[King Kong]

I never seen one B4, but how can they do that? The 300B is very hard
to drive in this way as input and driver tubes even with transformer??

[Patrick Turner]

On Feb 6, 6:10*pm, King Kong wrote:
I never seen one B4, but how can they do that? The 300B is very hard
to drive in this way as input and driver tubes even with transformer??

I myself don't much believe in 300B for anything. However, I have re-
engineered a VAC 70-70 amp which had 4 x 300B per channel, and after
my work it sounded just dreamy, proving yet again that PP directly
heated tubes could really perform outstandingly, when given a chance
by surrounding circuitry, so that there was not a hint of any bass or
treble inadequacy which many ppl believe is the hallmark of the 300B,
even though midrange is a glory.

Some years after that, I had a pair of 1996 Ming Da 100W monoblocs
brought to me for *quit-smoking* therapy and singing lessons. These
truly horrible Chinese amps had 6SN7 SET SRPP input tube, then 6SN7
LTP, then a pair of 300B in a balanced amp, and finally a pair of 845
in PP, operating with Ea at 1,230Vdc, and an anode RLa-a which was
half what it should have been.

The 300B driver stage was arranged with 3mAdc in each as the idle
current. It overloaded first and produced more THD than the OP stage.

After perusing all, I told my customer there was only one solution, to
gut the entire amp, and re-design/build the lot. The biasing method
for both 845 and each 300B used 4 dodgy Chinese pots. The amp was a
splendid effort at making a visually impressive amp, but was like the
most handsome young doctor you've ever seen, but who botches all his
operations.

I retained the tube line up, but re-arranged the PSU to give a far
higher B+ of around +600V using a "split rail", +450V and -150V so the
Vswing could be generate about 200Vrms from each 300B anode while
loaded resistively, 50k at each anode, and without a CT choke or IST
which is the usual way to get large Vswings using power tubes. When
finished, after a couple of hundred hours of work, these amps sounded
truly glorious, and my customer was mightily happy, so you *are
allowed* to use power tubes as a driver. The trouble is that 300B have
a gain of only 4, so if you need 120Vrms max to drive 845, then the
300B needs 35Vrms to drive its grid, so the LTP before 300B using 6SN7
needs 5Vrms to one input, and the SE input then needs about 0.3Vrms.
When 12dB GNFB is added, Vin becomes 1.2Vrms for the 85Watts into 8
ohms I could get using the 4 ohm tapping on the Chinese OPT.
If one uses EL84 instead of 300B, and with CT choke for DC feed, as
spelled out in my 8585 and 300W amp design schematics at my website,
200Vrms is still possible, but gain is about 18, so that ahead of the
driver stage you only need ONE more stage not two, which means better
amp stability with NFB.

For driving OP stages with 845, there is no grid current required, so
its actually easy; but you *do* want linearity at high signal levels,
and using 300B is doable as I mentioned above and could be improved
upon if the Ia is increased to say 20mA instead of the paltry 7mA, and
to get that to the anode the best way is via a choke with CT, with say
4k7 R between ends of choke winding and the anodes, to negate the open
loop phase shift effects of the choke shunt capacitance and shunt
inductance. The added 4k7 act to make a shelving network, try it,
you'll like it. I'm a great believer in choke feed to driver tubes
even in SET amps and where an air gapped choke is used and with a
series R, see my pages on SE55 845 amps with a pair of parallel 845
driven with 3 parallel EL84 in triode mode. http://www.turneraudio.com.au/monobloc845se55.html

Patrick Turner.

[John L Stewart]

Originally Posted by King Kong

I never seen one B4, but how can they do that? The 300B is very hard
to drive in this way as input and driver tubes even with transformer??

No problem. The driver can be bootstrapped to the OP section, just as it is in the McIntosh. There are two easy ways to do that as I showed in articles driving 6AS7/6080 in Glass Audio magazine Volume 2 & 3 in 1999. One version bootstrapped to taps on the OPT attached here. Another version uses split load resistors on the driver to get the correct FB.

+ve FB is about 2 db. One version of this amp using 2X 6080 developed an honest 30 watts of audio at very low distortion. Ra-a was 2150 ohms.

Cheers, John