RTTY RATT Radio Teletype
Home ] Up ]

 

 

 

 

 

 

The Start OF Amateur RTTY

We asked  Jim Haynes about the start, early societies  and publications...
 Here is what  he adds  tot the SMECC archive and shares  with us.- E.S.

 

 

 

 

Historical Notes on Radioteletype
Jim Haynes, W6JVE

1535 W. Cleveland
Fayetteville, AR
72701
jhhaynes@earthlink.net

 



R. A. Heising of AT&T contributed a paper to Journal of the Franklin Institute vol. 193, no. 1, January 1922, p. 97 titled ”Printing Telegraph by Radio.” This presents the results of AT&T experiments transmitting between New York City and Cliffwood, New Jersey, a distance of 25 miles. The experiments were done in 1919, using a wavelength of 450 meters (666 KHz). The operating speed was 180 wpm, using a four-channel time-division multiplex at 45 wpm per channel. He noted that the principal source of interference in NYC came from spark transmitters.


Federal Telegraph Co. conducted experiments on the West Coast USA circa 1920 using frequencies in the 35-70 KHz range. [RTTY Magazine, April, 1955, p. 7.] Their transmitter used a Poulsen arc, and therefore was frequency shifted rather than keyed on and off; however no use was made of the properties of FSK. It was simply too hard to key the arc on and off. The path was between Hillsboro, OR and the KFS site near San Francisco. The test results were not sufficiently good to recommend the system for revenue service.

John B. Brady received four patents on radioteletype

1,485,212 Filed Dec. 28, 1921 - Issued Feb. 26, 1924
1,523,377 Filed Aug. 14, 1923 - Issued Jan. 13, 1925
1,562,820 Filed Aug. 22, 1923 - Issued Nov. 24, 1925
1,563,958 Filed Dec. 28, 1921 - Issued Dec. 1, 1925

These patents are assigned to the Morkrum Co.; however Brady’s residence is given as Somerset, Maryland so he must not have been a regular  employee of Morkrum. All of these patents cover make-and-break keying of the transmitter. In the first of these he says that the system has been successfully operated between the U.S. Naval Aircraft Radio Laboratory at Anacostia, D.C. and the U. S. Naval Radio Research Laboratory, Bureau of Standards, Washington, D.C. Points stressed in the first patent are:

• Use of printing telegraphy, not requiring skilled operators
• System is ”substantially” secret in that signals cannot be copied by ear
• Code can be changed from time to time for increased secrecy.

The illustrations show a vacuum tube transmitter grid-block keyed by a relay connected to the keyboard contacts. The receiver has a regenerative detector followed by two stages of audio amplification. The final stage is tuned to a particular audio frequency, which is then applied to two paralleled triodes operating a relay in the printer circuit. An odd item is a series tuned circuit shunting the relay, with a switch to disconnect it. There is an illustration of a tape strip printer with a typewheel inside. He gives a hand-waving description of the Morkrum machine. He shows what look like single-magnet selector magnet coils, but doesn’t show what goes on between these coils and the rest of the machine; so it is not clear whether this is a machine with a single-magnet selector. The coils are identified as ”teletype coils”, showing that the word ”teletype” was in use at this early date.

Patent 1,523,377 shows a system in which teleprinter signals are transmitted by radio to any of a number of receiving printers. Circuits are shown for both make-and-break keyed transmitters and transmitters modulated with audio tones. A principal feature of this patent is use of signals at one frequency to transmit the intelligence and signals at a slightly different frequency to control the motors of the receiving printers. With the make and-break transmitter the start-stop keyboard key is arranged to change the transmitter frequency slightly. With the audio modulated transmitter the tones are generated by buzzers at two different frequencies, one for the message signals and the other for motor control. Hence the receiver may or may not need a heterodyne oscillator to produce an audio signal from what it receives. 

The receiver employs a sharply-tuned filter to detect the motor control frequency. The motor control detector drives a plunger-type solenoid relay which reverses each time it is energized, so that one signal turns the motor on and a second signal at the same frequency turns it off. The speed-governed motor uses electron tubes in its control circuit, the subject of another patent applied for by the same inventor. Presumably this is to reduce the RF noise generated by the motor governor. The message-signal-frequency channel of the receiver operates a relay which controls the teleprinter receiving magnet.


Points stressed are:
• One transmitter to any number of receiving stations
• Automatic (suggesting unattended) reception
• Motor control of receivers
• Selecting some groups of receivers to receive to the exclusion of others

The keyboard in the illustration looks something like a Model 14. The
printer is not shown.


Patent 1,562,820 improves upon 1,563,958 by using an AC-powered receiver instead of a battery model. Power for the Teletype selector magnet is obtained from the same AC supply, as is power for the printer motor. He notes that AC-powered printer motors generate a lot of RF noise and that DC motors are superior in that respect; but you can run the motor off AC if you want to.

He says the application is receiving news from a central radio station, stresses the simplicity and compactness of the arrangement for installation in business houses, banks, between warehouses and large stores. Suggests use of very high frequencies (10 meters) allowing directional antennas and confining the transmissions to a particular area. ”Teletype” is capitalized.

 Patent 1,563,958 shows radio receiver including a tone detector driving a relay which operates a Teletype printer. It goes into some detail about the selector mechanism. 

The Navy also did some experiments in air-to-ground radioteletype, documented in Scientific American, March 1923, p. 173. The illustration shows something like a Model 12 keyboard in the airplane and a Model 11 tape printer on the ground. The article states that the next experiment would be transmission from ground to air. The Bureau of Standards was also involved in this work, and perhaps Mr. Brady.


