The period 1918 onwards until the early 1920's [approximately 1922-3] saw many improvements and much innovation but much of it was in command/control of the W/T environment both for the operators and for the remote users of wireless facilities/circuits..  Spark and Arc transmitters were still widely fitted alongside an increasing number of valve sets and the chief way of communicating was still with Morse code, whether by damped waves [ Spark] , CW [Arc and Valve] or ICW [Valve]. In many ways, receivers had progressed faster than had transmitters, and by this time, the norm was a receiver with valves and heterodyning was common practice for the detector stages. When [and in the mid 20's] the heterodyning process was first applied to RN receivers to create a fixed frequency signal processing stage, it was called supersonic {because it was above AF range and thus RF still} and in the 1931 "Notes of W/T Sets" {BR222}, it was described thus:-

"The following names have been agreed upon by international convention for the frequency bands specified

Below 100 kc/s Low Frequencies L/F
100 - 1500 kc/s Medium Frequencies M/F
1500 - 6000 kc/s Intermediate Frequencies I/F
6000 - 30000 kc/s High Frequencies H/F
Above 30000 kc/s Very High Frequencies V.H/F

The terms "L/F", "I/F" and "H/F" which were previously also used in considering the theoretical working of a W/T set were found to cause confusion with the meanings allocated above.  For example, if Amplifier M11 is receiving Rugby Press on 16 kc/s this would now be defined as "L/F" and yet it is passed through what were known as "H/F" stages before reaching the note-magnifier.  Then again the so-called "I/F" stages in Amplifier M5 are actually tuned to a frequency which falls under the "L/F" category quoted above.

To avoid this confusion the following terms have been brought into use in this book and in the 1931 edition of the "Admiralty Handbook of W/T" when discussing the action of W/T sets:-

Radio Frequency R/F Frequency of the incoming or outgoing signal
Supersonic Frequency S/F Frequency after first detection in super-heterodynes and Quench Frequency
Audio Frequency A/F Frequency of  ICW interruptions or audible note

In general when referring to the class of W/T set or waves, where the term L/W and S/W were used while waves were still principally referred to by wavelength, the terms L/F and H/F may now be employed.  In this general sense L/F may be taken as embracing L/F and most of I/F, while H/F may be considered to include a small part of I/F as well as H/F.  V.H/F should be referred to separately.

Where two H/F Attachments are fitted to one set [as in the Type 47] the one dealing with the lower frequencies should be referred to as the H/F attachment and the other as the H.H/F attachment [Higher High Frequency]."

The following block diagram might have been used for instructional purposes in 1930
The receiver employing this technique became known as the Super-HET and by the year 1930 this was a well established norm.  Later on, the frequency spectrum was re-defined and the band "I/F" was no longer used.  For the RN, this released the term "I/F" for other uses and it supplanted the term "S/F".  The local oscillator {RF2} was mixed with the incoming signal {RF1} which produced sums and differences {RF2 + RF1/RF2 - RF1 plus the original two frequencies {RF1 and RF2}.  By using a simple LC filter [much later, crystal filters] the detector stage could be offered a fixed frequency to demodulate irrespective of the frequency of the incoming signal {RF1}. To put 'meat on the bones' of the above block diagram,  we will assume {a}  that the sum of RF2 + RF1 = RF3; {b} that of the difference RF2 - RF1 = RF4; {c} that RF2 is mechanically attached to the tuning control which dictates the frequency of RF1 and is fixed tuned to always be 500 kHz [say] above the frequency chosen for RF1.


Later on, and particularly for sets with a high front-end RF frequency, the double super-HET was introduced.  This involved heterodyning the off-air frequency twice resulting in having two [or more] IF stages before demodulation took place, and in cases like this, the modulation was always voice or voice frequency tones using A3 double sideband.

The new fangled communication of radio telephony  [R/T] {which had been known about for many a long year by this time} was still an emission for trial purposes only.  The earliest R/T transmitter in the RN was the Type 81 [1921] closely followed by the Type 83 [1923].  This is what the Admiralty Handbook of Wireless Telegraphy 1920 had to say about R/T communications * which the 1925 edition more or less echoed and wasn't at all sure about the reliability of voice/sounds/tones being transmitted through the ether as radio waves.

* Page 353 Chapter XIV - Remember, the theory is sound and well understood but the application is not yet mastered!

"458.  By "Radio-Telephony" [R/T] is meant direct communications by speech by wireless methods, whereas "wireless telegraphy" indicates the use of Morse code.
For Service purposes, R/T possesses special advantages over W/T in cases where high speed of communication and very rapid fulfilment or orders given is desired.  In cases where a message must be written down, R/T is not appreciably quicker than W/T when plain language is transmitted as W/T messages can be sent and received at a rate limited only by the speed at which the message can be written down.
The use of R/T is at present undeveloped, and a great number of problems remain to be solved, but there is no doubt that its range of utility for Service and commercial purposes will be very wide."

By the time the 1931 edition of the Admiralty Handbook of Wireless Telegraphy * had been published [which superseded the 1925 edition] R/T was being described as:-

* Page 675 Chapter XV

"Radio Telephone R/T consists in the transmission of speech, music or any audible sound to a distance by means of electro-magnetic waves.    It bears the same relation to wireless telegraphy as line telephony does to line telegraphy, and possesses the same relative advantages and disadvantages.  More particularly with reference to Service requirements these may be summarised as follows:-

[a]   All the normal advantages of oral over written communication e.g., actual saving in time, and the lessening in traffic that arises by the adjustment of minor points without the necessity of voluminous correspondence.
[b]   The ability to make and read Morse is not required.

