I have to ask about this idea of placing diodes on the PT HT leads to the Rect tube.

[trobbins]

diagrammatiks - I certainly disagree that fusing is the wrong type of protection part to use. My view is that it is emminently simple and effective in doing its appointed job - a primary side fuse (or equiv) must be in every amp due to regulations. Strangely RG started by saying they wouldn't protect anything in an amp, and then identified a few internal problems that could well be protected by such a fuse

One way to look at it is to get a broad view of how many times amps go back for servicing, versus how many amp-hours go on in the world (no not battery capacity ;-), and of those needing servicing then whether a protection-related fault was the cause, and how many had a blown fuse as part of that equation.

[diagrammatiks]

i'm certainly not saying to not use fuses. they should of course be used..

i was just saying that in light of the previous discussion, it's probably best not to depend on the fuse to do more then it's capable of doing.

[R.G.]

diagrammatiks - I certainly disagree that fusing is the wrong type of protection part to use. My view is that it is emminently simple and effective in doing its appointed job - a primary side fuse (or equiv) must be in every amp due to regulations.

Yes. As noted, it must be there to protect the rest of the world from failures in the amp. It does not protect the insides of the amp, except by accident.

Strangely RG started by saying they wouldn't protect anything in an amp, and then identified a few internal problems that could well be protected by such a fuse

Not strange at all - it's quite consistent when taken as written. You may want to reread that. Far from saying that they would not protect anything in an amp, I said they might protect something, as a side effect. Here's the quote, with emphasis added:

It sometimes happens that a mains fuse will stop one failure - say, a shorted rectifier or filter capacitor - from killing the power transformer, or causing a chain of destruction that kills almost everything. But it that happens, it's a happy accident.

An AC primary fuse is not put there to protect the internals of the amp. However, it does, by coincidence protect some of them in some few instances. As noted, it's a happy accident if this happens. Sometimes unintended side effects are good things. By luck, a lone primary fuse *may* save something in an amp from worse damage. But then it also may not. That good-looking blonde over there may think you're a great guy. Could happen. Another blonde did six years ago, therefore they all will, right?

But deciding that, once a happy accident has happened, one can count on it happening all the time is at best wishful thinking. A misquoted aphorism comes to mind. The race is not always to the swift nor the contest to the strong - but that's the way to bet.

[trobbins]

Maybe it's semantics - but depending on the location and character of the fault within the amp, a fuse would protect the primary side cabling and switch and transformer and DC cabling etc, eg. wherever the fault is downstream of the fuse. The character of the fault, the fuse, and the upstream impedances determine whether the fuse operates, and the level/extent of the current overload determines what parts/cables will survive undamaged.

Just saying "An AC primary fuse is not put there to protect the internals of the amp" is to my mind the misleading comment. Imho the act of protecting the internals is the reason it is there - it can protect against overcurrents causing insulation from overheating - the ac side of an amps internals are still 'internal' as I see it.

Ciao, Tim

[Enzo]

Well, I suppose this can go back and forth forever, but like RG said, the PURPOSE of the fuse is to prevent your house from burning down in the even of an amp failure. ANy protecting of the amp itself is purely a bonus.


I am reminded of some PA speakers from Klipsch. (We used to be a Klipsch dealer.) These speakers had fuse holders on the jack panels. Woofer fuse and horn driver fuse. The folks at Klipsch included a note. SOmething to the effect: "Many people have been asking why we do not fuse our speakers. We tell them it is almost impossible to protect speakers with fuses. But so many people were complaining, so we added fuse holders. We ship them with 20A fuses installed. We invite anyone who wants more protection than that to do their own calculations."

[R.G.]

Just saying "An AC primary fuse is not put there to protect the internals of the amp" is to my mind the misleading comment. Imho the act of protecting the internals is the reason it is there - it can protect against overcurrents causing insulation from overheating - the ac side of an amps internals are still 'internal' as I see it.

