diyAudio

diyAudio (http://www.diyaudio.com/forums/)
-   Analogue Source (http://www.diyaudio.com/forums/analogue-source/)
-   -   The "Muscovite" 6S3P Tube Phonostage (http://www.diyaudio.com/forums/analogue-source/213769-muscovite-6s3p-tube-phonostage.html)

kevinkr 2nd June 2012 11:22 PM

The "Muscovite" 6S3P Tube Phonostage
 
6 Attachment(s)
This project got started for no other reason than I needed another phono stage to use with my second turntable. I named it the Muscovite simply because it uses the Russian 6S3P-EV high frequency triode.

The design is built on a couple of repurposed Tektronix vacuum tube pulse generator chassis.

I thought I would have some fun with this design so it literally includes everything but the kitchen sink. The input stage is what I would term a mixed bias cascode with diode and grid leak bias. The second stage is a choke perverted version of the mu-follower which uses a combination of battery and diode bias.

It was also intended to be cheap and use parts I have in my stash.. (mostly)

The design has minimal PSRR and must run from a very quiet power supply - in my case tube preferred, but I suspect the Salas HV shunt regulator could be a good choice.

The gain is deliberately insanely high (>54dB) as I use a line stage with 4dB of voltage gain, and I wanted CD reference level output. It also had to be quiet.

Some technical information:

Gain: ~56dB
RIAA: +/- 0.1dB with trim and selected parts.
Input capacitance: Low, in theory 30pF is possible, with care 50pF or less is achievable.

Edit 2013-05-12: Please see later posts in thread for current HV supply and audio circuit designs.

kevinkr 2nd June 2012 11:25 PM

2 Attachment(s)
I am now using this with my TD-124/II/Schick/SPU set up and much prefer it to the one it unintentionally has replaced.

There are a couple of variants to the circuitry that I have not explored, but will share here for those who are interested.

I have included an alternative biasing arrangement for the mu-follower upper triode for those concerned about possible battery induced audio colorations. In practice I have not had a problem with this issue and was not looking to spend too much on this design so it's economical. No detectable downside so far.

I have also included a way to apply battery bias to the lower tube in the mu follower. This would eliminate potential distortion in the mu-follower due to modulation of the 1N4148 dynamic impedance due to variations in the audio signal current. Again I have not heard or measured a problem with this, and have not tried this approach and probably won't.

Note that there will be an Edcor plate supply transformer as I have ordered a custom for mine.

The filament transformer is an Antek AS/AN-0509.

With care this ought to be possible to duplicate for $300 or less with careful parts shopping. Mine cost a bit more than this, but has a small number of boutique parts in critical locations.

The tubes themselves are extremely plentiful and very cheap, they also perform rather well.

Salas 3rd June 2012 12:09 AM

Nice. Is R3 (U2) essential? Does it surely oscillate without? Ups input noise.

SY 3rd June 2012 12:11 AM

54dB is "insanely high"? Let's see, at 5mV nominal, peak is likely to be 20mV. Just because I'm a finger-counting mathematician, 60dB above 20mV is 2V, so 54dB is half that, 1V. Seems like it's not insane, it's optimum.

kevinkr 3rd June 2012 12:13 AM

Power Supply (Breadboard Version)
 
1 Attachment(s)
This is the power supply as it currently (and temporarily) exists. Note that despite my best efforts there are a few exposed high voltage points so this is definitely not recommended except in a lab situation..

Should note that as this phono stage design operates on 300V and the raw supply to the regulator is well over 500V that care in its construction and testing are required. (Construction, testing, and use at own risk.)

The design shown in the photograph utilizes a 1.5H choke, and manages a measured output noise and ripple level of 0.35mVrms (300V/35mA load/120Vin @ 60Hz) which is acceptably quiet, with the 5H choke specified in the schematic above the noise levels will be significantly lower if care in construction is taken.

SY 3rd June 2012 12:14 AM

Quote:

Originally Posted by Salas (Post 3046306)
Nice. Is R3 (U2) essential? Does it surely oscillate without? Ups input noise.

No, it really doesn't in any meaningful way. Cartridge DCR is likely to be ~1k and the noise adds as power. Given cartridge inductance and reduced shunting of the input resistor at HF, the resistor's contribution is even smaller than small.

kevinkr 3rd June 2012 12:17 AM

Quote:

Originally Posted by SY (Post 3046307)
Just because I'm a finger-counting mathematician, 60dB above 20mV is 2V, so 54dB is half that, 1V. Seems like it's not insane, it's optimum.

You're off one decimal place.. I get 2.5V out with 5mV in.. :D (60dB: 1000X)

Salas 3rd June 2012 12:17 AM

Why burning a huge drop? Utilizing an existing Tx reason?

kevinkr 3rd June 2012 12:19 AM

Quote:

Originally Posted by SY (Post 3046309)
No, it really doesn't in any meaningful way. Cartridge DCR is likely to be ~1k and the noise adds as power. Given cartridge inductance and reduced shunting of the input resistor at HF, the resistor's contribution is even smaller than small.

SY is correct, the noise contribution of that 100 ohm resistor is negligible compared to the noise resistance of the cartridge/transformer combo.

Salas 3rd June 2012 12:20 AM

Quote:

Originally Posted by SY (Post 3046309)
No, it really doesn't in any meaningful way. Cartridge DCR is likely to be ~1k and the noise adds as power. Given cartridge inductance and reduced shunting of the input resistor at HF, the resistor's contribution is even smaller than small.

SPU is extremely low DCR I thought.


All times are GMT. The time now is 02:51 AM.


vBulletin Optimisation provided by vB Optimise (Pro) - vBulletin Mods & Addons Copyright © 2014 DragonByte Technologies Ltd.
Copyright 1999-2014 diyAudio


Content Relevant URLs by vBSEO 3.3.2