cathode amplifier:
    Gain = muRa / (Ra + 2rp)
Consequently, a 6AQ8 will yield a higher gain; the 12AU7, a lower gain. Changing the input tube will, however, also require changing the noise cancellation voltage divider's resistor ratio. A potentiometer might be the best idea.
    The second circuit uses the ratio of the input resistor divided into the feedback resistor to set the gain, which must be less than the open-loop gain of the amplifier. Increasing the input resistor's value runs the risk of limiting the high frequency response of the amplifier because of the Miller effect capacitance. Nonetheless, I am sure that we do not need bandwidth to 1 MHz in a headphone amplifier.

Which reduces to Gain = mu / 2.
     Thus, the gain can be set by using different input tubes, such as the 12AU7, 12AT7, 12AV7, 12AX7, 12BH7, 12FQ8, 5751, 5963, 5965, 6072, 6211, and the E80CC, without having to change the values of the cathode resistors. For octal fans, the only two real choices are the 6SN7 and the 6SL7.
     The beauty of this input stage is that it must also halve the power supply noise, as it forms a ½ voltage divider. And halving the noise is what the output stage needs to see in order to ignore the noise at its output. The output tube can be any low mu, low rp triode. I like the 5687 because its linearity and its huge cathode, which allows for huge current draws. The downside to this triode it its 0.9 amp heater current draw. Other triodes that can be used are the 12AU7, 6XB7, 6BL7, 6922, 12B4, 12BH7, 12FQ8, E288CC, and the 6AC7 triode connected or 6V6 or 6BQ5 or 12BY7.
     I plan to use this circuit as line stage amplifier / headphone amplifier. When its function is driving the power amplifiers, the output coupling capacitor will be only a 1µF InfiniCap. When its function is driving the headphones, the output coupling capacitor will consist of the InfiniCap in parallel with a 40µF Solen polypropylene capacitor. The power supply will be kept in a remote box and will use a choke based filter rather than a voltage regulator.

Subject: Concerning "5687 black plate tube"
     Recently I came across a preamp schematic called "Pat's ultimate preamp." It talks about 5687 black plate tube. It also says: "I fine-tuned it in Tube CAD" Do you have any further details about this? I would like to build a tube preamp for  my dry sounding transistor guitar amplifier

Kai,
Norway

   The 5687 is mentioned in the answer to the last e-mail. I own about 40 5687s, but I do not know how many, if any, are "black" plate, so I cannot comment on their sound. Tube CAD is a tube circuit analysis program that I wrote. It contains 52 circuit topologies and costs $39. I use it almost every day in my tube design work and although it is not perfect, I recommend it anyone who builds tube projects. The best place to read about it is the GlassWare web site at http:/www.glass-ware.com .

    The circuit I am building today is shown above. A variation on this circuit was briefly covered in the last issue and in a previous letter in this issue. It uses an optimally designed White cathode follower as an output stage, wherein Ra equals the reciprocal of the triode's Gm. The input stage is a simple grounded cathode amplifier with an active load. By using the same triodes and the same valued cathode resistors, the gain is easy to compute:
      Gain = muRa / (Ra + rp + (mu + 1)Rk)
In this case,
      Ra = (mu + 1)Rk + rp
Substituting yields:
      Gain =              mu[(mu + 1)Rk + rp]           .
                   (mu + 1)Rk + rp + (mu + 1)Rk + rp

< PREVIOUS

NEXT >