The Story of HIGH COM
by Ernst F. Schroeder
1. Contents
The start of HIGH COM development
2. Introduction
This is the story of the HIGH COM (*) Compander (**). It is intimately linked to my time as electronics engineer at the Telefunken Basic Research Laboratory in Hannover from 1970 until 1983. The story is based on surviving notes, photographs and personal memory.
My thanks go to former colleagues Jürgen Wermuth and Hans-Jürgen Kluth, who have contributed facts, photographs and helpful corrections. I further have to thank Manfred Geske und Ingo Weishaupt, who have contributed photographs of their HIGH COM devices.
3. The very beginning
My time at Telefunken started on May 15th, 1970. Shortly after graduating from Hannover Technical University I was hired as electronics engineer for the Basic Research Laboratory. This was a laboratory specially set up for Professor Walter Bruch. It was not located on Telefunken premises in Hannover-Ricklingen, but since about 1968 at the other end of town, in Vahrenwalder Str. 215. I already knew Professor Bruch from Hannover University, where he held lectures on TV technology, but about his lab I first learnt from a paper by Werner Scholz in German electronics magazine Funkschau [1]. A small note stated that "The author is member of staff of AEG-Telefunken, Basic Research Lab, Hannover".
When I showed up for an interview at Telefunken headquarters, they tried to steer my interests towards the TV development department, but - apparently a bit disappointed - they finally let me go to the "Exots" in the Basic Research Lab.
My first year in the lab I spent with a number of smaller projects and with support work for other engineers. A regular and much loved, but sometimes also hated task for all staff was to man the demonstrations of the advantages of the PAL colour TV system. The lab was specially equipped with all sorts of devices necessary to demonstrate colour TV systems like NTSC, SECAM and PAL and their behaviour in case of noise and other distortions. PAL always won.
These demonstrations were given more or less regularly to all kinds of visitors. Every staff member had his defined place on a certain device, while Hans-Jürgen Kluth gave the master of ceremonies. He listened to the aural presentation by Prof. Bruch and quickly issued the appropriate commands: "PAL", "NTSC", "saw-tooth", "rotating phase" and so on. A standardized command for Günter Brandt, the operator of the 35mm film scanner, shortly after starting the 35mm film "Palettes of Mode", was: "more green!". From all these occasions I clearly remember the highly sovereign manner, in which Professor Bruch held his presentations in front of diversest kinds of audiences.
In the yard behind the lab building there still stood the Magirus OB van that had been used for the PAL demonstrations in South America in 1968. It had strong air conditioning and was a preferred refuge on hot summer days, even when other work was waiting. Sometimes even the ingredients for "Cuba Libre" found their way into that van. The photo shows me on the video mixer in that van, together with Prof. Bruch and my colleague Gunther Raschke.
Fig. 1: Professor Bruch, Gunther Raschke und Ernst F. Schroeder in the special PAL-OB-van
One of my first own projects in July 1972 was the investigation of a new electronic device, the bucket brigade circuit. With this variable analogue memory device for audio and video signals it was possible to realize variable signal delays. Around such a device I designed a circuit for cancelling wow and flutter in a simple cassette recorder, a Telefunken Stereosound TD. The following screen shot from an oscilloscope shows that wow and flutter were indeed greatly reduced. But the application needed some kind of reference or pilot signal on the tape. This was not feasible for a consumer type cassette recorder, but over all it was a good sample application for the new bucket brigade devices.
Fig. 2: Wow and flutter of a cassette recorder
original (below) and compensated (above).
Horizontal axis 50 ms/div, vertical axis 0.33%/div
In 1973 Professor Bruch travelled to the USA and visited the former labs of Thomas Alva Edison in Menlo Park. He returned with a chest full of old Edison wax cylinders. It became my task to capture the sound from these old cylinders. I took a simple Telefunken disc record player, mounted it vertically and had the workshop mount a conical cylinder on the turntable. Finally the arm was rotated by 90 degrees and the connections to the stereo cartridge were interchanged. Thereby the Edison hill-and-dale modulation could be captured as the monophonic output signal. While the lab was equipped with lots of video equipment, there was nearly nothing for audio. So I took some of my privately owned audio equipment to enhance the subjective quality of these recordings.
The most noteworthy of these recordings is that of the sound of Big Ben, the bell in parliament tower in London, originally recorded on 16th July 1890. This recording has been made with a very early Edison recorder driven by a spring motor. Therefore the recording is not very straight with lots of fluctuations.
