The Story of RATT - Part Two

by Godfrey Dykes


  When RATT 1 was first introduced at sea, Broadcast traffic to ships was duplicated and transmitted on RATT Broadcast and on Morse Code Broadcast too just in case the RATT system failed or that the operators could not cope. It didn't last long. The system proved itself worthy of a naval communications channel and the telegraphists were more than up to it, they considering the burden had been lifted from their shoulders. It was nice to get up and walk around the W/T Office, to have a cup of tea and (in those days anyway) a good puff on a cigarette. It wasn't known at the time, at least not by telegraphists, that the introduction of RATT would eventually lessen the need for operators to use Morse Code. As we now know, in the beginning it did lessen the need and went on to side-line the skill altogether, but few of us would have guessed that with the demise of Morse, the Royal Navy Radio Operator would perish altogether too.

  Throughout the 1950's and running parallel with other innovation (ICS for example - see the transmitter page) the Admiralty scientists and engineers had been beavering away at producing Fleet systems which were modern, but which above all else were common to each and every ship. By the mid to late 1950's, the Fleet was awash with piecemeal systems, adhoc arrangements, and all surrounded by a multiplicity of equipments. This 'throw it in and see if it works' approach was bad for every reason, covering the spectrum of operators employment/training through to BR's and stores, and of course affected users and maintainers alike. The example we have just seen, indicates that virtually each and every transmitter had its own RATT transmission equipment, whereas, on the RATT reception side, uniformity was achieved with the CV89A and its receivers the B40D and the B41 which nearly every ship was fitted with.

  Whilst the introduction of RATT was welcomed with open arms, in many ways it had shortcomings, and I don't mean technically, though it had some of those too. At introduction and for many years after, the Fleet speed was 45 - 45.5 Bauds, very much faster than had been Fleet CW Broadcast overall. However, all we had really achieved was to turn a Morse oriented Navy into a RATT Navy with little or no regard for the traffic load which was bound to increase; and did so! We were still an off-line Navy, and very soon that now under-employed telegraphist was 'knee deep' in coding and decoding signals on his off-line cryptographic machine, the newly emerging Adonis System, the KL7.

  All of us in those days were ready for a "tidy up" and just around the corner, a brand new system and concept was nearing its launch date. When it did come, it changed the very way we communicated and it gave us a management tool the envy of all navies. That management tool has grown, keeping pace with technology and, in one form or another, is still with us today helping the navy to be ultra efficient in its global communication responsibilities: how lucky sailors are today with their superb equipments ! Having said that, my friends and I can't help smiling when our modern sailors go away for a couple of months, come back bedecked with medals, £100,000 in the bank ex-pier-head- jump to a luxury Iranian Hotel for a couple of weeks, make phone calls home from the ship and send emails home. We were away for 18 months at a time (the two to two and a half year absences stopped after the Korean War in the very early 1950's), didn't get any letters unless we met an RFA and then weeks out of date, were in hostile waters and got our few medals 30 odd years later (! Suez Canal, Artic, PJM !), and as for emails and telephone calls*, you had more chance of having sex with a mermaid: oh, I forgot, we didn't have bar-b-q's in a warship either and the thought of getting drunk at a bar-b-q at sea and then having sex with a female crew member (CPO charged with first ever rape since women went to sea in 1990 and subsequently jailed for five years) - well even GOOD OLD FRED QUIMBY wasn't that clever!

