(Fritz Trenkle)

Wassermann Survey

FuG 402*

* We do neglect the suffix indications

Page intitated on 19 September 2015

Status: 13 May 2016

2 + 2a + 3 + 4 + 5 + 6 + 7 + 8ab! + 8c1 + 9! + X10X

(Rev)                 +  Y10Y  + Y10YB + Y10YC + Y10YD

+ Y11A + Y12

On 20 November 2015, Henk Peek visited me in the museum and I did present my PowerPoint Mammut-Wassermann planned for the Open Museum Day (21 November); we discussed some details which forces me to revise- and add some consideration on the existing Wassermann webpage:  Revision



My first approach was a previous website called: Mammut Wassermann query, where we started with putting a call for support on information on both Mammut and Wassermann. However, we have got used to - that hardly information is to be offered. Only a small group, but very dedicated men, were later providing the information-backbone, peu à peu.  

The Mammut Survey had been most successful so far, and have brought more to daylight than was expected and can be found elsewhere.


I do not pretend that this webpage will be as successful as has been the webpage on Mammut; but we might bring quite some unknown information. Most information we possess will be part of this survey, so that we can share it with you. Consequently, the succession of building up this page will be the result of what-is-when at hand; of course, also being influenced by what route my thoughts might take, sometimes.

During the creation of this web page, I encountered the problem that explaining its technology it often necessitate considering several details at the same time. I therefore have chosen to repeat information sometimes. For some it might be a nuisance, but from the educational point of view it can be very helpful understanding what an aspect is about. And just that is why we are here! 


I have to rely partially on materials used in the forgoing web pages; but also new British reports from wartime days. The foregoing Mammut survey was in the course of its unfolding survey being baptised: Eine Entdeckungsreise. A nineteen century discoverer did often not know what the next day would bring him, so did I - so many bits and pieces where on nearly daily bases arriving, that often virtually chapters should have been changed and/or reshuffled. This was ultimately impossible, and I decided to report upon what is new in succession of findings and/or arrivals.

Both Mammut, as well as Wassermann, might have constituted the biggest (complex) antenna arrays world wide. Neglecting the HF/DF Wullenweber system array.


Wassermann a GAF system

Which got the Allied codename Chimney


It is not our aim to provide details on the various Wassermann radar versions, like the M I ..IV and S ... series. These are technical details mostly translated in the dimensions of an antenna construction. Systematically, all relied upon GEMA (and Siemens) technology; as did, by the way, most German radar systems. With the exception of Telefunken and Lorenz products.

Siemens became engaged first, to what I know, on GEMA's demand. They did lack the engineering and scientific background from a major German firm. Siemens, then and now, being an industrial giant. They made it work! Also the once by GEMA antimated Jagdschloss radar concept necessitated Siemens & Halske's technical support to make it finally work.



What became necessary during the course of the war was to change over from narrow band- towards broadband operations. This was being caused by radar channel jamming and later followed by broad band noise sources (neglecting the Moonshine repeaters). When you cannot cope with these kind of signals, frequency change is a good option; as long as the new spectrum stays free of interference. Jamming became a kind of 'art'. The Germans anticipated, but were also keeping signals in the air which did not have a purpose than forcing the enemy to keep a great variety, of some times obsolete gear, operational (saturating his efforts). For those interested in this aspect, please notice Radar News 19, which is an Allied translation of a GAF radar periodical issued on 25 February 1945. Most interesting stuff, by the way. 

In the beginning of the many system conceptions, it was all narrow band application. The advantage, separation and coupling could easily be accomplished by means of so-called ¼ λ stubs. This is a ¼ wave length piece of wire or other conductive means. Its advantage, without bringing the proof, is, that impedance transformation occurs when one side of such 'line' is kept, for example, at a 'low impedance level - the opposite side is then constituting a high impedance v.v. Also the provision converting feeding lines from a symmetric into a non-symmetric behaviour v.v. I sometimes prefer to call these devices a 'balun' (balance-unbalance). But the ¼ λ technique baluns were (mainly) known as: Sperrtopfe*; normally converting from symmetric into asymmetric but the impedance ratio being 1 : 1; what a balun not necessarily is providing.    