A further unique recording is the attack signal blown by trumpeter Lansey, one of the surviving trumpeters of the battle of Balaclava in the Crimean war of 1854. He tells us that he blew exactly this signal on exactly the same bugle on October 25th, 1854 for the charge of the "Light Brigade".
Professor Bruch finally used his extensive collection on the history of audio and video storage for a series of papers in the German electronics magazine Funkschau [2].
One of my further tasks was the tracking of new developments in the TV field. In November 1974 I visited BBC in London and got to see a first version of CEEFAX. That was what we call Videotext today. The colourful presentation of text on a TV screen was new at that time and very impressive.
4. The telcom compander
Within the business unit “Weitverkehr und Kabeltechnik” of AEG-Telefunken there was a department working on professional electroacoustics and studio technology, the “ELA” in Wolfenbüttel, Lindener Str. 15.
There was a young electronics engineer, Jürgen Wermuth, who worked on low-noise microphone amplifiers. AEG-Telefunken at that time was distributor for Dolby Labs in London for their Dolby-A noise reduction system (a compander) for professional audio tape recorders. Jürgen Wermuth also worked with these companders and was not at all content with the technology and the typical problems that he had to deal with. That led him to try to develop a much better compander at home in his spare time. This he started around 1971 - 1972.
He tried to solve the following problems:
increased compander gain
no need for exact level adjustment
independence from tolerances
For this Jürgen Wermuth found the following solutions:
Compressors and expanders typically used field effect transistors (FET) for variable resistors and control elements. These had to be used in suitable symmetrical circuits in order to work with low distortions. For this there existed a solution by Rolf Röder from Sennheiser [3], but the use of RC-elements was not suitable for fast control. The solution found by Wermuth [4] only used operational amplifiers for balancing and was therefore much better suited for companders.
FETs as variable resistors show large tolerances. Wermuth developed a circuit [5] using twin-FETs (2 identical FETs on one piece of silicon) that delivered variable resistors with very low tolerances.
The use of the chain amplifier principle [6] made it possible to get a practically unlimited linear compander characteristic. This circuit uses a number of identical variable amplifiers, connected in series. When the audio level at the output of the last amplifier is held constant, automatically a linear compression characteristic is available at each intermediate amplifier. The corresponding expandor characteristic is attained by a slight change in the over-all circuit. Wermuth designed telcom c4 to use 3 identical variable amplifiers and thus a compression characteristic with slope 2/3.
The chain amplifier principle needs a special circuit at the output of the last amplifer to derive a control voltage for all variable amplifiers with the effect of holding the output voltage of the last one constant. For this Jürgen Wermuth developed a special circuit to linearly charge and discharge a capacitor via constant current sources [7].
Fig. 3: The chain amplifier principle with 3 amplifiers and the
resulting characteristics
At first the ELA management did not want to hear about Wermuth's superior solutions, but finally agreed to start a formal product development project. The result was the Telefunken-Compander with four separate frequency bands, therefore finally named "telcom c4".
Stephan F. Temmer, founder of Gotham Audio Corporation and distribution partner of ELA in USA, always looking for new and better solutions for audio technology, may have been instrumental in convincing the ELA management to market this new development. Naturally, the relation between ELA and Dolby labs did not get better when words about the new compander started spreading.
Finally in autumn 1975 the first presentation of the new telcom-c4 compander took place at the 10th meeting of German sound engineers (Tonmeistertagung) in Cologne [8] and in electronics magazine Funkschau [9].
Fig. 4: View of one of the first telcom boards (Photo by Jürgen Wermuth)
5. The start of HIGH COM development
In addition to his work on a professional compander, Jürgen Wermuth also had worked on a variant for use in consumer-type equipment like audio cassette recorders. He named it "RUSW" for "Rauschunterdrückungssystem Wermuth".
Finally, on April 23rd, 1974 Wermuth presented his "RUSW-200", coupled to a Telefunken MC 3300 cassette recorder, before members of the board, the head of product development and patent attorneys.
Fig. 5: View of the RUSW-200 prototype (Photo by Jürgen Wermuth)
Shortly after, on July 09th, 1974 Dr. Klaus Welland, board member for product development at Telefunken, put me in his car and drove the 50km from Hannover to ELA in Wolfenbüttel. That was a remarkable ride in many respects. I not only knew him as active radio amateur (his call sign was DL1MR), but also as fast driver. Well, I have survived.
