[ Home | Staff & Contacts | DIY & Tweaks | Listening tests | HiFi Playground | Music & Books ]

TNT PreAmble N - DIY Nuvistor Dual-Mono preamplifier

Balance exercise

[Italian version]

Product: TNT PreAmble N(uvistor)
Company: not for sale, TNT-Audio free DIY design
Approx cost.: 300-450 Euro, depending on components used.
Author: Giorgio Pozzoli - TNT Italy
Published: December, 2003

After the good results of the Preamble there was another component type I wanted to test in the single ended circuit topology I am working on.

Nuvistors were the last born in the classic tube world: designed when the transistors were already taking the lead, they are miniaturized thermionic tubes in a metal case. They look like very small metal cylinders (1cm=2/5" in diameter and less than 2cm=4/5" in length) with a ceramic bottom inside from which terribly slim pins come out. The pins are fortunately protected by the metal enclosure, that protrudes into two differently sized tags; these tags insert as keys and guides into the very small special socket necessary for these very special tubes.

In spite of the size, Nuvistors normally can cope with a rather normal Vb voltage; anyhow, most of the suggested working point are at rather low voltages, even though the suspect is that the producers were trying to convey the image of low voltage tube in order to make it a direct competitor of transistors in the same kind of applications and with a similar power supply voltage.

In fact the cost of a power supply unit increases very rapidly with the output voltage, so low voltages were mandatory to reduce costs. Unfortunately, tubes were badly penalized also by the necessity of a filament power supply and the limited life, and the battle was soon lost: Nuvistors just resisted a few more years where wide bandwidth and very low noise were required.

As soon as solid state devices reached the same performance levels, Nuvistors production ceased. This happened many years ago, and there is very scarce chances to see it start again. So they are available exclusively as NOS or used tubes. By the way, most of the still available components were recently been wiped out of the market, so that it would now really be difficult to think of any further industrial utilization of these tubes.

They were used normally in high frequency or very wide band applications (e.g. radio, tv, Tektronics oscilloscopes), but also in some very specific audio systems. And in spite of their scarce notoriety, they were used by major companies: for example Conrad Johnson (Premier 6 and Premier 7 pre-amplifiers, where nuvistors were used in the head pre-amp), McIntosh (MR65, MR71 tuners, in RF stages), Harman Kardon (Citation VII tuner again in RF stages) and also AKG (C-12A microphone, as pre-amp) and Neumann (U47 microphone, again as preamp in some variants: in facts the small size suggested to use them also as internal pre-amplifiers in these famous microphones).

More recently, they were recovered from the tubes' paradise, where all honest tubes one day will go (hopefully not before something better has been found, as often happens in our imperfect world...), by mr Michaelson of Musical Fidelity, who built a few very high quality products based on this technology. The Nuvista series are really famous in the audiophile world, and probably eliminated any further chance of seeing other industrial scale applications of nuvistors.

Following my never ending battle for the best sound at the lowest cost, I was mainly interested in them as a rather low cost alternate solution to tubes, as I had seen them presented in low voltage applications with very good distortion rates.

Unfortunately, the nuvistor data sheets show characteristics that are not so ideal in terms of linearity, and in facts the circuit did not prove so exceptional with regards to measures.

In any case the unit, mixing the best of tube and solid state, express and overall performance that I am not able to consider the result less than fully positive. And so here is the design.

The new unit

Most of the circuits used are exactly the same of the solid state Preamble. The original Preamble prototype was made up of three boards:

For Preamble N, I re-used the same input card and the original power supply as "high tension" Vb for the nuvistors, without any modification at all. All I had to do was to substitute the gain card with a nuvistor based one and to add a filament power supply.

Hence, most of the considerations valid for the solid state Preamble apply also to the Nuvistor version, and I am not going to discuss them again. Please have a look at the Preamble article for more info.

The audio circuit

Once again, the overall circuit disposition remains the solid state Preamble one: input selector followed by volume control and active stage.

The set of trimmers in series to each input remains in place, but as a matter of fact the have not proved necessary at all: you can probably get rid of them without major problems (unless the tube set you use is really poorly matched).

The input selector is always the same high quality model from Palazzo, with gold plated contacts, and the potentiometer is still the AURA D7A Model 1 stepping regulator. Any other implementation is possible, however.

The active stage is as simple as usual. A 7586 nuvistor in common cathode and automatic (cathode) bias configuration, without any cathodic bypass, drives another common plate (cathode follower) 7586 nuvistor tube.

The unbypassed cathode resistor limits gain and distortion. It introduces an amount of feedback, but it is local, intrinsic feedback, and as such it does not cause any problem.

Once more, the circuit is simple and works fine.

Only a couple of caveats. If you find big gain differences from one channel to the other try to match the two channels gain by measuring the output voltage with a fix sinusoidal input, while moving the nuvistors around the difefrent sockets and taking note of the output voltage. And if you can measure distortion, you should better do the same keeping distortion under control, as I found quite different values with different tubes; nothing really worrying or preventing the system from working, but why not try to minimize it, if it is possible. In any case I was not able to go lower then 0.7% @ 2V rms output in my configuration with the four tubes I had at disposal.

