|
|
||||||
|
|
||||||
|
A way
to get a lower noise and a lesser distortion from an amplifier is to inject, in
phase opposition, a fraction of the output voltage to its input. When the output
energy (voltage or current) is out of phase with the input signal of the
amplifier, it is said that it has a “negative feedback”. The greater the
feedback lower will be the distortion and less noise will be produced, this at
the expense of the gain.
One of the classical configurations of tube amplifier is the one called “cathode follower” (C.F.), i.e. Common Anode (Fig. 1 and 2). This “amplifier " has 100% negative feedback then leads to a very low distortion and having the noise generated by the tube in phase opposition with the input, also at its deletion or at least to its large decrease. Other features are its wide frequency response, high input impedance (grid) and a low output impedance (cathode). The only drawback is the low voltage amplification which is always less than 1, the maximum obtainable is approx. 0.98. The C.F. was (and is ) often used in Audio Frequency, even as an impedance adapter instead of a transformer, in some application for Radio Frequency (VFO buffer, instrumentation, etc.) but very little as a receiver RF amplifier. Why, given the excellent features of this configuration? Probably the absence of amplification has limited its employ.
The proposed circuit is in Figure 3, derived from that of Figure 2, it exploits the high input impedance of this version of the CF, combining it with a tuned circuit whose turns ratio provides the "gain". The high impedance lets take full advantage of the Q of the LC circuit and thus to obtain the maximum selectivity. This arrangement of the C.F., with the so called “Bootstrap effect”, allows manipulating more intense signals. In Figure 4 we can see the broadband version; the LC circuit has been replaced with a broadband transformer T1, whose impedance ratio is 1:9, then the primary/secondary turns ratio equals to 1:3. Thus if the source impedance is 50 Ω that of the load should be 450 Ω. The adoption of T1 is not, of course, for an impedance transformation, but to exploit its turns ratio. Consequently our signal will be "amplified" in voltage by 3 times (about 9.5 dB), this signal will not be "nicked" minimally thanks to the high input impedance of the C.F. Clearly the amplification is lower than that obtained with the tuned circuit but is more than sufficient. Note that at the input of T1 can be connected antennas with various impedances (within a limited range) without varying the "gain" too much (it also depends on the transformer type used), this is not to say that the antenna is matched! Simply, signals at the ends of the line are "processed", or amplified, by the circuit. The employ of antennas with different impedances to the input of T1, and the secondary "open" modify the original characteristics of the transformer, in particular its bandwidth that tends to be reduced. In one of the prototypes built I used a Mini-Circuits transformer type T9-1, which has given good results. The bandwidth from 3.5 to 30 MHz was fairly flat with a roll-off in the upper portion of the band. The amplifier alone (without T1) reaches 100 MHz. This amplifier utilizes a 6DJ8/ECC88 tube with the two triodes connected in parallel which permits to make it work at almost full capacity in order to bring the distortion to very low levels. In the electrical diagram it can be seen a 100 Ω resistor in series to the grids to avoid the occurrence of any self oscillations. Even the cathode followers can auto-oscillate! The 1 dB input compression point is found to be 10 dBm with a load impedance (ZL) of 50 Ω and with ZL=1 kΩ, reaches 20 dBm. Note that the values shown are those actually measured at the amplifier input (A-B). The sensitivity tests, with the amplifier followed by an HF receiver (SSB mode - IF BW = 3kHz), have led to the following results; a signal of -130 dBm was easily discernible, this throughout the bandwidth of the amplifier. Practical tests with improvised antennas have confirmed an excellent sensitivity. Output/Input isolation is more than 25 dB. This is an important aspect in its use as a RF preamplifier followed by a mixer in order to mitigate the possible feedthru of the local oscillator. The amplifier just described is an interesting circuit with a large dynamic range that deserves to be perfected. For example, by connecting a 450 Ω resistor to the secondary of T1 "stabilizes" the frequency response but to the detriment of the gain. You can get around using a transformer with a higher turns ratio. A similar version of the circuit (without T1) was used as a post amplifier (500 kHz) in the double triode passive mixer (Ref.1). |
||||||
|
References:
1-
www.radiopharos.it
N.B.
The contents (circuits, drawings, etc.) of this page belong to the writer and
owner of this website. You can use them for your pleasure and homebrew purposes. It
is forbidden to copy them and use them for commercial business without the
owner’s consent.
Copyright©
2014--2021 All rights reserved |
||||||