Lawrence Schmitt, an employee of Morkrum, received two patents


1,705,211 Filed Aug. 28, 1924 - Issued March 12, 1929
2,012,407 Filed June 13, 1932 - Issued Aug. 27, 1935


1,705,211:
This appears to be the original concept of frequency-shift-keyed radio-teletype operation. It shows teleprinter sending contacts, or a relay, connected to a radio frequency oscillator-transmitter to shift the frequency slightly. Another circuit shows an audio frequency shift keyed oscillator modulating a radio transmitter. The receiver has a pair of audio tuned circuits, one tuned to mark and one to space frequency, each feeding a triode detector that drives one winding of a polar relay. A differential milliammeter allows for accurate adjustment to eliminate bias. The inventor explains why his system is superior to make-and-break operation for teleprinter use in terms of not having the inertia inherent in a neutral relay. He also refers to radio transmitters used for Morse and having a ”compensating wave”. These were transmitters which were frequency-shift keyed because that was easier than make-andbreak; but the compensating wave (space) was not desired and was ignored. Indeed it had to be placed at a frequency far enough away from the intended carrier to enable it to be ignored. He proposes a shift of 250 Hz with spacing lower than marking frequency as an example, with a marking frequency of 230.50 KHz. The patent also mentions the possibility of a two-channel multiplex, although the exposition of this idea is not very clear. (Presumably it applies to amplitude modulating two audio FSK signals on a single radio frequency carrier, as there is no suggestion of the logic required to operate a transmitter shifting among four radio frequencies.)

2,012,407:
This is said to be an improvement on 1,705,211 above. More accurately it is an adaptation of that system for use with make-and-break keying. A shunt, consisting of a variable resistor in series with an inductor, is placed to produce a marking bias current through the relay equal to half the current when a spacing signal is being received. The space filter is tuned to the signal frequency, and the mark filter is tuned to a nearby frequency. The assertion is that static and noise will be received equally in both channels and will cancel out. This is aided by the inductor in the shunt circuit, which makes the shunt ineffective on transients so that they affect both windings of the polar relay equally.


A recent article by Mischa Schwartz of Columbia University [published in Antenna, the newsletter of the Mercurians, a sub-group of the Society for the History of Technology, October 2007] calls attention to a paper by Armstrong, ”Methods of Reducing the Effect of Atmospheric Disturbances”, Proceedings of the I.R.E., January 1928, p. 15 (with discussion on p. 27). In this paper Armstrong proposed something very similar to Schmitt’s patent - use of separate marking and spacing frequencies for telegraphy, and subtracting outputs from the two detectors. Like Schmitt, Armstrong makes the intuitive assumption that noise will affect both frequencies more or less equally and cancel out when the signals are subtracted. Note that in this application Armstrong was not yet using a limiter ahead of the mark and space filters, so that it is not true FM as he developed later. Schwartz then references a paper by Carson of AT&T, ”The Reduction of Atmospheric Disturbances”, Proceedings of the I.R.E., July 1928, p. 966. in which Carson argues that Armstrong’s scheme is not effective in canceling noise. Yet Armstrong had supplied ink recordings showing the superiority of two-tone operation. It seems that for a fair comparison the two-tone scheme should be compared with make-and-break transmission at twice the power, because of the 100% duty cycle of two-tone versus the approximately 50% duty cycle of makeand-break. Carson indeed suggested that an advantage might be had by transmitting the message twice so as to use the same power as the FSK proposal. I’m not clear whether he meant to transmit two copies on two different frequencies, or on the same frequency at two different times. He does note that the FSK transmission involves twice the receiver bandwidth of the single-channel transmission.


It seems to me that Carson’s critique of Armstrong’s proposal needs to be tempered by instrumental considerations. Copying Morse code signals by ear makes use of the operator’s training and the marvelously complex human auditory system to separate the signal from the noise. Copying with an ink recorder bypasses all that with a fairly simple electronic circuit, but still preserves some of the analog nature of the signal. Operator skills still come into play, in a reduced way. For reception by automatic printer the incoming signal must operate a relay, which eliminates all human ability from the process and substitutes simple electronics. FSK may be a greater help in this situation that is apparent from theoretical considerations of signal to noise ratio. The relay transitions in one direction when the marking channel signal exceeds the spacing channel signal, and in the opposite direction when the opposite signal conditions are true. With make-and-break keying the relay has to be biased somewhere between the noise level and the expected signal level.

It seems unlikely that Schmitt and Armstrong were aware of each other’s work. The main point of Schwartz’s article is that Armstrong went on to employ a limiter; and by that means the noise-reduction ability of FM was realized.


Gilbert Vernam of AT&T received patent 1,622,297 Radioprinting telegraph system Filed Dec. 29, 1924 - Issued March 29, 1927


The radio receiver is shown only as a block. An A.C. power supply is shown; and the printer is shown as having an A.C. motor. Like Brady, he suggests ”there may be a demand for a radio printer service in which the news matter would be broadcast from a central transmitting station to the newspapers subscribing for such service.” Make-and-break operation is used. The detector tube operates a polar relay which drives the printer magnets - a polar printer magnet is shown. An extra winding of the polar relay is connected to the secondary of a ”kick” transformer, the primary of which is in series with the operate winding of the polar relay. This is intended to insure quick operation of the polar relay.

Frederick G. Hallden received patent no. 1,864,303 assigned to Postal Telegraph for Radio Printing Telegraph System Filed April 27, 1928 - Issued June 21, 1932. 

The radio equipment is shown only in block form. The essence of the invention is to transmit each character three times, from a rotary distributor having three sets of segments. A character received the same all three times will be printed. If the character is not received the same all three times only the marking pulses which agree will appear in the printed character. ”It is well known that static can only cause marking pulses...” which shows that he is talking about make-and-break keying; for with FSK static would affect marks and spaces more or less equally. A lot of relays are used at both transmitter and receiver.


Austin Bailey and T. A. McCann of AT&T contributed a paper to Bell System Technical Journal vol. 10, October 1931, p. 601, ”Application of Printing Telegraph to Long-Wave Radio Circuits.” This was a paper presented at the 6th annual convention of Institute of Radio Engineers. Notes the need for better signal-to-noise ratio for printing telegraphy compared with aural Morse operation. It cites advantage of two-tone operation over single-tone. As with the Schmitt patent, two-tone operation is analogous to polar operation of a wire circuit in preventing bias resulting from varying signal strength. Discusses experiments conducted on a 60KHz carrier circuit between New York and London, using a power of 50KW. Directive antenna systems helped to combat noise. Further tests were conducted in 1930 between Rocky Point, Long Island, and Rochester, NY, a distance of 286 miles. Power was 700 watts.


