back to

To get more information contact me at:

Hi-End Class A-AB Hybrid Amplifier - the reference
started on November 30 st , 2010 


This amplifier does not born to be a normal hybrid but to create a my reference better than any other design.
The design of the amplifier stage and of the power supply are inspired to vacuum tubes amplifiers to follows the same characteristics in the sound but increasing the capacity to load any loudspeakers without problems.
I have got good result with the my previous Power Follower 99 and Hybrid 2010 but now I am want a step over and the first listen on one test channel confirm it.
There is an
incredible detail and no coloration on any frequency band, much more than any my other design.
In the output stage with the
Circlotron design , the single Mosfet and the CLC power supply for all stages will help to get the target result.
If I compare this project to others this is a not common design, also my PF2007 system is less neutral because need the output capacitor.
The my previous hybrid designs are not very cheap but these need a budget lower than the necessary to build a serious vacuum tube amplifier like my 300B PSE, 2A3 PSE or GM70/813.
In this new hybrid project the budget will be not considered and many time will be lost to search the target point.

UPDATE August 2012:

There is a prototype of this type of design developed in 1995 and published in 2004 in several editions by Yury Nagorny:


The unique problem of this design is the cathode capacitor on the first stage because it affect the sound but I cannot eliminate it.
It can be considered also an advantage because you can use this capacitor to modify the sound of your total system.

Some candidates for this cathode capacitor are:
1) Sanyo OS-Con - very neutral but a little bit too hard
2) ELNA Cerafine - very smooth, good to play normal CD
3) ELNA Silmic II - less smooth than Cerafine but still too much for me

The value of this cathode capacitor modify the low frequency band so it must be decided only after some measurements on the specific circuit (see the differences in another driver circuit).

In my last GM70 SE dc coupled amplifier with Fiat transformers I have got a good benefit using a led ( in the optocoupled) instead of cap/res. on cathode but here I cannot apply it because the low frequency band will be too much compromised.


The quality of the interstage transformers is very important in this design and you cannot use any other type without compromise the final result.
The normal Lundahl  LL1671/20mA is very good but you can ask for the
Amorphous strip core version used in my Hi-end headphone amplifier and in my DAC End.

The Spice simulation show an incredible slew-rate (20V in 200ns) because the mosfet are driven with a tube with a low internal resistance (near to 2Kohm) and an interstage transformer with a turn ratio 2:1 reduce it (4 times) this value (2Kohm / 4 = 500ohm) (here the .cir).


In my test 2SK1058  (Hitachi/Renesas) and ECX10N20R are much better than BUZ900P (Magnatec) as distortion value but the ECX10N20R (Exicon) are better than 2SK1058 because have the same distortion value but a better damping factor.

The minimum distortion with BUZ900P is at the bias point 30V DC and 900mA at any output level.

Using the 2SK1058 and the ECX10N20R the minimum distortion is at the bias point 30V DC and 1600mA and the THD level is 60-70% less than BUZ with the same input stage.

If we increase the power supply to 40 or 45VDC the ECX10N20R show a limit and for all the best bias point is 1000mA.

The 2SK1058 and the ACD100NSD have the same distortion and the same wide voltage range to get high power output. 

You cannot use IRF types and 2SK1529 in this project because these need a thermal compensation circuit.

In the simulation the output impedance is about 630mohm on all the frequency range with the mosfet ECX10N20R and about 1 ohm using the 2SK1058 and ECX10N20R.

It is a good value for a no feedback design with only one output device.

The my idea for this design is to use only common components easy to find so no custom transformer will necessary and 3 models of tubes will be tested.

This project could be considered ecological if compared to many other my design because with only 180w we have a two channels with an output power about 30w.



To get a good output distortion decay in any condition I have decided to use a single tube in this stage.
I have searched a tube with a very good linearity for large swing and low anodic resistance to drive the output stage without problems.
This  tube should have an amplification factor
greater than 40 to connect this amplifier directly to the CD player or phono pre-ampl.

To get 20w on 8ohm we need:
     P=V * I  and  I = V / R  =>   P = V * V / R  =>   V = sqrt (P * R ) = 12.5Vrms
so if we want a sensibility of 0.5Vrms we need
    Amplifier factor tube = 12.5 / 0.5 = 25

Using the LL1671/20mA with the turn ratio 4:1:1 => 2:1 we need a amplifier factor tube of 50.

Many of these tubes have a plate resistance about 2Kohm but only the D3A, 6C45 and 5842 have an amplification factor greater than 40.

D3A specifications in triode connection:

filament voltage: 6.3V
filament current: 315mA

Max plate voltage:  200v
Max plate dissipation: 4.2w
Max cathode current: 33mA
Plate resistance: 1.9Kohm
Amplification factor: : 77

6C45 specifications:

filament voltage: 6.3V
filament current: 440mA

Max plate voltage: 150v
Max plate dissipation: 7.8w
Max cathode current: 52mA
Plate resistance: 1.2Kohm
Amplification factor: 52
resistance in grid circuit: 0.15Mohm

Inter electrode capacitance:
input: 11
2 pF
output: 1.90.3 pF
transfer: 4.5pF

5842 specifications in triode connection:

filament voltage: 6.3V
filament current: 300mA

Max plate voltage:  200v
Max plate dissipation: 4.5w
Max cathode current: 38mA
Plate resistance: 1.7Kohm
Amplification factor: : 43



The most simple design to create a class A current amplifier is the mosfet.
This design is inspired to the
Circlotron US Patent n. 4229706 by James W. Bongiorno and to the Thorens TEM 3200.


I have design the power supply to follow a single ended amplifier style but you can use a common solid state design.
The CLC power supply skip any noise problem and it is superior to any pure capacitor design also if large value are used.
In the following simulations with PSU Designer II you can see the big difference in residual noise, with CLC type we get a pure sine wave so it mean low harmonics.

The Hammond 159ZJ (L=10mH Rdc=0.16ohm Imax=5A) used in this amplifier are not difficult to integrate in any amplifier because these have a compact size and these are not expensive.

There are many good capacitors to use in audio amplifier power supply but only few of these give hi-end performances:

For this project I have decided to use the Nichicon Super Through on output section and BlackGate (old stock) on vacuum tube stage because both these types will give extreme performances.

To increase the speed of the power supply has been inserted a 39uF Solen MKP in parallel to the electrolitycs.

It is not necessary a DC filament with slow turn-on but it increase the vacuum tube filament and it can be used also for the mosfet gate bias voltage.


This amplifier need a chassie with heat sink to dissipate many power also if it is not work in a pure class A operation.
The my choice is the HiFi-2000 mod. Pesante dissipante 5 units with 10mmm frontal panel got at Audiokit webshop.
On paper each lateral heat sink of this chassie have a capacity of 0.25 C/W so it will keep the output devices in a good safe area also when the environment condition are terrible 35 C.
Follows a good table to have an idea of the output power and relative heat sink necessary per each channel.

This layout optimize the signal path and reduce it to few centimeter.


IMPORTANT:  In all these measurements I have set the bias voltage for the minimun distortion, BUZ900P have this point at 30Vdc 900mA, the 2SK1058 at 30Vdc 1600mA and the ECX10N20R at 30Vdc 1550mA.












To optimize the circuit I have simulate it with Winspice.


see Amplifier End