  Radio telephone calls using SSB into the UK GPO Telephone System via Burnham W/T (near Bristol) were allowed when not in operational or exercise status, providing that the equipment and the W/T staff were available to set them up and thereafter engineer them on behalf of individual crew members, which meant that ultimately the Captain had to agree. It was a messy affair, with the W/T department raising official paper-work which was then sent to the pay office where the cost of the call (and quite costly too) was deducted from the mans next pay day. Additionally, though certainly a one-off, for ships on Beira Patrol (1960's/1970's) spending long periods at sea off the East African coast, each crew member was allowed 25 words per week home to his NOK free of charge to say that all was well wrapped up with much love etc. This imposed a sizeable EXTRA work load on the W/T staff, because each of these 25-word messages (200 or so in a frigate) had to be processed, then sent to Mauritius on HF RATT MRL (Maritime Rear Link). From Mauritius they were transmitted over a FX (Fixed Service) to Whitehall Wireless in London who reprocessed them, put them into OHMS (On Her Majesty's Service) envelopes and put them into the UK mailing system. Mind you, as the Radio Supervisor of HMS Rothesay in 1970, I was "called round" many times (for favours given) at least until it all stopped? "Called round" refers to being asked to visit a mans mess at rum time (tot time) when he would offer me a goodly drink from his tot glass as a thank you, and......In 1970, the tot stopped and the Admiralty withdrew the ancient privilege. Our incoming mail whilst on Beira Patrol was always a 'cool' event though not as frequent as one would have liked. It involved all our mail being diverted to Mombassa, and from there, a RAF Shackleton ASW (Anti Submarine Warfare) aircraft would over-fly our ship and drop the bags containing 'goodies' into the sea. This would be retrieved using our sea-boat, and the excitement of the 'bombing event' and letters from home were moments for morale lifting.

   In the RATT 1 section, we said that RATT Ashore in RN W/T Stations and Commcens had long been established by the mid 1950's, and obviously scientists and engineers would look at the terra firma systems when designing the sea going systems. They would also be picking the best ideas from the many American sets in our piecemeal equipment RATT jigsaw. This file, which shows the FSY and FSR systems, gives a very quick look at equipment fitted ashore to receive RATT signals - it has two pictures and a small amount of text Typical WT Shore Station FST Equipment.

   The first stage of the improvement was not as dramatic as the second stage, but it did improve RATT Broadcast reception with the CV89A many fold. Hitherto, and from the very beginning of RATT1, the only receivers in the RATT system were the B40D and the B41 used respectively for Wide RATT and Narrow RATT. Neither receiver could be left unattended for any length of time, and every now and again, the operator would have to adjust the Oscillator Trim control (on the B40D only) to make sure that the teleprinter page copy was free of 'hits' caused by inaccurate tuning. The B41, with its much narrower shift (85Hz) and very narrow bandwidth setting (200 Hz) was easier to tune and keep on frequency and LF frequencies were less prone to atmospheric interference than were HF frequencies. When COMIST (see Transmitter page) was introduced as the first stab at improving W/T communication, the main transmitter was the Type 640 and the main receiver became the CJK. The CJK was an accurate and synthesised unattended SSB receiver, and it would take over from the B40D as the HF Broadcast receiver (and indeed offer a new service of SSB reception for all other non-Broadcast requirements) which ships had never had before. The frequency range of the B41 and B40D together was from 15 kHz to 30Mhz with an overlap from 650 to 700 kHz which both receivers could receive. The CJK covered from 1 to 30 Mhz and therefore did not supplant the B41 as a Broadcast receiver for LF (RATT Narrow). The receiver CJK, unlike any other receiver at that time (or since), was a 'system' which had a receiver as part of the system ! Five individual components all fitted into a cabinet of goodly size with a glass-fronted hinged door, designed to be kept shut when access was not required in order to maintain the cooling air flow and the fire alarm sensor, made up the system. This included a basic non synthesised Racal R117 receiver; a separate synthesiser internally wired to the receiver; a power supply unit; a separate 100 kHz frequency standard which could be used, by external plugging, to run other devices requiring such a standard which were external to the CJK and an SSB Converter. These units, when interconnected, made up a very agile SSB synthesised receiver, and with the 640 gave the COMIST ship a full SSB capability, and with the GAA Keyer, an FST transmit facility. This enhancement to Broadcast reception from unattended receivers (HF anyway) and the ability to also receive SSB circuits put the final touches to "RATT 2A". Everything was still off-line and all the keyers were still in use - the GAA, GK185A, AN/SGC 1-A and 5AB/A, servicing the 640, the 691 and the 601 series respectively, with the odd 89Q still using the GK185A.


