Fuses do affect sound, the question is how much

[iwantobelieve]

Thanks. Could you elaborate on what exactly is the mechanism which causes the fuse to fail when too much current(?) is drawn through it?

I only realised while having this exchange that I don’t actually know. I had always assumed that it was the higher resistance of a relatively narrow gauge of wire/foil in a fuse which caused it to overheat and blow before anything else (including the AC cord), breaking the connection when too much current was drawn across it. It now seems like this assumption can’t be correct and that another process is at work. I’d be grateful to know, it’ll really annoy me until I find out now.

A Google of ‘how a fuse works’ or ‘why does a fuse fail’ produced nothing to enlighten me further, or I wouldn’t bother you folks here for an explanation.

 

[mansr]

Could you elaborate on what exactly is the mechanism which causes the fuse to fail when too much current(?) is drawn through it?

It heats up and melts.

 

[solderdude]

Fuse (electrical) - Wikipedia

en.wikipedia.org

https://www.littelfuse.com/~/media/electronics/application_guides/littelfuse_fuseology_application_guide.pdf.pdf

 

[SIY]

I had always assumed that it was the higher resistance of a relatively narrow gauge of wire/foil in a fuse which caused it to overheat and blow before anything else (including the AC cord), breaking the connection when too much current was drawn across it. It now seems like this assumption can’t be correct...

It is correct, but higher resistivity, not resistance. Resistivity in this context is resistance per unit length. A fuse might be, oh, 2 cm, whereas a line cord is more like 2 meter here-and-back. So you might get half an ohm from the line cord and only a tenth of an ohm from the fuse (numbers pulled out of the air for illustration), but that tenth of an ohm, and the power dissipated in it, is only distributed along 1/100 of the length of the line cord. Add in the lower melt point of the metal used and what you have there is a fuse! In olden times, people would often use a single strand of fine copper wire a couple centimeters long as a fuse (Futterman was notorious for that).

 

[Killingbeans]

Was going to write something about thermal dissipation, but SIY explained it in a better way

 

[Ingenieur]

Some things to consider:

System Z, basically what limits short circuit current at the receptacle, a good indication of transient response. I've measured these.
It works like this, measure no load voltage, apply a load, measure V and I, from V drop and I calcs Z. Typically 1000 A on a 120/20 A circuit.
At 120 V ~ 0.12 Ohm
How much can a 0.01 Ohm change if a fuse change that?

At peaks of 500 W the amp may draw:
PF ~ 0.3, eff ~ 0.5
500/(0.3 x 0.5 x 120) ~ 28 A
1,000 >>>> 28 A
Even if fuse delta cuts that in 1/2, moot.

Also, when operating at this 28 A, input Z is 4.3 Ohm, much greater than the fuse Z.

The C (Xc) of the power supply is >>> L (Xl)
time constant = RC
Assume PF = 0.3 while charging
Z = 4.3 Ohm
R = 0.3 x 4.3 ~ 1.3 Ohm (Xc ~ 4.1 Ohm)
At 100 Hz C ~ 400 uH
time constant ~ RC ~ 0.5 mSec, 2000 Hz
% C charged ~ 100 x (1 - e^-(time/RC))
1 TC ~ 63%
4 TC ~ 98% , 2 mSec or 500 Hz
The PS recharges 500/100 ~ 5 X faster

But PS are sized to only deplete 10% under max rated load, so only needs 10% charge: ~ 0.1 TC, 0.05 mSec, 20,000 Hz in our case.


Food for thought

 

[MakeMineVinyl]

Can you explain to me why the resistance in my AC cord is higher than the fuse in the plug, though?

Have you ever touched the AC plug from your vacuum cleaner or heater, and sometimes the AC cord itself after using it for awhile? You'll notice that it is warm - this is due to resistance of the contacts and cable which equals voltage drop. A power amplifier playing normal program material draws far less full-power current continuously, so the voltage drops are less. When a >1000 watt amplifier is tested at full power for an extended period of time, it's AC wiring will get hot too, as will everything else in the path such as the power transformer, rectifiers etc.

 

[mhardy6647]

It heats up and melts.

yup,
Under short circuit conditions, the fuse element usually sort of vaporizes.

 

[mhardy6647]

speaking of Ohm's law and resistance and resistivity and heat and conflagration and death and suchlike -- in the good old, bad old days, there were numerous, low cost, series-string "killer" radios built using 'resistor wire power cords" to balance the equation for voltage drop through the circuit, specifically across the series string of vacuum tube filaments heated off of the AC mains.

RadiolaGuy.com : resistance line AC line cords in radios

AC line cord problems in ealry AC/DC radios

www.radiolaguy.com

Nostalgia Air: Replacing Defective Resistance Line Cords

Nostalgia Air: Replacing Defective Resistance Line Cords

www.nostalgiaair.org

These resistance line cords were also popularly known as curtain burners.

It's a wonder anybody survived the first half of the Twentieth Century.

 

[MakeMineVinyl]

speaking of Ohm's law and resistance and resistivity and heat and conflagration and death and suchlike -- in the good old, bad old days, there were numerous, low cost, series-string "killer" radios built using 'resistor wire power cords" to balance the equation for voltage drop through the circuit, specifically across the series string of vacuum tube filaments heated off of the AC mains.

