EL84 Amp Design...Why Such High B+ Voltages?

[kg]

The other thing that's getting blurred in this discussion is that we're talking about two different types of tubes- beam power tubes (6L6 & 6V6) and pentodes (EL84s & EL34s.) Supposedly, beam power tubes are less prone to the screen current/dissipation taking off at clipping / Vg=0.

true--IF the beam tube is well-constructed. (that may have been the case for the majority of tubes in the "golden age," but QC ain't what it used to be.)

the g2 winding pitch is supposed to be identical to the g1 winding pitch, putting it in the shadow. take a look at figure 10.6:



also mentioned on that page is the effect of the space charge "virtual cathode" formed between g2 and anode. with the "parallel flow" of electrons formed in the spaces of the g1 and g2 windings, a high electron density exists in the g2-a space. that high electron density is negatively charged, and this forms a virtual suppressor field--a much more "perfect" field than that formed by a coarse g3 windings of a traditional pentode.

That's presumably why you'll see data sheet listings for AB2 connection of beam power tubes, but I've never seen or heard of anyone running pentodes that way.

as the text states, the BP tube is "in many respects superior to the conventional pentode." however, the ability to operate successfully while the control grid pulls current isn't one of them.

g1 current basically increases cathode current. no matter what kind of multigrid tube is in use, whichever element has the highest potential is what's going to "suck up" most of that cathode current. unaligned g1-g2 pitch (as in pentodes, and poorly constructed BTs) will make it more likely that g2 is going to get hit. however, running a pentode with low vg2 and into a low plate Z (keeping Vpmin to a higher value) would allow successful use (ie longevity) despite the presence of g1 current.

[riz]

kg, what book is that?

[kg]

that's karl spangenberg's vacuum tubes, the majority of which is available on my site:

Spangenberg

(i think there's a complete version in pdf form floating around somewhere on the net, but mine is higher res.)

[octal]

as the text states, the BP tube is "in many respects superior to the conventional pentode." however, the ability to operate successfully while the control grid pulls current isn't one of them.

g1 current basically increases cathode current. no matter what kind of multigrid tube is in use, whichever element has the highest potential is what's going to "suck up" most of that cathode current. unaligned g1-g2 pitch (as in pentodes, and poorly constructed BTs) will make it more likely that g2 is going to get hit. however, running a pentode with low vg2 and into a low plate Z (keeping Vpmin to a higher value) would allow successful use (ie longevity) despite the presence of g1 current.

This is very interesting. I still wonder why I've never seen published curves for the positive grid region for pentodes. (Are they out there and I've missed them?) Maybe the pentode manufacturers simply decided not to promote AB2 operation for whatever reason? My sketchy knowledge of tube history suggests that pentodes were originally a European thing and beam power tubes were an American thing, perhaps that has something to do with it- different design practices for different cultures?

[Jazz P Bass]

Alittle power tube history.
The Pentode tube came about as a way to solve the screen current problems of the Tetrode.
The Tetrode has some weird shit going on.

As the Pentode solved the issues of the Tetrode it also retained a few of it's difficulties.
The Beam Power tubes (Tetrodes by the way) came out on top.
The tube is considered easier to drive, less prone to distortion & more efficient than either the true Tetrode or the Pentode.

The Hi-fi guys ate them up.

Then there are guitar amps........

Pentodes & Beam Power tubes are a diffent construction.
They act & most importantly (for guitar players & such) sound different.
Cool.

[kg]

another interesting thing about pentodes: using g3 at some potential other than cathode.

check out figure 11.6:



the knee of the plate curve @ vg3=40 vs vg3=0 is SO MUCH SHARPER it's crazy.

perhaps on pentode amps that abuse the shit out of their screens, the answer might be to elevate g3, instead of increasing g2 DCR or lower vg2?

[Steve Conner]

A little more history: In the beginning there were only triodes. To get a decent amount of power from a power triode, you are pretty much obliged to drive the grid far positive and supply lots of grid current. This made the driver stage a real challenge. While the triode itself is very linear, the grid current is not, so the driver had to have a very low output impedance if the amp wasn't to generate lots of distortion.