   The introduction of RWA into surface ships, which more or less coincided with the introduction of ICS (see Transmitter page) had a profound affect on communicating to and from sea and upon the internal organisation of the W/T office. Just as in RATT1, 2 and 2A, there was a RATT Broadcast area and a RATT Tactical Communication area, but now with a control system which would allow total flexibility of the RATT assets which were based around a TONE system, and, upon ON-LINE working. From hereonin, the KL7 would be required for special handling signals only, and when it left us, the Literaliser (Outfit TLA), an integral part of RWA, took its place. With the new RATT control system came a new ships control system (the KMM for ICS and the KMP for ICS Mixed fits) and this did away with the old keying control lines of the KH system, moving just about everything to the MICROPHONE LINES, with of course some way of pressing the pressel switch to key the associated transmitter - a VOICE environment in short. Each separate part of RWA had its own on-line equipment, and for the Broadcast side this was a receive only 'Raleigh' System called a BID580, and for the Tactical side, a transmit and receive (synchronous or asynchronous) 'Orestes' System called BID660. RWA, just like Ratt 1 etc, was a DC based system which worked the teleprinter, and DC is a well known radiator. After going to all the trouble of fitted on-line crypto equipment, it wouldn't do to have 'sensitive' signals coming to or leaving the teleprinter, being radiated for all to print using a compatible DC signal receiver and a teleprinter. This was resolved in two ways. Firstly that the DC was converted as soon as possible to audio frequency (AF) and this was achieved by sitting the teleprinter on top of a box (the converter) with very short leads back to the teleprinter. Secondly, the new radiation field (from the short leads) had to be measured out to a distance where experts judged that the radiations were now innocuous, and at that point in any direction, an imaginary line was drawn inside which the area was declared as a no go area for any DC device: so things like telephones, which are totally DC except for the ringer which is AC (a charging and discharging capacitor), were not allowed neither were portable machines like the KL7 etc. Once established, this restricted area was occasionally re-measured and this was part of the overall TEMPEST measurement carried out routinely to maintain a known electronic security footprint. Such a restriction wasn't possible outside the ship, and it was possible for a van, for example, ostensibly delivery provision to the ship, to park on the jetty close to the MCO area of a ship, where, with a teleprinter and the 'right' kit, they could print off what was being printed on a teleprinter in the MCO. Circumspection was the order of the day especially when the ship was in a foreign port.

Before moving on to explain RWA in detail, I will mention that at this time there were two other RATT systems, namely:-


  Outfit RWB was exactly the same as RWA but without the on-line equipment, and this was fitted into some New Commonwealth navy ships. Outfit RWC was exactly the same as RWA but was fitted into submarines. This is a submarine RWC distribution panel and you can clearly see, as you will further down the page into the surface fleet distribution panels, the big sockets (carrying transmit and receive pins) and small sockets carrying receive only pins.

  From what we have said in RATT 1 etc, you will recognise some of the detail in the following description of the two BASIC circuits of RATT Broadcast and RATT Tactical Transmit/Receive in RWA.

RWA Tactical

  In this simple drawing, (below) which is drawn with double arrows indicating that it is suitable for outgoing (transmitting) and incoming (receiving), you will see the short lead I referred to above (shown as Figure 1 over to the left) fitted between the teleprinter 'A' and the Converter 'B'. These blocks are drawn so that you have some idea of how many units are involved in getting a signal from the teleprinter keyboard or autohead to the transmitter and vice versa back to the teleprinter to receive a page copy.

  The names given to these blocks are not the proper names, although, they do indicate the proper function. Now I am going to give them their correct names and after that we will have a look of what passes through each of these boxes. One of the best things about RWA is that what you see in front of you is good for all occasions for transmit/receive communications whether it be by HF, or UHF, FST, or A2RATT and, even Morse Code.

  The system as drawn is a little rigid and inflexible. Later we will add distribution panels which allows us to use any teleprinter with any BID660 with any TTVF(T) etc etc.

  Note the dash after each name in the table followed by the abbreviation for these names.

In BOX A (TP and TA)

There is a teleprinter which in this case (RWA), uses double current for receiving 80-0-80V DC and single current 6V 0V DC for transmitting. The square waves produced are MARKS and SPACES.

