Can I use a mains-rated fuse with a DC application?

[MarkR42]

Hi,

I'm designing a Thing (Robot electronic speed controller) for use with RC Lipo batteries which can put out a lot of current. It's a good idea to use a fuse to prevent a fire in case there is a fault.

I've checked various types of pcb-mounted replaceable fuses, and there appear to be several common sizes (20mm, 32mm), but I did wonder, can I use a mains fuse, for example, a BS1363 UK mains fuse which are commonly found in 10A and 13A ratings (I happen to want the fuse in about this rating anyway).

If I put the footprint on my board for a pair of 32mm fuse holders - and put them slightly closer together - presumably it will hold a BS1363 fuse (nominal length 1 inch / 25.4mm)

Is there any reason why it wouldn't work? The voltage will be about 12-24v

[Siwastaja]

DC is harder to break than AC, that's why you should always verify the DC rating (breaking voltage and current) from the datasheet.

This being said, ceramic sand-filled mains fuses that sit on your distribution box or in UK, in 240V outlet, will definitely be suitable for breaking 24V DC li-ion pack. Some may even have proper DC ratings defined (likely to 100-150V with slightly lower breaking current than at 250AC).

The key is these fuses are designed for breaking high-current circuits, the upstream fuses may be like 50A.

Not all 230VAC rated fused are guaranteed to be suitable, though. Some miniature fuses that are designed to be used inside a device which will be protected by those 10A/13A/16A "distribution box" fuses. Some are just a thin wire in a short glass tube filled with air. These will clear eventually but may produce too much heat and plasma when interrupting very high DC currents.

Generally speaking, 24V DC is still quite easy. Beware though if you go much beyond this. 100VDC coming from a low-resistance battery pack is very challenging.

[Rerouter]

Provided the fuse is rated for that DC voltage (almost all have a DC rating), then there is no issue,

Be aware the DC rating will be much lower than the AC rating, as its difficult to break a DC arc,

Be aware on your size scale you can fit blade fuse holders, and use resettable blade circuit breakers without issue.

[Gyro]

BS1362 standard mains plug fuses are designed to break at least 6kA on mains AC. They contain (the genuine ones anyway) fine Silica sand, which melts to quench the arc when the fuse wire melts.

They will definitely have lower breaking capacity on DC because of the lack or zero crossings in the waveform, but It's hard to say by how much. I am sure that they will be absolutely fine for reliably breaking a 12-24V DC circuit though (I would guess that up to 48V would be no problem). They ought to be much faster to interrupt the circuit than the ordinary bog-standard 'bare metal in a plastic case' automotive blade fuses, and probably cheaper too.

If you need to go higher than 13A rating, there are the BS1631 Consumer unit fuses (16.5kA breaking capacity). They are larger diameter and a bit more pricey though.

[Siwastaja]

You can estimate the short circuit current a li-ion pack provides as approximately 5-10 times the maximum rated current for the pack. An overrated "50C" RC pack, maybe just 3-4x. A carefully rated high-efficiency pack, maybe 10-15x.

If it is a 5Ah pack for example, it's very unlikely the short circuit current would exceed 1000A, especially if you have any extra wiring.

A sand-filled 250VAC 6kA rated fuse is likely rated at something like 100VDC 3kA. Even if there is no rating available, you can be sure it's OK for breaking 24V up to some 2kA.

The ease of interrupting just 24VDC is manifested by the simple design of the classical automotive plastic case blade fuses. Such design would not work with 250VAC, or at 100VDC for that matter. IIRC, such blade fuses are rated at some 30-50VDC and at that point, breaking current rating goes down significantly already so they kind of depend on better upstream fusing and/or circuit resistance.

[nightfire]

To consider: Which parts do you want to protect from damage/overheating? The ordinary household fuses are constructed to protect wires from overheating.
This means, that they allow current to flow for a certain time to heat the material to a point that is non-permanently damaging. After all, the copper (and some part of the insulator) have a energy/thermal capacity and will rise in temperature by absorbing some of the heat generated by the voltage drop over the line.

Do you want to also protect other, more delicate parts, you probably will need a more sensitive (read: faster) fuse- Here you have to define which parts to protect in which situation (slight overload or short circuit/crowbar event) and which time resp. melting integral you want to allow.

[Siwastaja]

All fuses are just to protect against fire.

The "fast" type of fuse which protects "more delicate" parts is a nice idea, but practically speaking, it does not exist.

For example, if an RC motor controller ("ESC") decides to put too much current through its MOSFETs due to design failure, it does that in a microsecond. Even the fastest available fuse type* takes way longer to blow (or orders of magnitude more I^2*t than the MOSFETs can take). Even if you designed an ideal e-fuse circuit that trips in a microsecond, it's usually still too late because even just the DC link capacitors hold enough charge to damage semiconductors. The only place for protection is in the control side of the switch-mode converters, motor controllers etc. itself.

*) so-called "semiconductor fuse", this misleading name actually means: "super-fast fuse designed to protect some very robust and large semiconductors like massive SCRs and diodes"

This is why "fast" types are usually not very practical; they can't protect much more than the slow types anyway, but nuisance fail easier due to capacitor bank inrush, for example.

