SETA - Single Ended Tube Amplifier

This is my first vacuum tube amplifier, original design by Andrea Ciuffoli

Such a topology is pretty difficult to implement, due to a severe input voltage swing request (over the power tube
grid). Such a large voltage can be obtained by the means of different solutions:

• Coupling Transformer, working also as voltage elevator. Requires a low plate impedance and sufficient power for the driving stage reducing the range of usable tubes.
• Inductive driver load, cannot work as an elevator, require a large output capacitor and a very low series impedance (i.e.: very low Driver Plate Resistance), does not perform very well in the lower range of the frequency spectrum.
• Enhanced Triode mode, big names (Tim de Parravicini) adopted this kind of topology. It has a good output swing, but the inverted usage of the grids generates a very high output impedance. such a situation forces the usage of a power tetrode/penthode for driving the output. Furthermore the stage linearity is pretty questionable.
• Shunt Regulated Push Pull, require a supply voltage much higher (double) than the output stage. It means that a second power supply is required.
• Single Ended Hi voltage, same as above, plus the need of a sufficient power triode.

• an inter stage coupling transformer with a ratio of 1:1.125.
• a very linear driver, the 6SN7 indirectly heated triode.

Since a low damping factor is very difficult to achieve with a vacuum state amplifier, it is often kept rather hidden,
nevertheless its bad impact over any loudspeaker system, especially for those with a complex impedance figure or
significant phase rotation over the frequency spectrum.

As output tube I am using a power pentode in triode connection instead of a direct heating power triode because the cathode follower configuration add itself a local feedback that increase the linearity of the output stage and is preserved the sonic characteristic of driver stage.

This project use a reference design vacuum tube power supply rectifier with choke input.
The main vantage of choke input rectifier is to skip any current peak generation on rectifier diodes that are present in normal capacitor input design.

The filaments are in AC and I have choice this type instead the DC because the normal DC power supply design with large capacitors generate very high current peak about 50A and the spectrum noise is amplified (the gain in this case 250V/6.3V = 40x so about 32dB) and go on high voltage secondary given terrible effect on sound.
 

The main results should match these items:

• power about 13watt rms with 8ohm nominal load
• very low distortion about 1% at the max power
• very high damping factor more than 15 with an output resistance near 0.6ohm
• very good low frequency response about 10Hz (-3db)

We have a first simple voltage stage with one section of the 6SN7 coupled by a capacitor, the uinque interstage capacitor of this amplifier, to next stage.
The voltage gain of this first stage is about 17x (22dB) and little chaneg of the anodic resistance should be done to optimize the sound and gain from 27K to 39K.
The second stage is loaded by .......

The Hybrid Graetz Bridge concept after a Fulvio Chiappetta paper, published by the Italian Magazine 'Costruire HiFi' a reference for the DIYers.
He tested and comparated this topology versus a pure tube vacuum state power supply design with a 300B output tube amplifier and there was no difference both while listening and measuring the output noise spectrum.

I built this power supply and the sonic result is very high even if it uses semiconductors the sound depends on the rectifier tube.

Infact the vacuum tube rectifier is in series with the solid state diode and the switch characterictic is of the lower speed device so the tube.

The 0.27 ohm resistors could be adjusted to have correct 5V on filaments of tubes and a value range from 0.18 ohm to 0.33 ohm should be good to test and match any rectifier.

To reduce a little the cost of this project without compromise the result is possible to use only one tube rectifier per both channel but the two secondary 11-13 and 12-14of the transformer must be paralleled to balance flux.

This must be considered as the last solution when cost and size are very important or as a starting point to test this amplifier.

Here the same power supply transformer of the reference version is used but in this configuration the high voltage secondaries are in parallel to get 250V instead 500V.

I am currently finishing up the final adjustments for the driver stage bias/load line. The amplifier has been tested with different output tubes and supply configurations, by two listeners (Andrea Ciuffoli and Francesco Pintavalle)
The overall sonic result is excellent : werll balanced and very natural.
Lower end does not resemble a tube amp so much, it is clear and tight, deep and defined. Effects of the good damping factor.

It is of interest the sonic influence of the supply tubes, where the usage of a 5R4 RCA gives impressive results in particular for the firmly focused soundstage.