WOT tube line-stage redux

A Line-stage With Output Transformer (WOT) was built back in 1998, a single 5687 tube per channel with output transformer (OPT).

However I failed to get rid of noise (EM pick-up and attenuator switching noise).  I lacked the expertise to optimise Grounding and to shield and/or orientate the output transformers (OPTs) properly.

Now, I have been fortunate to acquire a pair of Tango NP-126 Line Output transformers. So, I must re-visit the WOT Line-stage.

Diego Nardi-designed 5687 WOT Line-stage (with 6X5GT rectifier)

The original concept was simply one tube stage followed by the Output Transformer to drive cabling and the power amp input. A coupling capacitor is unnecessary.

Later, this was converted into a Berman/JE Labs Line-stage – with Type 76 tube + 6SN7/6J5 cathode-follower output, to drive cabling & the power amp input.

Berman/JE Labs 76-6SN7 Line-stage with external PSU

Now, a conversion back into an OPT-coupled line-stage, using the (ISO) Tango NP-126 Line Output transformers below…

(Top graph shows single-ended to single-ended connection.)

A 6SN7-WOT Line-stage (again, only one half of a 6SN7 used per channel).

(Later, I will also test a 76-WOT topology, but right now this configuration sounds wonderful!)

6SN7-WOT Line-stage – a temporary installation shown, with copper top plate removed).

Transformer-coupling offers the following advantages:

• Removal of a coupling cap (to my ear, ALL capacitors introduce a ‘sound’ flavour).
• Reduction of plate load resistance, more efficient plate circuit and PSU.

The new circuit schematic is below:

Notes:

• I have connected the NP-126 for 20k:1.2kΩ configuration (initially using some Kiwame secondary loading resistors that I have on hand).  Later I will test 10k:600Ω which, according to the above Tango datasheet, should give slightly better bass response.
• The 6SN7 (or 6J5) operating point is as suggested by A Direct Heating Triode Blog – very linear and low distortion.

• The PSU choke-input, LC cells are specified not only for adequately low ripple, but also keeping the RC time-constant not more than 15 milliseconds for the first LC cell.  Thereafter, RC time-constant increases from the first to last decoupling cell. (As recommended in the above design, to maintain a ‘punchy’ sonic response.)
• The DC heater supply is ‘elevated’ above Ground as previously (to minimise cathode-heater potential and any hum pick-up).

Correct phase:

A single-tube stage inverts ‘phase’, as shown when the 6SN7 plate is connected to the ‘start’ end of the OPT Primary winding :

Input (top), output (bottom) – 180 degrees out of phase.

However, with OPT Primary winding reverse-connected, output is in phase with input signal:

(Excuse some noise on output trace – my rough probe technique.)

Sound:

• When the signal is in correct phase, stereophonic effect is maximised, unsurprisingly.  The sound changes slightly, with more stereo image precision and ‘sharpness’, more treble brilliance, image width and ‘expansiveness’.
• Transformers (coils, inductors generally) deliver signals well in the lower frequencies, where the ‘power’ in music resides.  Dynamic ‘horsepower’ is much improved – rhythm and percussion, orchestral weight.  Dynamic ‘freedom’ and headroom seems ‘open’ and effortless.
• Some would point out that transformers have restricted bandwidth, especially high frequencies, – however, distortion (harshness) seems audibly lower than resistance devices (eg. resistors, potentiometers).  Smoother, more ‘coherent’, natural, it seems.
• These sonic effects are similar to what you hear with other coil devices – eg. Transformer Volume Controls, choke-input PSUs.

Advertisements