I have a picture of the Morkrum plant site, 1400 Wrightwood Ave., Chicago, when part of it was a vacant lot. This shows some antennas being erected for radioteletype tests with Lake Geneva, Wisconsin. The word from Walt Zenner, by way of Bob Reek, is that there were radio experiments between the Morkrum plant and Sterling Morton’s home in the western suburbs of Chicago. Also that when Walt arrived at Teletype, Schmitt was the plant engineer, responsible for utilities and the like. Nothing came of his radio work. I presume that when Teletype was an independent company it was scrambling for business wherever it could be found. After the purchase by AT&T it was viewed as an equipment manufacturer; matters involving transmission were on AT&T and later Bell Labs’ turf.  

It was reported in Telegraph and Telephone Age, October 1, 1932 that RCA Communications Inc. was using radioteletype between San Francisco and Hawaii, with plans to extend the use of RTTY in the immediate future.


A paper by Moore of RCA [Accuracy and Speed on Short-Wave Teleprinter Services; John B. Moore; Proceedings of the National Electronics Conference, Chicago, 1953, pp 927-934] confirms the use of radioteletype on the San Francisco to Hawaii Circuit and notes that it was not possible to use radioteletype on the Atlantic circuits at that time. Morse operation was faster, up to 300 words per minute versus 60. Time- division multiplex was employed starting about 1935 to allow teleprinter operation to compete with Morse. An error-detecting code was developed and used from 1939 until 1947. Post World War II a system called ARQ (automatic request for retransmission) was developed, based on the error detecting code. With this system a receiving station detecting an error in incoming traffic signals the sending station to retransmit the text that was received in error. Systems based on this principle continue in use today: for example the SITOR system used in marine communications. With microcomputers it is extremely simple to implement such a system. 

Moore goes on to note the increased need for record communication by radio during World War II, met by frequency-division multiplexing over single sideband radiotelephone circuits, and the development of frequency shift keying. RCA developed a two-channel time-division multiplex that was more tolerant of signal distortion than a single-channel teleprinter. 

Press Wireless, Inc. was also active in exploiting frequency shift keying for teleprinter operation over radio. A paper [Frequency Shift Radiotelegraph and Teletype System; Robert M. Sprague; Electronics, Nov. 1944, pp. 126131] notes that company’s use of FM for facsimile transmission followed by the use of FSK for telegraph transmission. This article shows that a limiter discriminator scheme was in use. Hence performance in the presence of noise and interference is governed by FM principles rather than by those of limiterless two-tone systems. He notes the adoption of 850 Hz as the standard for frequency shift, saying that gives the best compromise between signal to-noise ratio and bandwidth. There is some discussion of the bandwidth  requirement of make-and-break keying, which is considerably greater than than the theoretical requirement because of squaring of the waveform in the transmitter. I suspect that such a wide shift as 850 Hz was also helpful in coping with frequency drift of transmitters and receivers in that era. This conjecture is supported by their use of A-C coupling between the discriminator and the loop keyer, with a threshold correcting circuit in between.


Probably the best known RTTY equipment of World War II is the AN/FGC-1 diversity FSK converter and  its companion AN/FRR-3 diversity receiver. These units were manufactured by Western Electric using typical telephone style construction, apparatus mounted on flat plates, each unit occupying a 7-foot rack cabinet. Hence the units are largely for fixed-station use and on fixed frequencies. The Navy had some RTTY demodulators operating at the receiver intermediate frequency rather than at audio frequency.


Use of RTTY in amateur radio seems to have begin in the late 1940s in the New York City area. The reason seems to have been a confluence of amateurs having knowledge of military radioteletype during World War II and the availability of teleprinters. The New York Police Department had a network of Model 12 machines, which they had to replace because  maintenance parts were no longer available. News wire services were disposing of their Model
12 machines for the same reason. Amateurs were able to acquire these antiquated machines. They were put into operation on 2 meters using AFSK. The main reason was that FSK was not authorized for amateurs on the HF bands. A secondary reason was that the Model 12 was a prodigious generator of RF noise that made HF reception extremely difficult. Amateurs  trying to operate Model 12 machines on HF resorted to measures such as vacuum tube keying of the code magnets and the governed motors. 

For some reason, in spite of the huge amounts of radio equipment sold as war surplus, there was hardly any Teletype equipment. Perhaps the U.S. military needed to retain all its teleprinters as it phased out Morse operation; perhaps foreign governments outbid the surplus dealers; perhaps the surplus dealers just didn’t imagine any market for Teletype equipment. The New York hams did have access to a trickle of Western Union surplus, little of which was readily usable. 

The New York area operators tended to run high power on VHF. Perhaps there was some scatter propagation augmenting line-of-sight. John Williams W2BFD designed an AFSK converter that was duplicated by a number of the New York hams. Another feature of the New York area operation was timed autostart. Stations would arrange to have their equipment turned on for a few minutes at the beginning of certain hours, under control of clocks. During that time they could receive messages unattended. 

There was some experimentation on HF using make-and-break keying. FSK was allowed only on the 11-meter band until 1953, when privileges were extended to the CW-only portions of all the HF bands. Shift of 850 Hz and speed of 45.45 baud were required by the FCC rules.


Amateur RTTY suffered indifference, if not hostility, from ARRL. One reason was the bandwidth occupied if FSK were to be used. Why occupy nearly a kilohertz (using the required 850 Hz frequency shift) to transmit at 60 wpm when CW could operate almost as fast (assuming a skilled and talented operator) in a lot less bandwidth? The CW operators didn’t appreciate having their band segments occupied by the relatively wide RTTY signals. Another reason I think was that ARRL discouraged the use of any technology that the amateur could not build himself; that certainly included a teleprinter. In the same vein ARRL tended to ignore military surplus equipment. There was the official QST advertising policy that everything offered had to be new manufacture, not used or surplus. This situation did not change until RTTY achieved a considerable degree of popularity and ARRL member hams began to demand more coverage in QST. (And perhaps until  some of the older ARRL officials had retired.) 

