To fuse or not to fuse (on the secondary side of the transformer)?

[JohnRoberts]

I would use fuses in the +/- voltages. You could still have current flowing between the rails without gnd. And you'd change the impedance of the gnd path.

SSL used special 10R resistors on each pcb in the +/- voltages. These behave like a fuse, they open the circuit but don't burn. This practice makes sense to me for low current applications.

Michael

Fuses on every strip of a big console gets expensive so we used "flameproof" resistors, designed to fail open circuit without starting fires.

JR

 

[Tubetec]

Fusible resistors , I dismantled a big old Yamaha console , every channel had them .

The old Boss 9v pedal supplies use the polyswitch , its not effected by momentary shorts when pluging and unpluging but if you short the supply because you inserted a positive ground pedal in the chain its very effective at preventing the transformer from melting .

 

[abbey road d enfer]

Wouldn't a thermal fuse on the transformer itself avoid most of these scenarios?

Thermal fuses built-in transformers are non-resettable. If it blows, you just throw away teh xfmr. At least that's the way it is here.

 

[abbey road d enfer]

SSL used special 10R resistors on each pcb in the +/- voltages. These behave like a fuse, they open the circuit but don't burn.

Most british console mfgrs do that, Soundcraft, Soundtracs, Audient, A&H...
Often with different resistors for different feeds. The most common example is the headphones amp that has its own feed resistors independant of the rest.

 

[abbey road d enfer]

Just be careful if ordering, about current rating...some basic descriptions state the hold current, other the trip current, you better double check the datasheet to be sure about calmed spec. (I do the mistake once )

The actual choice cannot be made by just reading the specs. You're bound to hit the target -50/+100% out of whack. They are cheap enough to buy a dozen values and experiment.

 

[JohnRoberts]

Thermal fuses built-in transformers are non-resettable. If it blows, you just throw away teh xfmr. At least that's the way it is here.

resettable thermal breakers (like PTC devices) are generally to expensive and large to easily use inside transformers. One shot thermal fuses are smaller/cheaper, and more common.

I've shared this story before. Back in the 80s I designed a headphone amp at Peavey (HB-1) that was robust enough to drive loudspeakers and we routinely drove speakers in our booth at trade shows. While the HB-1 amp was robust enough to rock out all day long, the 1A wall wart had an internal thermal fuse that wasn't. The customer doesn't care if it blew up because he/she abused the product, the blame ends up on the manufacturer.

To protect the wall wart I added a R between the diode bridge and reservoir cap in the headphone amp. This R generated an intentional voltage sag under sustained power output that reduced average output power enough to save the wall wart.

JR

 

[abbey road d enfer]

resettable thermal breakers (like PTC devices) are generally to expensive and large to easily use inside transformers. One shot thermal fuses are smaller/cheaper, and more common.

And the end user pays dearly for this saving. I had to throw too many xfmrs that were otherwise perfectly good. It's even impossiblme to reach the wires and bypass the darn thing.
Not only the cost, but also the fact that you are stranded with a piec of gear that doesn't work and cannot be right away substituted.

 

[JohnRoberts]

And the end user pays dearly for this saving. I had to throw too many xfmrs that were otherwise perfectly good. It's even impossiblme to reach the wires and bypass the darn thing.
Not only the cost, but also the fact that you are stranded with a piec of gear that doesn't work and cannot be right away substituted.

Blame UL (or your local safety agency) for trying to prevent house fires.

I used a cheap 1A-16VAC wall wart in scores of SKUs, back before consumers rejected wall warts. If I had a product eating wall warts that was my personal design problem. Wall warts were cheaper/easier to get agency approval for tens of SKUs using the same already approved wall wart.

We replaced the wall warts last century with some small conventional transformers with line cords. Once a transformer/line cord set gets approved, you can just cut and paste that into future product agency files.

JR

 

[vintageampfixer]

once you've met local safety requirements (which ironically differ vastly from place to place) isn't it a case, to misquote Murphy's Law, where "a transistor protected by a fast acting fuse will protect the fuse by blowing first"?

 

[zamproject]

The actual choice cannot be made by just reading the specs. You're bound to hit the target -50/+100% out of whack. They are cheap enough to buy a dozen values and experiment.

I agree with that, just saying that if you think the thing trip at say 200mA but only hold (and trip at 400) you may see nice smoke downstream

 

[abbey road d enfer]

once you've met local safety requirements (which ironically differ vastly from place to place) isn't it a case, to misquote Murphy's Law, where "a transistor protected by a fast acting fuse will protect the fuse by blowing first"?

Actually, if you think of it, these regulations are not to protect the product/owner/user, but protecting the power distro.

 

[JohnRoberts]

UL is pretty focussed on customer safety.. Some EU regulations like power factor correction were about distribution capacity. But I am not expert about over there....