[a]   Where a message has to be written down, R/T is slower than W/T
[b]   Coded messages are difficult to pass owing to risk of phonetic errors, and plain language involves the risk of interception.
[c]   The interference caused by R/T is worse than that caused by W/T.  Conversely, interference exerts greater influence on a receiver adjusted for R/T than on one adjusted for W/T, and, in addition, the effect of any interference heard is of more serious consequence.
[d]   Listening-through is not possible with R/T unless the "single side-band" system is employed.

However, the 1920's, used for many experiments, did see the end years of that decade with an efficient R/T system although once again, used mainly at the V/HF frequencies and not for MF or HF working:  these continued to be used for CW and ICW only. What follows  is the NSR [Naval Staff Requirement] for W/T General dated January 1938.  There are several items within of interest. We have talked much about SPARK transmitters [see the PRE WW1 files] and its broadband emission causing considerable interference to nearby shipping. In paragraph 9 of this NSR we read that  as from 1936 - SDO 285/36 -  a Spark transmitter is fitted PURELY FOR DELIBERATE JAMMING OR ANTI-JAMMING USE.  Paragraph 12 of the NSR confirms that even as late as 1938, only a FEW SHIPS had a requirement for R/T and these did not include Capital ships other than aircraft carriers.  Note in Section A [Capital Ships] that the short distance set has moved away from the lower end of the HF band to upper end of the HF band/lower end of the V.HF band i.e., to 30 to 60MHz. Also in Section A, the Fire Control set frequency has moved up to V.HF which was also supplemented by the Spotting Top Aldis Lamp keyed remotely from the TS of the gun turret concerned.

Click to enlarge

Click here to read the 1938 NSR.

In the first paragraph above I mentioned the great improvements in the wireless telegraphy environment.  This involved power supplies, wiring/cabling in general, aerial and send/receive change over systems, office layouts and the remote user who in the early days, particularly of R/T, had to come to the wireless office to use the 'system microphone'.  This period said good bye to the long standing Silent Cabinet in which sat the receiving operator who also sent the Morse code outgoing messages.   All major ships were either built with or modified to have, a new suite of three offices for TRANSMITTING called [a] the Main Office which was also called No1 TR [transmitter room], [b] the Second Office or No2 TR and [c] the Auxiliary Office.  For RECEIVING they were called [a] the CRR {central receiving room which took over from the Silent Cabinet], [b] Second Office and [c] the Auxiliary Office.  In smaller ships [Flotilla Leaders and Destroyers] they had three offices also but their names were slightly different to those of larger ships. 

Equipment generally came as packaged items instead of the piece-meal assemblies [many of them self-build] of yore.  Mind you, this often resulted in  complicated apparatus [the Type 35S is a good example] and certainly the resultant equipment was on the large side.

Before we leave this period let us briefly look back to summaries the Royal Naval W/T transmitters from the very beginning.

We have now seen the four unique methods of transmitting W/T signals from the earliest of times right through the period leading up to the 1930's and because one of the methods discussed is a valve transmitter, right up to the end of the thermionic age!   In the 19th Century files and in the PRE WW1 files we have talked a great deal about the SPARK, the ARC and the HF Alternator methods and mention has been made here on this page and also in the earlier timed pages of valves. Now we can summaries these methods by highlighting their advantages and disadvantages vis--vis one another.


[a] The SPARK system = Damped Waves
[b] The ARC system = Undamped or Continuous Waves
[c] The Valve system = Damped or Undamped Waves at will
[d] The HF Alternator system = As for [b] above, but this method is not relevant to the Royal Navy


Method Advantages Disadvantages
  • Robust and durable
  • Faults easily cleared
  • Emits a wave which forces its way well through interference
  • Wasteful of power
  • Short range as compared with continuous wave generators
  • Interferes badly
  • Requires high insulation on account of initial peak voltages
  • Robust and durable
  • Faults easily cleared
  • Can be easily constructed to handle large powers
  • Slow in starting up
  • Present certain keying difficulties
  • High powered sets radiate harmonics badly
  • Unsuitable for use in Fleet as it is not possible to "listen through" for Admiral's signals, messages of distress etc.
  • High frequencies cannot be produced, the limiting value being 250 kc/s.
  • Radiates a very pure wave
  • Easy to key
  • Very suitable for R/T
  • Transmits damped or undamped waves at will
  • Quick in starting up
  • Valves are fragile and require frequent replacements
  • If faults develop, they are not so easy to trace as in other sets.
  • Radiates a wave which is very pure and free from harmonics
  • Easy to key
  • Very suitable for R/T
  • Suitable for high power working


  • Requires very expert supervision and maintenance
  • Its frequency cannot be varied so readily as in other systems
  • Its first initial cost is high
  • Only suitable for low frequencies.

And finally, the ionosphere or rather the road to understanding it !

The understanding of the propagation of radio waves and thus the ionosphere, was very slow on the up-take!

Looking for the progress made, four books of authority were consulted covering the period 1912 to 1930, and it was only in the last of these, that they got anywhere near to understanding the behaviour of electromagetic waves as we understand them today.

The 1912 and 1917 versions of the W/T Manual had virtually the same printed text on the subject.  This second paragraph of this thumbnail tells of the level of understanding just as WW1 was finishing.

Click to enlarge

In the 1920 version they had made some assumptions and some of the answers were appearing which would eventually decipher the unknown.  Here is that version as a pdf file  1920 em waves.pdf.