Simultaneous High & Low Frequency Amplification

ID: 588370
This article refers to the manufacturer: To the manufacturer

Simultaneous High & Low Frequency Amplification 
20.Sep.22 00:37
698

Gary Cowans (AUS)
Editor
Articles: 109
Gary Cowans

Simultaneous High & Low-Frequency Amplification

By P. G. A. H. Voigt.

Extracted from Wireless World December 10, 1921 Page 560-562

There are many circuits in which a valve will amplify both HF & LF currents simultaneously, thus obtaining with one valve an amplification for which two valves are usually used. This is a great advantage to those who have to fetch & pay for their filament current.

I have tried many of these double amplification circuits, and those given are among the best.

First, I shall explain the single valve double amplification circuit in Fig. 1, then I shall go on to the more complicated ones.

The HF oscillations are received on the aerial in the usual way by means of a variable inductance (shown fixed for clearness) and a variable condenser in parallel for long waves. The HF currents flow to the grid and to the filament through the blocking condenser, thus imposing an HF oscillation on the plate current. This oscillation produces oscillations in the tuned plate circuit similar to those in the aerial circuit but amplified. The oscillation here is passed to the crystal detector by means of the secondary winding, which should be very closely coupled to the primary, and wound or connected in the opposite direction as shown by arrows in the diagrams.

Since the secondary only carries the crystal and reaction condenser current, it may be wound with very fine wire. This has the advantage of reducing capacity effects. The secondary may have more turns than the primary.

The oscillation on reaching the crystal, causes it to pass a unidirectional current which will charge the blocking condenser between filament and earth.

The voltage on this blocking condenser will flow to the grid & filament & vary the plate current without affecting the HF oscillation, producing doubly amplified signals in the telephones.

Capacity reaction is used because in this instance it is as efficient magnetic but is much simpler to construct, as there is no need to make a variable coupling between any set of coils.

The capacity reaction condenser should have a maximum capacity not exceeding 0.0001 microfarad and may be easily made as follows: -

On a sheet of insulating material, a sheet of metal 5 cm by 5 cm is laid. This is covered by another sheet of insulating material, (i.e., 10 mil ebonite or waxed paper), and another sheet of metal 5 cm by 5 cm fitted with a flexible lead, and a long handle is laid on top so that the overlap can be varied. The minimum capacity must be small, and it should be possible to slide the sheets at least 5cm from one another. Better-looking condensers do not give better results but are more expensive.

The principle of capacity reaction is quite simple. Suppose at a certain instant the wave makes the grid positive, then the plate current increases and makes the plate more negative, if the transformer secondary is connected in opposition to the primary the change of plate current will make the crystal and one side of the reaction condenser positive, the charge on this side of the condenser will attract an equal negative into the other side, leaving an equal positive charge free to add itself onto the initial charge, thus increasing it.

In these circuits, a crystal not requiring a potentiometer, such as zincite, and bornite, or one of the artificial galena’s sold as permanite, rectarite, etc., must be used.

With hard valves, the grid may be made slightly negative. This reduces valve damping, and sometimes gives an increase in signal strength.

Fig. 2 shows how a valve detector can be used instead of a crystal, but the second valve is much more useful if used in one of the two valve double amplification circuits shown in Figs. 3, 4, and 5.

In Fig. 3 the second valve is coupled to the first by resistance coupling. The reaction condenser is connected to the plate of the second valve. This circuit works well at 600 metres, but like all resistance coupled amplifiers, its efficiency is low with short wavelengths.

The efficiency for wavelengths below 2, 000 metres can be considerably increased by shunting the resistance by a small condenser (0.002 µF) and putting in series a tuned circuit as shown in Fig. 4.

A switch is used to cut out the resistance condenser for “stand by” and the three tuned circuits used for “receive”. This circuit is wonderfully selective.

A further improvement in LF reaction is shown in Fig. 4.

By putting a second condenser between filament and earth, separating the negative of HT and LT by a variable resistance having a maximum of 5,000 Ω, and making the voltage drop in this resistance by means of a potentiometer, LF reaction can be obtained which doubles or trebles the signal strength.

Suppose the LF current made the first grid negative, then the first plate current would decrease and make the second grid positive, and the second plate current would increase to a much higher extent. The resultant through the HT and series resistance would be increased current. Hence the voltage drop along the resistance would increase and thus producing a higher negative potential on the first grid.

A suitable resistance wire would probably be too fine for a slider, and therefore as many tapings as possible should be used.

For the reaction resistance and the plate resistance, I advise one of the fine resistance wires, such as 48 S.S.C. Eureka, which has a resistance of 342 Ω per yard. About 150 to 200 yards would be required for a plate resistance.
When the HT is in order and the aerial earthed, the telephones should be absolutely “dead” unless the set is receiving some atmospherics or signals with the aerial earthed.

On long wavelengths, this method of amplification is difficult to work with, but for ordinary wavelengths the method is quiet and sensitive that I can receive the Dutch Concert (250 watts) on a four-foot square frame aerial indoors (top floor) with only 17 turns, using the circuit given in Fig. 4 in an open position in London.

Fig. 5 shows how transformers may be used for the inter-valve coupling; the HF transformer being tuned for preference.

Fig. 6 shows how a valve detector may be used with the two-valve circuit, the resistance coupling can be used between the valves instead of the coupling is shown, but is not so efficient, although much simpler.

The valve detector set will probably be very difficult to stabilise, and less time will be wasted in setting the crystal than fetching the extra current for a valve detector. Besides, the signals are loud enough with the crystal.

See follow-up article Wireless World May 27, 1922.

To thank the Author because you find the post helpful or well done.