Thursday, October 31, 2013

UX201A Sound Processor, Part 1 : Circuit

Hi!

In the Tube of the Month Article about the UX201A I already mentioned it's unique sound qualities. This is the first post of a series of articles about a device which I call a 'Sound Processor'. It uses a single UX201A triode per channel as active element. The purpose of the device is to present the sound of the tube itself in it's purest form with as little coloration as possible by other components in the signal path. This is the extremely simple circuit that does this:






As you might have noticed this circuit has no capacitor in the signal path at all. There is also no hidden one in the power supply since that is decoupled through a choke from the circuit.

To achieve this the circuit uses an enhanced version of filament bias. Filament bias effectively removes the need of a cathode bypass capacitor which would be in the signal path. It pulls the entire filament current through a resistor to ground which thus becomes very small in value (22 Ohm in this case). So small compared to the plate resistance of the tube that it can be neglected and does not need a bypass capacitor.

But the usual filament circuit still has a capacitor in the signal path on the B+ side. In case of a transformer coupled stage this is the capacitor from B+ to ground (or B+ to cathode in case of the ultra path connection).

In case of the circuit above this capacitor is replaced by a resistor, the 500 Ohm resistor from B+ to the filament. 500 Ohm is quite small compared to the plate resistance of the tube to which it is basically connected in series. I coined the name 'DirectPath' for this circuit since it DC couples the B+ side of the transformer primary to the filament. It also sounds very direct, removing the last capacitor from the circuit and thus revealing the tube sound in full.

This 500 Ohm resistor serves another purpose at the same time. It derives the filament voltage for the UX201A from the B+. Since the UX201A only requires 250mA filament current, this is doable in this way. Nothing is free however and this approach has a big disadvantage. The 500 Ohm resistor dissipates about 30W. This requires a massive resistor on a heat sink. Also the power supply needs to be quite substantial. It has to deliver 500mA for a stereo unit. Luckily the UX201A runs at low plate voltages of 135V max. In this circuit the plate voltage of the tube is about 125V. The B+ which is fed to the circuit needs to be somewhat higher to allow for bias voltage and the drop within the choke.

The 100k resistor in the circuit serves as grid to ground resistor to ensure the grid always has a defined potential. The cold end of the secondary of the output transformer is connected to ground through a 1M resistor to avoid any possible static charge build up in case the output is unconnected.





I have tested and used this circuit successfully many years ago. I am now working on a new implementation using Tango NP8 line out transformers. The assembly process and power supply will be covered in upcoming articles. Stay tuned.

Best regards

Thomas


20 comments:

  1. Hello Thomas

    Is your sound processor a pre or power amplifier or a completely different animal to be used in conjunction with these amplifiers?

    Thanks

    Pan

    ReplyDelete
  2. Hi!

    This is for example to be used between source and preamp. It will especially help with digital sources. It can also be built as line preamplifier

    Best regards

    Thomas

    ReplyDelete
  3. Forgive my ignorance, but does this circuit have gain or is it a buffer? Can you explain it's function further?

    ReplyDelete
  4. Hi Nick,

    the purpose of the circuit is to subject the signal to the unique distortion characteristics of this tube to alter the sound. The gain depends on the output transformer used. With the one I chose it has two output taps. One without gain and one with about 6dB gain

    Best regards

    Thomas

    ReplyDelete
  5. Hi Thomas,

    The 1M resistor is connected to B+ on the schematic, not ground. That must be a mistake...

    What is your experience with lifetime for these tubes and maybe also the rest of the family of small american DHT's of that era?

    ReplyDelete
  6. Hi Bjorn,

    you are right, thanks for pointing that out. As described in the text the 1M pulls the secondary to ground to avoid static charge. I updated the schematic.

    With these old tubes it is difficult to give a good answer as it is hard to tell in which condition the tubes are which you get. Even if advertised as NOS you can never be really sure until they come in sealed boxes which is very rare.

    I also found that with these old tubes you will have more often tubes fail due to broken filaments than loss of emission.

    I have not used a device with these for a long enough time to share any meaningful experience. I have used these tubes occasionally over the last 10 years.

    Best regards

    Thomas

    ReplyDelete
  7. Hi Thomas,

    Is the 7H/250mA choke 80% nickel type from Dave? Thank you.

    ReplyDelete
    Replies
    1. Thanks Thomas..

      If I was to use film or film/oil caps for the main psu filter instead of lytics, what minimum value would you suggest?

      Delete
    2. Hi!

      I published a suggested PSU with cap values. If you want to use other values, try what works for you

      Best regards

      Thomas

      Delete
    3. Thanks Thomas..

      If this was a linestage, I suppose the avc's would be within the same chassis but with an aluminum plate of some sort to shield them from the 7H chokes. Thanks for sharing the schematics.

      Delete
    4. Hi!

      The AVC can be in the same chassis or separate. Shielding should not really be necessary

      Thomas

      Delete
  8. How do you polarize the tube. I capacitor seems missing no ?

    Thank you

    ReplyDelete
  9. Hi! If by polarise you mean grid bias, that is through the 22 Ohm resistor. No there is no capacitor missing it is all explained in the post. The point was to get it capacitor less

    ReplyDelete
  10. Thomas, I am wondering about the output transformer, can it be gapped for no DC, it looks like it. I have these real nice old stock 0dc output transformers that i hesitate to use without a caps to protect them.
    best wishes, David

    ReplyDelete
    Replies
    1. No, the OPT carries DC albeit only a few mA

      Delete
  11. Four parts for an amp, we need to reduce that even more ;)

    Love what you did!

    ReplyDelete
  12. I have have a nice pair of UTC LS-27, how would they pair with the 01a'S? the primary is 15K

    ReplyDelete
    Replies
    1. I don't have the complete technical specs of that transformer to judge

      Delete
  13. Thomas, the transformer datasheet shows a 600ohm resistor typically across the secondary. Is this needed? I am a little confused. your schematic doesn't show the resistor.

    ReplyDelete