In BOX B (TT10)

These marks and spaces are converted as follows. For a MARK, Zero Tone and for a SPACE, 1kHz. Remember this ZERO TONE because it is important. There is a switch on the TT10 which says "Morse or RATT". In the Morse position, this 1kHz can be used to modulate a transmitter in the SSB suppressed carrier mode with 1 kHz offset, to generate an A1 signal (CW). This technique is known as A2J. See the thumbnail for a little detail of the TT10


The SPACE which has been converted into 1kHz is coded using a noisy diode whose random pattern would be repeated about every 200 years. It is then subjected to other cryptographic processes.


This 1kHz coded SPACE is used to send one of four oscillators to BOX E depending upon the switches and the plug-in pattern of the oscillators. There are two shifts available from the TTVF(T), 850Hz and 200Hz, and for 850Hz, oscillators 2125Hz and 1275Hz are used. For 200Hz, oscillators of 700Hz and 500Hz are used. We have already mentioned ARRANGEMENTS which are only applicable to FST, but for A2RATT we talk about the MODULATION ARRANGEMENT, where the highest tone is the highest frequency in the USB but the lowest in the LSB. Arrangement 1 is where the highest tone/frequency is the SPACE (Active) and Arrangement 2 is where the MARK is the highest. If we now chose an 850Hz shift Arrangement 1, you will see that the oscillator 2125Hz occupies the SPACE (Active) position and the oscillator 1275Hz the MARK (Inactive). So, when the 1kHz coded SPACE comes in it opens the gate and allows out 2125Hz. Remember our ZERO TONE? When BOX D recognises that there is no incoming tone from BOX B (and therefore BOX C) it releases the MARK (inactive) oscillator of 1275Hz. Had the shift been 200Hz, then the RED type would have been good for 700Hz and the YELLOW type for 500Hz.


As soon as a Tone is recognised, the SUR, in effect, presses the pressel switch to key the transmitter and to open up the microphone lines down which these VOICE FREQUENCIES will travel to the transmitter.

In Boxes F and G

The marrying of the plug and socket on the CCX upper and lower transfers the controlling voltages of the SUR to the transmitter.

  To maintain this arrangement, we must use the USB of the SSB transmitter set to Suppressed Carrier and offset down. This is the first time we have met anything other than 2975Hz and 2125Hz tones used throughout in RATT 1 and 2 at 850Hz shift. Obviously, it doesn't matter which tones are used as long as they are of the correct distance apart (the shift) and in the correct arrangement. All we need to do IN EVERY CASE OF OFFSETTING FOR AN FST SIGNAL, is to use the mean of the two tones given (by manufacturers design as much as by anything else) and use that as the offset. In this case our SPACE is 2125Hz and our MARK is 1275Hz, the mean being 1700Hz = our offset DOWN. If the assigned frequency given was, say 8933kHz and our description says that we are going to generate an FST Signal (F1) @ 850 Hz shift, arrangement 1, then are parameters are "F1/850/1/8933. The picture in the ether would look like this.

  If, in the picture above, Box H had been a receiver, the reverse of this picture would hold good in the case of the carrier. If the transmitter suppresses its carrier, then for the same TONES in the same Arrangement, the receiver must REINSERT the carrier in exactly the same place to successfully demodulate to feed the same tones back down the line to BOX A and the teleprinter. So the receiver has to be used in the USB and the Frequency Assigned has to be offset for WHATEVER TONES the receiving end wants to choose. However, in this case we are using a TTVF(T) for an 850Hz shift, so the only tones he can use are 2125Hz and 1275Hz. His dial set also will be 8931.3kHz. The SUR BOX E, for receiving, is a 'straight through' device, so it passes the two tones to the TTVF(T) BOX D. When it receives the SPACE it releases the 1kHz tone and when it receives a MARK it releases NOTHING. The 1kHz goes into the BID660 to be decoded and is passed as a 1kHz tone to the TT10 BOX B, which converts the 1kHz to +80V DC, and when ZERO TONE, -80V and printing takes place. We have mentioned the four oscillators in the TTVF(T) on the transmit side, so a quick word about the FILTER side of the equipment. The four oscillators were followed by four bandpass filters, one for 2125Hz, one for 1275Hz and one each for 500Hz and 700Hz. However, across on the receiver side there were four more filters, again for 1275Hz, 2175Hz, 700Hz and 500Hz. Occasionally these got mixed up and caused untold mayhem amongst some operators, although why I am not sure. There was a wide bandwidth set of filters and a narrow bandwidth set, the idea being that for Intra RN working where known stable transmitters and receivers were used, the narrow band were the norm for the receiver side and because the oscillators were 'state of the art' and highly stable the wide set would work satisfactorily in the transmit side. When working with the Air Force, the Army or with other navies it was sometimes prudent to swap over to the wide filters in the receiving side just in case of wandering transmitters.

