Phono Stage – Diego Nardi φ42

Diego Nardi published his φ42 (Phi 42) Phono Preamp in Sound Practices magazine, issue 14.

This is an all-tube (valve) circuit, designed without (global) negative feedback.P8180111-sm It can be built as ‘Standard’ version or as ‘SL’ or ‘SL Silver’ versions.

Nominal Gain is 42dB (‘SL’, ‘SL Silver’ versions) or 46dB (‘Standard’ version).

In 2000, I built the ‘SL version, with 12AY7/6072 input tube.

The ‘Standard’ version of the φ42 uses a 12AX7/ECC83/7025 input tube and different circuit values to suit.  (For the ‘SL Silver’ variation Diego specified more costly Black Gate and Audio Note silver foil capacitors.)OLYMPUS DIGITAL CAMERA Key design features:

  • All-tube, no-feedback (global) design.
  • Vacuum tubes specified for the  φ42 SL are – 12AY7/6072 (high mu, high impedance triode), 6SL7GT and 5687 triodes.
  • For Moving Coil cartridges, a Step Up Transformer should be added.
  • RIAA equalisation is ‘passive’ RC type, with some sub-sonic filtering – LF rolloff between RIAA and RIAA/IEC curves (later changed, see Construction below).
  • RIAA EQ is ‘split’ into 2 networks – 3180 µs (50Hz) + 318 µs (500Hz), followed by 75 µs (2122Hz) time constants –  in order to avoid high insertion loss and to maintain high dynamic headroom & low distortion.
  • The Power Supply employs a 6X5GT tube rectifier for high voltage, with twin-pi CLCLC primary filtering for low ripple (CLCRC in ‘Standard’ version).
  • Separate HT decoupling rails for each channel.
  • Heater supplies are SS-rectified 6.3VDC, with transistor-buffered regulators.
  • 2 separate heater supplies, for V1/V2 and for V3/V4, each ‘elevated’ to separate reference voltages – to avoid heater-induced noise.P8180108-crop-sm


  • Overall Gain at 1kHz is –  42dB (46dB ‘Standard’ version)
  • Input Impedance – 47k ohms (later, I installed loading sockets, so that input load impedance can be adjusted with RCA loading plugs – see photo below).
  • Max. Input – 500mV RMS (300mV ‘Standard ‘ version)
  • Output Impedance – 200 ohms (300 ohms ‘Standard’ version)
  • RIAA accuracy +/- 0.25dB (15-30kHz, load 47kΩ)P8180109-sm


  • Shinkoh Tantalum resistors (signal circuit) – Holco metal films used initially. (The 0.5W Shinkohs are dynamic and neutral sounding, transparent without the ‘hifi’ harshness of metal films. On the other hand, Shinkohs do not have a ‘vintage’ sound.)
  • AmpOhm aluminium & tin foil, paper-in-oil capacitors (RIAA and coupling) –  Jensen aluminium foil capacitors used initially.
  • Elna Silmic II electrolytic capacitors (cathode bypass) –  Black Gate Standard or N capacitors used originally. (I prefer the sound of Silmic capacitors – smooth, extended tonal range, sense of ambience, soundstage depth.)
  • Mills ‘non-inductive’ wirewound resistors (HT power supply).
  • F & T (C1) & Panasonic TSUP electrolytic capacitors (HT power supply filtering and plate decoupling) – Elna Cerafine (C1) used initially (similar sound ‘hard’ character as Black Gate Standard caps).
  • Vacuum tubes installed presently are – RCA 12AY7 – Tung Sol 6SL7GT – RCA 6SL7GT – SPL (Raytheon?) 5687 and RCA 6X5GT rectifier.

Phi42-schematic Parts-Audio Parts-PSU IMG_5069 Construction:

  • Hard-wired throughout, using 0.7 – 0.8mm solid strand silver and copper in the signal circuit.
  • Input Loading sockets were later installed adjacent to the input sockets – for RCA loading plugs (see photo above).
  • Tubes are located close to the input/output sockets – for shortest signal path.
  • Input tube 12Ay7/6072 uses a shielded socket.
  • Twin bus-grounding is used, Sakuma-style – using 2.0mm diam. copper wire.
  • A single, wood chassis has been used for aesthetic reasons – however there are no noise intrusion problems.
  • RIAA resistor R4 was increased to 180 – 220kΩ, to provide a slight bass lift (with slightly increased insertion loss) – for more accurate LF RIAA curve (ie. no LF filtering below 20Hz).


    (The HT decoupling caps C1 & C7 have been positioned directly over the tubes – so it looks busy in there!)

This is an excellent Phono Stage.  I am currently using this with a Koetsu Rosewood Signature MC cartridge, Fidelity Research FRT-4 Step-up Transformer and the JE Labs Linestage.

Sound is wide bandwidth and dynamic, with no complaints about tonal balance or timbre.  There seems to be excellent neutrality, with no obvious colouration or ‘hifi’ character.


  • The ‘Enhanced RIAA’ curve (Neumann Time Constant), was implemented as per Allen Wright’s recommendation – but the brighter tone was excessive, to my ear.
  • Elna Silmic II capacitors have beautiful tone colour, more ‘open soundstage’ – compared to Black Gate (Standard and N) caps, which now sound band-limited (LF & HF extension), ‘hard’, ‘forward’ and coloured, in comparison – to my ear.

9 thoughts on “Phono Stage – Diego Nardi φ42

    • Hi there – I am away out of town presently. Sorry, will get back to you when I return & check details. It’s been a while.
      Thanks for your interest & I like the look if your Lenco plinth ;-). Owen

    • Costs – sorry about delay getting back to you – in 1998, the cost of circuit parts, tubes & components was around USD 850-900 – excluding the wood & metal chassis. This was using ‘audiophile’ type passive parts.

  1. A very nice project indeed, nicely constructed. However, it isn’t really a “no-feedback design”. There’s feedback across R9a (if used) and there’s feedback across R11. Not that that is necessarily a bad thing. If we were to get really into the details it should be pointed out that there is also feedback across the anode load resistors too, because they cause the anode voltage (not the HT voltage) to drop as current flow through the valves increases. Is it purely academic? Maybe, but I don’t like to see “no feedback” claimed as a badge of merit when there is clearly feedback in the design. Feedback isn’t such a bad thing – you don’t need to be ashamed of it 🙂

    • Hi Andrew – thanks for your feedback.
      In Diego’s description, he refers to the idea of global feedback being used as a means of achieving a MC-capable, all-valve phono stage with low S/N..

      (Sorry I should have elaborated, for those who haven’t seen Diego N’s SP explanatory articles.)

      Of course all valve stages (common cathode, self biased) could be regarded as having local feedback in their normal operation.

      PS. Was pleased to see your ‘real world transformers’ SUT testing discussion on your Rothwell products webpage. For my own interest, what level of signal volts input can MC SUTs typically tolerate when testing like this, ie. without risk of saturation? I should put my own SUTs through bench tests like this.

  2. Pingback: Miyajima MC cartridges – fresh thinking | D a r k L a n t e r n

  3. Pingback: LCR Phono – Lounge Audio | D a r k L a n t e r n

  4. Pingback: 2A3PP amplifier – upgrades | D a r k L a n t e r n

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google+ photo

You are commenting using your Google+ account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )


Connecting to %s