# Placement of fuse on neutral wire. Why?

I have come across several occasions on this website where a circuit is shown with a fuse on the neutral wire rather than the line wire of mains power. For example

from this question.

I can think of good reasons why the fuse should be on the line side. If a fuse on the neutral side blows, and the building does not have ground fault protection devices, then current could still flow from the line to the ground. [Of course, if the fuse were on the line side, and it blows, current could flow between neutral and ground, but the voltages are much smaller] So, it would seem that a line side fuse offers more protection than a neutral side fuse.

On the other hand, I can't see any advantage to fusing the neutral side rather than the line side.

Of course some particular circuit might be fully enclosed in an insulating container. While that, at least somewhat, counters the safety argument for fusing the line side, it doesn't seem to provide any benefits to fusing the neutral side rather than the line side.

So, my question is, am I missing something? Is there any reason to fuse the neutral side rather than the line side? Or are those who designed these circuits just missing an opportunity to make their circuits possibly safer with no compensating benefit?

So, my question is, am I missing something?

Yup.

You're correct: a blown fuse on the neutral means the circuit is still live rather than a blown fuse on the hot where the circuit would be dead. It is always better to fuse the line side.

Q: How do you tell which side is hot?
A: We can't.

Graphs and schematics are nice but the piece of the puzzle you're missing lies with J2. A simple power connector, part number 770W-X2/10, is an IEC 320-C8, Non-Polarized receptacle. It doesn't matter which side the fuse is one, they're both quasi-line-side. If the connector were plugged in such that the top was 'L' and the bottom was 'N', great. However, it's equally likely that it is flipped and the top is 'N' and the bottom is 'L'. There's no real benefit to be gained here since we don't know.

As for why the designer didn't just draw the fuse on 'L' -- I don't know. Maybe to make you question your knowledge on fuses.

Or are those who designed these circuits just missing an opportunity to make their circuits possibly safer [...]?

Using a polarized connector they could guarantee the line side is fused. For clarification, this would also require that the plug is polarized as well. Not all countries use polarized plugs. Although it is "Texas" Instruments, they design and sell globally.

I guess a bigger point here is that we're misunderstanding the circuit as a whole. What exactly is this circuit for? It's a 50/100W Audio Power Supply but if we pull it up on TI we can see that this circuit is part of a product that is for "testing and performance validation only, not available for sale". The PDF further labels it as a reference design. So, we do need to take TI's disclaimer into account here:

These resources are intended for skilled developers designing with TI products. You are solely responsible for (1) selecting the appropriate TI products for your application, (2) designing, validating and testing your application, and (3) ensuring your application meets applicable standards, and any other safety, security, or other requirements.

It is not TI's intention that you go plugging this in around the world amazed that it works everywhere, rather, you decide that it is going to work with 120 V in North America, and you would adjust J2 to be something like an IEC C14. If you were to use this circuit at 220 V in Europe and design it around a CEE7/16 Europlug, you'd want to add another fuse since that's basically two hots.

• So in this case, the labeling of TP2 and TP4 as "L" and "N" is a bit misleading. Jan 15 '21 at 5:21
• As far as the unit is concerned, there is no earth as it is double insulated. If someone opens the unit up and touches the live circuit, then the gfi will trip. Regardless of whether the fuse is intact or blown and be in on either conductor makes no difference to safety - to fuse is to protect the wire, not for someone touching what they shouldn’t Jan 15 '21 at 5:26
• Do note that as in some countries plugs are symmetrical, there are no polarized plugs. Jan 15 '21 at 6:22
• FWIW, my impression is that dual fuse holders (L and N) are now standard. Jan 15 '21 at 13:19
• I would put it on N in the schematic to remind people that even when its blown the circuit is live Jan 15 '21 at 16:55

I can see that your circuit is a section of a flyback power supply. Certain things give that away and, the general rule is that a fuse is needed to prevent imperfect line-side components forming a short and creating a fire. The fuse is there to protect infra-structure (to reduce the risk of fire) and it is irrelevant to which of the incoming AC wires it is connected. It does its job perfectly well in either line.

If a fuse on the neutral side blows, and the building does not have ground fault protection devices, then current could still flow from the line to the ground.

Given that the design (see link further below) uses a proper flyback transformer that isolates the output from the incoming AC, it's got nowhere to flow other than at minor sub-mA leakage levels.

It looks like the circuit comes from here and also from this question.

On the other hand, I can't see any advantage to fusing the neutral side rather than the line side.

Given that the input connector (770W-X2/10) is unpolarized: -

Then it is likely that the fuse placement will sometimes appear on the neutral and sometimes on the live i.e. not being able to see any advantage is also not being able to recognize that the input connector is unpolarized.

