10 amp fuses - quick moan

[williamwright]

So what's your problem with Himberg?

Bill

[Scott]

Okay then: on that basis, when the system started there were 1A, 2A,

3A, 5A, 7A, 10A and 13A fuses. Does this mean there were seven different amperages of cable in use? If not, I think your theory falls.

[Scott]

I wish I had known this when I was looking for red and black PVC cable.

[Owain Lastname]

Big metal thing that's supposed to perambulate around the villages every half-hour (except after 5pm, on Sundays, if the driver's got covid, if the bus has broken, if there's a y in the day or an a or u in the month) before ending up round the corner from Screwfix :)

Owain

[Animal]

thousands of amps in fact

[ARW]

Greater than 12kA at my house supply.

[ARW]

OK, so you chop into a 6A 1mm lighting circuit by accident.

What is the limiting factor on the fault current that passes through your snips and the 1mm cable?

[4587Joey]

No it doesn't because there are also fuses in the appliance itself

[Animal]

lol. This topic comes up over & over & folk are seldom willing to learn.

[Scott]

Are you saying the fuse in the appliance itself is interchangeable with the plug fuse and could be 1A, 2A, 3A, 5A, 7A, 10A or 13A? Personally I have never seen an appliance in my entire life that contains a cartridge fuse that can be replaced using a standard plug fuse.

[Scott]

Well, I think I learned at primary school that appliances that draw very little current such as lamps and radios were fitted with 2 amp or

3 amp or 5 amp fuses and items drawing large currents such as radiators and washing machines were fitted with 13 amp fuses. No-one suggested looking at the flex instead of the appliance. Obviously the manufacturer would need to fit a flex that was adequate for the current or else it would heat up and in these days I think you would notice the smell of burning rubber.

[John Walliker]

Well, it gets more complicated when a lot of devices are connected with IEC (C7, C13 or sometimes C15) mains connectors, so the plug is independent of the appliance. Also, most products are designed for use in many countries, so they have to be safe when there is no mains fuse in the plug and the wall socket is expected to be protected at 16A. What is really odd though, is that if you have a new product which uses an IEC mains inlet safety tested at an approvals lab they insist that you provide a sample of the mains cable that you intend to supply with the product and they test that even though the customer might use a completely different cable. Also, many products do not have an internal fuse even though they do not rely on any external protection apart from that provided by the fixed wiring which might be up to 32A. They rely on intrinsically short-circuit proof mains transformers. Adding a fuse is a disadvantage because that is an extra component that has to have a certification paperwork trail.

John

[John Rumm]

With historic kit, the logic does not apply.

With stuff confirming to current standards (say < 30 years old), then the device itself *should* have its own fuse if it needs additional overload protection. (the exception to this rule usually being marginal extension leads - because there is no "device" as such - just wire that might be coiled up).

If the device has its own fuse, then the only thing the fuse in the plug must do is provide fault protection for the flex (since you can't rely on the 32A MCB in the CU to necessarily do that)

They should be - although possibly not guaranteed on some of the more budget end of the chinesium variety.

[John Rumm]

At a minimum it must be able to delver in excess of 160A of fault current if it is to be able to operate a B32 MCB in the magnetic (i.e. "instant" part of it's trip curve). Hence why there are rules about maximum earth loop impedance at the far ends of circuits.

(remember also that fuses and MCBs etc don't limit fault current - they just limit the duration of the fault)

[John Rumm]

When the system started the standards were different to now. You may still find some old appliances that have potential overload conditions, and don't have internal protection. Hence the "right" fuse is required for those.

The protection required is not just a reflection of the CSA or the construction materials of the cable, but also the earth loop impedance at the position of the fault. So even a relatively substantial flex could be a fire risk if too long.

For modern kit, an adequate fuse is required (i.e. if it will draw a 7A continuos load, then fitting s 3A one will not work), but generally it is designed to remain safe even with a 13A fuse fitted, even if the "right" one would be 3A. i.e. the stuff needs to be engineered to assume that the user is clueless and might stick in a 13A fuse without giving it a second thought.