  In the early days of RWA, when the 601 series was still 'treading the boards', to make the GK185A and the 5AB/A FSK Keyers compatible with RWA, two extra devices were brought to bear. One was the GK198 for the GK185A and the other was the GK199 for the 5AB/A. The GAA keyer for the 640 went out of service almost immediately after RWA was rolled out. Within a few years all the FSK Keyers had gone and RWA was 'king'!

Now lets look at some of the equipment for real on the Tactical Side.


All these unit were connected together using various distribution panels.

  Inside this file you will be able to use the zooming tools to get a good look at each block RWA Blocks The file also shows the Broadcast side but we have yet to cover that.

  These are the distribution panels which were used for the flexibility of the system. The whole system was duplex and on a typical Tactical Bay like the one indicated in the pictures above with BOX A to BOX H in them, one could print a circuit, say on 6330 kHz and simultaneously send traffic on the autohead sitting on that bay on a transmitter tuned to say, 8669 kHz but without a teleprinter page copy/record of transmission. Normally, when using the keyboard the page copy was mechanical and when using the autohead, the page copy came from the sidetone of the BID660, or, if that was not in circuit, from the sidetone of the transmitter. Sidetones and receiver outputs were on the same line R+ R2 pair but the receiver always overrode the transmitter. R+ R2 with Microphone line M+ M2 always switched out the USB whilst R3 R4/M3 M4 always switched out the LSB.

  This first distribution panel connected Tactical Teleprinter Bays to multi-pinned sockets, and here the possibility of eight teleprinters are shown but only two bays, 3 and 4 are shown marked. A short flexible connector with a male-plug each end of different sizes, would connect a teleprinter to either a TTVF(T) which was not associated with a BID660, or to a DP(T). In our example, we will plug from Bay 3 to DP(T)1 - big socket above the marking. This takes the 1kHz SPACE tone from the TT10 (sitting directly beneath the teleprinter on bay 3) to a second distribution panel shown below.

This diagram below, is, let's say, DP(T)1 (see name plate in centre), and on this panel all the plugs and sockets are the same size namely big.

  Our 1kHz SPACE is now sitting and waiting on the top row second from left socket. Two typical cables are already shown connected, but we will take a third lead and plug it between DPTP(1) and SB4 No1 IN. SB4 is another name for the BID660. This action puts the plain language 1kHz SPACE into the BID and across to the right, also top row, the coded 1kHz SPACE is waiting. A fourth lead is plugged from SB4 No1 OUT to the socket on its left, namely into TTVF(T) No 1. This action sends the coded 1kHz SPACE to the TTVF(T) where it triggers the SPACE/ACTIVE tone and sends it to its tied SUR. Of course, when the teleprinter is not active (it is IDLE) the other tone, the MARK, leaves the TTVF(T) for the SUR. The SUR culminates as a plug on the CCX lower and it is pulled up, out of its housing, and married with a socket on the CCX Upper going to the USB of that transmitter. To this same socket (there are four of them for each sideband) and via a different route called the ROX (Receiver Output Exchange), the output of a receiver can be connected to the transmitter to form a complete signal circuit of transmit and receive. If the transmitter and the receiver are on the same frequency (simplex), then receiver muting is switched on to protect the receiver when its associated transmitter is transmitting. If they are on different frequencies (duplex) then the transmitter is switched to key in and out a band stop filter (EZ) during transmission which is placed in the aerial line going to the receiver.

