US1668653A - Amplifier - Google Patents

Description

May s, 192s. 1,668,653

J. M. MILLER AMPLIFIER Original Filed March 15, 1921 atroz neig Patented Mey 8, 1928.

UNITED STATES PATENT OFFICE.

AMPLIFIER.

Original application led March 15, 1,921, .Serial No. 452,542. and in Great Britain March 14, 1922.

Divided and this application filed August 21, 1928.

This application is a division of my coplpnding application, Serial No. 452,542, filed arch 15, 1921, for amplifiers My invention relates to an amplifier, particularly one using threev electrode vacuum tubes serving to amplify alternating currents or voltages having frequencies such as those used in radio signaling; or what is termed a. radio-frequency amplifier. In particular it relates to the means for obtaining and passing on an amplified voltage from one stage to another in such an amplifier, and one of the novel features thereof resides in the provision of an amplifier circuit of a character which gives a high amplification throughout a desired or wide range of fre quencies such as are used in radio signaling.

Other features will appear in the following description, reference being had to the accompanying drawings in which:

The figure is a diagrammatic view of one form of my amplifying system, which is merely for the purpose of illustration and is not to be considered in any limiting sense.

Reference characters a and b indicate respectively the usual receiving antenna and the secondary receiving circuit. The manner in which the energy is collected from the radio signaling waves constitutes no part of the present invention and it will be understood, of course that any type of collector, such as the loop collector, or the radio compass loop, may be substituted therefor.

In the well known manner, a radio signal produces voltage variations between the grid g and the filament 7' of the first amplifier tube (l. These voltage variations produce variations in the current in the circuit between the filament f and the plate 79.

These current variations produce amplified voltage variations across the induetance .coil e. which coll. as will be described heremafter; has a high inductancc. and low capacity. Such a coil offers a high impedance to these current variations regardless of their The high resistance indicated by reference Serial No. 658,560.

character fr', maintains the grid at approximately the potential of the negative terminal of the filament, but does not reduce appreciably the voltage Variations applied to the vgrid and filament.

The action of the second tube is similar to the first, and stili, higher amplified voltage variations are obtained across the coil e', which may be similar to coil e. If desired, further similar stages of amplification may be used, but in the particular circuit I have shown, the third tube (Z2 is represented in a well known manner as a detector tube, in which case the condenser c2 should be of suitable capacity, and resistance r of proper resistance and connected to the positive terminal of the filament.

The plate circuit of the detector tube may contain the customary telephone receivers t, as shown, or the input circuit of an audio frequency amplifier. In lieu of the detector tube, as shown, any other means of utilizing or detecting the amplified radio frequency oscillations may be employed.

The main feature of my invention resides in the employment of a coil having a suitably high inductance and low electrostatic capacity. Such a coil may be conveniently obtained by a concentrated winding of many turns of' fine wire in the so-called pan-cake winding, that is, a winding having a crosssection of small breadth and large depth;

or by the use of a number of such windings in series with each other but spaced apart mechanically so as to vreduce the electrostatic capacity. It is to be understood that my invention is not to be limited to any particular form of winding. asl the essential point is that the winding be of high inductance and low capacity.

A form of winding suitable for obtaining these desirable characteristics is that illustrated as coil e between the first and second stages of the amplifier, in which 1 and 2 are circular discs preferably of insulating material spaced by a suitable member to form a deep-grooved spool, the groove holding a Winding 3 of many turns of very fine enamcled wire, the two ends of the winding being brought out through holes in one of the discs to form terminals 4 and 5. The actual dimensions and size of wire used for a coil having the necessary characteristics to satisfy my invention are hereinafter recited. The small diameter of the turns of the winding and the most unusual fineness of the wire are largely responsible for the exceptionally lovv distributed capacity. wWhile the small-diameter of the turns operates against obtaining large inductance, this factor is many times overcome by the extremely large number of turns concentratedly arranged, the size of the wire permitting the large number of turns, their concentration, and further having a large inherent self-induction in proportion to the mass of the wire.

