4.3 Input stage of an integrated vacuum tube amplifier

The input stage of an integrated vacuum tube amplifier is basically a voltage amplifier, which was already discussed in Chapter 3:.

An example of input stage is shown in Figure 27. A resistive load Ra is connected to the anode of the vacuum tube. The output voltage is taken at the anode itself, before the load. The grid bias voltage for the input stage is obtained by using self-biasing with a cathode resistor Rk at the vacuum tube cathode. The cathode resistor is bypassed by a capacitor Ck to reduce the local negative feedback, produced by the cathode resistor, and to increase gain. Cathode biasing and local negative feedback details were already discussed in Section 3.6.2.

Figure 27: Basic schema of the input stage of an integrated vacuum tube amplifier.
The input stage is basically a voltage amplifier. It takes the input signal, coming from an external source, and applies it to the grid of the vacuum tube. Cathode bias is generally used in this stage, and the cathode resistor is generally bypassed by a capacitor to reduce the local negative feedback and increase the gain.

Next example shows how to draw the loadline, set the quiescent operating point, and chose the cathode resistor for self-biasing the input stage.

Example 12: Loadline and bias of the input stage

Suppose, for instance, the input stage is built using a 12AX7 vacuum tube, the high tension V+ is 300 V, and the load Ra is 220K Ohm.

When the vacuum does not conduce, the anode to cathode voltage is 300V. In the theoretical case that the vacuum tube does not offer any resistance, the anode current is 300V/220K Ohm=1.35mA. By connecting these two points, we obtain the loadline plotted in Figure 28. A good operating point is identified by the red spot. It corresponds to a bias current Ib of 0.65mA and an anode to cathode voltage of Va 160V. This can be obtained with a grid bias voltage of -1.5V. Since we are using cathode biasing, the grid is at ground level and we have to elevate the cathode voltage to 1.5V, by computing an appropriate value of the cathode resistor Rk. By using the Ohm law, we have that Rk=Vk/Ib=1.5V/0.65mA=2.3K Ohm. The closest standard resistance is 2.2K Ohm, which is a good approximation.

The bypass capacitor Ck has the purpose, as discussed in Section 3.6.2, to reduce local negative feedback and increase gain. Small capacitance values increase gain just for high frequencies, high ones increase gain also for low frequencies. For instance, in our case, using the calculations discussed in Section 3.6.3, we determine that a value of 100μF is sufficient to bypass and increase gain at all audible frequencies.

The grid stopper resistor Rg is used to block very high frequencies that can enter the circuit and parasitic oscillations, by forming a low-pass filter with the internal vacuum tube capacitance. As we already said, values around 47K Ohm are generally used with 12AX7 vacuum tubes.

The use of a grid stopper resistor is particularly important at the input stage of an integrated vacuum tube amplifier. We are at the very beginning of the amplifier stages and signals that have to be amplified are very small. All noises, interferences, parasitic oscillations are here significantly amplified through all the remaining stages. For instance, consider that the wire connecting the input jack to the gird cannot be generally very short, for practical assembling issues. Therefore, it acts as an antenna, capturing electromagnetic interferences, which must be blocked before being amplified.

The potentiometer Rv is used to control the volume of the amplifier, that is the amount of input signal applied to the grid of the input stage vacuum tube. The potentiometer also acts as a grid leak resistor and forms, with the coupling capacitor Cc, an high-pass filter that blocks the unwanted low frequencies. The coupling capacitor Cc also isolates the input stage from possible DC coming from the external input source.

Example 13: Coupling capacitor of the input stage of an integrated vacuum tube amplifier

The value of the coupling capacitor Cc, can be computed using the high pass filter formula. Suppose the Rv potentiometer is 100k ohm, and we want to stop all frequencies below 7 Hz. We have:


Figure 28: Loadline and operating point of the input stage.
Loadline and operating point obtained with a 12AX7 vacuum tube, with an anode load of 220K Ohm, and a cathode resistance of 2.2K Ohm.

