Some connectors of electrical system in Europe is unpolarized - line and neutral are connected at random. As a consequence of the potential uncertainty on which connection is live or neutral many models of power entry modules have DPST switches disconnecting both conductors of the entry lead from the load.

A typical power inlet module (Schaffner FN9280) is an example of a circuit where both power lines disconnected by the switch.

Schaffner FN9280 power inlet

Image source: Schaffner - FN9280

Schematic from Schaffner datasheet FN9280

Image source: Schaffner - FN9280 datasheet

\$\endgroup\$
7
  • 1
    \$\begingroup\$ D Duck - Hi, Please note the site rule which requires that when a post includes content (e.g. text, image, photo etc.) copied or adapted from elsewhere, that content must be correctly referenced. As a minimum, for online material the source webpage or PDF etc. should be linked (see that rule regarding references for books / articles etc.). In order to help you, I found what I believe to be the source links for the copied images & added them. For the future, please remember it's your responsibility to do that :) Thanks. \$\endgroup\$
    – SamGibson
    Commented Sep 30, 2023 at 22:08
  • \$\begingroup\$ Some European power connectors (eg Type C) are unpolarised and result in it not being predictable which wire is live (phase) and which is neutral. But that is the exception, not the rule - consider eg Type G. Indeed the photo illustrates a C13 connector which has well defined live (phase) and neutral connectors, as also shown on the wiring diagram. It's inaccurate to say "The electrical system in Europe is unpolarized - line and neutral are connected at random." - Neutral vs Live (Phase) are distinct, they are kept separate on most connectors, and neutral electrically connetced to ground. \$\endgroup\$
    – abligh
    Commented Oct 1, 2023 at 9:32
  • 1
    \$\begingroup\$ @abligh Isn't type F the most common type in Europe? It is not polarised, so you cannot tell which connector of the C13 end will be live when the the other end is plugged in. Type G is only used in the UK and Ireland (within Europe). \$\endgroup\$ Commented Oct 1, 2023 at 13:16
  • 1
    \$\begingroup\$ The German plug is unpolarised, the Italian plug is unpolarised. The French plug is physically polarised but AIUI the french and the Belgians can't agree on which is the "correct" polarity so it may as well be un-polarised. So appliance designers have to assume that live/neutral may be swapped. \$\endgroup\$ Commented Oct 2, 2023 at 14:43
  • 1
    \$\begingroup\$ And rules aren't always consistent. Lighting in particular seems to get away with things that would be unacceptable in almost any other context. \$\endgroup\$ Commented Oct 2, 2023 at 17:22
0
\$\begingroup\$

I travelled a lot installing equipment so had to get used to all manner of supplies. There is a lot of variation, none of it unsafe BUT!

This is catch all:

Use switches that disconnect all wires. all domestic cables have a neutral so in theory that does not need switching but don't chance it! Someone may have crossed the live phase with the earth. Cable colours vary from country to country so don't chance that either. Ask! There are also some odd ones like 120/120v out of phase to make 240v. That must be dual switched. Biochemistry labs can be a nightmare - especially in the USSR.

Always carry a DVM and a test screwdriver

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

You asked if a load required a double-pole single-throw disconnect?

You betcha. My 220v air compressor, which uses a 220v motor (two magnetic windings, one for each phase) to drive compression. It requires (and uses) a DPST switch because leaving one hot line active, the unit would try it's darndest to fill itself with air when the tank pressure fell below the pressure-switch cut-in pressure. That would quickly damage the motor — if the motor would turn at all. (If it didn't, the active magnetic windings would rapidly burn themselves out. If it did, the stress of driving a compressor expecting both active windings would also burn it out, just a bit more slowly.)

It's worth mentioning that a great many appliances that require 220v service simply don't require switching. Ovens, stoves, furnaces, air conditioners... But, then again, they're all DPST switched at the breaker box, so I suppose you could say all 220v loads must be DPST switched.

\$\endgroup\$
2
  • \$\begingroup\$ The questions was really in the "single circuit" context. Multi phase motors are really multi circuits. I didn't make a note of it because multi phase answers might still be useful for other users. \$\endgroup\$
    – dtech
    Commented Oct 2, 2023 at 3:57
  • \$\begingroup\$ @dtech I've never heard of a single-circuit 220v problem in "Real Life" because the voltage across the two phases is zero. So long as two phases are involved, multi-circuit is all you have. \$\endgroup\$
    – JBH
    Commented Oct 2, 2023 at 13:12

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.