[xavier60]

I recently went hunting for a DC rated fuse. First thing I noticed was the crazy high prices for most fuses.
I settled on this one,
https://au.element14.com/schurter/8020-0609-h-pt/cartridge-fuse-time-delay-50a/dp/2983535?st=50%20amp%20fuse

[nctnico]

All fuses are just to protect against fire.

The "fast" type of fuse which protects "more delicate" parts is a nice idea, but practically speaking, it does not exist.

From experience I agree with this. Fuses only protect against fire. And not always... One thing to watch out for are inductive loads. When the fuse interrupts the current the reverse voltage spike can create an arc which then continues to conduct and causes a fire. Seen it happen with my own eyes with a system using voltages around 48V and fuses rated for DC use.

[xavier60]

All fuses are just to protect against fire.

The "fast" type of fuse which protects "more delicate" parts is a nice idea, but practically speaking, it does not exist.

From experience I agree with this. Fuses only protect against fire. And not always... One thing to watch out for are inductive loads. When the fuse interrupts the current the reverse voltage spike can create an arc which then continues to conduct and causes a fire. Seen it happen with my own eyes with a system using voltages around 48V and fuses rated for DC use.

That is my concern also. There is a series 30uH choke and the supply is 74V battery. I have a fuse on each line and 100uF of bypass capacitance between the the choke and the fuses so that the voltage kick from the inductor will be slowed enough to allow the fuses to interrupt. Hopefully.

[Siwastaja]

Remember that a circuit consisting of a DC supply + series inductor is also known as "welder".

A classic non-inverter welder is a transformer and rectifier; from 3-phase supply it's basically DC even without capacitors. Welding transformer has leakage inductance on purpose, or an explicit inductor is used. The inductor is there so that there is a boost converter action: short the output by touching the stick / mig wire to the ground, the inductor charges up, at the moment you lift a high voltage is generated, initiating the arc. Then the relatively small DC voltage around 30-50V is enough to keep the arc going!

Imagine what you can do with a welder. Then think about how difficult it is to design a fuse with good DC ratings. The sand inside is critical.

[JohanH]

Another aspect of fuses and DC is that for very low currents, the resistance of the fuse might become a problem. E.g. if you'd want to have a 200 mA fuse in a small DC circuit, it will have a typical resistance of 8-9 ohms (regular 250VAC 5x20 fast glass fuse). That will for sure impact circuit design. So using fuses for low current DC isn't typically done (or use a larger fuse).

[wizard69]

As others have said you need to read and understand the manufactures data sheets for a specific fuse.   You are right to want to protect from fire as that is a real risk, but getting suitable protection is not a given.    If this is a commercial product it might be worth looking for professional help so that you can meet regulatory requirements.   By the way fuses have to be rated to be able to handle the short circuit currents (interrupt rating) that the source might supply along with the proper voltage rating.

As for the fuse and where you put it, I'm not a big fan of mounting high current fuses/holders on printed circuit boards.   Fuse holders, especially the cheap spring type, have a habit of failing thermally.   Even panel mounted holders can have suspicious ratings and I blame that on imported stuff that clearly can't deliver what is stamped on the package.   So what ever you do for a fuse holder make sure it is robust enough to do the job, and buy from a reputable manufacture.   I only mention this due to seeing multiple failures on purchased hardware, often AC line in, where the holder couldn't handle even a fraction of the load it was rated for.   In some cases we ended up sending brand new instruments back to the manufacture.   So what I'm saying is buyer be aware and stay with reputable suppliers.

One issue with fuse holders, on PC boards, is that they can be significant heat generators.

It may be worth considering automotive type fuses but these are commonly supplied with a low voltage rating.  ATC fuses for example are rated for 32 VDC at 1KA interrupt rating.   If they can meet your safety needs automobile fuses have a couple of advantages, they are easy to find and fuse holders come in a variety of types including inline and PCB mount if you want that.   Your big problem is figuring out what the short circuit capacity of your battery is to determine if it is under the interrupt rating.   Here is an ATC data sheet: https://www.eaton.com/content/dam/eaton/products/electrical-circuit-protection/fuses/bussmann-series-supplemental-fuses/automotive-blade-type/bus-ele-ds-2009-atc.pdf.   You can get some automotive style fuses in 58 volt ratings designed for 42 volt systems but you still have the same current limitations.   You would need to determine if the fuses meet all of your other technical requirements.   You will need to look up the specific data sheet for the version you might what to use but some of these automotive fuses are UL approved for "control circuit" usage.

[sam_sam_sam]

Have you thought about an electronic fuse type device they do make them and or you can build one it use a shunt and a micro controller with a relay to control the current

I have used them in the past and for the most part they work very well

But what you want to do is to size it based on your inrush current so this would take some time to figure out what your specifications are going to be and then what your normal current would be and configure your micro controller program for these two specifications

For example

When I troubleshoot ( ATX or ) switching power supply I use an incandescent light bulb and a electronic fuse controller and it function very well

There is one drawback and that it sense the inrush current a little to well at times and can be very anointing to deal with but I would rather deal with this than let all of the factory smoke all over the place