It was a high level meeting at ELA in Wolfenbüttel, to which also Jürgen Wermuth and myself were allowed to attend. The result was that within AEG-Telefunken naturally Telefunken in Hannover would be responsible for a consumer-type variant of the new compander system and from now on the further development would be done by myself. As a start in August 1974 I received detailed papers and further information from Jürgen Wermuth, as well as a first design for an analogue integrated circuit.
It is interesting to note that exactly while this meeting took place, a letter arrived from Dolby labs in which they cancelled the distribution agreement with AEG-Telefunken, withdrawing the sole distribution rights for Dolby-A devices.
Meanwhile Professor Bruch had transferred the direct management of the basic research laboratory into other hands. My new boss was Dr. Gerhard Dickopp, who had been transferred from Berlin to Hannover together with part of his staff. In Berlin he had been instrumental for the development of the Telefunken-Teldec video disc TED. That TED video disc consisted of a thin foil onto which analogue video and audio information was mechanically stamped and also mechanically read out. It used a special method to store colour video signals called "TriPal" that had been developed in Hannover. I also had contributed some ideas to the way the audio signal was recorded on the TED video disc.
I immediately started with the compander development. The initially proposed circuit contained the basic building blocks of telcom, but had only two chain-connected variable amplifiers and used only two separate frequency bands. To further simplify the implementation, only the higher frequency band was companded. The frequency at which the two bands overlapped was practically constant.
After a short while the initial design had been changed in a way so that a much more advantageous compander with a sliding-band characteristic emerged. Such a characteristic was also used by the Dolby-B System, already introduced in October 1969. What was not used was the separation into "main path" und "further path", characteristic for the Dolby-B system [10]. The total gain in signal to noise ratio by the new compander was about 20 dB, about twice that of Dolby-B. In June 1975 the first patent application [11] for the new compander system was issued. In due course further applications were worked out.
Fig. 6: Block diagram of a sliding-band compressor with two chain-connected amplifiers
Fig. 7: The sliding-band frequency-response of the compressor in Fig. 6
In early 1976 a further remarkable event took place in the Telefunken basic research laboratory. Some of the engineers had come together over a cup of coffee and were fooling around with crazy ideas on TV technology. As we also worked on video recorders, a problem was how a recording could be started on time, even if the desired program was delayed for some reason. Benno Jahnel, one of the engineers who had come from Berlin to Hannover, had the revolutionary idea and turned it into a patent application: The TV program should not convey the actual time of day in some coded way, but the wrong time information, namely the intended start time of the program item. The subsequently issued patent [12] led to the VPS-system in TV transmissions and also to considerable license income for Telefunken.
Finally, in October 1976, I delivered a paper before the FKTG meeting in Freiburg on the professional telcom c4 compander and the still nameless compander for consumer application, jointly written by Jürgen Wermuth and myself [13]. I still remember very well, that Elmar Stetter, representative of Dolby labs, was not amused when he heard about this new competition.
For actual application in consumer devices it was absolutely necessary to realize the compander in an analogue integrated circuit. In early 1977 I therefore travelled to AEG Labs in Ulm and to the AEG-Telefunken semiconductor factory in Heilbronn. There I met Dietrich Höppner und Kurt Hintzmann. The latter finally transferred the initial design by Jürgen Wermuth together with the newly developed improvements by myself into an analogue integrated circuit [14].
When the first samples were available and worked as intended, I received an invitation by a German broadcast station (Hessischer Rundfunk in Frankfurt). It was end of September 1977, when Joachim Bublath made a feature broadcast on audio companders including me and my first sample device.
6. Nakamichi joins in
Again, Stephan S. Temmer stepped in. His contacts to AEG-Telefunken ELA in Wolfenbüttel had probably transported the news about a new consumer-type version of telcom being developed in Hannover. Very probably he had given this information to the company building the undisputed best cassette recorders of the time: Nakamichi in Tokyo.
It did not take long until the complete top management of Nakamichi, led by founder Etsuro Nakamichi, appeared at Telefunken in Hannover and had me demonstrate the new compander system. They were very impressed but at the same time disgusted when they realized what kind of audio equipment the poor development engineer had to use.
Immediately a selection of the finest Nakamichi equipment was sent to Hannover: N1000 and N600 cassette recorders, pre- and power amplifiers, the special T-100 Audio Analyzer, and two large loudspeakers. Compared to the typical Telefunken Hi-Fi equipment at that time, this really was something special that clearly influenced the further development of the new compander.