  The title is quite clear on this distribution panel below. Notice the flaps. Either the large sockets are used (transmit and receive) or the small sockets below (receive only), but not both together. This panel was rarely used there being enough spare capacity on other distribution panels. As its name suggests, it picks up remote user line which are not W/T Office Users. Here TO1 (Tactical Operator 1) could be a bridge or an operational room teleprinter, and Link 14 was a much used OPS ROOM data link along with Link X (or 10) and Link 11 but they were CAAIS and ADAWS high speed computer links and not RATT based like Link 14.

RWA Broadcast

  The RWA Broadcast side introduced greater frequency diversity into reception and out went the B41 for Narrow Band LF RATT reception and in came the CJD receiver. Out also went the CJK, replaced by the CJA, meaning that the B40D and the B41 were no longer required as main line receivers. However, whether it was a staff requirement problem or that somebody didn't really understand W/T equipment or what, but they was a glaring gap between the beginning of the new receiver CJA/CJC and the end of the new receiver CJD which no receiver could fill. Was it better to procure a receiver which could fill that gap, or to keep our old pals the B40D/B41 (which could, and did) and give them a face-lift ? The latter choice was decided upon, and the old receivers got a new play mate, the FAZ. From much personal experience it really was a waste of money, for while the B40D (in particular) got lots of use usually in the thick of it, centre stage in the MCO on the General Purpose Bay, rarely ever, apart from training, was the SSB facility used or needed. I suppose that the demise of the CV89A was the greatest 'culture shock', after all, what would the men on watch watch now that their little oscilloscope was no longer. The CV89A was relieved by the TTVF(B) Terminal Telegraph Voice Frequency (Broadcast), and once set, the damn thing did everything by itself. The CV89A had always had an input of 2.55kHz ± Shift, either from the B40D or from the CJK (mainly) but now the TTVF(B) could have any offset it wanted. Whilst this was true, always assuming that the tones selected were 850Hz apart, all one had to do to do ones duty, was to set it on Channels 11 and 16 and offset 2.55 for USB, or 'to ring the changes' on say, Channels 6 and 11 every other Sunday afternoon remembering to offset 1.7kHz. In those, and in other cases also, the page copy of the broadcast would have been perfect. The multiplicity of combinations and the relevant offsets and bandwidths necessary, became one of the reason for the Technical Instructors in the Signal School to get together for a 'pow-wow'.

  This is what it looked like and immediately following is the basic block diagram of the equipment. Note that there are two identical units with one common power supply in the middle. The whole thing is called a TTVF(B) which comprises of two TT20's. It was the simplest of things to operate and in most ships, spend the majority of its life SET on the same settings which were applied when it was first set to work.

  Looking at the diagram above you can see that the story, however told, always starts with the position selected on the 'A' Channel Oscillator and the 'Z'/FSK Channel Oscillator followed of course, as in any receiver (RF or AF) by filtering to keep out interference. The setting of these switches dictates what offset is applied (using Two Tone) to the SSB receiver. For WIDE shifts there are 16 possible frequency tone inputs to either the 'A' or 'Z' channels on each separate TT20, and these are marked 1 - 16 on both channel switches. Additionally, there are two other switch position 'X' or 'Y'. These are used for NARROW shifts (200 Hz) and VERY NARROW shifts (85 Hz for surface Fleet, and 50 Hz for submarine Fleet) and are valued as follows - on the 'A' Channel X=500 Hz Y=700 Hz : on the 'Z' Channel X=1000 Hz. Apart from these latter positions, channels are separated by 170 Hz, starting with position 1 on 425 Hz and 16 on 2975 Hz. Just in case you want to apply your own settings here is a full list for the values.