In the absence of much information in the ARRL official journal QST, most of the published information available to amateurs came through other channels.

• The publications of Amateur Radio Teletype Society, also known as
V.H.F. Teletype Society, of the New York area.
• RTTY, the journal of the Southern California RTTY Society.
• A RTTY column in the magazine CQ, first written by Wayne Green W2NSD. Wayne was also for a time editor of the ARTS bulletin. Later he became editor of CQ and the RTTY column was taken over by Byron Kretzman W2JTP.


The ARTS publications came out rather irregularly and seemed to just fade away as time went on. RTTY ran from 1953 through 1966 as a labor of love of Merrill Swan W6AEE. Then it was sold to a more commercial publisher and continued into the late 1990s under various publishers. By then there were other digital modes in use; and RTTY operation was usually fully electronic rather than using a mechanical teleprinter. RTTY articles in CQ continued until 1965 or so. QST began giving major emphasis to RTTY about that time as a result of demands from the membership.


There was also personal contact, through organizations and one-on-one meetings, of RTTY-savvy amateurs with those wishing to get started. Nearly every large population center had an RTTY society in its area.


As noted earlier, the first teleprinters used by amateurs were primarily the obsolete Model 12 machines, and an assortment of other oddities. The telephone companies had a policy of destroying any machines surplus to their needs, to prevent their being used in competition with their charged for services. Amateur organizations entered into protracted negotiations with AT&T and its subsidiaries and eventually achieved agreements that surplus machines could be sold to amateurs. These sales all required that the buyer sign a waiver of any non-amateur use for the machine. In the mid 1950s the telephone companies began releasing quite a few Model 26 machines. These were page machines originally used in Bell’s TWX service. They were made in the late 1930s in an attempt to have a lower-cost light-duty machine that could replace the heavy-duty Model 15 in TWX service. The expected economies were not achieved; so after World War II AT&T decided that Teletype could quit supporting them with maintenance parts and that they would be phased out of service. A trickle of Model 15 machines came into amateur hands as well. Well-heeled amateurs could obtain the latest Model 28 machines, the new Bell System and industry standard, principally for individuals who refurbished them from junked equipment. One prominent refurbisher was Ray Morrison, W9GRW. Later there was Model 28 equipment being surplused by the telephone companies, probably as a result of the transition to ASCII equipment.


The standard commercial and military method of generating FSK signals through the 1940s and 1950s was to use a 200 KHz L-C oscillator with reactance tube modulation to vary the frequency. The desired output frequency was obtained by heterodying a crystal oscillator with the 200 KHz oscillator, and in some cases by frequency multiplication from a lower frequency. This method of modulation was fairly linear, so the equipment could be used for FM facsimile transmissions as well as for FSK. Amateurs used a much simpler arrangement, a diode which could be biased on or off to switch a slight reactance into a variable frequency or crystal oscillator to shift its frequency. Varying the bias allowed varying the shift, albeit not linearly. 

The standard commercial, military and amateur method of receiving FSK signals through the early 1960s was the FM limiter-discriminator circuit, followed by a trigger circuit to provide solid mark and space signals to the teleprinter or other equipment. Unfortunately the usual amateur designs were somewhat faulty, based on some misunderstandings of filtering. Don Wiggins W4EHU, a professor at University of Florida, attempted to set things right with articles in RTTY in 1960 and 1963. He explained why any narrow filtering needed to go ahead of the limiter, and why the discriminator should be linear far beyond the mark and space frequencies. Sprague had stated this principle in his 1944 article but did not explain the reasoning behind it. 

Some Post Office workers in England published a paper in 1957 on non-FM detection of FSK signals, treating mark and space as a frequency diversity pair of on-off keyed signals. In a sense this was a reversion to the early ideas of Armstrong and Schmitt, but with more involved ways of combining the two signals. An amateur terminal unit without a limiter had been described by Gates in RTTY in 1954, but had attracted little attention. The idea hit the amateur community in full force about 1963. Thereafter it found its way into FSK converters designed mostly by amateurs but marketed to commercial and government customers.


As electronic technology has advanced so spectacularly in recent years the mechanical teleprinters have fallen by the wayside, except for nostalgia reasons. First we had active filters and video terminals based on personal computers. Then came digital signal processing on specialized DSP engines; and then personal computers became powerful enough to do digital signal processing without extra hardware on the side. New modulation methods are continually being invented, all offering some improvement over two-tone FSK. Baudot RTTY continues to be used in amateur radio, but only in two niches: contesting and DX (distant station) chasing. For other purposes the newer modulation methods generally offer much better performance than RTTY.

Disclaimer: The above is not the result of an exhaustive search of patents
and literature pertaining to radioteletype. Rather it is material I ran across
while searching for other things. The above should be considered a work-in progress. J.H.

Editors note -  How sad there model 12s stacked up like cordwood at one time and now nary one to be found.   SMECC DESPERATELY needs any model 10, 11 or 12   teletype  gear  for our corner of the display that ...  alas just  hold a photo on one presently!

Please contact Ed Sharpe - Archivist at  info@smecc.org  

 

 

ARTS -   Amateur Radio Teletype Society

History and Bulletins

From the  Jim Haynes, Red Wilson and Ed Sharpe collections at SMECC

 

Bulletins

9  10  11  12  13  14  15  16   17  18   19  20

21  22   23  24  25   26  27  28  29  30-31

32  33  34   35  36  37  38  39  40   

41  42   43   44  45   46  47   48  49 50-51

52  53  54          bulletin-letter488  Intro  

Reprint early bulletins printed on tty paper 

 

 

 






 

Although ARRL did not  publish as much on  Radio Teletype as many of the other publications  did (as noted by Jim Haynes in the above article) John Evans Williams  had an early QST article on Radio teletype history you can read  HERE




 

WANTED!  More info and photos  of this early era!  
Please Email Ed Sharpe Archivist for SMECC at    
                        info@smecc.org
 

 

 

HAL HISTORY
HAL Devices - HAL Communications Corp.

