Add an HT Fuse?

[syscokid]

While preparing for a 5E3 build, it brought back a question that I’ve been meaning to ask for a long time. Why do a lot of guitars amp circuits do not have an HT fuse, while others do have them?

If I’m convinced that an HT fuse is a good thing for a 5E3, or any other amp for that matter, I would need to place the fuse inline between the rectifier and the main’s filter, correct? Fuse needs to be rated at 500mA? Slow-Blow, or not so slow-blow?

IMG_1254.png

[Stevem]

A HT fuss will cover the whole DC side of the amp, but you mainly have filters that can go bad and output tubes.

Filters will give you notice before they go south enough to pop the main fuse, but output tubes do not when they short .

In light of this I would be more inclined to place in line with the OT primary center tap a 150 ma fast blow Fuse.

[R.G.]

Why do a lot of guitars amp circuits do not have an HT fuse, while others do have them?

I believe that it's an economic decision based on trading off the cost of a fuse, holder, and wiring plus some number of warranty calls on the fuse setup, versus the warranty savings of the small percent of amps in the field that get protected by a HT fuse. There are lot of subtleties in this, as it's a trade of certain cost on every amp versus how many will be protected in the field and the intangibles of what the users think of the added reliability - if they think of it at all.

Marshall as a brand historically had a rough-and-ready design process, as well as users that tended to use them all-up-at-11. Fender's experience was different.

If your engineering (or raw luck!) is good, you can pick a fuse that will blow at the right time to save your OT and possibly your filter caps and/or rectifiers from an output tube gone mad. There are ways to do this, but it can get complicated.

If I’m convinced that an HT fuse is a good thing for a 5E3, or any other amp for that matter, I would need to place the fuse inline between the rectifier and the main’s filter, correct? Fuse needs to be rated at 500mA? Slow-Blow, or not so slow-blow?

I personally would put it between the first filter cap and the OT, not between the rectifiers and filter cap. The relevant questions are (1) what are you protecting? (OT is a good thing to protect IMHO); (2) what failure will cause the death of what you're trying to protect? (3) what percent of the time when the particular death-situation happens does your protection function correctly? Putting the fuse after rectifiers protects the rectifiers and ... maybe,., the PT secondary against a short after the fuse. Not a bad idea, as the PT is often more expensive than the OT, but its better to put fuses in the PT leads themselves, not after the rectifiers. That way a failing rectifier doesn't kill the PT without even blowing the fuse.

The most likely need for a HT fuse is a shorted output tube or maybe a loss of bias letting the outputs look soft-shorted.

Setting an actual fuse rating is tricky as well. Fuses are not rated at the current where they will blow. The rating is where they will definitely not blow. A 1A fuse will carry 1A semi-forever. A 1A fuse carrying 1.2A may take a looooooong time to blow. But to shorten up some of the doom and gloom; The fuse rating should be about the same as the max DC current rating for the OT. Failing that, it ought to be about the max DC current rating of the PT. Failing that, buy an assortment of likely looking fuses and start with the smallest one. Play loud and hot and see if the fuse pops. If it does, move up one notch. At some point, the fuse will stop blowing and that's your ideal rating. The process shows that for your amp, your tubes, and your playing, the fuse is just enough to carry the load. It's possible to short-circuit (!?) this process a bit by using a current meter to find out what the max DC draw is for your amp at maximum warp drive, then stick in a slightly bigger fuse.

Yeah, slow blow.

[Stevem]

Just to be 100% clear, my reasoning for what I posted is grounded in many decades of replacing power transformers in amps that have only a mains Fuse.

In terms of a vintage amp that’s seen 40 plus years of use ( effects of heat accumulation) and the situation of repetitive blowing of the mains fuse while tracking down a bad output tube over and over and over just does in the PT that otherwise would have lived on for decades more.

[trobbins]

As RG indicates, adding a fuse is not a straight-forward task, and I'd suggest is quite a technical effort to appreciate what faults can be protected against and how the added fuse would respond. If you've had to replace PT's, then there may be incentive to do such a technical assessment when adding a fuse.

The linked article goes through the technical issues and processes to choose a HT fuse in a valve amp. Each amp is different, and I'd suggest each amp requires its own assessment, and that would include clone's being subtly different from an original 5E3, as parts and values are not the same. Happy to help those who want to go through a technical assessment, if they have the incentive to identify an exact schematic and parts and make some rudimentary measurements like winding resistances.

https://dalmura.com.au/static/Valve%20amp%20fusing.pdf

Ciao, Tim

[ViperDoc]

I'd be interested in commentary regarding fusing the rectified DC line vs. both HT legs of the secondary (AC).

[R.G.]