  The STANDARD shift at HF is 850Hz. If you divide 850 by 170 (the distance in Hertz between channels) the answer is 5. Thus, notwithstanding the selection made, the switches are ALWAYS 5 switch-position apart. If you select position 1 then you must also select position 6 : if position 6, then 11, and so on. If you take the first option, 1 and 6, the offset on the receiver will be (1=425 + 6=1275) plus 2 = 850 Hz, and for the second pair, 6 and 11, the offset is going to be 1700 Hz. Had you been able to visit all our warships of the 1960's/1970's period, you would have found that the VAST MAJORITY of ships would have chosen this second option of 6 and 11, not just at the time of your visit, but for all times when reading a 850Hz shift broadcast. So why all the choices ?

  Suppose you were tasked to read (copy) three broadcasts, each keyed with its own separate crypto stream and from the same Operational Area Commander, ONE when joining FORCE 'A', shifting to TWO, when joining FORCE 'B' and to THREE when crossing a given Latitude. The parameters are F1/200/1/8598.9, F1/850/1/8599.66 and F1/850/2/8600.85 respectively for the three broadcast. A 'trained eye' can see straight away that this is a multi-channel frequency broadcast, but others may miss that point and re-tune the receiver for each event. What the controlling transmitter did to effect this operational requirement was as follows:-

  If the transmitter can do it, so too can the receiver. In this case the receiver would also have a dial set of 8598.3 kHz, the USB will be used and the AF filter of 300 to 3300 Hz would allow through all the various tones required. This receiver, as previously shown, would be connected at the ROX (upper to lower) to an RWA Broadcast Selector Switch (BSU) sited above the TTVF(B). For Broadcast ONE, the BID580 would be loaded with the relevant keying material, the TT2O selector switches would be on A=Y Z=X, the Two Tone/FSK switch to TWO TONE, Normal/Reverse to Normal, Wide/Narrow to Narrow. The necessary connection through the RWA broadcast distribution panels would be made and after BID synchronisation, a page copy would be achieved. At the first shift to Broadcast TWO, all that would be required would be that the BID580 keying material would be changed, the TT2O switches would be put to A=9 Z=4 - nothing else. After a re-synch, the same teleprinter would print out the new broadcast. For the final shift to Broadcast THREE, change the BID580 keying material and put switch A to 11 and Z to 16 : re-synch and off you go. From this description, you can see how useful it would be for a ship carrying the Flag to have a TTVF(T) with lots of choices of tones to transmit instead of as originally configured, just two for 850Hz and two for 200Hz - predictably, it did change, and a brand new TTVF(T) was issued which looked exactly like the TTVF(B) which was also re-engineered. That is one reason for having the many choices. There are few, if any, others.

  The rest of the TTVF(B) drawing is of interest because it shows you the gated 1kHZ output to the BID and the VERY NARROW band route (85Hz and 50Hz shifts) where the shift leaves the receiver sat astride a 1kHz note (BFO output). A typical receiver might have an IF of say, 500 kHz ± signal, and to detect this, a frequency of either 499 kHz or 501 kHz would be used to beat with it to produce that 1 kHz tone. Thus, for a 50Hz submarine shift, the output frequencies are 0.975 kHz and 1.025 kHz (plus/minus 25 Hz centred on 1 kHZ). The job of the FSK Detector in the TT20 is simply to tell the bi-stable trigger which is the SPACE and which is the MARK ignoring the centre frequency of 1kHz, and if the receivers BFO switch had been set to '+' when it should have been set to '-' a simple throw of the Reverse/Normal switch will correct the ARRANGEMENT, and all will be well.

  One of the benefits of FST using two tones is that there is an inbuilt guard against 'selective fading'. Each tone (Mark or Space) when in the ether is transmitted on a different frequency, and although only 850Hz apart, it is possible that one frequency could fade whilst the other not. Providing the Terminal Telegraph Voice Frequency (T or B) receives ONE of the transmitted tones as a good strong signal, the system will continue to work because the missing tone will be replaced (at its appointed time) by NOISE, which will act as though it were purposely transmitted. Obviously, this would be a transient, and were this condition to worsen, a frequency change would be ordered.

Remembering how we introduced the tactical side of RWA, here is a picture of the basic building blocks for the Broadcast side.