By Bill Henry, President HAL Communications  - Retired

 

HAL history goes back to 1966/67. We started as "HAL Devices", a partnership, but formalized in 1972 as "HAL Communications Corp.", an Illinois Corporation. From the start we were hams and graduate students at the University of Illinois In Urbana. Until last year, HAL was owned by one or more of the original founders.

The name "HAL" was chosen by the first HAL Devices partner, George Perrine. George said he picked the name because it is "one letter ahead of IBM". Turned out to be a good choice and easy to remember. It also does not seem to have any "nasty translations" into any other language. There was not and never has been any connection to "Hal", Arthur Clarke's intelligent computer in the screen play "2001" - later published as a book. We were very surprised when the movie hit the Coed Theatre on campus and discovered that the movie's computer had our name. We never had any problems with that similarity - "Hal" for the movie and "HAL" (all caps) for our small company. But, from time-to-time, we did have issues with others trying to use "HAL". That resulted in us paying lawyers. The offenders folded or eventually went out of business. I've also had issue with others reprinting several of my technical articles under their own names. I am sensitive to that and over the years paid a lot of money to attorneys to protect HAL's documents.

HAL Devices:

HAL Devices started in 1966 as a loose partnership of 3, 4, or 5 graduate students at the University of Illinois in Urbana, Illinois. We bought parts from various sources and then resold them at hamfests. The big items were the Fairchild 900 RTL chips - 8 pin round plastic packages. We bought ours in 100 lots from Semiconductor Specialists in Chicago and then resold them at hamfests for a price midway between list and the 100-lot price. Our first "product" was a double-balanced modulator circuit board that used HP hot carrier diodes. That was based on an article in Ham Radio Magazine. We also sold CW keyers - the 111/211/311 Keyer, then the 1550, and finally the 2550. The numbers used to mean something but I have no recall now. My activity started with the RT-1 TU that I designed for use on USAF MARS. It worked but was not manufacturable or easily serviced - live and learn. The ST-6 was of course our first big deal product and it continued for almost 10 years. The RVD-series of video display generators put us on the map. Terminal products continued to dominate sales well into the 1980's - until the IBM-PC and its clones came along. HAL Devices evolved as our products and customer base increased. Our "facilities" included George Perrine's unheated garage, my basement, Paul Tucker's kitchen, and finally 2000 sq. ft. on Locust Street in Champaign. Jim McNabb, WA9YLB, was our first full-time employee, complete with a paycheck, payroll deductions, and all that stuff. The assemblers still did piece-work but now had a place to work other than their kitchens or basements. We soon hired several high-school students to help with bagging parts, packing orders, and testing of equipment.

HAL Communications Corp.:

On January 1st, 1972, George Perrine, Paul Tucker, and I met at in Champaign to discuss the future. The meeting started at 8AM and lasted until NOON the following day, 1/2/1972. With pizza, donuts, Big Mac's and many pots of coffee, we hammered out the organization that became HAL Communications Corp. As a result of the meeting, HAL Devices was sold to HAL Communications Corp. with each of us receiving equal shares in the new company. HAL Communications quickly out-grew the 2000 sq. ft. facility in Champaign and in 1973, we purchased an old grocery store building in Urbana. That building served us well until we built the present facility on Kenyon Road in 1980. In July, 1976, George Perrine moved on and Paul and I purchased his interests. In 1983, Paul also moved on and I became the sole stockholder and owner. On Dec. 7, 2012, I sold my stock and interest in the company to Barrett Communications in Perth Australia. Joe Wittmer (my son-in-law) continues as Managing Director for Barrett.

That's the short form of the "history". The list of products is long - several hundred, I think. Some were successful, many were not. We had some really great ideas - like the video display generator, ASR terminal, MSO (mailbox), fancy modems, and of course CLOVER. Other ideas were not so great! You win some and you lose some. I had a flirtation with doing MIL-SPEC products and for 5 or 6 years it was very successful and profitable. But I almost lost my shirt when a big contract was suddenly dropped. Learned my lesson - never bid on another government contract after that (1995). Over the years, I counted over 300 different names in my employee files. Two people who started with us in 1972 are still there - Phyllis Costa, Assembler, and Mark Prather, Manufacturing Manager. As I mentioned above, Joe Wittmer continues as General Manager. None of us at HAL ever had much money. The company was always "leveraged" - heavily leveraged in some years. Growth depended on selling something. I was approached by "investors" and venture capitalists but never accepted their money - or control. I witnessed several friends go through that with generally bad results. We did make use of bank loans but that became increasingly difficult after 1993. The best plan for a small business is to not spend any more than you have! (Easy to say now.)

I thoroughly enjoyed the job. There are not many occupations where you get to pay yourself to do exactly what you always wanted to do. I had never planned to retire - "I'll die at my desk" was my response when asked. BUT - Barrett came up with "an offer I couldn't refuse" and here I am. In hindsight, I see that I should have planned for this. I really miss the place. "Retirement" is OK but gets boring at times. I still wake up in the wee hours worrying about business things - like I had done most mornings for the last 45 years. "Getting old is not for the faint of heart"!

73,

Bill Henry,

K9GWT


 

 

HAL  PRODUCTS  TIMELINE   -   Historical Reference 
This list is current as of 2013 - For Historical Reference ONLY!