Initially Nakamichi had planned to incorporate the telcom c4 compander into their N1000 in an attempt to build the ultimate high-end recorder for compact cassettes. But AEG-Telefunken ELA wanted telcom c4 only in professional applications and did not agree to a license. Instead, they talked about the consumer version just under development. As Nakamichi had seen and heard the first version in Hanover and received two working samples in July 1977, they finally gave in and waited for the IC to become available.
On June 4th, 1977 a license agreement between Nakamichi and AEG-Telefunken was signed.
7. HIGH COM is born
A new cassette tape recorder was under development at Telefunken and it was clear that it would not have Dolby-B but the newly developed compander instead. But meanwhile Telefunken management had developed some fear that the similarity between Dolby-B and the new compander could lead to long and costly patent litigations. Even while it seemed quite obvious that the new compander did not violate patents by Dolby labs. Therefore I got the task to investigate other variants of companders with ingredients of telcom. Meanwhile the IC was nearly ready, so what could be done at all?
The history of companders for audio signal goes back to at least 1925 [15]. The first companders had worked on the complete audio signal spectrum as a whole, so-called broad-band companders. Most of them had been found less suitable, this was also the case with later developments by EMT [16], Burwen [17] and dbX [18].
Since about 1937 it was known [19] to cut the audio frequency range into a number of separate ranges or bands and to apply separate companders to these bands. The design as multiple-band compander has probably contributed to the success of Dolby-A [20] and was also used by telcom c4. For a compander in consumer-type applications this would have meant some quite costly expenditure.
But there we had the outstanding principle of linearization by chain-connected amplifiers, a special generator for control signals, high-end Cassette tape recorders like Nakamichi 1000 or the new one under development at Telefunken, what could happen at all? The frequency dependent parts for the sliding-band effect were easily removed. The marginal masking of noise in a simple broad band compander was attacked with an additional pre/deemphasis, and the dreaded noise tails behind abruptly ending signals were attacked with a special generator for control signals [21] that exhibited a dual time constant switchable between "slow" and "fast".
Both the static and the dynamic behaviour of this broad band compander were intensely tested with all kinds of music and especially with the attack sound of a triangle. I still have this triangle in my possession
Fig. 8: Attack signal of a triangle sound, played back from a cassette recorder,
above without compander, below with HIGH COM.
Only a very small distortion can be seen at the start of the signal.
Fig. 9: Photo of that very triangle I still own
On November 28th,1977 I presented both versions, sliding-band and broad band, with their advantages and disadvantages to the management of Telefunken product development.
The result of development work was considered OK and the decision fell in favour of the broad band version. In order to start production, the IC first had to be adapted, which amounted more or less to a complete redesign. There may have been some harsh words between Telefunken management in Hannover and AEG-Telefunken management of IC factory in Heilbronn.
In February 1978 there appeared a paper by Dr. Dickopp and myself in Rundfunktechnische Mitteilungen (RTM) [22]. The paper described the new "Telefunken Compander", both the professional telcom c4 and the two variants of the consumer-type compander, the sliding-band and the broad band version.
A further development happened in early 1978: End of August 1978 the first Hi-Fi exhibition should take place in Düsseldorf. This would be run in direct competition to the well established biennial “Funkausstellung” in Berlin. German Hi-Fi manufacturers did not like that idea and they really did not want to spend twice the money on two similar exhibitions. Therefore they all agreed to abstain from participation. But all the young competitors from far-east gladly accepted.
Now Telefunken had their new cassette tape recorder with direct drive ready and also the new compander. This was a good reason to change ones mind and to take part in Düsseldorf Hi-fi exhibition on short notice. A crash program for preparations was started. At that time there were only a few functional compander ICs that still had to be augmented externally for the broad band version.
Therefore I quickly built some PCBs with the compander circuit and Bernd Wiedenroth from product development married them to the new cassette recorder, the TC750.
Fig. 10: Frontal view of a Telefunken TC750
Next a name had to be found for the new system. After some thinking and searching the name "HIGH COM" was agreed upon. The exact spelling contains a half-blank space between "HIGH" and "COM". This was something that electrical typewriters were able to produce at that time (much better than desk top publishing software today). But this was really difficult to communicate and is still difficult today, even with modern DTP software. Also a logo was designed that looks a bit like magnetic recording heads:
Fig. 11: HIGH COM logo on the cassette lid of a Telefunken HC1500
The efforts that Telefunken undertook for this Hi-Fi exhibition were enormous. The Telefunken display occupied nearly a complete exhibition hall and with tons of sand an absolutely sound-proof room with air condition was built in the back part of that hall. There the top equipment of Telefunken was on display, especially the TC750 with HIGH COM.
