How to read the fuse charts (I^2t)

[Siwastaja]

First row in in "Clearing Time Characteristics" at ambient temperature and rated current, shows minimum clearing time as 4 hours, while maximum clearing time is not specified (it is not being tested).

That's kinda weird way to rate fuses, though. I think it's more usual to guarantee the fuse never blows at rated current.

The problem with this Bourns part is, the provided data does not allow the part to be designed in at all. There is no tabulated rating to tell the designer which current the fuse can be used at. Curves are only typical. Tabulated values show min/max, but the current rating where Min time is infinite (or years) is missing.

This is a good reminder to everyone: always read the datasheets instead of assuming. Conventions in electronics are weak. But this also teaches us: while you really can't assume, you have to assume anyway, because the data provided by manufacturer is not sufficient to use the product.

[Alti]

That's kinda weird way to rate fuses, though. I think it's more usual to guarantee the fuse never blows at rated current.

No, it is not more usual.

Traditionally Old World characterizes fuses that never trip at rated current while 'merican designs are not guaranteed to operate at rated current indefinitely. Those Bourns fuses from opening post are made to comply with UL 248-14 standard so it would have been surprising if these "guaranteed the fuse never blows at rated current". This "weirdness" is not a standard but just a tradition from some Edison era. That is one of the reasons you cannot replace 1A 'merican fuse with 1A IEC fuse - fuse standards do not overlap.

If you want more "weirdness" - Littelfuse Fuseology Selection Guide.

According to the IEC 60127 Standard, a one ampere-rated fuse can be operated at one ampere. A one ampere-rated fuse made to UL/CSA/ANCE 248-14 should not be operated at more than .75 ampere (25% derated — See RERATING section).

Anyway, weirdness or not weirdness - that is a separate subject.
Main problem is that if a datasheet does not state some property holds, then it simply does not.
This is a generic rule, nothing specific to fuses.

[Vovk_Z]

Hi. Did you bother opening the pdf from opening post?

First row in in "Clearing Time Characteristics" at ambient temperature and rated current, shows minimum clearing time as 4 hours, while maximum clearing time is not specified (it is not being tested).
Your response: this fuse must not melt at rated current.
Then contact Bourns and tell them they make lousy fuses as some burn in fifth hour at rated current.

Hi! I haven't seen this exact pdf. I just say typical glass fuse behavior (as I know it). Possibly something changed in different types of them. I don't want to dispute with Bourns.  I wish they were more specific about 100% behaviour but ok.

[Nusa]

But one must remember fuses are a safety devices, and the value picked should be higher than the expected normal draw of the device in question, but low enough that they'll blow before you get flames. Anyone who sizes the fuse to be exactly the same as the expected draw of the device has literally designed in a problem. Especially if they haven't tested in worst possible conditions (high ambient, sagging/over voltage, heavy duty use, etc).

[Siwastaja]

Anyone who sizes the fuse to be exactly the same as the expected draw of the device has literally designed in a problem.

As Alti explained, clearly in the US. of A, yes, which I was not aware of!

As per IEC standards, one can design a fuse to carry nominal rated current indefinitely - no problem.

This is just an arbitrary choice someone has made. Both ways of thinking are valid, but the USA way would require the fuse manufacturer to separately specify the current which can be used (equivalent to the European In, just under different name), and as the Bourns example shows, they don't always bother doing that (and then you need to assume some "derating" factor)! With our European fuses, this is simply the rated current, so one number less to specify.

Note this has nothing to do with fuse construction itself, curve shape or anything like that. Just a maybe ~25% discrepancy in nominal rating.

What to call "nominal" is always a tad problematic. For example, tantalum capacitor manufacturers define nominal voltage as something which will destroy the product in short time, so that significant derating is always needed, while aluminum electrolytic capacitor manufacturers chose nominal voltage as something that the part can be exposed to indefinitely.

[Nusa]

Being aware of the relevant standard is beside my point, although of course you should read the datasheet. My point is that you should always leave some overhead. If the question of how long it will last at the exact rating becomes important, you've failed at fuse selection and made product that may not be reliable. Which ends up costing money and reputation on the customer service side.

[Siwastaja]

Being aware of the relevant standard is beside my point, although of course you should read the datasheet. My point is that you should always leave some overhead. If the question of how long it will last at the exact rating becomes important, you've failed at fuse selection and made product that may not be reliable. Which ends up costing money and reputation on the customer service side.

I prefer the mindset of doing actual worst-case analysis, instead of leaving "some margin", but sure, the "some margin" way is used when there is not enough time, expertise or information to do proper worst-case analysis. It's important to note the "some margin" way is more risky, and failures happen when the assumed amount of derating was not sufficient.

In actual reality, combination of best effort worst case analysis and some additional derating is needed.

But IEC-style rated fuses do not need any extra derating. They are fine when used at In, but of course then you make sure you don't exceed In.

I think this is an important distinction. You need to understand why you are doing things. Some components require extra derating, some do not. This goes in addition to the derating you need to add due to the uncertainties of the rest of the circuit (e.g., tolerances of voltage supplies, load currents, etc.), and your generic "safety margin" which you apply as a last step. If you get too comfortable just adding a decent amount of generic margin and call it a day, components which require a lot of extra derating could catch you off guard.

[Nusa]

Certainly you should apply intelligence when selecting overhead. What I said is that not having any overhead is a problem. Because there's really no such thing as walking the line in real life...you're either under it or over it, not exactly on it.

[Alti]

But IEC-style rated fuses do not need any extra derating. They are fine when used at In, but of course then you make sure you don't exceed In.

If we talk about IEC 60127-4 (closest to UL 248-14 of Bourns from opening post) then test is specified at 1.25In for 1h min., in contrast to  1In for 4h min. for 'merican UL 248-14.

If you know that silver link melts at 960oC and it is more than certain a fuse is designed to barely survive that at 1.25In for 1h, keeping it at 600oC even for a day smells like asking for trouble. Does it guarantee "infinite" operation at 1In? I doubt it. It looks like it is further from the border line than 'merican but in practical application when you include elevated ambient temperature and thermal cycling, I wouldn't hang 1A on a IEC 60127-4 fuse with In=1A for "infinitely long". These are not optimized for dealing with 1In but for small size and decent reaction for short-circuits (i2t).


Then there are serious IEC 60269 fuses that are well known from industrial applications but are also quite common in residential. The most popular miniature are IEC 60127-2 that some call "glass fuses". Both of those standards are characterized at 1.5In for 1h minimum.

[Siwastaja]

Then there are serious IEC 60269 fuses that are well known from industrial applications but are also quite common in residential. The most popular miniature are IEC 60127-2 that some call "glass fuses". Both of those standards are characterized at 1.5In for 1h minimum.

I am thinking about these; they come in variety of form factors and are used extensively in households, industrial and also equipment. Google image searchs shows the typical form factors: https://www.google.com/search?q=IEC+60269&source=lnms&tbm=isch

These fuses are routinely used at In. Grey area is between In and 1.5*In.

[Vovk_Z]

Does it guarantee "infinite" operation at 1In? I doubt it. It looks like it is further from the border line than 'merican but in practical application when you include elevated ambient temperature and thermal cycling, I wouldn't hang 1A on a IEC 60127-4 fuse with In=1A for "infinitely long".

As for me, I haven't seen fuses used with In actual current. We usually choose them as 1.5-2.0x from the working current because we have to deal with inrush transients.