Neil Young's "Magic" 5E3

[Mike K]

Supposedly the "proper" PT spec for current is 80 mA - I won't say who told me that, but they have a good relationship with Fender. But because this amp is so wildly popular to build, there's variants all over in both current and voltage spec.

For me, I don't want to cook my nice NOS 6V6s, so 14W, 350V max Va-k is what I aim for. They sound good there and I see no reason to run them any harder.

[Helmholtz]

But if my own amp sounded like that I'd think it was broken.

^^That.

[Mike K]

Oh, regarding core saturation. I'm told if that happens the amp goes flat and bass drops out the floor. I don't think they actually "saturate" in that sense.

I think what happens with a larger core is you either get less bass rolloff due to higher inductance, more clarity due to a better efficiency i.e. less distortion, or just more volume, and you are compressing the highs so it gives the illusion of a bigger, tighter bass response. I'm sure it's different based on every design but going to a bigger core isn't always better.

What I tend to notice just off the cuff is: 1) more volume - but not just in the sense it's louder, also more dynamic. 2) usually more clarity in that the harmonics sound more sharply defined rather than blurred together. I'd wager if you looked at a spectrogram you'd see more sharply defined resonances.

Conversely I notice when you run a small core very hard, you sometimes get what we call "fizz", which just seems to be less defined high frequency harmonics and sometimes it just sounds "blurry", again, less defined harmonics in some range. I've used some damn tiny cores (the size on a small choke) and run them till they smoked, and this is my observation. And yes, you can smoke small core OTs with an inductive or resistive attenuator. Not recommend.

[Chuck H]

Ok. So a higher spec PT or even one with a lower DC on the HV secondary will provide more voltage under the highest output conditions. Does this qualify cooking the 6V6's? Probably not for some new models. As to NOS... I have limited experience but I've lost a couple of them plugged into stock SF Champ circuits that have stupid high voltages (not all do) and ended up using current production tubes in those amps to avoid failures.

[Chuck H]

And yes, you can smoke small core OTs with an inductive or resistive attenuator. Not recommend.

I'd actually argue that a good attenuator is easier on the amp than an actual speaker. Speakers have wildly varying impedances. A resistive load is probably the least likely to damage an OT. So why do people think that attenuators are bad for amps? Well, because there's more cords and devices involved creating opportunities for failures and user error. Plugging an amp into an attenuator means more stress on the output jack, another cord involved, the possibility of plugging things in backwards, another device that can be faulty (the attenuator) and more possibility that something might not be connected. Not to mention that players that only rarely ever cranked their amps without an attenuator are now cranking them ALL THE TIME. So the amp spends more time at full output. All this adds up to a greater failure rate when attenuators are used. But it's NOT the attenuators fault and they are NOT harder on amps than speakers are.

JM2C on that. Coming from a guy that uses an attenuator almost full time so I'm "biased".

[Mike K]

I have a Weber 50W mass. Nothing was plugged in wrong. I watched the sparks fly from the OT and the volume fade as I was playing. Asked a supplier and they said that attenuators do that. I could never get any good physics as to why, but smaller cores are more susceptible to this failure.

[Mike K]

Most newer 6V6s can take much more voltage than rated. Some older version can too. I tend to run 6V6s according to the datasheet, although I know there are a lot of guitar amps that do not. YMMV there.

[Chuck H]

Most newer 6V6s can take much more voltage than rated. Some older version can too. I tend to run 6V6s according to the datasheet, although I know there are a lot of guitar amps that do not. YMMV there.

One of the problems I see with currently manufactured tubes is that the data sheet for them is quite obviously just a regurgitation of vintage specs and example conditions. With the actual specs for those tubes being different (modern 6v6's being capable of higher performance and el34's being less robust, for example). So IMHO we don't actually have data sheets for new tubes at all.

[Chuck H]

I have a Weber 50W mass. Nothing was plugged in wrong. I watched the sparks fly from the OT and the volume fade as I was playing. Asked a supplier and they said that attenuators do that. I could never get any good physics as to why, but smaller cores are more susceptible to this failure.

I don't doubt you had this experience. And as far as I know only Weber sells Weber attenuators. So I have to assume that by "supplier" you mean Weber. If so then this is a stupid and poor way to handle the matter. If true, meaning that their own experience is that their attenuators cause failures "sometimes" then they need to either redesign them or improve QC and testing. Because the reality is different.