For our demonstrations we needed music in top high quality. For this I travelled to Berlin, where Professor Martin Fouqué, the leading sound engineer (Tonmeister) of Teldec let me copy some original recordings that had been recorded with telcom c4. These were then copied to CrO2 compact cassettes using HIGH COM.
The resulting audio quality really was remarkable. What our audiences liked best was a recording of "Also sprach Zarathustra" with Sir Georg Solti.
Of course, some technical measurements were also part of our demonstrations. We used a real-time 1/3 octave spectrum analyzer from Brüel & Kjaer (Type 2131) to optically display and compare the resulting gain in signal-to-noise ratio.
Fig. 12: Display of compander gain on a real-time 1/3 octave analyzer,
left: Dolby B, right: HIGH COM
dark: Noise spectrum without, bright: noise spectrum with compander
Between 18 and 24 August 1978 Bernd Wiedenroth and myself have given about 16 presentations and demonstrations on HIGH COM of 60 min each. In addition and “on the fly” two in English and two more for the press.
During a press conference on August 17th, 1978 it was announced that Dr. Dickopp, then manager of the Telefunken basic research lab and therefore my boss, had accepted a call to University of Duisburg.
After the successful introduction at Düsseldorf Hi-fi exhibition, the Telefunken management decided that for further promotion of this system a dedicated HIGH COM manager was needed. The choice fell on Dr. Thuy from Telefunken in Berlin. He came to Hannover and moved into the former office of professor Bruch. Meanwhile I had occupied this room and installed equipment for acoustic experiments. This was one of the larger rooms in the otherwise small building and I had put my equipment directly in front of the large picture that showed the young engineer Walter Bruch in 1936 as TV camera operator in the Berlin Olympic stadium. It looked like there was no other suitable room in the building. Well, I did not move out and Dr. Thuy got used to my experiments …
Finally, end of September 1978 the first functional samples of the HIGH COM IC, designated U401B, were delivered by AEG-Telefunken IC factory in Heilbronn.
In that same month a paper by Dr. Dickopp and myself appeared in Funkschau [23], in which we pointed out that the assessment of stationary properties of companders should be carried out with noise signals and 1/3 octave real-time analysis.
8. HIGH COM II
Contacts to Nakamichi prevailed, but engineers there still thought they would get something like Dolby-B, but with much better properties. But by now Telefunken had definitely decided to go with the broad band version.
On September 20th, 1978 Mr. S. Takai, vice president of Nakamichi, came to Hannover. The result was a big and typical misunderstanding. Nearly all he said was "hai" and therefore everybody thought that now Nakamichi would agree to the broad band version. Only after his departure the situation cleared up. Of course, Nakamichi did not agree. And what now?
After some thinking a solution was found that even came a bit nearer to the initial intentions of Nakamichi: Based on the IC developed for the HIGH COM broad band compander a split-band compander would be developed specially for Nakamichi.
At that time I personally was working in complete overload mode, as the first 50 working samples of the U401B ICs had to be tested and PCBs had to be built. There was no other support available, it was single-handed work of a single person. In addition test procedures had to be developed for the IC production. And there was only one person who exactly knew everything about HIGH COM ...
Therefore Werner Scholz, a year-long colleague from the basic research lab (the one who for PAL-Demonstrations always had to man the PAL-, NTSC- und SECAM-Decoder rack), got a short introduction and the task to develop a two-band and three-band split-band version of HIGH COM. He did this in his usual precise and inventive way [24], so in April 1979 we had functional samples that nearly fulfilled the expectations. Nakamichi favoured the two-band version, probably more on cost grounds, but they were not really content with the over-all sound quality. Of course, I knew the resulting quality and finally added a small but important detail, an additional pre/deemphasis in the lower of the two frequency bands [25]. Finally a solution was needed for a further small problem: someone had to travel to Tokyo and Nakamichi Labs. Well, they didn't search for long.
From 06-13 May 1979 I visited Nakamichi Labs in Tokyo. They had specially flown in a young engineer from Los Angeles, who was native Japanese and spoke good English. He was there to look after me. For several days Harron K. Appleman and myself sat until deep into the night and twiddled with the parameters of the two-band compander, until the "Golden Ears" of Nakamichi were finally content with the results. It was quite a sensation when they found out that this engineer from Germany was able to read colour code and wanted to use the soldering iron by himself. Finally all were content and HIGH COM II was born. Nakamichi only had to design a product, which they did in their usual very professional way.
