Tubes that can handle 500VDC on screens, or tame the power supply??

[Mick Bailey]

Agreed, the power rating is a paper maximum. When you think though we're generally dropping say 30v to 40v at maybe 500mA peak, then 20W is about what the dissipation will be at full volume when the tubes are drawing maximum current. That's not factoring RMS. A TO220 gets hot in this application and needs careful consideration about mounting. It does actually work OK in practice, given that most amps aren't run to maximum output. It's not a device selection I personally like to use because of the issues discussed and when I've experimented with TO220 packages with an amp into a dummy load they get way hotter than a TO247 when pushed to maximum output. That's how I settled on that particular package even though people often use TO220s. I just don't like the hotspot they create.

In reply to Gregg, The circuit works in the B+ line or CT as per R.G.s original article. I use either method depending on the PS configuration and convenience of access.

[bob p]

I might be in the minority on this, but I don't think that clamping the screen sounds good, especially if you aren't also clamping B+. IMO you need to regulate both of them or neither of them to get the best sonic results. Clamping one while letting the other sag just doesn't work well.

[bob p]

Had to Google that.

Some of the cultural references may not be recognizable outside of an urban slum in the USA, which is the marketing niche where Tbird became famous. It was a cheap screw-cap fortified wine. And when I say it was cheap, I mean really cheap. Like 60-cents for a bottle. It had a high alcohol content (like a Port) and a cheap price, which made it popular with skid row bums.

I grew up in an urban area where Tbird was popular. When I was riding my bike as a kid I had to be careful about taking shortcuts through the alley as it almost always meant a flat tire, caused by the shattered remains of a Tbird bottle. Tbird was one of the famous "bum wines", which is probably why Chuck mentioned it in his post. It was so famous in the 'hood that people even wrote songs about it:

Tbird came around when the wine purity laws in the USA were relaxed in the 50s to allow "flavored wines" and alcohol-fortified products to be sold as "wine." That allowed the cheap vineyards to dump their low grade products by blending them, and adding flavorings and distilled liquor to bring the alcohol content up to about 20%, like a Port. The combination of low price and high alcohol content made Tbird popular in urban areas. The bums liked it because it was a cheap drunk, and the poor people liked it because it was a cheap mixer. I remember going to the grocery store as a kid and seeing Tbird on the shelf with packages of lemon Kool-Aid next to it. And there was the catchy jingle on the radio:

What's the Word? Thunderbird!
How’s it sold? Good and cold.
What’s the jive? Bird’s alive.
What’s the price? Thirty twice.

Mixing tbird with lemon Kool-Aid was pretty common in the neighborhood. It was always regarded as a cheap mixer. I have to confess, as a kid we used to sneak Tbird and Kool-Aid at a friend's house. We did the same thing with Mogen David and 7-up at another friend's house. Even if you cut them in half, you still had 10% alcohol content, and one glass was enough.

Tbird was always regarded as a cheap wine, but in the 70s they tried to make the brand more upscale with the grapefruit juice "Shake 'em Up!" campaign in the commercial that Juan linked. I confess ... I've also had Tbird and grapefruit juice "shakers", but that was something like 50 years ago when I lived in the 'hood ...

Gallo was one of the most successful companies at selling cheap wine. Some of their most successful brands from the 50s-70s were Thunderbird, Ripple and Boone's Farm. They all had a screw cap. When the market got tired of these products, the same low grade wines ended up being marketed in a more upscale fashion as carbonated "wine coolers" in the 80s. Same juice, different blend.

[Chuck H]

Juan, If he likes the way the amp sounds now, but just wants it to stop eating power tubes then maintaining the existing preamp voltages might suit the end result of the effort. Submitted for consideration only.

[Gtr0]

Pardon my ignorance on the topic... But what is the nature of the relationship between the screens and plate anyway? I *thought* the screens purpose is to better control the flow of electrons from the cathode?

How is it then does an amp like the silver jubilee cut out the screen resistor connection.. effective turning the tube into?? A triode?

Also how are amp manufacturers getting away with such his voltages on the screens as well? Obviously the SLO and similar amp the Dual Rectifier??

