VM Tubes: 4 - TWTs and BWOs
VM Tubes: 4 - TWTs and BWOs
Foreword
A very well done and synthetic article on traveling wave tubes is already available thanks to Professor Rudolph Dietmar. Unfortunately it is in German and I am not able to translate it. For this reason I have decided to prepare this incomplete and confused English overview of TWTs. I hope that somebody could align it to the original article and complete the missing information.
TWTs
TWT (Travelling Wave Tube) appeared in 1946, thanks to the work of Rudolf Kompfner at the Clarendon Laboratory, Oxford. Kompfner moved to the Bell Telephone Laboratories where the device was further improved by John R. Pierce, who also left a simplified theory. To amplify TWT uses the interaction between an electron beam and the signal propagating along a transmission line, usually a helix. Its internal schematic is given in the following figure.
.jpg)
Fig. 4.1 – The TWT in the figure has the electron gun on the left. The generated electron beam travels to the collector through the helix, terminated at both sides by short coupling antennas. The beam is then captured by the collector at the extreme right. The signal propagates in the helix in the same direction of the electrons at approximately the same velocity, as in the draft 4.1(b).
The figure above shows one of the very early models, with a beam of about 8 mA through the helix up to the collector. An external magnetic field parallel to the beam is applied to keep it focused throughout the length of the helix. Input and output probes are coupled to input and output waveguides. The signal propagates from the input coupler probe through the helix up to the output coupler as in any transmission line. Depending upon the geometry of the line, usually many wavelengths long, we can imagine several signal maxima and corresponding minima moving forward along the helix in a movement that recalls the advancing of a cork-screw. The functions corresponding to the bunching and the catching of a klystron here are more or less continuous over the length of the tube. At any bunching around slow-moving electrons, some energy is lost by the beam and is transferred to the propagating signal. Bunches progressively increase as the signal does, moving in the direction of the output coupler.
Fig. 4.2 – Model of the TWT proposed by Pierce. The helix is here represented as a lumped-constants line and the signal moves forward in the direction z; successive arrows, indicated as i, represent the convection current of the electron beam. The current (J) impressed from electrons to the propagating signal is given by the relation on the right.
Fig 4.3 – Early TWT prototypes at Bell Telephone. The lower tube is mounted into the waveguide input and output transitions, with the focusing coil between them.
Helix is the simplest transmission line capable of interact with an axial beam, but other structures can be used to slowly propagate the signal: disk-loaded coaxial lines, helical waveguides or waveguides with transverse slots. TWT operates with no resonating cavity. It is therefore by nature a wide band amplifier. Bandwidth in the order of some gigahertz and power in the order of some kilowatts are easily obtainable from these devices. Since the mid fifties Bell claimed 500 MHz bandwidth, resulting in the capacity of amplifying and transmitting signals modulated with 11,000 voice conversations or 12 TV programs and 2500 conversations. In the late 1954 RCA announced a helix-coupled TWT with 2 GHz bandwidth. TWT amplifiers found an elective application in microwave communication links, as satellite communication, where still today are used.
TWT structures were also investigated for UHF television tuners. Robert Adler from Zenith Radio proposed some miniature solutions that could operate between 100 and 1000 MHz. Their principle was anyway slightly different. Actually they looked and operated like bundled distributed amplifiers.
.jpg)
Fig. 4.4 – In the miniature TWT signal enters from the base, travels through the helix and is available on the top. Right, a cross-section view of electrodes. The tube operated in a magnetic field. Electronics, October 1951.
.jpg)
Fig. 4.5 – Pictures of two British Standard Telephone (STC) TWTs, with details of electron gun and of helix terminations. Top, the flying leads W4-2GC type and (b) the octal-based W7-2D.
Fig. 4.6 – Philips 55340 has contact rings for the connection of the helix to external waveguides and integral finned radiator on the collector.
Fig. 4.7 – This CPI VTC-6369 is a modern TWT intended for ground to satellite communication systems. It operates with 15 KV, 847 mA beam.
Backward Wave Oscillators (BWO)
Of course in a transmission line the RF field can propagate in either directions. The BWO uses a regressive wave, a wave moving in the opposite direction with respect to the electron beam, the operating principle being represented in the figure below.
Fig. 4.8 – The RF field is propagated in the transmission line from right to left, with a phase velocity slightly lower than the velocity of the electron beam, that moves from left to right. On the left energy is transferred from the line to the beam, on the right it returns back to the line, each transfer adding 90-degree phase lag.
Oscillations take place when total phase shift is an integral number of cycles and loop gain equals the unity. For a given geometry, frequency is related to the velocity of the beam that depends upon the beam voltage. BWOs can be tuned over a very wide frequency range, up to three or even four-to-one ratio, just adjusting the beam acceleration voltage. Here are two examples of BWOs.
Fig. 4.9 – 4.9(a) CV2393 continuously tunes from 7.0 to 11.5 GHz, generating 120 mW minimum output. 4.9(b) Watkins Johnson SE-215A can be electrically tuned from 2.0 to 4.0 GHz.
Reducing the beam current, and hence the gain, just below the oscillation start-point, the structure operates as amplifier. In this case the signal is applied to the right coupler, the one close to the collector, and the output is taken from the left coupler. Just like any other amplifier using regeneration, backward-wave amplifiers are characterized by high gain, up to 40 or 50 dB, and very-high Q.
BWOs are often referred to as ‘carcinotrons’. I do not know the origin of this use since the radix of the word recalls something like a crab, from the Latin ‘cancer’ that means crab. For this reason I guess that it was the name given by some manufacturer to a round and flat shaped variant of BWO. I found two patents for toroidal TWTs that could have originated a carcinotron. Probably, before it could enter in common use, the term was a trade mark of CSF for some line of toroidal backward-wave oscillators.
Fig 4.10 - Toroidal TWT, patent 2,620,458 issued to Percy L. Spencer and assigned to Raytheon Mfg. Co. (Electronics, July 1953, pages 251, 252).
Fig. 4.11 - Ultra-Short Wave Transmitting and Amplifying Tube, patent 2,633,505, issued to Alfred Lerbs, Paris France, and assigned to Compagnie Generale de Telegraphie Sans Fil. (Electronics, December 1953, pages 258, 259).
_________________
Fig 4.12 - 50 GHz BWO, Bell Telephone Laboratories. Electronics, October 1953, pages 135 to 137.
References
1) Traveling-Wave Tubes, J. R. Pierce, Van Nostrand – New York
2) Principle and Applications of Waveguide Transmission, G. C. Southworth, Van Nostrand
3) Radio Enginnering Handbook, K. Henney, McGraw-Hill
4) Electronics, April 1951
5) Electronics, October 1951
6) Electronics, July 1953
7) Electronics, October 1953
8) Electronics, December 1953
9) Electronics, November 1954
10) Electronics, November 1957
To thank the Author because you find the post helpful or well done.