The pentode was invented to break this tradeoff. Here was a tube that could generate lots of power with high gain and no current at all. Watts and THD per dollar were much improved over triode amps, heaps of unwanted Western Electric 300Bs were banished to the dumpster, where in a rational world they would have stayed.

So, I guess the tube makers couldn't imagine why anyone would want to drive the grid current that they worked so hard to design out.

And as stated the beam tetrode was invented to offer the same advantages without paying royalties to Philips.

[Jazz P Bass]

The reduction of interelectrode capacitance of the triode was another reason for the design of the tetrode/ pentode/ beam tube.

[Gregg]

perhaps on pentode amps that abuse the shit out of their screens, the answer might be to elevate g3, instead of increasing g2 DCR or lower vg2?

I think I've seen this trick in a Traynor amp.

[kg]

as far as i remember, the traynors actually did the opposite: tied g3 to raw bias supply.

this thread seems to agree: Bias on suppressor grid? - AMPAGE Archive

i wonder if dai h was right about steve c.?

[J M Fahey]

Early Mesa amps (the first ones) were of course able to and meant to be cranked (ask Carlos)

I was backstage when Carlos played Buenos Aires in the early 70's and was amazed not by his using 2 tiny Mesa Markż1? but by his having another two by them to be able to switch "in seconds, when (not *if*) he blows one of the main ones" (as told by his assistant !!!!).
I could not believe it, my experience with "American" amplifiers way back then was mainly with Fenders and Ampegs which seemed to last forever.
Another thing that I found even more amazing was the row of speakers with *huge* magnets (Altec? / EV SRO?), out of their boxes and face down on tables, complete with screwdrivers by them , waiting also backstage. Clearly they were meant to be replaced onstage, *during* the show if necessary.
So, in a nutshell, I can confirm that those early Santana Mesas *did* blow often (and they blew their speakers too).

[Chuck H]

Ha ha... I could have guessed. Tone like that comes at a price. Those early "Boogie" amps were just hopped up Fenders. Later "Mesa" amps were clones of the first ones. Collectible now Of course Carlos was playing through some really nice guitars and had the sustain points marked on stage (I can make any amp sustain if I'm playing a Yamaha SBG and standing in the right spot). I do think it was a great tone though.

Serious shame on Randall Smith if those early amps were prone to failure under their intended use. The one I gave away LIVES cranked and just keeps rockin' so I guess he figured it out. Of course it's heeavily modified. One of the first things I did was take out the "dyna watt" circuit, rip out all the treble bleeders and change the goofy input circuit. The stock amp felt unresponsive and sounded dull. Nice speaker though. I changed it out a couple of times and always go back to the original. I think it's very much like an Eminence 1058, but a little better IMHO. Cheaper too.

P.S. Yes, some Traynors had a funky -V on the screens thing. Never seen an elevated G3 on one. Interesting idea. Not saavy enough to get it other than to guess it must reduce output as a consequence of increased reliability. Just reducing screen voltage (as also suggested) will do that. But would there be a tonal difference???

[Steve Conner]

KG: Yes, I remember doing that. The screen current reached a minimum at Vg3=0, and increased both ways.

[Gaz]

Really, at "0"? That's surprising actually. I'm gonna try that myself in an amp where I've got the screens hooked up to the bias supply.

[redelephant]

The pentode was invented to break this tradeoff. Here was a tube that could generate lots of power with high gain and no current at all. Watts and THD per dollar were much improved over triode amps, heaps of unwanted Western Electric 300Bs were banished to the dumpster, where in a rational world they would have stayed.

Hehe, I'd like to see u post this at diyaudio.com

My history books claim the tetrode - then the pentode, were born to get rid of the miller capacitance that exsists between grid and plate in triodes. The developement those days was in radio and the need for faster tubes that can operate in high frequencies. The fact that they also were easier to drive was just a bonus. (Cascode triodes and u pretty much have a pentode...)

ooops...sorry, off topic...