IMHO, it's not misleading; it is a statement of my personal understanding of electrical safety standards. I have a long experience with these, but each design is a matter between the designer and their testing agency.

The standards don't say "put in a fuse". They say rather that the unit must not emit flaming debris or cause an electrical hazard (like making the chassis live to the AC line) when tested with any worst-possible scenario failure of a single part that the tester can cook up, and that can be a failure of any part, including the AC wiring, switches transformer primary, transformer insulation, or anything on a secondary. As a practical matter, that means a fuse (or rarely, circuit breaker) is the first thing that the live side of the AC power line touches inside the amp. That's the place where it's best equipped to stop flames from escaping, flaming liquids from dripping out, and electrical lines shorting to the chassis.

So as a matter of necessity and purpose, any safety testing done on an amp design will completely ignore whether the designer was clever enough to also make the AC mains fuse protect the power transformer (or anything else). If it does, great! But they won't even look at that. A fuse that somehow protects part of the amp but lets it fail safety testing will not be sold.

So for any amps you design, I recommend you first set it up so the power line fuse stop the fires and electrical hazards. Then, for extra credit, try to make it also protect other stuff. That's cool. It's good if you can.

But I can promise you that in any commercial amp you buy (leaving aside "boutique" amps whose designers may never have heard of IEC60650, and semi-boutique amps which have never been safety tested at all) the AC mains fuse was sized to pass safety testing, not to protect the amp innards. The failure of the power transformer itself is very often the worst-possible failure scenario that is chosen.

Perhaps it is semantics. IMHO, the reason the fuse is there is that it's de facto required by safety standards. I can promise you that if it were not, some bright MBA-droid would be selling amps without them. If you want to think that "well, stopping fire and flames *is* the same as protecting something in the amp", that's OK.

I like to simplify things down to binary decisions: do I have to put in X, or can I do without it? Will X meet the needs of the design? I effectively have to have an AC power line fuse for safety reasons. I *don't* have to have one to protect transformers, tubes, caps, resistors and so on. And more importantly, it is too complex and conflicting to find and procure one single fuse in only one position in the circuit that will protect all things. The fuse has a primary job to do; there are other things it may also do, but none of these by itself can be allowed to conflict with the first job for the fuse. So why is the fuse there? It's not to protect everything, or even most things. It's nice if it does, but not mandatory.

I like happy accidents and fortuitous side effects. But I don't count on them. I don't think it's misleading to say "don't count on happy accidents or have blind faith that they will happen."

I'm obviously looking for a way to state this more clearly. If someone asks "How do I pick an AC power fuse that will protect my amp from damage?" the only correct answer is "You can't; the value of fuse that protects part of it will let other parts cook to ashes. The value of fuse that is guaranteed to protect it all will not let it work anything like normally."

Notice that the problem here is that the question is about one fuse, in one position in the AC mains. Using a number of other fuses, sized to protect other parts of the amp, is a good idea, and by carving the protection problem up into smaller ones, you can eventually reach a place where the decision the fuse has to make (i.e. "is it too hot in here, or is it just me?) can be made on a single conductor where nothing past the fuse dies at any current it will carry.

I'm not trying to be argumentative, just clearer. AC fuses *HAVE TO* stop fires and shock hazards. They may also protect some of the rest of the amp from damage. But protecting the rest of the amp is not the main reason they're there. They're there as a last resort when something inside is already so damaged that nothing else will stop the oncoming disaster.

[g1]

R.G., your comments on fusing all the secondaries and also the main primary fuse being more of a code requirement got me thinking of the CSA version of the JCM800's we had up here in Canada. There was a separate fuse board inside with H.T., bias, and 2 or 4 heater fuses (50 or 100W).
At least I was told that was for CSA approval and now I'm wondering why they went about it that way and what were the extra requirements that this would have solved.