Such a Winding has a high electrical impedance over a wide range of frequencies, and this range, by suitable design, can be made to cover a portion or all of the range of frequencies customarily used in radio communication. And it is because this impedance is high relative to the plate circuit resist-ance of the vacuum tubethat the use of such a coil in an amplifier circuit leads to a high amplification throughout a desired or Wide range of wireless frequencies. This, of course, also assumes that the vacuum tube has an amplification constant considerably greater than unity, as is customarily the case. Y

As an example, I have constructed an amplifier in which the inductance coils Were Wound With No. 44 A. W. G. enamelled copper Wire and Were of the pancake type, having a cross-section of coil winding of onesixteenth of an inch in width and about five-eighths of an inch in depth, the inner diameter of the Winding being three-eighths inch. The coils had about five thousand turns of Wire and a. direct current-resistance of three thousand ohms. The inductance of these coils Was about half a henry, While the coil capacity was only two and a half micromicrofarads. The impedance of such a coil is primarily determined by its capacity at the higher frequencies, being high When the capacity is loW, and by its inductance at the lower frequencies, being high When the inductance is high. Thus, at a frequency cor` responding to a Wave length of 200 meters (1500 kilocycles), the above coil has acapacity reactance of 40.000 ohms, While at a frequency corresponding to a Wave length of 20,000 meters (15 klocycles), it has an inductive reactance of 45,000 ohms. termediate frequencies, the impedance Will be higher than these values. Such a coil will give a high order of amplification over this entire range Which covers the extremes of practical operation in radio communication.

To obtain the best results at the shorter Wave lengths, it is desirable to use vacuum tubes having a low capacity between the 'tube electrodes. in particular a low capacity between the grid and plate of the tubes`ince the tube capacities in effect introduce a capacity Which is in parallel With the capacity of the coil, and appreciatingthis refined At inof course appreciated that for all Wave lengths every precaution should be taken to keep down distributed capacity of leads and other connections whichwould bring a capacity in parallel with the coil. Using such tubes and the above described coils, I have built an amplifier which gave good amplification throughout the range of frequencies corresponding to Wave lengths from 200 meters to more than 20,000 meters. With the particular tubes used, the voltage amplification throughout the major part of this range was practically equal per stage to the amplification constant of the tube.

p An advantage of my system of amplification is due to the fact that the value of the inductance of the coil can be limited to such a value that the coil has a high impedance over the entire range of radio frequencies and Will therefore lead to a hi h amplification of alternating currents an voltages of these frequencies; but at the same time will have a relatively low impedance for frequencies of the order of a thousand per second, or less, and hence variations of current or voltage of thesel frequencies Will not be amplified. The current and voltage variations caused by variable contacts, unsteady batteries, variable tube action and the like which would tend to cause distributing noises in the telephone receiversare either amplified not at all or to a slight degree. For example, theabove described coil would have an impedance not much in excess of its direct current `resistance at a frequency of five hundred cycles per second, and with vacuum tubes of the usual characteristics, variations of currents or voltages of this orde of frequency would not be amplified at a Another advantage of my system of amplification is that practically the Whole of the plate battery voltage is applied to the amplifier tube. circuit of the vacuum tube, and on account of the loW ohmic resistance of this coil relative to the direct current resistance of the tube, practically the Whole of the plate battery voltage will be applied to the tube Without serious drop of voltage in the resistance of the coil.

While I have illustrated two stages of amplification, it is obvious that one stage or any desired number of stages may be employed. Further, the amplificationsecuredi by the means described may be utilized in various Ways other than that shown.

Having fully described my invention, what I claim as new and desired tol secure by Letters Patent is:

1. A coil of pancake form Wound with enamelled Wire of a fineness approximating No. 44:.American wire gauge and having a depth of Winding measured radially of the The inductance coil is in the plate4 order of ten times the breadth of winding measured in a direction normal to the radial line and parallel to the axis of the coil whereby the coil has the characteristics of a large inductance and a relatively small distributed capacity.

2. A pancake coil structure having the characteristics of the minimum inductance` of the order of one henry and an optimum value of distributed capacity of the order of two micro-microfarads comprising turns of enamelled wire of a neness approximating' No. 44 American wire gauge wound tightly on a form of insulating material and having a mean diameter of windin of the order oi one inch and breadth of win in of the order of one-sixteenth part of an inc.

3. As an article of manufacture, a coil of pancake form comprising turns of enamelled wire of a tineness approximating No. 44 American Wire gauge wound tightly on a form of insulating material and having an inner diameter of the orderof three-eights part of one inch and an outer diameter of the order of one and Ve-eights inches and a breadth of winding of the order of onesixteenth part of .an inch.

JOHN M. MILLER.

Images