A letter to my private mailbox reminds me that on a now forgotten conference some fellow was pulled up by an old-timer for a clearly wrong statement. "But," explained the perpetual noob, "I said that before when Andre was on the board, and he didn't object, so it's gotta be right, innit?"
So, lest it go into street myth because no one countered it, we shouldn't let misinformation stand without immediate correction.
Quote:
Originally Posted by Carl 
6SN7s are a flawed driver for 300Bs as they don't have enough anode current to drive them fully
|
To anyone with extensive 6SN7 *experience* Carl's flat statement about 6SN7 rings false. Perhaps Carl is confusing the 6SN7 with the 6SL7, which is a 1W pre-amp tube with less current capability than the 7.5W 6SN7.
But, to knock this street myth in the making on the head, let's investigate whether 6SN7s "have enough anode current to drive 300B fully". A good 300B transformer extends to 40kHz or so. A 6SN7 requires a fraction over 8mA to drive a 300B to 40kHz. Let's run it another 25 per cent over that just to be certain we kill the Miller capacitance. So call it 10mA at 200V at -5V bias. No one who has measured that topology at the output of the 300B still thinks the 6SN7 is limp.
The 6SN7 is actually a small power tube, rated at 5W per half, 7.5W with both halves in service. It is also very sturdy and so can be used at the full theoretical spec without reducing its lifespan appreciably. So we can bump up the current to 15mA at 250V and still get 200Vptp of clean swing with both sides in use. Or we can parallel the two tubes in the envelope for 30mA.
So, whoever originated this story that Carl retails about 6SN7 not cutting it as drivers for 300B needs to study the spec sheet for the tube and write out 100 times on the blackboard: "The 6SN7 is a 7.5W driver tube, not some limp little 12A?7." For perspective, the 30mA available from a 6SN7 is half as much again as you need to drive 845 to the full frequency extension of the best output transformers, never mind 300B. The 6SN7 is a driver tube that leaves hairy footprints.
There is also nothing more linear than a 6SN7 that can be used as a driver. That's refinement as well as beef.
I imagine that this story about the 6SN7 that Carl spreads originated with new boys on the block trying to make their mark by attacking the orthodoxy; it's an old story, trying to find something new to put your stamp on a niche. But the 6SN7 is such a revered orthodoxy because it is indispensible to low-noise sound; indispensible literally means that there is no uncompromised replacement.
Full disclosure: I use even heftier tubes as drivers in my personal amps, 417A to drive 300B, trioded EL34, 300B or other kilovolt tubes to drive kilovolt transmitting tubes -- but when, rather than a personal preference, I publish a "standard good" amp such as the T44bis "Populaire" (on my netsite, URL under my signature), it sure as God made little green apples uses 6SN7 drivers to get the right curves and harmonics and the quietest sound on the planet. Anything but a 6SN7 will by definition give inferior results; that isn't an opinion but a truism known to everyone who actually measures the noise of his amp and its composition. With electrostatics, any shortfall in linearity will be more disturbing than with lesser speakers; count on it; that is another reason to use 6SN7 with the best amps, destined for electrostatic use.
Furthermore, contrary to the subtext in this discussion (Carl's axiomatic assumption of using maximum current on the driver), there are good reasons (besides not electrocuting yourself with your headphones) not to go for the maximum current available on the driver but only for the *right* amount. The right amount is determined by your speakers, and in the case of Stax earspeakers excessive current on the driver stage in the amp can very easily be the cause of screechy treble. As it happens, 6SN7 implemented with modern conservatism as distinct from cookbook conservatism naturally fall into a mode of operation which avoids screeching earspeakers. The linearity, the correct current capability, and much good experience with the tube, are three of the reasons why I chose 6SN7s for the output tubes on the dedicated Stax driver amp (actually a family of amps with various power supplies) I shall publish when I get around to writing the accompanying text.
Andre Jute
Visit Jute on Amps at
http://members.lycos.co.uk/fiultra/
"wonderfully well written and reasoned information
for the tube audio constructor"
John Broskie TubeCAD & GlassWare
"an unbelievably comprehensive web site
containing vital gems of wisdom"
Stuart Perry Hi-Fi News & Record Review