Fig. 13: View of a Nakamichi HIGH COM II device (Photo Manfred Geske)
9. HIGH COM Promotion
After the Düsseldorf Hi-Fi exhibition many representatives of consumer equipment manufacturers came to Hannover and received a treatment with HIGH COM demonstrations. This was pushed even more by the activities of Dr. Thuy. Many times I had to travel to one of these manufacturers.
In July 1979 a collaboration started with the Institute for Broadcast Technology (IRT) in Munich, that would last for many years and in some years let me travel to Munich at least once a week. From that date on we had continuous discussions about the application of companders in FM-broadcasting. The IRT had already done research on this topic and had tested the Dolby-B compander [26].
Early in 1980 the AEG-Telefunken semiconductor factory in Heilbronn finally had the application document [27] ready for the HIGH COM IC U401B. All the information in this document had been prepared in Hannover. After U401B and U401BG they were meanwhile talking about the U401BR, and actually there have been three slightly different versions of this integrated circuit:
- U 401 B
- U 401 B with yellow dot or U 401 BG
- U 401 BG with blue dot or U 401 BR
On February 7th, 1980 at 21:45 a TV feature by Westdeutscher Rundfunk WDR was broadcast, „Hobbythek” by and with Jean Pütz. In this TV broadcast Jean Pütz presented HIGH COM intensively and finally showed the Hobbythek-version of an attachment for tape- and cassette-recorders with HIGH COM: HobbyCom. The small device used the same modules that were also used in all new Telefunken cassette recorders, and the surrounding circuitry had been designed by Telefunken product development. As typical for Hobbythek, the device was available to the public in kit form.
Fig. 14: View of a HobbyCom device (Photo Ingo Weishaupt)
For the time between 4th and 18th May 1980 Telefunken organized a HIGH COM promotion tour in Japan. Participants were Dr. Klaus Welland, board member for R&D, and Rolf Schiering, director of sales, and someone who had to do the work. The reader may guess who that was.
A presentation suite at Tokyo Hilton was prepared and demonstrations were held for representatives of 34 companies.
Fig. 15: A HIGH COM presentation at Tokyo Hilton
The second Hi-Fi exhibition in Düsseldorf took place from 19th until 24th August 1980. This time Telefunken received the German Hi-Fi Award for development and introduction of HIGH COM. Of course, only board members were allowed to attend the ceremony and party.
It is not known, where this trophy finally went to, but at least a photo remained.
Fig. 16: German HIFI-Award 1980
The inscription reads:
Der Firma Telefunken für die Entwicklung des Rauschunterdrückungssystems HIGH COM verliehen vom HiFi-Magazin KlangBild
Even the local press took up the story. On September 5th, 1980 the Hannoversche Allgemeine Zeitung HAZ published a half-page article under the title "Rauschender Fluß der Töne wird im Wehr gereinigt" (Rushing river of sound is cleaned in the weir). In the first paragraph the author tries his hand at prophecy:
"The brick building at Vahrenwalder Str. 215 may perhaps some day be decorated with a commemorative plaque. Here, the audiovisually educated descendants will learn, revolutionary technologies were invented in the sixties and seventies, revolutionary for eye and ear - in this house Walter Bruch developed the TV colour system PAL, and his young staff member Ernst Schroeder later helped the tape cassette creep out of its infancy."
Good idea, but after Telefunken moved out in 1983, the building stood empty and was demolished in the 1990s. So, no walls left for a plaque. And after all, PAL had not been developed here, but in a basement laboratory in Hannover-Ricklingen in Göttinger Chaussee.
At the end of 1980 Telefunken reported that more than 25 companies had signed license agreements and that more than 35 companies worldwide were system partners with HIGH COM.
1980 also was the year in which Dolby Labs announced their Dolby-C system, a sliding-band compander that used two of the ICs normally used for Dolby-B. This compander also reached an improvement in signal-to-noise ratio of 20 dB.
10. HIGH COM in Beijing
AEG-Telefunken meanwhile had seen some drastic changes. After record losses in 1979 Heinz Dürr was appointed CEO on February 1st, 1980. For the first two weeks in January a large sales exhibition had been planned to take place in Beijing. Telefunken as part of AEG-Telefunken had the honour to present consumer electronics and especially PAL colour TVs. Also a technical colloquium had been planned and a paper on HIGH COM had been announced.