• "Of course some particular circuit might be fully enclosed in an insulating container. While that, at least somewhat, counters the safety argument for fusing the line side, it doesn't seem to provide any benefits to fusing the neutral side rather than the line side." Jan 15 '21 at 15:16
• No, I was going to write more, but forgot that "return" is not allowed in a comment. Jan 15 '21 at 15:18
• More precise would be that the fuse protects a) the line cord which can't handle the same ampacity as wall wiring, and b) the unit itself from becoming an 180W heater and starting stuff on fire. Jan 15 '21 at 19:12
• The fuse is there to protect infra-structure (building wiring) Absolutely not! It would be absurd for the building to rely on devices connected to it to protect its own wiring. The building wiring is protected by circuit breakers in the distribution panels. Fuses protect the equipment you find them attached to, not the power source they feed on.
– J...
Jan 15 '21 at 20:39
• @Harper-ReinstateMonica Andy's in the UK, so even that's not true - the Brits' venerated BS1363 plugs are themselves fused to protect the appliance cord. In the US, appliance cords have to be able to take the full circuit current. Smaller gauges are allowed because derating is different outside the wall. Fuses inside the device only protect the device. Just like a circuit breaker can't (fully) protect anything upstream, neither can a fuse fully protect an appliance cord upstream. A short between TP2 and TP4, for example...
– J...
Jan 15 '21 at 20:41

## El NEC prohibits fusing neutral

Mind you NEC does not bear on equipment, that's the domain of Underwriter's Laboratories and its various product standards.

However in building wiring, fusing neutral is forbidden. The only exception is if the if neutral has common trip with all hots, which really means we're talking about circuit breakers, not fuses.

The reason is what your intuition says: the equipment is left floating at "hot" voltage, and fusing neutral does nothing to arrest a hot-earth fault or prevent chassis from being energized and shock people if it isn't earthed.

## With cord-and-plug connection, you often have no choice.

Look at a polarized NEMA 1 socket, which has a tall neutral. Or NEMA 5 or UK BS1363, provided the socket has blocks to prevent reverse insertion*. Those assure neutral is where the installer thinks it is.

However, some countries have non-polarized sockets - consider the Europlug or the Schuko. If you're marketing into those countries, you simply have no choice. You have a 50/50 chance of either leg being the hot one.

Having 2 fuses would be bewildering and counterproductive to the consumer, and to blow simultaneously you'd need a 2-pole circuit breaker that would add considerable cost to consumer appliances. So the standards simply call for the machine to be insulated for any leg being hot. Assuming standards are even followed - a real question on equipment with no NRTL mark and a self-certified CE mark when the maker/distributor is outside the EU.

* Example of smart sockets guarding ground pins to prevent reverse-polarity insertion, or failing to :)

Sometimes an appliance can have a non-polarized plug, so you can plug in into wall sockect in any orientation.

Therefore by the time an appliance is connected to mains, there is no guarantee which input is live and which input is neutral, so it really does not matter which wire has the fuse.

The appliance mains parts still need to be isolated from touching such as a plastic case (e.g. laptop PC, mobile phone), or grounded metal chassis (e.g. desktop PC).

• A non-polarized plug (or receptical in this case) does give the end user the option of plugging in either way. However, that seems a rather paltry "benefit". Jan 15 '21 at 15:26
• @MathKeepsMeBusy it's one of the reasons we have USB-C now. It's nice having options. Jan 15 '21 at 15:46
• @Arsenal, I plug and unplug my phone charger daily. Devices that are plugged into my wall sockets, hardly ever. Jan 15 '21 at 16:10
• Yeah, but USB is infamous for having some sort of registration problem where "correctly oriented but misregistered" felt exactly like "incorrectly oriented". Thus the "try, flip, try, flip, try" we've all experienced with USB. Jan 15 '21 at 19:24
• @MathKeepsMeBusy I always unplug the charger from the wall when not needed. 0.3 W don't matter much but it's not really an effort doing it and it can't go up in flames while I'm not around. And if everyone is doing it, it saves quite a bit overall. Jan 16 '21 at 17:44

In case the fuse holder is accessible to touch? (due to obsolete design or some defect) ... according to discussion on this home improvement website

Plz don't take this possible reason as a recommendation for fusing only N. If any doubt, fuse both L and N, and use an insulated fuse holder designed to disconnect before fully pulled out (which I believe is now a standard function in double-fuse-holders).

• This seems to be the most reasonable explanation so far. However, I note that one of the answers claims that: "In the US if you fuse or switch only one wire it has to be the hot wire." Jan 15 '21 at 4:38
• @MathKeepsMeBusy I think Pete's right. Most of my older instrumentation includes an externally accessible, replaceable fuse holder with a screw-cap. You just open it up, drop out the old fuse, replace it with the new one, and cap it. Putting that on the hot could very well be dangerous as one inserts the replacement fuse with their finger pushing it into the spring-loaded holder. I can easily imagine some people failing to unplug their instrument before replacing a fuse. It's a question of what's the dominant safety issue.
– jonk
Jan 15 '21 at 6:05
• @Math Keeps Me Busy -- yes, I was pretty skeptical of this answer myself, at least for present day equipment. But it might still be an important point for lab stuff cooked up by grad students etc. // When needing to build mains powered equipment I have almost always used off-the-shelf "power-entry-modules" (filter+fuse+switch), and these seem to always have polarized plug, some clever plastic interlock to avoid exposure, AND double fuse, both L and N. Don't see any reason not to do the same. Jan 15 '21 at 13:31
• Although I think in some cases, as in the link you provided, the fusing of the neutral is intentional, and the rationale is related to humans touching the fuse, I think that for the example circuit I posted, C Lange's answer is right. So, awarding him the answer, but find yours valuable too. Jan 15 '21 at 16:07