RadiolaGuy.com : resistance line AC line cords in radios

AC line cord problems in ealry AC/DC radios

www.radiolaguy.com

Nostalgia Air: Replacing Defective Resistance Line Cords

Nostalgia Air: Replacing Defective Resistance Line Cords

www.nostalgiaair.org

These resistance line cords were also popularly known as curtain burners.

It's a wonder anybody survived the first half of the Twentieth Century.

Never knew about that. Truly scary.

 

[mhardy6647]

Yeah. Us "old radio" guys do, occasionally, run into one of those. The thought of it's terrifying -- as if the thought of the transformerless, AC-DC series string circuits hadn't already been terrifying enough

 

[egellings]

How much would a competent line fuse decrease dynamic range in an amplifier? A fraction of a dB? Would it matter all that much?

 

[MakeMineVinyl]

I would be surprised if it was even measurable, especially considering the 800 pound gorilla of losses elsewhere in the incoming power system. If somebody is suspicious that the fuse in their amplifier is 'compressing' dynamic range, they're always free to replace it with a much larger value fuse or a dead short for that matter. Safety would be compromised of course so I wouldn't recommend it.

 

[Ingenieur]

How much would a competent line fuse decrease dynamic range in an amplifier? A fraction of a dB? Would it matter all that much?

Let's compare it to the wire from your panel to receptacle. Assume 40' LF of 12 AWG ~ 0.14 Ohm
Ignoring cord, connections, etc.
5 A fuse 0.021 Ohm
Assume 2 A draw
Vdrop feeder ~ 0.28 V
Vdrop fuse ~ 0.042 V
-16.4 dB or 0.3%

BUT, it is mostly resistive and has little affect on transient response.

https://www.eaton.com/content/dam/eaton/products/electronic-components/resources/data-sheet/eaton-gma-time-delay-glass-tube-fuses-data-sheet.pdf

 

[egellings]

A fuse in the power line of the amp would have little if any discernable effect on SQ outside of a microscopic amount of max output power loss. A fuse in the speaker line could have an effect on the sound quality caused by sticking a tiny amount of resistance in series with the amp & speaker. A speaker fuse that is inside the amplifier's feedback loop would have no effect on SQ at all.

 

[iwantobelieve]

Thanks guys, with your comments and some more research I think I'm more enlightened on this now.

I have large active speakers with 4x 150W per speaker plus associated electronics, so probably 700W each in total, as an estimate, at max power, so it's good to know the mains fuse isn't limiting anything.

 

[Holmz]

A fuse in the power line of the amp would have little if any discernable effect on SQ outside of a microscopic amount of max output power loss. A fuse in the speaker line could have an effect on the sound quality caused by sticking a tiny amount of resistance in series with the amp & speaker. A speaker fuse that is inside the amplifier's feedback loop would have no effect on SQ at all.

I totally thought it was all “super BS”, but at least the speaker fuse is a non linear effect amd it is on the higher current speaker side.
So I can see that on a speaker side, then maybe a case can be made… The mains side I still see as “Super BS”.

 

[audio_tony]

I totally thought it was all “super BS”, but at least the speaker fuse is a non linear effect amd it is on the higher current speaker side.
So I can see that on a speaker side, then maybe a case can be made… The mains side I still see as “Super BS”.

A fuse will only become non linear (very) close to the point of rupture. Before that, it's simply a piece a wire with a miniscule (and irrelevant) amount of resistance.

Rotel were one of the early designers that incorporated the (speaker) fuse into the feedback loop of their amplifiers, and this would cancel any distortion introduced by the fuse - but as I stated above - any distortion would only become relevant if the wire in the fuse was heating sufficiently (putting it close to the point of rupture), and playing the music loud enough to bring the fuse close to the point of rupture would likely be putting a lot of stress on the speakers (and one's ears!). Most fuse manufacturers provide resistance specifications at 'n' current.

As it happens, for several years now, Rotel have been using relays in the speaker protection circuit and staying with fuses in the power supply (in their class a/b models anyway).

As for mains fuses - if any device is stressing the fuse close to the point of rupture in normal operation, that fuse is probably slightly underrated for the required current. However, this would still have no discernible effect on the the sound quality, and there would only be a trivial loss of maximum power as the fuse resistance increased, causing a (very small) voltage drop across it. At this point the fuse is probably very close to rupture, so it wouldn't matter anyway.

 

[SIY]

Rotel were one of the early designers that incorporated the (speaker) fuse into the feedback loop of their amplifiers

Hafler, DH-200, late 1970s. My first published scribbles in the audio world was a letter to the editor (Holt, in those days) in Stereophile about that very thing.

 

[mhardy6647]

Hafler, DH-200, late 1970s. My first published scribbles in the audio world was a letter to the editor (Holt, in those days) in Stereophile about that very thing.

Not that I'm stalking you or anything (!) but... when I read something like this... well... it piques my curiosity...and... the scholar in me kicks in.

https://worldradiohistory.com/Archive-All-Audio/Stereophile/80s/Stereophile-Vol-7-No-2.pdf

I do hope that you don't mind.