Separate names with a comma.
Discussion in 'Glowing Bottle Tube Amp Forum' started by theprofessor, May 26, 2017.
Going from a 5U4 to a 5V4 will RAISE the B+, not lower it. Did you mean to say a 5Y3?
Professor, you're a gentleman. I appreciate your polite and respectful stance when (Annie Hall reference) Marshall McLuhan pops up from behind the signboard outside the movie theater. But the guy in the movie was a pompous know-it-all. As Mr. Kelley himself suggests, I don't think you were being that at all. Of course, he's a gentleman too. I need a better metaphor: New Yorker cartoon: You go to a mountaintop and suddenly the guru parachutes in.
Back at the amp, I do wonder if all of us vintage amp types should own a variac. Rectifiers, as BobbyZ and others have noted, have several effects on tone beyond just altering B+, and also as noted, different examples within and across series can sound quite different.
On a note I hope is relevant, this also reminds me of a question MuchXS often raised here, of why Leo loved to torture his '60s 6V6s (IIRC, he invoked a silver-tongued RCA rep). Although Much was (is, I should say) a big fan of controlling B+, thousands of those amps have been played for decades. And although it ain't an elegant solution, if you start burning through 6V6s, the JJ version is a brawny Slovak who laughs at voltage...
Oops. Meant to say 5ar4 to 5v4. According to all the voltage drop tables, it shouldn't have been that extreme on mine, but it was.
Thanks, King Fan! When Mr. Kelley responded to my second round of queries, he said:
He mentions a variable transformer in particular. This product qualifies as one: http://shop.vintagesoundworkbench.c...E325.p3plqscsfapp002?productId=1&categoryId=1
I was interested in this, but I am less interested after trying a lower voltage rectifier in my SF DR. I didn't like the sound as much, so I assume that I wouldn't like the sound if I dropped the voltage at the line instead of at the rectifier. It does make me wonder whether there would be a difference, however, in lowering the voltage at the line with a spec rectifier (5U4), versus lowering the voltage (and introducing more "sag") by using a different rectifier like a 5R4.
Yes, those JJ's don't even break a sweat. I still have the pair that came with the SF DR when I bought it. But I think that well-made vintage tubes can handle the voltage, too. My Tung-Sols should be fine, but time will tell. I think that Mr. Kelley has been known to tout Sylvania 6V6GTA's for their larger plate structure. And the JJ's sure as heck have some honkin' plates!
Assuming the amp is running at the correct line voltage, and it is biased at the maximum possible (negative) bias voltage without the appearance of crossover distortion at the onset of clipping, the power dissipation of the tubes can be reduced by lowering screen grid voltage.
This is most easily accomplished, as was previously suggested, by increasing the value of the screen grid current limiting resistors.
One can also reduce screen voltage by running the screens in both tubes from the same dropping resistor. This doubles the voltage drop, though in such a case, the resistor power rating must also be doubled. I believe there are several advantages to doing this.
Surely in a push-pull configuration, the two screens would be out of phase and alternativly reaching the higher/lower current draw situation, therefore the combined current draw and voltage drop would only be double at the quiescent state. The power rating is normaly taken from the highest current draw when the bias hits the -0V, but as the other tube at this time will be at its lowest current draw, a bit less than the quiescent, the actual difference at maximum current draw from that of a single screen resistor will not be that much different. Mind you....that being said, it never does any harm to opt for a larger safety margin.
I guess one advantage of a shared resistor is that the current draw over the resistor is more constant, therefore also the voltage drop, keeping both the control-grid bias levels more level, .....maybe this could cause less intermodulation distortion....it would also generaly lead to a more stable power supply.
Screen voltage drop is too important to power tube overdrive tone to use a shared screen resistor. Use one per tube to decouple and keep the mojo sweet.
Wikipedia: "Zener reverse breakdown is due to electron quantum tunnelling caused by a high strength electric field."
Once the quantum tunnel (gate) opens the voltage stays constant. That's why they're used so often in power supplies. It's Roswell technology that no one truly understands
Thanks Rob, drops the voltage just like a resistor, but unlike a resistor and ohms law, The voltage drop is unaffected by the amount of current. It's behavior seems almost digital like.
I would like to know more on that. I thought it was good to have the screen voltage lower than the plate voltage, for tube life, but what's the effect to the overdrive tone? I've messed with it a little and not sure what I heard.
That's a good write up and description. I see how the tone and break up characteristics are greatly affected by the screen, and the voltage on it.
I am under he impression that people don't tweak screen voltage for tone purposes as much as for tube life.
Might want to do a bit of checking on something like that first - if it's anything like a Variac it may not be isolated from line voltage, requiring an isolation transformer in front of it as well.
The 1977 DR schematic calls for 120v, anyway, and I assume that applies to my 1976 as well (I've included the image below; I don't know if it's easily visible; it says 120V / 60Hz. So if I'm running between 120-122v A.C. (which is all I've seen at my home, since I've been checking in the last couple of weeks), I'm running at the spec'd line voltage. So I guess I don't need a step-down transformer. Plus, @Silverface : I don't want you to torture me with 24 straight hours of Styx's Greatest Hits as a punishment.
Dude, there will be payback - I just shot hot coffee out of my nose!!!!
I always considered the signal voltage cancellation a plus because I don't want signal on the screen grids anyway. I recommend choosing the best value resistor for your application. There's nothing magical about the number 470.