4.3.1    Directly Coupled concertina

In the Example 9 we determined that the quiescent cathode voltage of the concertina was 100V. In order to obtain a bias voltage of -1.4V, with respect to the cathode, we had to bring the grid voltage to 100V-1.4V=98.6V, by using a voltage divider. The coupling capacitor had the main purpose of isolating the grid voltage from the anode voltage of the input stage. In fact, in Example 12 , the quiescent anode voltage was 160V. The coupling capacitor, basically, had the purpose of isolating the 98.6V quiescent grid voltage, of the concertina, from the 160V quiescent anode voltage, of the input stage. In this way, just the AC signal was allowed to go from the anode of the input stage to the grid of the concertina.

Figure 29: Directly coupled Concertina.
In a directly coupled concertina, bias voltage for the phase splitter grid can be taken directly from the anode of the input stage. In order to do that, loadlines and operating points of both the input and the splitter stages should be chosen so that, in the quiescent state, voltage at the anode should be below the voltage at the cathode of the splitter of a value corresponding to the wanted grid bias voltage. This allows eliminating the coupling capacitor between input stage and phase splitter, and the voltage divider, needed to set the grid bias. Eliminating the coupling capacitor is particularly relevant, since less components along the signal path always improves sonic quality of the amplifier.

However, in many cases, it is possible to set the quiescent operating points of both concertina stage and the input stage of an integrated vacuum tube amplifier, so that the quiescent anode to ground voltage of the input stage is exactly what is needed at the concertina grid. This allows directly coupling the input stage and the concertina stage. The input stage quiescent anode voltage is used to bias the concertina, thus eliminating both the coupling capacitor and the voltage divider. Eliminating those components, not only makes the schema simpler and cheaper. It also improves sonic quality of the amplifier. In fact, remember, that all components along the signal path slightly degrade the audio quality of the amplifier. The resulting schema of the Directly Coupled concertina is that shown in Figure 29.

Example 14: Biasing for directly coupled Concertina

Suppose we use the configuration from Example 9 for the concertina stage. We determined that the needed grid to ground voltage is 98.6V, to produce a grid bias voltage (grid to cathode voltage) of -1.4V. Correctly configuring the input stage, and directly connecting the input stage anode to the concertina stage grid, without coupling capacitor, can accomplish to this.

With an input stage high-tension voltage Vi+ of 300V and a load of 220K Ohm we obtain the input stage loadline depicted by the violet line in Figure 30. With a grid bias of -0.7V, we obtain an anode to cathode voltage of 98V and a quiescent current of 0.9mA, as depicted by the blue spot in the figure. The needed grid bias of -0.7V can be obtained, using the cathode biasing technique (see Section 3.6.2). Choosing a cathode resistor 820 Ohm we obtain a cathode elevation of 820 Ohm∙0.9mA=0.73V, which is close to what needed. The anode to ground voltage of the input stage is 98V+0.73V=98.73V, which is also very close to the voltage that we need at the grid of the concertina stage.

Using these values, we are able to directly couple input stage and concertina splitter stage, saving components, costs, and improving sonic quality of the amplifier.

Remember that we have also to consider the AC loadline, to check that the concertina operates linearly when handling a signal. Suppose that the power stage grid leak resistors values are 180K Ohm. These resistors, parallel to the anode and cathode resistors of the concertina, give approximately 65K Ohm and produce the AC loadline depicted by the green line in Figure 30, which shows that the concertina operates in a fairly linear area.

Figure 30: Loadlines and operating points for directly coupled Concertina.
Suppose we set the high-tension voltage of the concertina to 380V, and the anode and cathode resistors both to 100K Ohm. We obtain the DC and AC loadlines depicted by the red and green lines in the graph. Suppose we set the quiescent operating point to 1mA current. This can be obtained with a grid bias voltage of -1.4V. With 1mA quiescent current, the cathode voltage of the concertina is 100V, so the grid voltage should be 100V-1.4V=98.6V. In order to direct couple the input stage and the concertina, the input stage anode voltage should be set to this value. This can be obtained by setting the input stage high-tension voltage to 300V, the load to 220k, and the grid bias to 0.7V. In this way, we obtain the loadline depicted by the violet line, and the wanted quiescent anode voltage, as depicted by the blue spot.

Proudly powered by WordPress | Theme: Baskerville 2 by Anders Noren.

Up ↑