[trobbins]

RG - well said and explained. Not many here would have been forced to trawl through 60950 or harmonious equivalents. It is semantics - the standards beget the use of a mains fuse (because it is the cheapest way to pass the standard). My aim was to say that given a fuse is in the amp by defacto, then as we're mostly targetting the optimisation or benefits of what is already in place then there is sometimes the opportunity to improve protection of expensive internal parts by some extra effort (eg. design selection of parts or mods).

[Enzo]

g-one, the Canadians have more strict rules, at least in that regard. I was once in the arcade industry. Well I suppose I still am, though inactive. Our arcade games often came with 15 or 25 foot power cords so they could be placed where outlets were not close. But Canada had some sort of regulation that power cords could not be longer than 6 feet. And if I recall, you couldn;t use extension cords, had to mount an outlet within the 6 feet. Or some such. And that explained why some game makers only shipped with short cords. They didn;t want to make two versions of the game cabinet.

I see some amps with the extra fuse board. And some where the fuse positions are right on the main board. And in those cases, I see the amps heading to the USA with jumper wires in place of the fuse clips. Not a kludge, the silk screen has both fuse clips and jumper wires with a note to the effect of USA this and elsewhere that.

[soundmasterg]

Is there any reason why circuit breakers aren't used? Are they too expensive, o rnot as fast to blow as a fuse?

I was also wondering why the automotive style blade fuses aren't used? Do they not have a high enough voltage rating or something?

Greg

[trobbins]

In general cbs cost more, especially if a pcb fuse can provide the same compliance attributes. And they are physically larger. And that's not even getting into in-rush coping curves (D-curve here).

Auto fuses are obviously a good fit for their intended application, given their predominance. I suggest there has been no 'need' to attempt to make them cover another generic application such as AC mains rated. Their rating doesn't cover mains AC. Any bit of wire can act as a fuse, but companies can only use rated parts

[R.G.]

RG - well said and explained. Not many here would have been forced to trawl through 60950 or harmonious equivalents. It is semantics - the standards beget the use of a mains fuse (because it is the cheapest way to pass the standard). My aim was to say that given a fuse is in the amp by defacto, then as we're mostly targetting the optimisation or benefits of what is already in place then there is sometimes the opportunity to improve protection of expensive internal parts by some extra effort (eg. design selection of parts or mods).

Yep. That's it. You have to use a mains fuse. If you can cover other issues, great. But don't beat yourself up if you can't cover everything, and don't think you *have* to only use one thing for protection.

We had a saying back in my prior life. You and I were in violent agreement.

[diagrammatiks]

was just coming here to ask about circuit breakers and fuses.

is there a fundamental difference between the two? let's say a 1a cb and a 1a slo-blo.

[trobbins]

Some comments are that I'm pretty sure most simple cb's for low current AC are just thermal trip - and so it is relatively easy to compare current time curves and appreciate what the difference can be. CB's can come with different time-response curves, just as fuses can, but they would be not so common and probably at a price premium. The curves are an 'average' response (aka tube curves) and have large tolerances and are influenced by ambient temperature (which are not often given). Fuses can 'age', and cb contacts can 'age' and I guess mechanism quality can vary, so they both have operational issues. It's probably not a great idea to use a cb as the 'on-off' switch for a device that gets turned on and off quite a lot, and some CB's may not be specified for use as an isolation switch (?).

[Enzo]

Auto fuses would not be good for this. When the fuse is out, the females for the blades are exposed where your fingers can touch them. That is why the old bayonette style fuse holders with the short twist cap have been replaced by the similar thing with the long extended cap. With the cap off, the fuse sits far enough down inside that you cannot touch it.

Other than that, fuses are fuses, what matters is the voltage they can break, the current they can carry, and what it takes to blow them. What kinds of terminations are on the end are merely to convenience the designer.

Any fuse maker publishes curves for their various fuse types. Plotting current versus time to open. SImilar curves are published by circuit breaker makers. Most applications we might have in guitar amps are not very demanding and don't have a ton of critical criteria associated with them, so a breaker would generally work. But if you do have a demanding application, then study the spec sheets and see what products fit your needs.