The preparations were extensive, and shortly before it was to start properly, the available funds were slashed. For Telefunken there remained the exhibition of TV sets and the paper on HIGH COM. The booth staff was slashed to a minimum, so that only one person was left: me.
We were about to start on December 28th, with a Chinese 747 from Paris to Beijing. Paris airport had heavy fog, so the incoming flight was redirected to Frankfurt. In the night before I received a telephone call and was just barely able to catch a flight to Frankfurt and then the 747 to Beijing. Then we had a stopover in Qatar, one of the emirates. It looked as if Doha airport had only been opened in the middle of the night for just this single flight. Today this certainly is a completely different airport
After arrival in Beijing we were transferred to the friendship hotel, a building from the time of the great Russian-Chinese partnership that just had passed by. The building for our sales exhibition was built in a similar style. Of course, there was a Chinese restaurant but also an international restaurant. The food was the same, the difference was between chop sticks and cutlery. After all, both places were well heated.
The exhibition schedule was exhausting and the hall was barely heated, but the general mood was good. I was assigned a team of 4 Chinese people, who spoke a little German and English and were very proud of that. The audience was thrilled by the consumer devices and they literally ripped the Telefunken catalogues out of my hands. Finally I had to strictly ration the few remaining ones.
Fig. 17: All AEG-Telefunken exhibition staff in front of the Beijing exhibition centre
Notable for me was a typewriter shown by AEG-Olympia, which with its ink-spraying technique could produce Chinese characters on paper. The operator only had to memorize a special coding of the typewriter keys when he wanted to produce Chinese characters.
Here I had my first encounter with a PC. As far as I can remember, this was a portable TRS-80, which one of the teams of AEG needed for their presentation. Of course, they were very proud of it and also demonstrated that one could use it to play games.
My lecture on companders, especially on HIGH COM, was well received. For about one hour I had to answer very precise questions. Apparently they had prepared themselves very carefully. I still remember the dress code of the professors and academics: Blue, unironed overalls and glasses with broken glass. The party officials were also easy to recognize, their clothes were ironed.
In the March 1981 issue of the Chinese magazine "Audio Engineering" there appeared an article on HIGH COM, which was based on my presentation [28].
Fig. 18: The author and team at the Telefunken booth
Fig. 19: First page of article on HIGH COM
Of course there was also a social program. We visited the relevant sites, the Forbidden City, Summer Palace, Ming Tombs, and of course we also went to the Great Wall at Badaling. It was still very cold, so we did not get out of our thick coats there, even for lunch in a restaurant.
Still the best memory for me is a walk across the frozen lake at the Summer Palace, in freezing cold and with a deep and yellowish red sun shining through the smog.
Fig. 20: The Great Wall at Badaling, January 1981
11. Even more companders
In early 1981 the problems at AEG-Telefunken and Telefunken in Hanover became more and more visible. I no longer had defined tasks and the basic research lab did not have a director any more. I started to look around for my own projects.
German consumer electronics company Grundig used an audio compander in its Video2000 video cassette recorder, built from discrete components. After some discussions the AEG-Telefunken semiconductor plant in Heilbronn was awarded the mandate to develop an integrated circuit, which also covered other functions. I got the contract for the development of the compander circuits. The project was completed successfully and my work at Telefunken was paid for by Grundig.
Then, in autumn of 1981, CBS in the U.S. introduced a compander for vinyl records, the CX-system [29]. From January 1982 on I developed an integrated circuit for the CX compander in collaboration with the designers in the AEG-Telefunken, semiconductor factory. From 5th to 9th April 1982 I went to CBS in Connecticut to get the finished circuit (U2141B) tested there. I was well received at the hotel, but was then held there for a day because of a severe snow storm. All the traffic and all public activities had stopped. The following day then I met Dan Graveraux, head of the CX development team at CBS. In a pleasant atmosphere we tested the circuit and I won the coveted OK. This time my work was paid for by the AEG-Telefunken semiconductor factory in Heilbronn.
In late 1981 the IRT in Munich finished their investigation of HIGH COM FM, the modified version for FM radio [30]. The result was cautiously positive, they recommended to conduct some field trials.
12. The end
In mid 1982 the AEG-Telefunken board was looking for a settlement with creditors before declaring bankruptcy.
It became apparent that the time of analogue signal recording and thus also the time of companders had passed. In March 1983 the digital compact disc developed by Philips and Sony was introduced into the market.