 

 

 

 

 

 

 

 

Design

End

List Price

Model Number

Description

Date

Date

Each

 

 

 

 

 

DBM Kit

Dual Balanced Modulator (HAL Devices)

1967

1972

$10

111 Kit

Morse Code Keyer (HAL Devices)

1967

1969

$30

211 Kit

Morse Code Keyer (HAL Devices)

1967

1969

$50

311 Kit

Morse Code Keyer (HAL Devices)

1967

1969

$75

RT-1 Kit

RTTY Terminal Until (HAL Devices)

1967

1969

$50

ST-5 Kit

RTTY Demodulator (HAL Devices)

1968

1975

$75

ST-6 Kit

RTTY Demodulator (HAL Devices)

1968

1980

$100

ST-6 Wired

RTTY Demodulator (HAL Devices)

1969

1975

$195

AK-1 Kit

RTTY AFSK Oscillator (HAL Devices)

1968

1980

$50

W3EFG Kit

Slow Scan Converter (HAL Devices)

1968

1969

$65

1550

Morse Code Keyer (HAL Devices)

1969

1974

$95

ID-1A

CW Identifier (HAL Devices)

1969

1974

$150

RVD-1001

RTTY Visual Display (HAL Devices)

1969

1970

$495

RVD-1002

RTTY Visual Display (HAL Devices)

1970

1972

$495

RKB-1 Kit

RTTY Keyboard (HAL Devices)

1970

1972

$195

MKB-1 Kit

Morse Code Keyboard (HAL Devices)

1970

1972

$195

FLI-1Kit

Floating Loop Interface (HAL Devices)

1970

1972

$50

RVD-1005

RTTY Visual Display (HAL Communications Corp)

1972

1974

$795

DKB-2010 Kit

CW/RTTY Keyboard

1972

1975

$495

DKB-2010 Wired

CW/RTTY Keyboard

1972

1975

$795

XTK-100

Crystal Controlled AFSK Oscillator

1972

1975

$95

2550

Morse Code Iambic Keyer

1974

1980

$195

ID-1B

CW Repeater Identifier

1974

1978

$195

VDS-3000

Video Display System

1974

1975

$495

ST-6000

RTTY Modem

1975

1990

$595

Morse-A-Verter

CW to RTTY Converter

1975

1976

$495

RVD-1005A

RTTY Visual Display

1975

1979

$595

DS-3000 KSR

RTTY Visual Display Terminal, Keyboard Send/Receive

1975

1980

$995

MCEM-8080

Microcompter Experimenter Module

1975

1978

$495

MCEM-VDU

MCEM Video Display Module

1975

1978

$395

MCEM-DACI

MCEM Dual Audio Cassette Module

1975

1978

$95

MCEM-7KMEM

MCEM 7K Memory Exapansion

1975

1978

$195

MCEM- POWER

MCEM Power Supply

1975

1978

$50

DS-3000 RO

Receive Only Video Terminal

1976

1978

$595

AQS-6860

6860 Modem for Grain Quotation System

1976

1979

$95

ST-5000

RTTY Demodulator

1977

1990

$295

DS-4000

Visual Display Terminal -KSR, ASCII Code

1977

1980

$695

DS3280

Bank Video Terminal

1977

1983

$995

SignaCheck

Signature Verification Terminal

1977

1978

$1,995

PS-3100

Line Printer Controller

1977

1982

$995

System 8200

Hobby Computer using MCEM Components

1977

1979

$1,995

PS-3200

Line Printer Controller

1978

1982

$995

AQS-RO

Video Display System for AQS Quotation System

1978

1984

$1,095

DS-3281

Bank Video Terminal

1978

1984

$995

VDB-8024

Video Display Board

1978

1980

$995

DS-3100ASR

RTTY Video Terminal, Automatic Send/Receive

1978

1985

$1,995

ID-1000

CW Repeater Identifier

1978

1985

$495

PS-3350

Line Printer Controller

1979

1982

$995

DS-2000KSR

RTTY Video Terminal, KSR

1979

1985

$795

PS-3800

Line Printer Controller

1979

1982

$995

Super Duper

Contest Checking Terminal

1979

1979

$995

PS-3850

Line Printer Controller

1980

1984

$995

PS-3900

Line Printer Controller

1980

1984

$995

KB2100

Keyboard for CT2100

1981

1990

$295

DS2050KSR

RTTY & CW Video Terminal, KSR

1981

1985

$995

CT-2100

All Mode RTTY Terminal & Modem

1981

1983

$995

NCT-RO

Video Terminal for National Computerized Trucking

1981

1983

$995

CNS Ticker

Data Collection Terminal for CNS (AQS)

1981

1984

$995

CWR-685A

All Mode Portable Terminal - Telereader

1981

1982

$795

CWR-670

All Mode Portable Receive Only - Telereader

1981

1982

$495

MPT3100

Message Processing Terminal with DS3100ASR

1981

1988

$2,495

DS-3285

Bank Video Terminal

1982

1985

$1,495

RS-2100

RTTY Oscilloscope Tuning Indicator

1982

1985

$495

CWR-6850

All Mode Portable Terminal - Telereader

1982

1986

$995

CWR-6700

All Mode Portable Receive Only - Telereader

1982

1986

$595

IF-200

CNS Interface for IBM PC Jr.

1982

1984

$295

PS-3910

Line Printer Controller

1983

1985

$995

ARQ-1000

SITOR Error Correcting Terminal (for CT2200)

1983

1986

$695

DSK-3100

Disk Interface for MPT-3100

1983

1985

$995

M1700

Internal RTTY Modem Card for ARQ-1000

1983

2012

$500

LP-120

Loop Power Supply & Interface

1983

1984

$495

IF-500

CNS Interface for IBM PC Jr.

1983

1984

$395

CRI-100

RTTY Interface for IBM PC - Telereader

1983

1986

$195

CRI-200

RTTY Interface for IBM PC - Telereader

1983

1986

$295

IF-600

CNS Interface for IBM PC Jr.

