August 1944 Radio-Craft[Table of Contents]
People old and young enjoy waxing nostalgic about and learning some of the history of early electronics. Radio-Craft was published from 1929 through 1953. All copyrights are hereby acknowledged. See all articles from Radio-Craft.
Signal multiplexing was originally performed using a rotating mechanical device with commutator contacts. Such a contraption suffered from a number of limitations including contact wear, noise, speed, size and weight. Probably the most limiting were contact wear and switching speed. Bulk could be accommodated because back then everything was bulky. Standard vacuum tube switches were eventually used to build multiple (n) input / multiple (m)
National Union Type 6324 Radial Beam-Switching Commutator Tube (Lamps & Tubes website photo)
New Development is an Electronic Commutator
By I. Queen
All illustrations courtesy of Bell System Technical Journal
Fig. 1 - Drawing shows the focused beams. Rotation of these beams by a magnetic field is the tube's secret of operation.
Fig. 2 - How the focused beam looks (above and below). The electron paths were made luminescent by putting a small amount of gas into the tube.
An entirely new type of tube with unusual possibilities has been developed and is already in operation in New York City, in a multiplex signaling system. This radial beam electron tube is remarkably simple in construction, requires no focusing arrangement, is small, works on low voltages and has high efficiency. It is used as an electronic commutator device and has recently been described in the Bell System Technical Journal by A. M. Skellett.
The cathode of this tube is held vertically and is surrounded by a cylindrical anode structure, as shown in Fig. 1. If each anode is at the same positive voltage with respect to the cathode, the anode current will, of course, be equally distributed among them, and each anode will receive approximately only 3% of the total cathode emission.
If a magnetic field (such as shown by the Harrow) is applied, the electron beam will be directed in only two diametrically opposite directions (Fig. 1), parallel to the applied magnetic field. If the uniform magnetic field were made to revolve, so would the two beams, so that such a field could serve not only for focusing but to provide rotation. In this application the two beams will contain approximately 90% of the cathode current!
The action of the beam is well shown in Fig. 2. For these photos the cathode was actively coated in only two opposite spots. Note how the electron beam twists in following the applied magnetic field.
A convenient rotating magnetic field is furnished by the stator of a two-pole poly-phase A.C. motor. The tube is simply inserted into this stator in place of the usual armature. An unwound stator of this type is shown alongside an experimental tube in Fig. 3. The loss in a typical stator is under three watts, making the entire set-up highly efficient.
The tube in Fig. 3 is constructed with 30 anode elements. Each element is really a pentode tube, containing a control grid, screen, suppressor grid, along its path. Fig. 4 shows the construction. Note that only one beam is emitted, the other being suppressed.
Fig. 3 - The radial beam tube and its field magnet.
Suppression of one of the beams is accomplished in several ways. At any instant the anodes on one side of the tube may be maintained positive and on the other side negative, this polarity rotating with the magnetic field. It may also be done by means of the suppressors. Still another method is to use an odd number of anode elements. Then when the beam falls on an element on one side it will fall between two elements on the other.
The maximum cyclic speed is approximately 10,000 per second. Since no mechanical wear results and no inertia is present, this tube makes an ideal rotating commutator.
An early system of multiplex telegraphy used a mechanical rotating commutator which switched in each communication channel for a small portion of the total time. This system proved impractical because of mechanical difficulties. The present tube will, of course, eliminate these difficulties. Incidentally, this type of communication does not require the elaborate filters of the carrier system, now in wide use.
Two tubes of the radial beam type have been operated successfully in New York for experimental signaling. A neon tube in the anode circuit of each element was used as indicator. The only amplification provided was that of the tube itself! Both transmitting and receiving tubes were connected to the same source of 60 cycle A.C. so that automatic synchronism of corresponding anodes was obtained. Much will probably be heard of this tube in the near future.
Fig. 4 - Internal construction detail.
Posted August 21, 2014