Saturday, March 12, 2011

Some Early-Modern RF Alternator History

The above postcard bears a photograph of the Grossfunkstation at Nauen, Germany as it appeared in 1941. A combination reflecting pool/transmitter cooling pond is located in front of the main radio station. The building was designed by a prominant achitect of the day; Hermann Muthesis. German industrial architecture of this period is still celebrated today.  

By the mid 1920's the RF alternator had emerged as the most efficient means of generating high-power VLF energy. By then even the last hold outs had replaced their arc transmitters with alternators. 

Telefunken installed 400kW magnetic frequency-multiplied RF alternators at Kootwijk Radio and at Torra Nova near Rome. They installed a 150kW alternator at Prado del Rey (Madrid) and a 100kW station at Funabashi in Japan. Telefunken stations were also erected Malabar and Monte Grande. The last great alternator of this type was the 500kW station built at Nagoya, Japan.   

Both the installation and the operating costs for a 200 to 400kW RF alternator represented a huge expenditure. Despite the ever-increasing demand for commercial international wireless services, as the shortwaves gradually became available the primary workload for the VLF alternators shifted from commercial to military operations. In his, History of International Broadcasting, Vol. 1, James Wood remarked

"Paradoxically, the cable telegraphy, whose demise had seemed so clear two decades earlier in 1906, outlived long-wave wireless telegraphy. This was attributable to two factos: the huge costs associated with high power, long wave installations, and the increase in commerical traffic from 1920 onwards, which gave a larger market." 

Indeed, the operational cost ratio for several kilowatts at shortwave using vacuum tubes as compared to several hundreds of kilowatts needed at VLF would have spelled the demise of the big alternators by 1930, had it not been for two saving graces.

Firstly, the transmissions from high-powered VLF stations suffer less disturbance due to changes in propagation than those at shorter wavelengths. 

Secondly, it had been discovered before the First World War that submarines submerged to a depth of 2 to 3 meters could continue to receive the signals from these VLF "powerhouse" stations. 

Following the First World War, the station at Nauen, Germany was run by Transradio A.G. (Radio Trans-Ocean Ltd); with Telefunken as the majority shareholder. Despite the worsening economic situation in Germany, the station was steadily upgraded. A model of the station as it appeared in 1920 may be seen here and here.

In 1931 this charming photograph was taken from the heights of the transmitter towers at Nauen (not a safety belt in sight!). In January of that year the Reichspost took over control of the station. The VLF RF alternators were refurbished in 1937. 

In a series of articles titled, Funkgeshichte, Dr. H. Richter described the VLF transmitters that were used for U-Boat communications in the early years of the Second World War 

"Eine große Bedeutung für die fernführung der deutschen U- Boote hatten auch die Längstwellen, die von den Booten mit ihren Peilrahmen- Antennen auch noch unter der Wasseroberfläche empfangen werden konnten. Am Anfang des Krieges stand dafür nur der Sender Nauen auf der Hauptbetriebswelle für U- Boote (Wellenlänge 18130 m = 16,55 kHz) mit einer Leistung von 300 kW zur Verfügung. Nach der Besetzung Frankreichs kamen noch drei weitere Längstwellensender dazu, St. Assise mit 300 kW, Croix-d´Hins bei Bordeaux mit 360 kW und Kootwijk (Niederlande) mit 120 kW. Der Sender Nauen und die beiden französischen Sender waren Maschinenfrequenzsender, der holländische ein Röhrensender." 

Up till the occupation of France, the RF alternators at Nauen were the main VLF transmitters available to the Kreigsmarine. Later, two French RF alternators as well as the Kootwijk Radio transmitter at Apeldoorn in the Netherlands were pressed into service. Herr Richter states that the later was a vacuum tube transmitter. However, I wonder about this given the Dutch 400kW VLF RF alternator was said to have remained in serviceable condition until the Germans themselves destroyed it upon their retreat.

In his book, Funkführung der U-Boote in der Praxis, Arthur O. Bauer describes the workings of the passive frequency-multiplied Telefunken RF alternators that were used to issue orders to the U-boat fleet

The RF alternators at Nauen are said to have been dismantled by the Soviets at the war's end. Whether they were ever placed on the air again remains a mystery. Fortunately, the Soviets decided not to follow through with their original plan to blow up the grand, old Muthesius building. The structure is used to this day as a radio transmitter.

The previously mentioned Telefunken alternator at Yosami, near Nagoya, Japan was saved from the scrap heap early-on for the same reasons that the German alternators remained in service throughout the Second World War. Originally installed for the purpose of commercial oversea wireless signaling, this largest of all RF alternator stations was nearly obsolete by the time it went on the air in 1929.

The station was eventually taken over by the Japanese Navy and used primarily for communicating with their submarines until the end of the war. I believe the callsign of this station was JNI3. After the war the US Navy took control of the station and changed the callsign to NDT. The US Navy used this station for its own submarine communications from 1950 until 1993.

Happily, after 60 years of on-air operation one of the two alternators - along with its ancillary equipment - has been moved to a permanent museum. In May of 2009, the IEEE commemorated the station with a dedication ceremony.

JA2DJN has posted some wonderful photos of the well-conserved Yosami transmitter here. Be sure not to miss the complete schematic diagram of the station which may be found here. The photo of the RF output feed-through insulator at the rear of the original building reminds me a giant spark-plug!