This is a typical fit and each block shown is necessary for the functioning of the system.


There is a teleprinter which in this case (RWA), uses double current for receiving 80-0-80V DC. Notice the DC Link between the teleprinter and the TT11 sitting directly underneath it. This configuration limits the amount of DC radiation.

In BOX B (TT10)

TT11 Converter. Converts the 1kHz plain language tone to 80-0-80V DC



2 in number BID 580 on-line machines. Only one is shown connected but normally both would be used. It decodes the 1kHz on/off SPACE condition.


TT20 No 1 Forming one half of the TTVF(B). This is set on positions 11 and 16 to receive tones of 2975 and 2125 Hz. The TT20 passes a coded 1kHz tone to the BID 580 on every SPACE condition received.


TT20 No 2 Forming one half of the TTVF(B). As drawn, it is set on positions 6 and 11 to receive tones of 1275 and 2125 Hz. Had this switch been in the other position (i.e., downwards) for the LF receiver, then FSK would have been set on the System Switch, and the Z/FSK Selector switch would have been set on position 'X'. The TT20 passes a coded 1kHz tone to the BID 580 on every SPACE condition received.

In Boxes F

A bank of 6 two position switches allowing the maximum selection of twelve receivers. Each TT20 (see Boxes D and E) has a switch. Normally, only two switches are wired giving four combinations into two TT20's. In this case switch 1 (the upper switch) has only one receiver married to it whilst down below, switch 2 can select either HF or LF reception. Switch 1 (up and down) feeds the AF tones to TT20 No 1. Switch 2 (up and down) feeds the AF tones to TT20 No 2.

In Boxes G

ROX(L) Receives AF tones via plug and socket action.

In Boxes H

ROX (U) Passes AF tones via plug and socket action.

In Boxes I

HF Receiver 1 (CJA) - Demodulated output 2550±425Hz shift for example - receiver offset is 2.55kHz.

In Boxes J

HF Reciever 2 (CJA) - Demodulated output 1700Hz ±425Hz shift for example - receiver offset is 1.7 kHz.

In Boxes K

LF Receiver (CJD) - Demodulated output 1000 Hz ±85Hz shift - receiver is not offset but switched out as BFO plus.

The RWA Broadcast distribution panels look like this

This (above) is the BSU. The first two position switch (Line 1 and 2) feeds one TT20 and the Line 3 and 4 the second TT20.

  The encoded 1 kHz leaving TT20 No1 is fed, via a cable (as shown below) across to the right hand side socket marked SB No 1 'IN' meaning the input to the BID 580: on this panel all the sockets and thus plugs are small. The now decoded 1kHz SPACE is passed, from the left hand top socket via a plug-in cable to DPT(B) No 2 which arrives on this socket

bottom, second from left. Finally, a last connector is placed across the sockets DPTP(2) and, say, Broadcast Bay 1 which routes the 1kHz plain language SPACE to the TT11 for conversion to DC.

  I will use the next picture as a summary. It is of an aircraft carriers fit and shows the distribution panels, the BID660 and BID580's, the teleprinter outfits TGA's and TGB's, the autoheads (6S6). Whilst not a massive fit, it nevertheless takes some good management to make sure the full fit, Broadcast and Tactical, performs well throughout. Have a look at it and see if you can trace the beginning and end of each chain of events. Don't be thrown by seeing CV89A's - in the very earliest of fits for RWA there were some 'left overs' but take it as read that these would have almost certainly been 4 TT20's (2 TTFV(B)'s). For TT(B) read TT11, and for TT(T) read TT10

  After the end of our Museum period (approximately early 1980's) the AT System (Automatic Telegraphy) RWA, RWC etc came to an end, and in its place there were two AT Sub Systems - one for the equipment like RWA and one for the Message Handling. Although I don't know too much about the new systems, I'll wager a bet that most of the ideas came from our time, from the 'RW.' series and from switching/storage equipment like T.A.R.E., just as many of the RWA ideas had come from RN W/T Shore Station/Commcen fits.

Good bye and I trust that you now know a little bit about our systems of over forty years ago.