back to

To get more information contact me at: 

Capacitor less circuits - working in progress
Dec 2017


In this article the result of the studies and test developed with my friend Marco Ferrari Bortolini regarding a voltage amplifier stage without an output capacitor.
The capacitors have a big responsibility in the sonic performances and to have a capacitor that does not affect the sound  you have to spend a lot of money especially if you need to use a value over the microfarad.
A good review about interstage capacitors is the article published on Humble Homemade .
The my favorites are only the Audyn Cap Reference, used in my last hybrid amplifier, and Auricap used in my KT88 Push-pull.      
If you have low budget use Solen with Auricap bypass or Superior Z-Cap (red type).

The initial ideas to eliminate the output capacitor coming from 2 products:

The VTPA The Last is a vaccum tube preamplifier using a divider consisting of 2 resistors connected after the first stage to reduce the signal on the grid of the second stage to a value near to 0V.
In the Thorens the first stage drive a pentode on g2 grid to create a level shifter.
As will be show below both this configurations have a big loss of voltage gain, near to 50%.
If we accept transistors on the signal path the best method to create a good level shifter without limits is a solid state current mirror like my DC coupled Circlotron but this design can be considered a dirty solution by vacuum tubes purists .
The following circuits show in this article could be used like pre-amplifier, headphone amplifier, driver for output tubes, driver for solid state current amplifier like my Power Follower or 100W Hybrid amplifier.
For any use a delay circuit will be necessary to wait some seconds before connect this stage to the output device to wait the warm-up of tubes and and the zero leveling of the output through the DC servo.



We have stater to create a valid first stage with exceptional characteristics.
It is based on the D3a (Philips - Siemens) used in the triode connection and with anode choke to get the lower distortion and max output voltage swing.
I love this tube, in fact it is used in many of my projects, its peculiarities are the high amplification factor (about 70), the low internal resistance (about 2Kohm) and very low distortion.
Another important advantage of this tube is the sound quality determined by the manufacturers Siemens or Philips.
A good article about the use of alternative to D3a is on the good Bartola Valves article.

Follows a table to compare the distortion of some tubes in the same configuration.
Tube model Pd max (W) Ri (ohm) Va (V) Ia (mA) Rk (ohm) Voltage gain (dB) THD 30Vrms load 30K THD 40Vrms load 30K THD 30Vrms load 15K THD 40Vrms load 15K
D3a triode 4.5 1900 150 24 60 37.0 0.4 0.6 0.6 0.7
EC86 2,2 5600 165 12 110 34.5 0.7 0.9 1.2 1.7
E180F triode 3.0 2700 165 15 110 33.2 0.9 1.2 1.5 1.9
6C45 7.8 1100 148 25 43 32.5 0.3 0.4 0.4 0.6
5842 4.5 1700 150 25 60 34.2 0.8 0.9 1.0 1.2
6S3P-EV 3.0 5000 160 15 110 31.4 0.6 0.8 1.0 1.3

Here the pinout of the tube used in the following tests.



It is clear that the cathode capacitors creates a resonant RCL cell with the internal resistance of the tube and the inductance of choke.
The frequency response plot show 2 different result changing the capacitor on cathode.
To eliminate this commonly used RC group for the auto bias my friend have elaborate a bias circuit powered by the filament.

The frequency responce is not flat until 20Hz but acceptable and it can be improved increasing the current dissipated on LM317.
Using this values the circuit dissipate the following power:

I(lm317) = 1.25 / R = 1.25 / 2.2  = 0.56A

V(Rk) = 3.3 * 0.56 = 1.87V

P(lm317) = V * I = (6.3 - 1.87) * 0.56 = 2.5W

P(Rk) = V * I = 1.87 * 0.56 = 1W



Follow the method used in the Thorens TEM 3200 by Frank Blöhbaum.

This design lost 45% of the original voltage gain.
In order to have no dc on output it is necessary use a variable negative power supply with a dc servo loop like the type used in my DC coupled Circlotron.
Also a delay circuit is necessary to wait some seconds before connect this stage to the next.



Follow the method used in the VTPA The Last - Vacuum Tube Preamplifier by Bartolomeo Aloia

This design lost 60% of the original voltage gain.


This design lost 50% of the original voltage gain.
In order to have no dc on output it is necessary use a variable negative power supply with a dc servo loop like the type used in my DC coupled Circlotron (see schematic on previous section).
Also a delay circuit is necessary to wait some seconds before connect this stage to the next.



Follow the new method using a choke on grid to reduce only dc voltage.

This circuit does not lost voltage gain and the distortion is very low but there is a big problem on low frequency band also using a second choke with high inductance.
The R1 (22Kohm) create with the 270H a high pass filter so Ft(-3dB) = (R + Ra) / ( 2 * 3.14 * L ) = 15Hz.
If we reduce the R1 to 6800ohm the problem is reduced but the power to dissipate in this resistor will increase to P = Va1^2 / R1 = 2.1W and we are near the limit of anode choke.



Follow the new method using a Constant Current Source to negative power supply.
Using this method is not possible use a variable negative voltage regulator so another method will be used to create a DC servo loop.

This circuit is better than the LEVEL SHIFTER WITH RESISTOR and CHOKE because there is no problems on low frequency band.
Now we are near to the best configuration but in order to have a perfect low frequency band has been changed the input stage.

This circuit until 31Vrms is very similar to D3a version but it increase the low frequency band.



Follows a method to eliminate the anode choke and preserve the low distortion.

The 100ohm resistor on V1 anode is not necessary but important to measure anode current.

The DC servo is a variable current generator controlled by dc output offset, here a simple circuit.



Follow the last method using a variable current source to negative power supply and two costant current sources.
It is possible replace the R33 with 3.3ohm an LM317 current generator like see in the previous circuits.
This new design need 6 power supply: +135V, -135V, +270V, +15V, -15V and 6.3V.


Here the last schematic and the relative pcb with optional cascode configuration for current generators.

Here the final product with the vacuum tube 6S3P-EV , the op-amp NE5532, Caddock 20Kohm MP915, trimmer 500ohm and the mosfet IXTP01N100D.

Isolate the mosfet from heatsink because these are connected to ground.



It is necessary a protection circuit with a power-on delay like this one to eliminate the DC on output during the first minute.