JR

 

[abbey road d enfer]

UL is pretty focussed on customer safety.. Some EU regulations like power factor correction were about distribution capacity. But I am not expert about over there....

JR

Yes, it was more a sarcastic comment than a real critical analysis.

 

[rock soderstrom]

Overall productive thread, thanks for that! The vote is still somewhat extendable.

resettable thermal breakers (like PTC devices) are generally to expensive and large to easily use inside transformers. One shot thermal fuses are smaller/cheaper, and more common.

That's probably true, but as I said I've also seen them a few times in some transformers.

The German Wikipedia also sees it that way:

"They are often built into the windings of small mains transformers, so they are thermally coupled and make such transformers intrinsically safe(r)."

The advantages and disadvantages are described there like this:

"Self-resetting fuses are more expensive than conventional fuses and fuse resistors, but create increased utility value by avoiding total failure or service."

For me, there is the added advantage that they take up less space on the PCB than fuseholders, at least for my currents.

I think this time I'll go with this type of protection.

 

[Newmarket]

Actually, if you think of it, these regulations are not to protect the product/owner/user, but protecting the power distro.

Not sure about that in terms of product design / specs. Power Factor Correction requirements relate to transmission losses. But other short circuit / surge type conditions are essentially dealt with at the "Board" ie the Consumer Inlet. Although tbf I am only really familiar with UK/EU requirements. And they have recently been updated wrt UK "IET Edition 18".
I do understand that UK is something of an exception in that the mains plug itself is fused (replaceable cartridge) as a safety precaution.

 

[Newmarket]

Most british console mfgrs do that, Soundcraft, Soundtracs, Audient, A&H...
Often with different resistors for different feeds. The most common example is the headphones amp that has its own feed resistors independant of the rest.

Yes. Add DDA (ie pre- Audient and Dave Dearden is ex-Soundcraft).
IIRC 10R NFR (Non Flammable Resistors) in every individual opamp voltage rail feed. With 22u (?) Cap' to 0V. ?
So also adding RC filtering.
Meaning that any individual opamp fault drawing excessive current should (hopefully) not pull the whole rail down as the NFR should have gone open circuit.

 

[trobbins]

My comment would be that this thread on fusing is related to ss pre/amp equipment, and should not be linked to valve pre/amp equipment as that type of equipment typically has quite a different set of aspects to consider for the topic of fusing.

Another comment is that large manufacturers may well go to quite some technical and economic effort to choose a solution to over-current protection - just looking into such equipment and seeing a part like a flame-proof resistor does not tell the whole story, and a diyer or boutique manufacturer taking up any such approach may be missing out on the actual balance of reasons why a large manufacturer chose such a path.

My first response about the circuit in the thread's original post is that it needs to be technically assessed for over-current, as well as the power transformer. Just adding a fuse, or an NFR or polyswitch without a good effort to be aware of all the technical consequences is a bit like blindly pinning a tail to a donkey.

Following on from that response I would also say that this type of thread is good for discussion, but the idea of voting for a simplistic outcome like whether to fuse, or not, is prone to be misinterpreted by many as a vote of confidence for fusing (or not) any and every circuit with scant further technical assessment for their own circumstances.

 

[JohnRoberts]

How did you vote?

JR

 

[trobbins]

I didn't vote, for the reason that imho any voting outcome could be misinterpreted by many who visit this thread. If I had my 2 cents, I would prefer to see posts that technically looked at an example situation and tried to design and then test a particular over-current fault scenario, and people then commented on that.

For example the original circuit in post #1 could be put through some assessment and testing. That would need the OP to indicate if the circuit was constructed and what they had tested so far, including details on the PT and 317 heatsinking and design loading, and whether the circuit could have further testing done on it and if that could be extended to stress type testing that had the risk of damaging parts.

 

[moamps]

Ordinary fuses are quite inaccurate and in my opinion only serve to prevent a major accident. Their value in a circuit depends not only on the current consumption of that circuit, but also on the impulse currents during switching caused, for example, by the magnetization of the mains transformer and the charging of the electrolytic capacitors in the power supply.
Additionally, if they are placed in the dual power supply shown in the first post, special precautions should be taken to prevent the fault of one fuse from doing more damage than protection. If only one fuse blows, it can happen that the second power supply stays working and that there is a power supply of circuits with only one power supply and the consequent occurrence of a huge DC offset, irregular operation of the circuits and the occurrence of far greater damage. Imagine a large console where only one rail + or -17.5V is left to work.
Such power supplies must have circuits that will turn off the other rail too (an example are Soundcraft power supplies). A similar problem can occur with power output amplifiers, so they must be fitted with some kind of protective circuit that will disconnect the amplifier from the speakers if a DC offset occurs at the output.