* Although, in the Wassermann manual used for it: Symmetriertopf

A Sperrtopf concept, according a wartime conference paper

Derived from what is already on our website: ZWB- Berichte, please look for Roder's contribution

By the way, these kind of 'baluns' were maintained in the Mammut system too. Impedance ratio 1 : 1 and being equal to the coaxial cable impedance; mostly in the order of, say, 60 Ω to 70 Ω . Phased antenna arrays constitute often rather comprehensive radiator (dipole) systems, where correct signal phasing is essential. Many baluns do not always provide information: what line-contact is fit onto what cable- or line phase. The above shown Sperrtopf does provide exact line phase.

The consequence, when frequency changes - also the performance of this kind of balun or Sperrtopf will deteriorate. The consequence - when (narrow band) phased arrays being involved, is that beam-steering can no longer be controlled accurately.  


However, this HF technology was widely adopted, but, the disadvantage comes in when the applied frequency does not obey to ¼ or ½ λ - caused by change of operational frequency. For it new antenna design was necessary. But systems fit with the old technology could not be simply converted into a wide band system. The entire antenna concept needs to be changed.


I mistrust the information found in books on the many spectra at which Mammut should have been once operated. From a British A.D.I. report of late 1944, we are being told that for Mammut the operation frequency should be maintained within 100 kHz, at an operational frequency of 125 MHz. In those days, with the application of self-oscillating power transmitter stages, quite a task (performance).


The front cover of A.D.I. (Science) Report No. 1

(Crown copyright, courtesy Phil Judkins)

This report is not containing gold weighted information, but also did fall into a technical trap, due to the editor - who apparently was not knowing all the many technical facts, he might not even have understood it fully. He might have written this report without having had knowledge how the so-called Kompensator (compensator) apparatus did really look like. We certainly do, in contrast!

However, we will learn - that the latter might have relied on what he knew about the Wassermann compensator technology instead.


Preventing to keep you much longer away from intriguing matters - I would like to explain why and how my contribution being accomplished.

As in most of my technical contributions, technology is the core of what I would like to explain; you will become part of my reflections and/or misperceptions; inevitable in surveys like this one.

Explaining technology needs some technical explication, and we have to confront ourselves with - for some of you - new fields of technique and its implications.

My main aim is to let our website grow into a huge source of information. Information necessitate also education in the field in which someone is not acquainted. World War Two (WW II) radar technology must be regarded being at a 'stone age' level, compared to the state of the art today; where high speed computing onboard flying platforms is daily practice. Some high performing systems can cope with several hundred separated threats at the same time.

Therefore, there is no way around than to confront yourself also with technology in a broader sense.

One can only grasp what something is about, when you understand it (at least a bit). Our generation might be the last one who are standing with one foot on the wartime remains and also with the other one on what is common knowledge nowadays. The next generation have to rely on what have been saved from the past - accompanied with some kind of explication. 

After a night sleep I came to the conclusion that - for better understanding of what the various Wassermann derivates is about, the we should have, at least, a brief look at it.



This information was found in the periodical: Frequenz 1955/10

by H.J. Zetzmann

Zetzmann apparently was engaged at Siemens & Halske already during wartime days. Thus we are relying on someone who knows from his own experience where he is talking about.

Genuine text: Die verschiedenen Antennen von Großraum-Suchanlagen des Typs "Wassermann" im Vergleich zur Antenne des "Freya"-Gerätes

The different Wassermann types versus Freya's apparatus dimension of 1937/39

It is interesting to know, that in fact the Freya and Seetakt radar concept was maintained during the entire war. In other words - Wassermann, as well as Mammut constituted a big version of a Freya system. Where the main difference is to be find in the complexity of the antenna array and auxiliaries.