Another fine photo collection may be found here. I especially like the wrist-diameter Litz wire inductors set into those gorgeous, teak-wood frames.

The Telefunken E-378S "All-Wave" receiver provides a good example of the type of equipment that was used to receive VLF transmissions in this time period. This regenerative receiver was nicknamed "Brotkiste," or "breadbox," for obvious reasons. These receivers were standard issue on Kreigsmarine ships and U-boats up to the end of WW2. For example, the E-378S receiver can be located (#10) above the operating table on the starboard side of this Type-XXI, U-boat radio compartment. The above two links are taken from the web sites that may be found here and here

The E-378-S also had a place of prominance in the radio room of the Graf Hindenburg. Please click here and here.    


  1. Very interesting Mike. I have always been fascinated by VLF and ever since "discovering" some details about the SAQ 17.2 Khz transmitter I have spent much time following links and references such as the ones you have provided to similar such installations from that time period.

    I have made several attempts to copy the SAQ transmissions but up til now have only been partially successful. The last transmission in January I was finally able to "see" a very faint trace with the occasional identifiable CW characters on the waterfall of Spectrum Lab but otherwise could not hear a thing. Some more changes are in order for my VLF receiving setup so that perhaps next time I will be able to "hear" one of their transmissions. I received a very nice QSL card for my efforts. The operators at Grimeton take great pride in keeping this old transmitter on the air and in their occasional transmissions.

    Your web posting always make for interesting reading, thanks!

    cheers, Graham ve3gtc

  2. Thank you Graham,

    Yes, Mr. Alexanderson's alternator at Grimeton is an engineering marvel as well as a thing of beauty. As with the others, this machine would have gone to the knacker quite a long time ago had it not been found useful for submarine communications. We're in debt to the good people in Sweden who have worked to preserve this important historical artifact.

    Very FB on your efforts to receive SAQ over here, Graham. If I have my way, you'll eventually be able to work an RF alternator on the amateur radio bands. I can hardly wait!

    Mike, AA1TJ

  3. Great stuff Mike -a wonder-full glimpse of our early history and a great perspective when we compare your modern QRP versions of that technology thriving over great distances today!

  4. WOW, fascinating stuff. I had no idea that such efficient multipliers could be built with saturating reactors. Your work got me thinking about the problems of building an alternator to drive your proposed 9x multiplier to 100 KHz. One consideration is the accuracy that could be achieved in the placement and size of the holes or slots in the rotor. Any variations would show up as phase noise in the carrier signal and could make for a rough note unless carefully controlled. Keep up the fine work.

    Paul W2IOG

  5. This comment has been removed by the author.

  6. Hi Paul,

    Thanks for your perceptive comment. Indeed, there is a historical precedent for your concern. A frequency-multiplied RF alternator was employed as a broadcast band AM transmitter in Munich back in 1926. Complaints about the rough sounding carrier started immediately. An investigation by the manufacturer, C. Lorenz A.G., turned up a number of possible culprits: a mechanical resonance in the alternator housing, eccentricity in the bearings, and inaccuracy in the alternator pole placement. The transmitter was replaced by a vacuum tube model before a final determination of the cause(s) had been made.

    I take some comfort knowing that manufacturers of modern-day stepper motors have an incentive to strive for angular regularity in the poles. But given that tolerances are finite, angular irregularity is to be expected; something frequency multiplication will only accentuate.

    Once I have my stepper-motor (working as an alternator)running as part of a PLL it will be interesting to look at the 11kHz output signal with a spectrum analyzer.

    Part of the excitement that comes from playing with this sort of thing is knowing that failure is a real possibility. The practical limit using an alternator may only be 500kHz. Whatever it is I hope to find out.

    BTW, my initial worry was that core losses would have rendered the frequency multipliers impractical already by 3.5MHz. This concern surely has something to do with my excitement over seeing them now operate at 14MHz with an efficiency that I'd not dared dream of.

    You know, I talk about the necessity of looming failure, but in another sense you can't really fail when you're doing something that you love. I mean, even if the frequency multipliers had fizzled out by 500kHz, I'd still be left with all this great history. And while I hope to push this project as far as possible, I'm nearly ecstatic to have come this far. It certainly has been an interesting ride.

    I appreciate your very kind thoughts, Paul. Thanks so much for your message. I'll see you on the bands, OM.

    Mike, AA1TJ

  7. Wow again. I never thought of a stepper motor. My idea was to make an aluminum disk with an array of small holes near the perimeter and a short section of steel rod pressed into each hole, flush with the surface of the disk. A permanent magnet could be used for bias with the output coil wound on an armature between the magnet and the disk. It would have been a great reason to buy that precision rotary table for my vertical mill that I have wanted for a while now. The stepper trumps that, though.

  8. Aha...a radio amateur/machinist! Very FB Paul. My little 7" lathe and 5" Perfecto metal shaper are currently resting beneath my radio workbench.

    I have several radio projects that I'm dying to make a start on once I have my machine shop setup again.

    Mike, AA1TJ

  9. Yes, I love projects that combine electrical and mechanical design. I have a 10" South Bend lathe and a small Elgin VM5 vertical mill
    both of which I bought at a farm auction many years ago. They were built in the late 40's but still give good service.

    Paul W2IOG