[J M Fahey]

In no particular order:
* thanks for the Sociologycal/Anthropological data, something which always interests me.

Among tons of other things, I have read quite a lot about Skid Row which is a unique phenomenon with its own (unwritten) laws, didn´t know the brand Thunderbird, but would have instantly recognized the generic "muskie" , as well as its users: "winos" and even their "starting a Corporation" (pooling coins and spare change so as to buy one or two bottles and then share them).
Plus the infamous "flophouses" where just *imagining* the smell makes me retch.

Oh well, poor people, I´m certain nobody drifts there on his own will

EDIT: almost forgot: my own simplified explanation is this:

* the basic tube is the triode.
Plate attracts electrons with a strength proportional to its voltage.
That makes electron flow (current) VERY dependant on plate voltage.
Datasheets show that, curves are inclined, with current increasing/decreasing a lot depending on plate voltage.

* personal explanation, others may differ: screen which is always at a certain high voltage, strongly attracts a lot of electrons.
Most of them miss it because being a fine wire grid it occupies very little physical space ... so most electrons miss it and overshoot.
Then it´s easy peasy for plate to capture them, net result is that current is much higher than in a similar triode, and quite independent of plate voltage, so lines are *almost* horizontal.

* tying screen to plate kills its unique effects and turns it into little more than a plate extension.

Under same conditions, a triode strapped pentode will provide less power because maximum available current is greatly reduced.

[pdf64]

... But what is the nature of the relationship between the screens and plate anyway? I *thought* the screens purpose is to better control the flow of electrons from the cathode?...

The effect of the screen grid g2 is to 'de-couple' plate voltage from plate current.
See chart at top of p6 http://www.mif.pg.gda.pl/homepages/f...93/6/6L6GC.pdf
So, all else being equal, plate voltage can vary over a wide range without plate current being affected.

... How is it then does an amp like the silver jubilee cut out the screen resistor connection.. effective turning the tube into?? A triode?
...

Yes, the low power switch re-configures the power tubes as triodes; the current limiting resistors, eg R55 & R56 https://drtube.com/schematics/marshall/2554amp.gif are still in circuit.

[Gtr0]

Most of them miss it because being a fine wire grid it occupies very little physical space ... so most electrons miss it and overshoot.
Then it´s easy peasy for plate to capture them, net result is that current is much higher than in a similar triode, and quite independent of plate voltage, so lines are *almost* horizontal.

Hmmm... to better clarify what I meant (I was on the road during the last post, so it was by phone) and you may have answered is what is their relationship voltage wise as well as physically. So why is it that the screens would come from the next B+ node and see only a very tiny (usually only a handful of vdc at most) v drop. So in relation to the plates they should only be slightly less in vdc as to appear somewhat negative in relation to the plates and highly positive to the cathode?

But to be completely honest, I can't see the picture here as to why that would make the amp so much louder... only unless the high voltage of plates AND screens would attract that much more electrons...


Then, meanwhile back at the ranch...
As if on cue, I had a screen resistor go up in flames some days ago. Bastard still read 1K after removing it though.

Once I replaced it, and the tubes for well roundedness, it has been fine. But that voltage is still high, though I have yet to order a 10 watt resistor in the 180 ohm zone to bump the voltage down. My EL34 tube supply has been stretched beyond limits. Time to stock up again

BUT I did order a new custom tranny, this time 345-0-345 instead of 360-0-360 - so that's at least a start.

[Jazz P Bass]

I think in reality, the screen grid, being between the cathode and the plate, is the anode of the circuit.
The electrons reach the plate because of the wide open spacing of the screen grid.
By lowering the screen voltage you are throttling back the dissipation of the tube.

[eschertron]

I like that explanation ^^^

Think of the tube as a setting for a slasher movie.
The electrons (having been boiled off the cathode) are simply milling about (space charge) as if they were teenagers with nothing to do. The high positive potential of the screen grid is calling "psst. Hey! C'mon over here!" so the potential exists for the electrons to travel to the screen. BUT WAIT! IT'S A TRAP! The wide spaces between the screen grid elements allow the electrons to whizz past and be collected by the "monster", a.k.a. the plate electrode.