The attenuator I use is of my own design. It's an inductive load with a resistive distribution between the actual speaker and the dummy load. In fact all "inductive" attenuators do this except perhaps the Marshall power break which seems to be predominantly inductive. I have even built attenuators INTO amplifiers that can be activated by a switch on the back of the amp to prevent any possibility that a less appropriate load or any user error could damage the amp.

[Mike K]

I don't doubt you had this experience. And as far as I know only Weber sells Weber attenuators. So I have to assume that by "supplier" you mean Weber. If so then this is a stupid and poor way to handle the matter. If true, meaning that their own experience is that their attenuators cause failures "sometimes" then they need to either redesign them or improve QC and testing. Because the reality is different.

The attenuator I use is of my own design. It's an inductive load with a resistive distribution between the actual speaker and the dummy load. In fact all "inductive" attenuators do this except perhaps the Marshall power break which seems to be predominantly inductive. I have even built attenuators INTO amplifiers that can be activated by a switch on the back of the amp to prevent any possibility that a less appropriate load or any user error could damage the amp.

Again, this is a touchy subject.

I'm a mechanical engineer by trade and worked in the auto industry for some time. WTBS, I can't comment on how any given engine failed under load. And even if I was a warranty engineer I might not know all the specifics.

I'm perfectly fine with anyone taking any risk they want, but I don't want to make any claims. I only know the information I was given by the OT supplier. The failure looked to be an inductive spike which broke the dielectric and shorted the OT. I didn't have it analyzed but I was told by said supplier they can tell modes of failure by tear down inspection. I'm assuming an attenuator failure looks like an open circuit/no load failure.

If you're an engineer who works with a lot of volume of transformers and analyzes different failure modes, you might have some authority to speak on this. Other than that, I don't want to go down this rabbit hole. I've decided, that for my purposes, an attenuator is not for me. If I was to use one, I'd be sure to make sure my OT was overdesigned for my amp loading at full tilt if that's where I was going to use it.

[Helmholtz]

The attenuator I use is of my own design. It's an inductive load with a resistive distribution between the actual speaker and the dummy load. In fact all "inductive" attenuators do this except perhaps the Marshall power break which seems to be predominantly inductive..

There is no fully "inductive" or "reactive" attenuator.
While the Marshall Power Break uses a stepped autotransformer for attenuation, the load seen by the amp essentially stays constant (and resistive at middle frequencies).
It's a clever design providing good speaker impedance emulation.

[Chuck H]

There is no fully "inductive" or "reactive" attenuator.
While the Marshall Power Break uses a stepped autotransformer for attenuation, the load seen by the amp essentially stays constant (and resistive at middle frequencies).
It's a clever design providing good speaker impedance emulation.

I tried one once at a Guitar Center decades ago. I thought it was pretty snazzy.

[Helmholtz]

Excessive and undamped inductance due to load or OT leakage inductance can cause high voltage spikes with clipped signal.
May result in arcing e.g. between winding layers, carbonizing the insulation and leading to shorted wire turns.
An open primary is unlikely.

But we can't know if the same would not have happened with a speaker load and sustained fully clipped output.

In the end it's the OT insulation which failed.

[Mike K]

Excessive and undamped inductance due to load or OT leakage inductance can cause high voltage spikes with clipped signal.
May result in arcing e.g. between winding layers, carbonizing the insulation and leading to shorted wire turns.
An open primary is unlikely.

But we can't know if the same would not have happened with a speaker load and sustained fully clipped output.

In the end it's the OT insulation which failed.

Certainly.

And I don't know if there was a momentary open circuit on the secondary, but it didn't sound like it (I assume I would have heard some break in sound). The actual failure wasn't abrupt, it was like someone pulled a fader.

Why I'm suspicious is because the supplier told me this is a common failure they see and I run pretty much all my OTs wide open all the time with no failures. The only failure I've ever had is with that attenuator.

I don't know if it's horsehockey but I also asked about built-in attenuators and how they handle warranty on that, and they say they will not and cannot guarantee them. Perhaps it's just overprotective warranty but I don't blame either the OT or the attenuator. I took a risk and I had a failure. I knew that it would not be warrantied going in and I ignored that.

[Chuck H]

I'm perfectly fine with anyone taking any risk they want, but I don't want to make any claims.

That's defines a smart policy.