[sts9fan]

So do the experts think fuses on speaker outputs can protect the speakers or are the too slow?

[NagysAudio]

There are no fuses on the speaker output in amplifiers. There is usually a slow blow fuse (main fuse), usually located together with the IEC inlet. And then there are the fast blow rail fuses.

[sts9fan]

IF a fast blow fuse was used at the speaker outputs could/would
this protect your speakers?

[NagysAudio]

In short answer, yes.

[andyr]

Amps without fuses = ignorance. If the amp has no fuses on the rail voltages and something should happen to the amp, your speakers will see a DC voltage of 60-80 volts. This means that the woofers will certainly blow as they don't have a protection capacitor and the tweeter will probably survive.

Absolutely agree with you. 

FUSES HAVE ABSOLUTELY NO EFFECT ON SOUND.

They do (have an effect on sound) IME. 

Regards,

Andy

[andyr]

Let me take this one step further. I wouldn't make a piece of a equipment for myself that was a shock or fire hazard, and I wouldn't do that to anyone else either.
It's your call.
Dan Banquer

But, Dan ... if there's a mains fuse in place - which is mandatory to gain electrical approval - surely there is no shock/fire hazard?

Whether or not there are fuses on the DC rails is merely a matter of protection - probably not of the output transistors (as the fuses are probably not fast enough) but, hopefully, for the speakers?  However, I have no doubt the sonics are better without them. 

Regards,

Andy

[JohnR]

Just wondering - if one were concerned about protecting speakers, wouldn't something like this (either built into the amp or in a separate box) be the best solution?

http://aussieamplifiers.com/lsp1.htm

[Dan Banquer]

So do the experts think fuses on speaker outputs can protect the speakers or are the too slow?

Yes, a properly rated fast-blo fuse either at the output of the amps or the input of the speakers will protect your speakers. The key is in the rating. Most fast-blo fuses are rated for double the rated current for five milliseconds, so a 3 amp fast-blo fuse will pass 6 amps for 5 milliseconds. For the DIY crowd, setting the right fuse will be dependent on speaker efficency, room size and your personal tastes for how loud you like to take it.
I am going to provide a general outline below.
High efficiency speakers (92 db/spl and up): two to two and one half amp fast blo fuse.
Medium efficency speakers (86 to 88 db/spl and up): two and one half amp to three amp fast blo fuse.
Low efficency speakers (86 db/spl and lower): 4 amp to 6 amp fast blo fuse.
Please note that this is a general guide.
Also please note that some fuse manufactuers my have a different fuse rating so check the manufacturers data sheets.

Hope the helps;
Dan Banquer

[sts9fan]

Thanks Dan. 
I think I am gonna build some fuse boxes into my speaker cables just in case one of my projects gets funky.  At least my speakers will survive.

[Dan Banquer]

Thanks Dan. 
I think I am gonna build some fuse boxes into my speaker cables just in case one of my projects gets funky.  At least my speakers will survive.

As a cost effective alternative, you could install a couple of the bulkhead fuse holders in the back plate of the amp. Do you need a link?
Regards;
   Dan Banquer

[turkey]

There are no fuses on the speaker output in amplifiers. There is usually a slow blow fuse (main fuse), usually located together with the IEC inlet. And then there are the fast blow rail fuses.

In addition, some amps (Dynaco, Hafler, and AVA spring to mind) have fuses on their outputs to protect the speakers.

[ohenry]

Uninsulated wiring?      Here's a quote from the history piece:

Obviously, Teflon is nice, but it is expensive. Even better is nothing at all, and so Mr. Dunn embarked on this approach, ending up with a large gauge tinned copper wire used to carry all the power supply currents, output currents, and a really rugged grounding system...

The result, when one peers into an NVA amplifier, is like nothing you've seen before. Large hunks of tinned copper wire float at various heights around the amplifier, making it truly unique.

[Dan Banquer]

"But, Dan ... if there's a mains fuse in place - which is mandatory to gain electrical approval - surely there is no shock/fire hazard?"
Fuses really have little to do with shock hazard they are predominantly for fire hazard.
Secondly, engineers have developed over time various protection schemes to limit or prevent damage due to assorted forms of failure, but the discussion of this is far outside the capabilities of this forum. If you wish to look into this further you might want to look at some of the things National Semiconductor or Crown has done for audio. In addition, the computer you use has some very sophisticated techniques.
Regards;
   Dan Banquer

P.S. Personally I don't think just a fuse at the AC input is enough for any piece of equipment, and there is a huge majority of engineers and safety people who would agree.

[toobluvr]

Amps without fuses = ignorance. If the amp has no fuses on the rail voltages and something should happen to the amp, your speakers will see a DC voltage of 60-80 volts. This means that the woofers will certainly blow as they don't have a protection capacitor and the tweeter will probably survive.

This statement is 100% spot on.  It is exactly what happened to my amp and speakers recently......right down to which driver survived.   More than a month has passed and the stink of burnt voice coils still lingers in the air.

Parts and therefore amps can always fail.  That is not my real beef.  But I'm really peeved that the amp has no SPC.

[Kevin Haskins]

Obviously, Teflon is nice, but it is expensive. Even better is nothing at all, and so Mr. Dunn embarked on this approach, ending up with a large gauge tinned copper wire used to carry all the power supply currents, output currents, and a really rugged grounding system...

The result, when one peers into an NVA amplifier, is like nothing you've seen before. Large hunks of tinned copper wire float at various heights around the amplifier, making it truly unique.