Professor, if you would like to hold a 'sidebar' discussion on WHY the difference between 12W and 14W plate dissipation came about, contact me directly.
Why not tell all?
Surely the RMA release 201C, from 3 nov.1958 , that I posted in #25, showing and stating that everything is the same, and that the tube had not changed , only the rating system, is the sole reason...or do you have an explination as to the reason why they changed the chosen datasheet rating system designation?
I was just looking through releases 201 and 201B where slight changes are present, but do you know the 201D from RCA? They wanted to change the transfer function of the 6V6, but it was rejected so 80 years on, we still have much the same tube as Ken-Rad first presented to us.
If the 201D wasn't rejected, then we could be having wonderfull debates about how the RCA's aren't "real" 6V6GT's.
You outlined it well, here's my brief summary...and addition (at bottom):
pre-1942: Nothing specific was stated on data sheets prior to 01-Apr-1942.
post-1942: "ABSOLUTE" values were stated after 01-Apr-1942.
1943: "DESIGN CENTER" values were 'added' to updated data sheets starting in 1939; 'required' for all tubes in April 1943...per RMA
1959: "DESIGN-MAXIMUM" values were stated after April 1959...per EIA.
Thus, tubes introduced prior to 1942--such as 6L6 and 6V6, etc.--all had their specs "adjusted" but there was NO actual physical change to the tubes, ie: numbers change only.
However, later, SOME tubes had a physical change made in their plate material that enabled them to dissipate MORE heat which (a) increased their plate dissipation values and/or (b) enabled them to use smaller glass envelopes. The change in plate material (a) originated in April 1949 when GE engineers developed a "ply" metal-forming process, (b) which GE applied for a Patent for in 1951, and (c) was granted to GE in October 1954. [W.H.Kohl, Material and Techniques for Electron Tubes, 1960, Reinhold Press, pp. 191-193] This patent was later shared with RCA, Sylvania, Westinghouse, and Tung-Sol.
• RCA introduced their 5881 in 1954 (for use as servo-driver in B52), it employed the new 5-ply "Copper-Cored Aliron" (Aluminum-Iron-Copper-Iron-Aluminum), allowing a 20% increase in plate dissipation (23W) over the 6L6GB (19W) that it was developed from [stated in 2/57 RCA tube manual].
• Tung-Sol introduced their 5881 to HiFi in September 1950.
• Tung-Sol introduced their 6550 to HiFi in November 1954.
• GE introduced their 6L6GC in Jan/Feb 1960 utilizing their 5-ply plate material; suffix "C" appended due to this change, Ppd now 30W and Psd 5W. [Ham News, Jan/Feb; Vol.15, No.1; pp. 1 & 7.] *
At some point just about ALL power tubes began using the multi-ply plate material (patent had expired)...that's why most later year (post 60's) were basically all QC-derivatives of MIL-SPEC production: (1) Fully meeting spec = MIL-SPEC tube; slightly out of spec = Industrial tube; marginally out of spec = Commercial tube; fully out of spec = crushed/destroyed.
* http://n4trb.com/AmateurRadio/GE_HamNews/issues/GE Ham News Vol 15 No 1.pdf
P.S. - notice the "Unretouched photo" and caption on page 7: "...showing a 6L6-GB (left) and a new 6L6-GC (right) with five-ply bonded plate material, each operating at a plate dissipation of 80 watts. The 6L6-GB has "hot spots" on the plate, bright orange in color, while the plate of the 6L6-GC shows only a uniform dull red, due to the superior heat dissipating characteristics of the new material."
Doubt that any current 6L6GC-labelled tubes would survive 80W plate dissipation very long.
Here is some early stuff I ran across. Back in Roosevelt's day.......
Then the following year. from the RMA again......Radio-Engineering-1934-04
Early in 1933 the Radio Manufacturers Association (RMA) rationalised the US valve numbering system. In 1934 they started sending out questionairs to all the manufactuures to start the standardization, such as this very early release of theirs #6
In 1934 the RMA (Radio Manufacturers Association) began to allow developers and manufactures of electron tubes to reserve a designation for a proposed new type of tube they where evolving. When the technical specifications and parameters where established, and working examples could be produced of the resulting development, spec-sheets of the new tube would be submitted to the RMA. The new tube on the "Electron Tube Registration List" would be issued an official release number, and the relative spec-sheets were distributed to RMA members, who where then free to develop compatible tubes of their own design under the same designation, so long as the registered specifications and parameters were followed.
Such classic tubes that appeared after this slow restart of the tube production included the release #64 the metal tube 6L6 march 1936 which is still with us, and closly followed by the much loved 6V6G, release #96.
It seems the Department of Commerce, Bureau of Standards push for the electronic industry standardization in November 1941 might well have been the reason for more detailed limits to be included on the data sheets. But I see that the earliest 6V6 data does say "max."
...and now you know why I wanted a "side-bar" with the professor since the answer to his question was a meandering "...yes, but also because of..."
...and, no, I never knew of the RMA #201D from RCA. I got most of my information from my employers, Hughes and Raytheon, AFTER they'd already closed their vacuum tube manufacturing facilities by raiding their Tech Libraries before they dumped the documents too. Unfortunately, my IEEE membership and access expired when I retired, so I only have what I accumulated over 28 years of searching.