1984

1986

$395

PC1000

Plu-in RTTY Modem for IBM-PC

1984

1985

$795

SPT-1

Spectra Tune Tuning Indicator

1984

1988

$95

ST-8000

High Performance FSK Modem, Tunable

1985

2000

$2,995

MSU-4

SIO Interface, Modem Sharing Unit

1985

1990

$195

WX-1000

Weather Data Interface and Collection Terminal

1985

1990

$995

AMTOR-10A

Amtor Interface for CWR-6850 - Telereader

1985

1986

$495

ARQ-200

SITOR Terminal for Portable Marine Use

1986

1990

$995

LP-1200-1

Loop Power Supply & Interface

1986

1990

$795

HFCS-1000

HF Communications Simulator (RACAL Receiver)

1986

1990

$4,950

RMX-3100

Dual Port Switch for MPT-3100

1986

1988

$1,495

ARQ-1000A

SITOR Error Correction Terminal

1987

1995

$1,495

DS-3200

Bank Video Terminal

1987

1990

$1,995

Alert-1

RTTY Message Alert for US NAVY

1987

1989

$149

RPC-1000

RTTY PC Interface Card

1987

1989

$795

MRI-1000

Morse RTTY PC Interface Card

1987

1989

$795

RPC-2000

RTTY PC Interface Card

1987

1989

$795

STI-1000

Spectrum Tuning PC Interface Card

1987

1989

$795

ST-7000

Modem for use with AX.25 Packet Radio

1988

1989

$295

SCM-1000

Split Channel Modem for Rockwell

1989

1989

$1,495

LP-1210

Loop Supply - 10 Channel, Military, 100K Hrs MTBF

1989

1992

$4,950

PC-AMTOR / PCI-3000

AMTOR/RTTY Interface Card for IBM-PC

1989

1993

$995

SPT-2

Spectra Tune Tuning Indicator for PC-AMTOR

1989

1993

$149

ARQ-200B

SITOR Terminal for Portable Marine Use

1990

1993

$1,495

ST-6100

FSK Modem

1990

1990

$1,495

ST-8000A

FSK Modem, Military, 50K hours MTBF

1990

2012

$5,500

CLOVER-1 (Summer Clover)

CLOVER Terminal - Version 1

1991

1991

$995

PC-CLOVER / PCI-4000

CLOVER Modem Card for IBM-PC

1992

1995

$995

HFCS-2000

HF Communications Simulator (Remote Access)

1993

1995

$3,495

RVM-100

Radio Voice Mail Interface

1994

1996

$395

KFS SIO

Interface Card for KFS Marine (Globe Wireless)

1994

1995

$995

P38

CLOVER/RTTY Interface Card for IBM-PC

1994

1996

$795

PCI-4010

CLOVER/RTTY Interface Card for IBM-PC

1994

1996

$1,295

ARQ-1000B

SITOR Error Correction Terminal

1995

2012

$1,495

DSP-4100

CLOVER/RTTY Modem, Stand-alone

1995

2000

$1,295

LI-4100

Telephone Line Interface for FAX4100

1996

1996

$495

FAX-4100

FAX-Over-Radio Interface

1996

2000

$1,495

HFCS-2000A

HF Communications Simulator (2Ch., Remote Access)

1997

2000

$4,000

DXP-38

CLOVER/RTTY Modem, Stand-alone

1997

2011

$395

LP1203

3 Channel Loop Supply, MIL-810B, 20K Hrs MTBF

1999

1999

$2,000

FAX-4110

FAX-Over-Radio Interface

2000

2006

$1,495

DSP-4100/2K

CLOVER-2K/RTTY Modem, Stand-alone

2000

2002

$1,495

W5XD Keyer

CW Keyer for WriteLog Company

2001

2003

$149

DSP-4120

CLOVER-2K/RTTY Modem, Stand-alone

2002

2003

$1,495

Clarinet Merlin

CW Decoder for SPAWARS (NAVY), 20K Hrs MTBF

2003

2003

$2,925

DSP-4130

CLOVER-2K/RTTY Modem, Stand-alone

2003

2005

$1,495

DSP-4131

CLOVER-2K/RTTY Modem, Stand-alone

2005

2006

$1,495

FAX-4120

FAX-Over-Radio Interface

2006

2012

$1,495

DSP-4132

CLOVER-2K/RTTY Modem, Stand-alone

2007

2008

$1,495

BRT-4133

CLOVER-2K Modem Card, for Barrett 2050 Radio

2007

2008

$793

DSP-4134

CLOVER-2K Modem, Stand-alone

2008

2012

$1,495

FXD-4100/2K

FAX-Over-Radio Interface with CLOVER Modem

2008

2012

$2,250

BRT-4135

CLOVER-2K Modem Card, for Barrett 2050 Radio

2008

2012

$793

DSP4200

CLOVER-2K Modem, USB Interface

2010

2012

$1,595

M4200

CLOVER-2K Modem, USB Interface, Ruggedized

2010

2012

$2,750

 
 
 
 
 
 
HAL ARTICLES  IN QST 
   HAL                        2550/ID Keyer                       Sep-76 
   HAL                        CWR 6850 Telereader RTTY/CW Te      May-83 
   HAL                        DS-3000 Video Display Terminal      May-77 
   HAL                        DS-3100 ASR Video Display Term      Apr-80 
   HAL                        FYO Key                             Dec-76 
   HAL                        MCEM-8080 Microcomputer             Dec-76 
   HAL                        ST-6000 Demodulator                 May-77 

(We are putting together a list of any and all journals -  drop us a note if you have a list!)

 
 
 

        T2eC16NHJHQFFhJ-ZtBR5vtlPEpw60_57.jpg (485915 bytes)

 

 


(From a PHONE/TTY ad  in  the Deaf Jersyite)

PHONE-TTY In New Jersey resold   HAL Communications products to the Deaf and Hard of Hearing
This  would have been a typical offering  n the early 1970s and in conjunction with a phone TTY modem.

 

 

 

 

 

ESSCO 

RTTY TUs  starting in the  1960s
Also Modems and Terminals for the Deaf and Hard of Hearing
CLICK THE IMAGE ABOVE TO READ ABOUT ESSCO...

 

 

 

 

 

 

 

 




 

(Wayne Green Editor or ARTS had his mother  help illustrate!

 

 

Let Play.... NAME THAT TU! (or it's designer!)

Had a nice visit with Larry - W0OGH and he turned us onto this
TU it hooks to the IF of a recv.