The reason it's a trap is because for large portions of the signal cycle, the plate is actually at a relatively low voltage compared to the screen. The plate voltage varies with the rate of electron collision, but the screen collects fewer electrons and doesn't vary significantly from the B+ voltage at its node. So the electrons are still highly attracted (you know how teenagers can be) to the screen/plate despite the - on average - lower plate voltage. They "see" the continuously high screen voltage.

Triodes (and triode-strapped pentodes) do not exert the continuous higher attractive force, and so do not deliver the higher currents, because the electrons do "see" the - on average - lower plate voltage.

[J M Fahey]

So do-nothing teenagers are attracted to Movie Theater by a bloodcurling Bela Lugosi movie ... and when there they are attracted by the *real* Monster.
Yes, I think I get it

[bob p]

The Blob is a great analogy.

But that trailer has been colorized. Oh, how I hate it when they do that to B&W movies.

[Gtr0]

The plate voltage varies with the rate of electron collision, but the screen collects fewer electrons and doesn't vary significantly from the B+ voltage at its node.

BINGO - there's the line that triggered the lightbulb.

Thank you!!

[cerrem]

The Screen current is more closely resembles a hyperbolic Cotangent function ..... ie, the screen current will climb as the plate load swings more plate positive... and decreases to it's idle current of a few mA as the load line swing back down towards the idle state...then into cut-off....when the other side of the Push-Pull is already conducting within the overlap angle...
Marshall amps from the early 70's run about 500V plate and 498V screen at idle .... 1K screen dropping resistors... choke is roughly 110 Ohms..
With the 1.7K plate load and running the amp at full tilt with Siemens EL34's..... the power supply will sag to roughly 430 Plate and 400V on the screens... of course your mileage mat vary depending on the PT's regulation..
The screen current for each EL34 at idle will be about 6mA ...the screen during clipping is up about 32mA RMS ...this includes the Duty Cycle since the screen does not conduct the full degrees....roughly 225 give or take degrees in Class AB1...
The typical NOS EL34 had 8W MAX of Screen Dissipation... What size screen resistors are you using ??
One time I had seen a Laney Super Group amp have similar issues where some of the El34's were blowing up with white internals... It wound up being internal arcing of the tube that damaged the tube.... It only happened on a certain speaker cabinet... So I measured the fly-back voltages up to 1.1kV ...this arced internal to the tube between pins 3 and pin 2 heater....since heaters were at ground potential ....

[Gtr0]

The Screen current is more closely resembles a hyperbolic Cotangent function ..... ie, the screen current will climb as the plate load swings more plate positive... and decreases to it's idle current of a few mA as the load line swing back down towards the idle state...then into cut-off....when the other side of the Push-Pull is already conducting within the overlap angle...
Marshall amps from the early 70's run about 500V plate and 498V screen at idle .... 1K screen dropping resistors... choke is roughly 110 Ohms..
With the 1.7K plate load and running the amp at full tilt with Siemens EL34's..... the power supply will sag to roughly 430 Plate and 400V on the screens... of course your mileage mat vary depending on the PT's regulation..
The screen current for each EL34 at idle will be about 6mA ...the screen during clipping is up about 32mA RMS ...this includes the Duty Cycle since the screen does not conduct the full degrees....roughly 225 give or take degrees in Class AB1...
The typical NOS EL34 had 8W MAX of Screen Dissipation... What size screen resistors are you using ??
One time I had seen a Laney Super Group amp have similar issues where some of the El34's were blowing up with white internals... It wound up being internal arcing of the tube that damaged the tube.... It only happened on a certain speaker cabinet... So I measured the fly-back voltages up to 1.1kV ...this arced internal to the tube between pins 3 and pin 2 heater....since heaters were at ground potential ....

For all intents and purposes, it is basically a plexi/2203 output section, with one exception. 1k-5W dropping resistors, also choke is in the same area, 115 ohms if I recall correctly. However, the exception here is that I have a 2k plate load, not 1.7 like a typical 100 watt marshall.