Ha... I assume it has some sort of insulation like an inductor.   The above description may sound cool but one of the issues with power supply currents are cast magnetic fields.   The reason you twist wire is to control the size of that field and prevent in from influencing low-level pre-amplification signals.    The problem with a lot of audiophile lore is that it is engaged by people without a basic background to understand what electrically is occurring when you make such tweaks.      It is easy to convince yourself that a change is an "improvement".   As humans we hear what we want to hear sometimes.   

[Kevin Haskins]

Just wondering - if one were concerned about protecting speakers, wouldn't something like this (either built into the amp or in a separate box) be the best solution?

http://aussieamplifiers.com/lsp1.htm

Looks good to me.   The best solution is when it is designed into the amplifier from the get-go.     

[Kevin Haskins]

"But, Dan ... if there's a mains fuse in place - which is mandatory to gain electrical approval - surely there is no shock/fire hazard?"
Fuses really have little to do with shock hazard they are predominantly for fire hazard.
Secondly, engineers have developed over time various protection schemes to limit or prevent damage due to assorted forms of failure, but the discussion of this is far outside the capabilities of this forum. If you wish to look into this further you might want to look at some of the things National Semiconductor or Crown has done for audio. In addition, the computer you use has some very sophisticated techniques.
Regards;
   Dan Banquer

P.S. Personally I don't think just a fuse at the AC input is enough for any piece of equipment, and there is a huge majority of engineers and safety people who would agree.

There is also a breaker on the circuit.   But the code has multiple levels of redundancy.   The idea is to provide the shortest return path for the fault.   If you have a high voltage short to the chassis you don't want the easiest path to be through a body touching the chassis.    You want it to be back through the safety ground.   The flow of current once a fault occurs is designed to pop the fuse via the flow of current and removing AC line voltage from the equipment. 

The other redundancies for non-safety grounded equipment are providing multiple layers of insulation between line voltages and the chassis.    The idea is that if you breach one layer of insulation there is another.   This is Class II safety insulation practice and it is more complicated to pass.    I think designing for both, double layers of insulation, good distance between live exposed parts and the chassis along with securing all connectors and wire AND providing a safety ground are best-practice.    It is just a matter of how much safety you are willing to trade for economy and/or imagined sonic improvements. 

[Kevin Haskins]

So do the experts think fuses on speaker outputs can protect the speakers or are the too slow?

Depends on the fuse and the speaker attached to it.   If you have itty-bitty woofers without a cap in front of them the thermal capability of the coil is small.   You basically have a fuse of some rating in series with the coil and current flowing based upon the DCR of that circuit and the voltage of the power supply rails.    The item with the lowest thermal handling capability will pop first.   You should design that to be the fuse but the reality is that the designer of the amplifier doesn't necessarily know what is going to be hooked up to the amplifier. 

The better way is to have a circuit that detects DC on the output of the amplifier and immediately opens a relay, either on the loudspeaker output or on the power supply rails.   It is relatively easy to design something that works in mS so you don't fry ANY loudspeaker attached to the output and you lift the rails of the device and somehow indicate to the user that a fault has occurred.   You also don't have to use a low impedance fuse which effectively increases the DCR of the loudspeaker cables.    You use a relay that has very good contact area and high current capability.   There are plenty of relays designed for AC mains voltages and current ratings >20A.   

[Dan Banquer]

Depends on the fuse and the speaker attached to it.   If you have itty-bitty woofers without a cap in front of them the thermal capability of the coil is small.   You basically have a fuse of some rating in series with the coil and current flowing based upon the DCR of that circuit and the voltage of the power supply rails.    The item with the lowest thermal handling capability will pop first.   You should design that to be the fuse but the reality is that the designer of the amplifier doesn't necessarily know what is going to be hooked up to the amplifier. 

The better way is to have a circuit that detects DC on the output of the amplifier and immediately opens a relay, either on the loudspeaker output or on the power supply rails.   It is relatively easy to design something that works in mS so you don't fry ANY loudspeaker attached to the output and you lift the rails of the device and somehow indicate to the user that a fault has occurred.   You also don't have to use a low impedance fuse which effectively increases the DCR of the loudspeaker cables.    You use a relay that has very good contact area and high current capability.   There are plenty of relays designed for AC mains voltages and current ratings >20A.

There is one particular problem with the relay approach. Typically relays have somewhere between 12 and 16 milliseconds before they fully engage. With a fast blo fuse the time before the fuse opens can be much shorter if the current is appreciably higher than the fuse rating. Please feel free to check fast blo fuse ratings for more information.
Regards;
   Dan Banquer

[Kevin Haskins]

There is one particular problem with the relay approach. Typically relays have somewhere between 12 and 16 milliseconds before they fully engage. With a fast blo fuse the time before the fuse opens can be much shorter if the current is appreciably higher than the fuse rating. Please feel free to check fast blo fuse ratings for more information.
Regards;
   Dan Banquer

Lets assume so.... then the question becomes how quick is sufficient.    If you have the full power supply rail on the loudspeaker for 20mS is it enough to pop a coil?    Lets say DCR of the coil is 3.6 Ohms and the power supply rail is 80V.     Simple Ohms law has 22A of current through the coil assuming the power supply can hold up at that and the Re of the coil doesn't change.   We know that DCR will rise as the coil heats but probably only in the range of 1-2 ohms up to 200 deg C so lets ignore it and consider worst case.   

Power in watts is I^2 * R  (I in A & R in Ohms) so we get 1742W (P = 22A * 22A * 3.6Ohm = 1742W) into the coil, ignoring all other losses.   For 20mS that is 35W that has to be dissipated by the coil.    Knowing what I know about coils that might be an issue with some tweeters but we have calculated a worst case situation making highly conservative estimates and ignoring the fact that DCR changes with heat and other losses.    Any woofer coil will easily handle that and that is our "worst case".   


Do you see any problem with that analysis?