Effective April 1st, 1983 Thomson in Paris took over 75% of Telefunken from AEG-Telefunken holding. The headquarters of Deutsche Thomson-Brandt GmbH (DTB) in Villingen were located on the premises of the previously acquired former HiFi manufacturer SABA. This seemed to close the fate for the development department in Hannover. With vigor and against strong resistance a move to Villingen was initiated. But Erich Geiger, the charismatic leader of DTB, recognized in time the potential that the company might loose. He changed his mind and founded the DTB laboratory for digital systems in Hannover on the old premises of Telefunken.
Here some important and lucrative developments on video and audio codecs, close relatives of the compander, would emerge. But that's another story.
13. Literature
[1] Scholz, Werner: Schmalbandige Farbfernsehübertragung mit einfachen Mitteln, Funkschau 1970, Nr. 4, pp 109 - 111
[2]
Bruch, Walter: Von der Tonwalze zur Bildplatte, Funkschau, 1977 Nr. 24 until
1979 Nr. 10
[3] Röder, Rolf; Sennheiser: DE2161905, application date 14.12.1971
[4] Wermuth, Jürgen: DE2403756, application date 26.01.1974
[5] Wermuth, Jürgen: DE2531475, application date 15.07.1975
[6] Wermuth, Jürgen: DE2406258, application date 09.02.1974
[7] Wermuth, Jürgen: DE2403799, application date 28.01.1974
[8] Wermuth, Jürgen: Dynamik-Erweiterung durch neuartigen Studio-Kompander, 10. Tonmeistertagung Köln, 19.-22.11.1975
[9] Wermuth, J.; Temmer, St. F.: Dynamikerweiterung durch neuartigen Studio-Kompander, Funkschau Nr. 18 (1975) pp 571 ff
[10] Dolby, Ray M.: US3665345, priority GB 21.07.1969
[11] Wermuth, J.; Dickopp, G.; Schröder, E.: DE2529031, application date 28.06.1975
[12] Jahnel, Benno: DE2661056, application date 02.04.1976
[13] Schröder, E.; Wermuth, J.: Ein neues Kompandersystem – Grundlagen und Einsatzmöglichkeiten, paper presented on FKTG meeting in Freiburg, on 05.10.1976, published in: FKT 30 (1976) Nr. 12, pp 9 – 11
[14] Höppner, Dietrich; Hintzmann, Kurt D.; Schröder, Ernst F.: Monolithisch Integrierte NF-Kompander, Wiss.-Ber. AEG-TELEFUNKEN 52 (1979) 1-2, pp 97 – 104
[15] Mathes, Robert; Western Electric: US1757729, application date 13.03.1925
[16] C.R.: Kompander verbessert Magnettonkopie, radio mentor Nr. 4 1965 pp 301 - 303
[17] Burwen, Richard S.: Design of a Noise Eliminator System, Journal of The Audio Engineering Society Vol 19 (1971) December pp 906 - 911
[18] Blackmer, D. A.: A Wide Dynamic Range Noise Reduction System, db, the Sound Eng. Magazine, vol 6, pp 54-56, Aug/Sept 1972
[19] Doba, Stephen; Bell Labs: US2173472, application date 22.06.1937
[20] Dolby, Ray Milton: An Audio Noise Reduction System, Journal of The Audio Engineering Society 1967 pp 383 ff
[21] Schröder, E.; Wermuth J.: DE2830784 und US4321482, application date 13.07.1978
[22] Dickopp, Gerhard; Schröder, Ernst: Der Telefunken-Kompander, Rundfunktechnische Mitteilungen Vol 22 (1978) Nr. 2, pp 63 - 74
[23]
Dickopp, Gerhard; Schröder, Ernst: Meßverfahren für Kompander,
Funkschau 1978, Nr. 17, pp 29 - 32
[24] Scholz, Werner: DE2856045, application date 23.12.1978
[25] Schröder, Ernst: DE2919280, application date 12.05.1979
[26] Mielke, E.-J.: Einfluß des Dolby-B-Verfahrens auf die Übertragungsqualität im UKW-Hörrundfunk, Rundfunktechnische Mitteilungen Vol 21 (1977) pp 222 - 228
[27] U401BR, AEG-Telefunken Semiconductor Information 2.80
[28] HIGH COM, Audio Engineering 3 (1981) pp 30 - 35
[29] Handbook CX Low Cost Expander Model E-1016, CBS Technology Center, Stamford, CT, revised August 2, 1981
[30] IRT Technical Report 55/81, Prüfung eines modifizierten HIGHCOM-Kompanders für den Einsatz bei der RF-Übertragung im UKW-Hörfunk, 30.12.1981