PT protection - one more question

[bluesbear]

Well, the PT in my son's big amp went south (Marshall Plexi clone with Hot Switch). This is the one amp I didn't install a MOV across the primaries. I won't skip that in the future!
I know some of you fuse wires from the PT. I've never done that so I don't know which wires to fuse or what size fuse(s) to use. A discussion on this subject would certainly be beneficial to me... and I suspect to others.
Fuses are cheaper than transformers!!!
Thanks!
Dave

[tubeswell]

A couple of R.G. Keen's immortal amp mod articles (Pt 2 on PT fuses, Part 3 on MOVs)

[kagliostro]

Only one more thing

Never fuse a Bias supply

Kagliostro

[bluesbear]

Wow! A lot of work... but worth it! I'll do it.
Thanks!
Dave

[bluesbear]

Okay, I've been studying this info and I have a few questions.

1) Why would you add up the tubes draw to figure the fuse rating? Wouldn't it be easier to just set the fuse for the PT winding's rating? For instance, the filament winding on this PT is 6A. Instead of fooling around, why not just fuse it for 6A? Likewise, although I use a 5AR4, the recto winding is rated at 3A. Same question.

2) Should the F2 and F3 be equal? The article never mentions F3 but it would make sense that it would be the same as F2.

3) The diagram of fusing in mod #2 shows a capacitor before the B+ fuse. What does it do, why is it there, and what should it be?

Thanks,
Dave

[kagliostro]

3) The diagram of fusing in mod #2 shows a capacitor before the B+ fuse. What does it do, why is it there, and what should it be?

I assume the first capacitor is an electrolytic and I think that is shown there instead after the B+ fuse as to avoid inrush current to blow the fuse too easily

Kagliostro

[tubeswell]

1) Why would you add up the tubes draw to figure the fuse rating? Wouldn't it be easier to just set the fuse for the PT winding's rating? For instance, the filament winding on this PT is 6A. Instead of fooling around, why not just fuse it for 6A? Likewise, although I use a 5AR4, the recto winding is rated at 3A. Same question.

Good question. seeing as how in most cases the Pt windings are actually able to handle a bit more than the rated draw. But I guess if the PT winding is poorly rated and the fuse didn't blow when it needed to then you might lose the winding?

2) Should the F2 and F3 be equal? The article never mentions F3 but it would make sense that it would be the same as F2.

Yep, altho' if you have the diodes, you probably don't need the fuses.

3) The diagram of fusing in mod #2 shows a capacitor before the B+ fuse. What does it do, why is it there, and what should it be?

I believe that's merely the reservoir cap for the B+.

[eleventeen]

1) Why would you add up the tubes draw to figure the fuse rating? Wouldn't it be easier to just set the fuse for the PT winding's rating? For instance, the filament winding on this PT is 6A. Instead of fooling around, why not just fuse it for 6A? Likewise, although I use a 5AR4, the recto winding is rated at 3A. Same question.

It's a matter of one's theoretical approach. In this case, it would work fine. You'd like these protective items to protect BEFORE the thing they are protecting overheats. You want the fuses to blow at a clearly "over" condition, not necessarily at a "maximum allowable" condition. Heater fuses are somewhat of a grey area. When cold, heaters can and typically DO use a tremendous amount of current for a very short time. 

Suppose your tranny was rated 50 amps heater current. Would you advocate selecting a 50 amp heater fuse? I think not. What about 10 amps? By the time your Deluxe Reverb tube-complement is using 10 amps on anything but a momentary basis, something is big time wrong. 

2) Should the F2 and F3 be equal?
Yes.
The article never mentions F3 but it would make sense that it would be the same as F2.
Yes.

3) The diagram of fusing in mod #2 shows a capacitor before the B+ fuse. What does it do, why is it there, and what should it be?

A power supply charging the biggest of the big caps for the first time, on the first charge cycle, could produce an inrush current well in excess of normal operating conditions. If that first cap were located after the fuse, the fuse would have to be rated at high amps to avoid blowing the fuse all the time on startup/inrush current. If it were then that big, then it wouldn't have much protective function in normal operation. This is the general challenge on all of these things: To select a fuse rating that doesn't blow on first turn-on/inrush current, yet still protects, and protects BEFORE smoke. 

[catnine]

 I saw this site , in fact years ago I posted this same info on this site . The last two builds I put a fuse between the B+ and the res cap . If you have a low powed SE amp I could not find a fuse lower than 1/2 amp slow blow and 1/4 amp fast blow whcih both are still above the 100 mA rating of the PT . I suppose one could be found . I am not going to bash the idea of all these fuses . I don't even know what the deal is with a new amps warrenty . Say a tube blows and shorts out and takes out the PT is it the production amps builder problem or the tube manufacturers?

 Looks Like I've been lucky so far based of the last statement of this artical.

 I suppose it's worth adding all the fuse protection and diodes . Just what sort of fuse holders are there other than the plastic based single screw mount or the in-line type that twist together or the chassis mount like the line fuse uses. What ever you use it's got to be near the PT leads .

 So lets say I wanted to do this mod with fuses here are my two builds , where am I to place them? The top photo is my champ you can see I have a B+ fuse on the far left side yet the rest of the amp is packed and the PT leads are short . . The other is my 6G2/5E3 build I suppose I could line fuse holders up in the inside of the left of the chassis to the left of the PT , there is a B+ just to the right of the PT .