I made quite a lot of calculation based on the curves, and also quite a few tests, using other nuvistors, for example 6CW4 or the 8056 (that is specifically a low voltage tube), but I was not able to get anything better. I also tested a configuration presented on a paper review a few years ago as giving a 0.07% distortion at full output, but I measured more than 1.5%... never believe to what you have not measured yourself (and, I would add, never trust too much even your own measures: mistakes are very easy...).

The output is coupled as usual through a very high quality polypropylene capacitor. I used a 10uF cap, but the value is not so crytical (anything from 1uF on can do, the higher the stronger bass you'll get).

Even for nuvistors, special care is suggested for the power supply on the audio board: two paralleled capacitors, a 3300uF electrolytic (more of a reservoir than a real filter component...) and a 0.47uF paper in oil, are placed near the nuvistors.

[TNT PreAmble N Line Stage schematic]

The power supply

The "high" voltage power supply is exactly the same used in the solid state Preamble: I just quote form the previous article.

"The connection to the mains is through a detachable mains cable, connected to an IEC socket on the unit back panel. The socket unit contains also the mains switch, the two fuses (one for each mains lead), a mains filter.

From this point on, the unit is completely dual mono: there are two transformers and two PSUs, one for each channel. Each power supply unit is very simple because of the fact that only one supply voltage is required, given the circuit topology, exactly as in SE tube designs.

The transformers supply 30V RMS. You can obviously use a 15+15V transformer provided you are able to correctly make a series connection of the two secondaries (which is highly expected, if you are going to address this project...). The rectifiers are ultra fast discrete diodes, type BYV27-100. This helps reducing the switching noise.

They are followed by a set of capacitors, ideally polypropylene for the lower values and electrolytic for the higher ones. These provide a first filtering of the mains frequency ripple.

Then comes a MOSFET series voltage regulator. The configuration is usual, but here the values used are rather special. The time constant of the network at the gate of the MOSFET is in fact higher than 2000sec... that is, the units takes something like 3 hours to reach 95% of the final voltage!!!

The effect is a terrible smoothing of any ripple. The power supply becomes very very silent."

Given the fact we have tubes in place, it is not suggested to keep the unit up all time, due to their limited life (in any case huge, of the order of 50000 hours, for nuvistors with respect to other tubes). So there is at least one real drawback: the unit takes a few minutes to get operative, and it is not able to express its best performance before some time.

Normally the powering up is very smooth, and no special bump comes out of the loudspeakers if the amp goes on during Preamble N power up phase: so no special protection has been used. There is in any case a potential risk: the large current flowing in the output capacitors at power up. With my power amp, which has a the output protected against DC (if any DC is detected at the output, the output is disconnected), the transient currents are temporarily detected as DC voltage, and protections are operated. In any case, I do not believe this transient could really damage a loudspeaker.

[TNT PreAmble N HV PSU schematic]

One more PSU have been added for filaments (one for both channels). This again is a regulated PSU with a topology very similar to the high voltage PSU one.

The transformer is a a 15VA unit with two 12V secondaries, connected in parallel. The rectifiers and capacitors are the same type as above stated.

The time constant of the regulator is also less heavy than the analog PSU one, but there is a specific arrangement that allows the unit to get to 90% of final voltage in a very short time. At power up, C308-9 are completely discharged. The voltage across C302-C307 increases very rapidly, while C308-9 start charging slowly, so the drop across R302 tends to increase, but as soon as it reaches the Vf (forward voltage, typically around 0.7V) of D305, the diode lets the current flow through, rapidly charging up the C308-9 capacitors; as soon as the difference in voltage between the central contact of the potentiometer and C308 becomes lower than the Vf of D305, currents flows only through R302, and the circuit starts working as a high constant regulated PSU.

Note that a similar solution could also be applied to the HV PSU, but while in the filament PSU the Vf of diode is such a percentage (>10%) of the final voltage as to reduce enormously the probability that impulsive disturbs coming from the power line get through the diode to the audio subsystem, in the case of the HV PSU the drop would be only less than 2%, which in my view could possibly not be enough.

The potentiometer must be set for a 6.3V output.

[TNT PreAmble Filament PSU schematic]

The implementation

Again, I am not able to do anything better than quoting from my previous article.

"All the circuits are mounted on single side printed circuit boards. As usual, the copper side is used as ground plane and all the components are mounted on this side. Have a look at the assemblying instructions for more information.

The internal connections, between the boards and from these to the RCA connectors) are in rather thin, solid core, silver plated copper wire. When the connections are longer, I used isolated solid core wire from an UTP Cat.5 network wire.

Follow exactly the line stage schematics in wiring the input pins. The input jacks are isolated from the back panel, all the cold leads of the input jacks of each channel are connected togather. All ground leads of each channel meet only in one point, which is the ground star center shown in the schematic.

If possible, even the connections between R105, R108, R109, R110 and ground should be implemented using separate leads going dorectly from the resistor to the ground star center.