Can we find who designed it or where it appears in a journal or...

Maybe it is just a ONE OFF.. but it is neat! Any hints appreciated!

And many thanks for the addition to the TU collection Larry!

Ed Sharpe archivist for SMECC - KF7RWW

 

RTTY TU?
SMECC has this at the museum....  WHO MADE IT?!

 

 
 
 
 

Jim Haynes talk on  the question  which is
 better the   ST-8000 or the Dovetron  TU?

 

The ones with CRT displays are, I think, pretty rare. It's my
understanding that a time came when CRTs were impossible to get at
any reasonable price, so they went to the LED displays.

There seem to be many variations of the black faced model with LED
display. Some have the regenerator/speed converter, some have
Baudot-to-ASCII converters, some have 60ma loop and some don't,
some have everything on the main board and others have additional
board piggy-backed on the main board.

The Dovetrons seem to have got poor ratings from some users.

I have only one, and found a construction error on a PC board that makes
it perform much worse than it should. But I haven't yet compared it with
any of the others I have after fixing the error.

The original design of the MPC (multi path corrector) is I believe what
is embodied in Hank Scharfe's patent 4,013,965. But apparently that
did not perform well, because later models have all that stuff pulled off
the PC board and there is a plug-in daughter board with a "binary bit
processor" of some complexity. But apparently that didn't work well
either, so there is a later version of the binary bit processor that is
just a threshold corrector. The earlier ones detected mark and space
separately and tried to deal with multipath echoes where mark and space
signals are both present simultaneously. This last one simply combined
the detector outputs with a threshold corrector and doesn't treat the
signals separately after that point.

So it would be good if you could keep a selection of them with these
different modifications and run them in parallel to see how they
stack up against one another.

In the HAL ST-8000 which is generally regarded as the best of the
hardware demodulators there is a multipath handling circuit, but I
don't know how effective it is. They detect mark and space separately.
When the two detectors agree, that it is either mark or space, that is
what it puts out. When the two detectors disagree it might be that
it is neither mark nor space, or that it is mark and space at the same
time. Then they set a latch that preserves the last known-good signal.

This could be considered hysteresis - the signal is held at whatever was
good until the opposite condition is considered good. The Sprague paper
back in 1944 advocated hysteresis for limiter-discriminator demodulation,
but I haven't seen anything about how effective or useful it really is.
And some hysteresis is added by putting the signal through a low-pass
filter before making the mark/space decision.

One published amateur design incorporated anticipation, the opposite of
hysteresis. As soon as he saw mark going away he flipped the output to
space, and vice versa. It would be interesting to build a demodulator
with a switch giving no hysteresis, hysteresis, and anticipation and
see which performs better under various conditions.

There is presumably a government specification somewhere which led to
the development of similar demodulators by several different companies.
Dovetron, HAL ST-8000 and Electrocom 440 series all have independently
tunable mark and space frequencies, achieved by different means, and
MIL-STD-188 output instead of loop current, and other similarities.
And the Frederick 1280.

The outstanding feature of the Dovetron is that, while the frequency
calibration of the knobs is only approximate, it's very easy to tune in
a signal of arbitrary shift. As I recall the ST-8000 (I haven't had
one in my hands) is rather harder to tune. The Electrocom has two
options - one lets you set the frequency and the shift by separate knobs,
and the other lets you set the two frequencies by separate knobs. But
the way the tuning indicator works it is very hard to tune in signals
with the knobs. And the Frederick has you enter the mark and space
frequencies on a keypad.

Someone suggested that the real purpose of these things is to tune in
one channel of a frequency division multiplex carrier system. In that
case the frequencies are pretty well known and stable, unlike random
FSK RF signals, so the tuning indicator needs to show you only that you
are or are not in tune, not how to get in tune. The old crossed-ellipses
CRT display has the virtune that you can see which direction you are off
frequency and tune the receiver to correct it. Zero-center meter displays
are not very helpful since the average of typical FSK signals is not zero.

I'll put in an aside here. Some WW-II military stuff, like the BE-77 line
unit has a zero center meter and you do adjust for a zero reading while
receiving repeated space-bar characters. This signal doesn't have a zero
average at all, but the meter is intentionally biased to read zero on
specifically that signal.

I've always wondered why somebody didn't do a peak-reading zero-center
meter so that correct tuning would give a zero reading. (The average of
the peaks would be zero even if the average of the signal isn't.)

There's Irv Hoff's meter tuning indicator where you tune for highest
meter reading and least meter wiggling - I find that pretty easy to use.

Another good one is in the amateur PK-232 and some other amateur equipment
where a single LED bar indicates mark and one end and space at the other,
so you just tune to center the brightest spots on the LED.

The Electrocom 440 series has a CRT tuning indicator, but it's practically
useless because they multiply the deflection signal by the detector
output. So you get either a vertical bar or a horizontal bar or nothing,
which doesn't tell you which direction you are off frequency. I've got
a modification in mind that would make the display more conventional and
useful, but I haven't tried it yet - just bypass the multipliers and
display the filter outputs. The Frederick 1280 has parallel vertical
bar graphs for the mark and space. That's no worse, I guess, than the
crossed LEDs of Doveton and later HAL, but it's not very easy to tune.

Now I bet that's more than you wanted to know.


jhhaynes at earthlink dot net

 

 

 

 

 

 

 

 

 

 

 

 

 

Everyday we rescue items you see on these pages!
What do you have hiding in a closet or garage?
What could you add to the museum displays or the library?

PLEASE CONTACT US!

===================

DONATE! Click the Button Below!


Thank you very much!

===================

Material © SMECC 2007 or by other owners 

Contact Information for
Southwest Museum of Engineering,
Communications and Computation 
&
www.smecc.org

Talk to us!
Let us know what needs preserving!


Telephone 
623-435-1522 

Postal address 
smecc.org - Admin. 
Coury House / SMECC 
5802 W. Palmaire Ave 
Glendale, AZ 85301 

Electronic mail 
General Information: info@smecc.org