[bluesbear]

I can see why the F2 and F3 would be overkill with the diodes. The big problem I see is that 99% of players (probably me) would never know the recto tube has failed. The best answer is probably just to go with SS rectification. A nice big fat 50 watt resistor bolted to the chassis with a heat sink would be close enough for just about anyone to be happy. I've done that before actually but it was a tiny little amp with no room and it just got too hot in there, even with a fan. It worked, though.
I've seen amps (expensive!) with the B+ fuse on the back panel, next to the main fuse. This seems like a good idea.
"Would you advocate selecting a 50 amp heater fuse?"
Of course not but what kind of PT would have that? I only buy PT's made for tube guitar amps so that isn't really an issue. I can't imagine ever building an amp that needs more than 6 amps for the heaters and 3 for the recto. I don't do reverb or trem and, except for my son's plexi and a bass amp, I only build 5 to 20 watt amps. Even my son has wised up and is using a 15 watt amp, miked. He was amazed the first time he took my 15 watt 5F6a and actually got to turn up a bit. I just don't need that many tubes.
Thanks, everyone. I love having a place where you can actually get answers... from people you can trust!
Dave

[bluesbear]

Okay, I've been studying and simplifying. With SS rectifier, fuses aren't necessary there. That's 3 fuses out! The B+ fuse can be on the back panel. The MOVs aren't a problem... except I've only used one per amp. He advocates that one but adds 4 more. They're small and at $.28 each, I won't kick.
One more question, though. I thought that by using the 100 ohm resistors in the filament circuit, they acted as a fuse. Am I wrong about that?
Thanks,
Dave

[sluckey]

One more question, though. I thought that by using the 100 ohm resistors in the filament circuit, they acted as a fuse. Am I wrong about that?

Those resistors are used to provide a ground reference to the filament circuit. Doing so will reduce filament injected hum in the tubes. They don't 'fuse' anything.

[bluesbear]

I've been looking through some of my very early research. I believe the 100 ohm resistors actually do act as a fuse. The resistors will burn before the transformer windings. That was why I started modding my old amps in the first place.... but my memory is a bit iffy (not old age; I was born that way!). According to my notes, I considered the balanced ground reference a happy side benefit. It was the winding protection that got my soldering station turned on.
Thanks, I just ordered the MOV's. I have everything else on hand.
You guys are my heroes!
Dave

[sluckey]

I believe the 100 ohm resistors actually do act as a fuse.

No way. A fuse must be in series with the load (filaments in this case). These resistors are in parallel with the load and in fact create an additional load on the PT filament winding. It's a small load, 6.3/200 = 31.5mA, but a load just the same. If one or both those resistors open up for some reason, current will still flow thru the filament circuit.

Now, there is some discussion about the merits of using 100 ohm resistors versus using a real CT. Assume you have a circuit that has a real center tap that is grounded on the filament winding. Now suppose you have a short between plate and filament of a output tube. Since the filament winding (big wire) has a real center tap that is connected to ground, you will have a very low resistance path from plate (B+) to ground. The HT winding is small wire and the excessive current will likely smoke the HT winding. In this scenario, using 100 ohm 1/2 watt resistors instead of the real center tap would likely prevent damage to the PT. The small resistors would burn up before the PT was damaged. Maybe! In this case, the resistors would be preferred to the real center tap.

Perhaps the above discussion is what you're remembering?

[Fresh_Start]

Great thread!

Tubeswell - I hadn't saved those Immortal Amplifier articles before but have now.  Just searched Premier Guitar's website and printed to PDF.

Using the MOVs on the OT primaries & center tap makes much more sense to me than the zeners I've seen in some commercial amps (e.g. Blues Jr.).

Cheers,

Chip

[jjasilli]

Late to the party, but another way to "fuse" the power tubes, often used in hi-fi amps of yore, was to use a 10R 1/2W resistor under the cathode of each power tube.  This would have near -0- effect on bias; and: a) the resistor would act as a fuse in a high over-current situation; and b) the resistor could serve double duty as bias sense resistor, though it moves the meter reading over by a decimal point.  

Curiously, though my Stromberg-Carlson PA amps used these 10R resistors, they were made switchable to read bias. This defeated their value as fuses when shorted for normal operation; and makes you freak-out reading 540mA intead of 54mA of current when switched-in to take a bias reading!  

[bluesbear]

"Perhaps the above discussion is what you're remembering?"

This has to be something I got before I knew this board existed but you could be right about the concept and my understanding of it.
Thanks,
Dave

[terminalgs]

for fusing B+,:

It seems like a 1A slo-blo is too big.    If normal draw is <200ma @ 360VDC, for example,  the point of failure for a 1A fuse would be 360W, right? (which is a dead short).   If all you are defending against is a dead short, then wouldn't you want to not use a slo-blo if you are going to wait until you have a 1A DVC draw?

If the 1st filter cap is on the transformer side of the B+ fuse, then inrush isn't a big deal, right?   is a guitar amp B+ even  really considered to be a 'high inrush' power circuit? (when i think of high inrush, i think of DC motors...).  It seems like if the first filter cap was moved to the other side of the fuse,   at 300+ VDC,  inrush would never exceed 1A.

I haven't done this on an amp, so my questions originate from the point of view of trying to understand the problem, not from the point of view of experience being critical.

Also, if you fuse B+,  why don't you also fuse the filaments?  Is it that a cap on the B+ side is so much more likely to fail closed than a filament is?  (I've never heard of a filament shorting, either in a tube or a light bulb...)

[jjasilli]

I've never heard of a filament shorting, either in a tube or a light bulb - filaments can short to the cathode inside a tube, then to ground.  Merlin among others recommends fusing.  Hoffman notes that virtual center tap resistors also serve as fuses.

I have fused ea power tube at the cathode to ground connection.  If full signal draw is 200mA; then double that.  Closest choices are probably .315A or .5A. Go with .5  I don't think slo blo or fast blo makes much difference - unless fast blo fails in normal operation.  Trial & error will prevail. The trannies are more robust than the fuse of correct size.  The trannies might get hot, but even a slo blo fuse should melt before a tranny does.