Special care must be taken to minimize ground loops. With a dual mono construction this is not so easy: the configuration shown below is the one that works best according to my tests, but it depends probably on the mounting structure."

[TNT PreAmble global schematic]

Please remember to connect the ground plane of the Nuvistor board to the ground star point: it reduces slightly the background noise (which is very low, lower than -96dB scale of my instrumentation...).

The sound

As usual, I am not going to review the unit at all, I know I could not be objective.

Just to help the reader to decide if it is the case of going for Nuvistors or stick to solid state, I will say that the sound is in both cases extremely precise, controlled, neutral (frequency response is flat to one hertz...), and does not lack at all in dynamics.

Also (but I should better say especially, given the technology used) for Preamble N I can say that its sound is very different from what is normally considered "tubey" sound. Nothing more distant from Preamble N presentation: tight, clear, very fast, authoritative, controlled sound; the only characteristics shared with normal tube systems is the absolute transparence, the absence of any hint of grain, the sparkling high frequencies and the wide and precise soundstage.

In facts, the major difference between the two sister units are the slightly more detailed and naturally sparkling high frequencies, and the slightly more transparent and musical sound: it is, in a lower scale, as in best tube systems: music seems to come out of the speaker and immediately start dancing around, light as a breeze.

What is lost to tubes in terms of sheer musicality is anyway gained in weight and presence, so that result is as convincing and authoritative as to constantly produce "air guitar" effects: the unit is able to convey this involving presentation, while preserving at the same time a perfect control, without that hint of constrained or forced quality that happens to feel, more than to hear, with solid state.

But it must be really clear: the differences are not large at all, and even though I cannot say thay are difficult to perceive, for sure a good to very good level overall system is required to be able to get aware of them.


The design of this pre-amp has been one of the most demanding tasks I had to face with. The working point selection have been reviewed and challenged many times, as the Nuvistor type to be used.

The result is in any case a pre-amp that pays back for all the efforts spent.

Spent up to this point, at least, I'd better say. Yes, because this is just the first battle of the war, you damn'd and loved Nuvistors...

Part List

Line stage - une unito for each channel

Code Value Rating Notes
C101 3300uF 50V Electrolytic
C102 0.47uF 50V Paper in oil
C103 10uF 50V Polypropylene
R101 47k 1/4W 1% Holco
R102 47k 1/4W 1% Holco
R103 47k 1/4W 1% Holco
R104 47k 1/4W 1% Holco
R105 1.2k 1/4W 1% Holco
R106 33k 1/4W 1% Holco
R107 1k 1/4W 1% Holco
R108 120k 1/4W 1% Holco
R109 120k 1/4W 1% Holco
R110 22k 1/4W 1% Holco
R111 220k 1/4W 1% Holco
VR101 500k 1/4W 1% trimmer
VR102 500k 1/4W 1% trimmer
VR103 500k 1/4W 1% trimmer
VR104 500k 1/4W 1% trimmer
VR105 potentiometer   see text
U1 7586   Nuvistor
U2 7586   Nuvistor
SW101 Selector   2 ways 6 positions
  Pin RCA Isolated, gold plated 12 required

HV PSU - one unit for each channel

Code Value Rating Notes
C501 1nF 50V Polypropylene
C502 0.47uF 50V Polypropylene
C503 3300uF 50V Electrolytic
C504 3300uF 50V Electrolytic
C505 3300uF 50V Electrolytic
D501 BYV27-100   Superfast Diode
D502 BYV27-100   Superfast Diode
D503 BYV27-100   Superfast Diode
D504 BYV27-100   Superfast Diode
R501 3.3k 1/4W 1% Holco
R502 22k 1/4W 1% Holco
R503 625k 1/4W 1% Holco
R504 10k 1/4W 1% Holco
M501 IRF830   Power MOSFET
TR501 PSU Transformer 30VA Prim:220V Sec:30V (or 15+15V)
  Mains Filter with IEC socket with power switch and fuse

Filament PSU - one unit for both channel

Code Value Rating Notes
C301 0.47uF 25V Polypropylene
C302 4700uF 25V Electrolytic
C303 4700uF 25V Electrolytic
C304 4700uF 25V Electrolytic
C305 4700uF 25V Electrolytic
C306 4700uF 25V Electrolytic
C307 4700uF 25V Electrolytic
C308 4700uF 25V Electrolytic
C309 0.47uF 25V Polypropylene
D301 BYV27-100   Superfast Diode
D302 BYV27-100   Superfast Diode
D303 BYV27-100   Superfast Diode
D304 BYV27-100   Superfast Diode
D305 1N4148   Signal Diode
R301 1R 5W Power resistor
R302 150k 1/4W 1% Holco
R303 4.7k 1/4W 1% Holco
VR301 20k 1/2W Multiturn
M301 IRF610   Power MOSFET
TR301 PSU Transformer 15VA Prim:220V Sec:12V

© Copyright 2003 Giorgio Pozzoli - www.tnt-audio.com

[ Home | Staff & Contacts | DIY & Tweaks | Listening tests | HiFi Playground | Music & Books ]