Hence why on appliances with 0.5mm^2 flex they are usually only fitted with a maximum of 1.8m of the stuff (that ensures the loop impedance is low enough to give a large enough fault current to open the fuse quickly

- and that controls the final conductor temperature of the flex to keep it below a point where it could fail in a spectacular way).

[John Rumm]

Mostly true, although there are appliances that draw relatively low nominal currents, but have a high inrush current, and so need fusing at a higher level than the nominal power consumption would indicate. (many CRT TVs would pop a 3A fuse even if they only drew < 200W in operation)

There is adequate and adequate. What might be adequate for normal operation, might not withstand the fault current caused when someone sticks a chair leg down on the flex near the appliance and causes a dead short. The take home point, is that under fault conditions the nominal load of the appliance does not factor into it at all. It all comes down to the total round trip resistance though the supply to the fault.

With fixed wiring such concerns have been addressed by specifying a maximum permitted earth loop impedance at the furthest reach of the circuit. For a circuit with a B32 MCB that is specified as 1.37 ohms.

That means that even at the extremes of the permitted supply voltages, you will get a prospective fault current close enough to the the lower threshold of the 5 x In "instant" trip current of a B type 32A MCB (i.e.

160A) to clear a fault quickly.

So you can show by calculation that the circuit cables will withstand that fault current for long enough for the MCB to clear the fault by applying the adiabatic equation[1]:

s = sqrt( I^2 x t ) / k

s = sqrt( 230^2 x 0.1 ) / 115 = 0.63mm^2

So you need at least that area of copper to ensure it fails safe. Since a ring circuit wired in 2.5mm^2 T&E has a CPC of 1.5mm^2 (or twice that with both legs acting in parallel) you have plenty of area.

Now think about what happens when you have an appliance with only a

0.5mm^2 flex plugged into that socket. If it were not for the fuse in the plug, the flex will likely fail. However fuses have the advantage of opening faster than MCBs at high fault currents. So say you have 2m of flex with a round trip resistance of around 18 mOhms/m, you add 36 mOhms to the 1.37 of the circuit, giving 1.41 ohms. So a fault current of 230 / 1.41 = 163A.

If you go find that on the fusing curve for a 13A BS1362 fuse:

You get a worst case opening time of around 0.06 secs.

So, S = ( 163^2 x 0.06 ) / 115 = 0.35^2mm

Just over half the area that would be required without the fuse, and comfortably less than the 0.5mm^2 available.

Now imagine someone has extended the flex on their curling tongs so they can be plugged in in the bedroom, but reach the bathroom. Now you can get cases where the PFC falls, and the fusing time increases, and a 13A fuse will now not clear a fault fast enough, but a 3A probably will. So there will be some classes of device where the combination of long flex and high inrush mean it is impossible to protect them adequately.

[1] Full detail here:

It can be enlightening to play with the numbers for difference scenarios and different PFCs.

[John Rumm]

The joys of living next door to a sub station, reassuring to know that many domestic MCBs can only guarantee to interrupt 6kA :-)

(still you probably have spare BS88 incomer fuses to hand in case something goes really pear shaped!)

[John Rumm]

It depends on the appliance and the type of fuse it uses. In most cases not usually, since many will have 20mm glass cartridge fuses internally and not BS1362 cartridge fuses.

The glass fuses are fine for overload protection for the appliance, but can't protect the flex due to its position, and won't have the breaking capacity to clear a likely fault current.

I do have some devices with internal BS1362 3A fuses. Some extractor fans, and IIRC one audio amp with a fairly substantial toroidal transformer.

Keep in mind that most appliances will be sold in other 230V countries that don't use plug uses at all, and hence the flex will need to be safe when protected by only a 16A MCB.

[Mark Carver]

It's not my theory m'lord. You need to take your case up with the residents at

2 Savoy Place, London, WC2R 0BL

[charles]

I suspect those who came up with all those fuse values are no longer there.