I'd be interested in commentary regarding fusing the rectified DC line vs. both HT legs of the secondary (AC).

It goes back to that "what are you trying to protect?" thing.

A fuse in the rectified DC line can only stop current flow between the rectifiers and whatever is after them, presumably both the filter caps and the amp that eats current from the filters. So a short (especially a soft short, high current without being big enough to make the fuse blow quickly) in a filter cap, or the OT, or output tubes can pop the fuse if conditions are right.
A rectifier can fail open, shorted, or soft short. Open rectifiers may not pop the fuse, it just makes ripple higher in the amp by converting the power to half wave rectified. (Note that in this and other stuff here, there is a long tail of asterisks and footnotes for special cases.) A shorted rectifier shorts across half or all of the PT secondary. None of this current goes through the rectified DC wire to the caps and rest of the amp, so it can't possibly pop the fuse. PT can then overheat and die.
The center tap of the high voltage is also in series with the DC line, just on the negative side of it. A fuse on the CT has the same effect as a fuse in series to the first filter cap, as the same current flows through it. So this does not protect the PT from rectifier faults.

A fuse in series with both halves of the HT secondary limits the current from that half-secondary winding. This catches all of overcurrents from the capacitors, amplifier, and rectifiers. So fusing both sides of the HT winding protects the PT against rectifier faults too.

Protecting the filter caps and/or output tubes may need different schemes than protecting the HT secondary. Tubes losing their bias make for the dreaded soft short, too much current to be sustainable, and maybe too small to make fuses act quickly. This can kill the filter caps through ESR overheating if it's not big enough to kill the rectifiers or PT, or pop fuses if they are installed. In general, all protection schemes need to be designed with thought about (1) what, exactly, conditions and failures are covered, and (2) is the product of the likeliness of the fault(s), possibilities of annoyance at false positive protections, and total repair cost of ignoring the failure?

I once designed a "tube saver" to continuously watch the current through individual output tubes and open the DC power to the tubes (actually, the cathode lines) when the circuit decided that it was too much for too long. Same idea as fuses, actually, but with the advantage that the threshold was very predictable and settable, and the circuit remembered which tube went mad and lit an LED so you could figure things out. Worked. Did not protect against failing rectifiers or caps at all, of course. But then output tubes fail more often than caps and rectifiers. There's a reason they're in sockets.

[pdf64]

Especially if there's a valve rectifier, I prefer to add a HT fuse in the CT return to 0V common of the HT winding.
That seems to be more common in early 60s amps, such as JTM45 or AC50 https://el34world.com/charts/Schematics ... x_ac50.pdf
In this arrangement, F500mA works fine with my JTM45 and JTM50 based builds, even when heavily overdriven for prolonged periods.
I used a fast acting fuse in the hope that it would respond faster to eg a shorted valve than a time delay type.

My thinking was this offers a good balance of a single HT fuse protecting as much of the HT supply and power amp as feasible.

This schematic is broadly applicable, minus the standby http://marstran.com/66JTM45Schematic.gif
Winding resistances
Mains transformer for both 45 and 50:- 240V primary 7.2R, 700V secondary 128R
Choke 45:- 380R
OT 45 primary impedance 6k6:- 160R

I'll have to update the JTM50 choke and OT (384-179 3k4) resistances later.

With suitable series protection / back up silicon diodes added to the GZ34 anodes, and a suitable resistor connected across the OT secondary as per the AC50 (eg 470R 5W dirrctly across the 16ohm winding, then significant collateral damage resulting from most failure modes seems to be have been mitigated?

[trobbins]

A quick PSUD2 assessment of a JTM45 with 350-0-350V secondary (effective 80R winding per side), with GZ34 into a 32uF filter with 5k simple load for 88mA continuous supply indicates a 250mA T IEC fuse in the CT link should have some margin to cope with a hot-start, and is operating at about 75% of continuous rating. That fuse margin should allow up to 120mA cranked loading, and/or a 60uF cap, and possibly an old filter cap with significant initial forming current.

Using a UL rated fuse is technically fuzzy, and likely has to be a 315mA slow.

Having a hotter bias than 40mA, or filter cap with substantial initial forming current, or a high mains voltage, or a higher cranked current may need the next highest current rating like 315mA T IEC.

Using a PSUD2 assessment has a risk, but does seem to indicate that a 315 or 400mA Slow fuse (UL rated) would provide better protection for fault situations when compared to a 500mA F fuse.

[Roe]

I've tried running a superlead with a 500ma ht fuse instead of the normal 1a. it ran for many hours without blewing

[syscokid]

All the replies have been duly noted, and has become food for thought for any future adventures… Thanks!