Another photo incorporated in the same paper


The left-hand Wassermann type M I used vertical antenna polarisation

The right-hand photo shows type M IV fit with horizontal polarisation

Why the difference?

Please look carefully at the right-hand photo, the top section shows vertical dipoles, used for IFF (I.F.F. or Freund-Feind-Kennung). This provision is lacking at the left-hand side photo. Thus polarisation change had been accomplished, because the IFF relied upon a system technique (FuG 25a) and was adopted for most aircraft; operating a vertical antenna rod underneath the fuselage.

To most readers not yet known - there are two different Wassermann related antenna techniques, those being fit without height-finding and the one with height finding facility.

Let us therefore repeat the photo shown at the start of this new webpage:


From left to the right: type M I, in the centre type M IV and on the right-hand side the version we should focus our attention at

(Frequenz 1955/10)

When you ever encounter a picture of a Wassermann antenna please look first whether it has a cabin about half way the antenna mast.

This 'cabin' was used to cover the Siemens type Wellenschieber or Kompensator (compensator) of a quite unique design. Its purpose, was to steer virtually the antenna beam (between 0 and +20 degrees). Hence, elevation information is being provided by means of a compensator; azimuth was for all Wassermann derivates provided by antenna rotation.

Inside the cabin was placed a compensator apparatus, also known as Wellenschieber; or 'wave-mover'. Not a bad choice. The compensator apparatus was remote controlled by the system operators; which should be considered later.


We have in our Mammut Survey already noticed what a compensator is about.

Briefly, in my wording:

In case of no beam elevation - the upper and lower antenna array sections being fed with equal signal phase. When such a compensator device being 'steered' into a certain direction, the signal phase will arrive 'earlier' at one antenna group and the opposite array being fed retarded accordingly. This is what an 'antenna compensator' is doing. An example of  the consequence is shown below.


In this case, the lower antenna array field being supplied earlier and the upper section fed with an equal phase shift but retarded. Resulting in a positive tilt angle of the value "

(Frequenz, Bnd  9 / 1955  Nr. 10 pp 351-354)


This is the raw schematic as to how the Wassermann type M IV antenna being interconnected

(Fritz Trenkle and Frequenz 1955/10)

Please notice first the vertical equilibrium constituted by drawing a line between the points A - B

Here is where the compensator governor* is grasping.

* Will be dealt with in due course

Consider also the long stretched vertical line, this is the equilibrium left and right; applied for accurate positioning in the azimuth system plane.

Split switch relay g is a device that continuously alternates between the left- and right-hand antenna side. As to steer the horizontally (azimuth) a bit a phasing loop e being implemented (Umwegleitung = phasing loop). The result is that the beam being steered alternating a bit to the left and right.  Such technique was known in German 'A/N Verfahren'. This latter expression originates from other (historical) navigational beam steering, where in Morse symbol A = .- and N = -. when both are just in the correct phase to one another A and N forming a constant signal (output). Such technique was also used for the Lorenz Blind-Landing signals. Azimuth beam centre was when the left- and right-hand signal sides being equal; beam split being 2.5 degrees to either side. The disadvantage of such technique - is that you loose signal strength. Because when both antennae sides being interconnected in a correct signal phase, both antenna signals adding to one another, which is resulting in a higher signal level. But, the latter example making accurate azimuth steering less precise.

What just popped up in my mind, is that this A/N technique was active for both transmission as well as reception.

We luckily possess such a split beam relay device   



The connectors left and right are connected onto both compensator arms or moving sliders, albeit through a matching circuit. The towards us facing coaxial connector (symbolised h in the foregoing drawing)  being fed onto the T/R stage

The U-shaped coaxial system (Phasing loop = Umwegleitung) on the top constitutes a delay-line, symbolised by e in the previous drawing.


For more detailed information on this device Exhibits-details new


A T/R stage in German terminology called: Simultangerät

In such a circuit the transmitter and receiver being fed onto the same antenna system. When transmission is on, the receiver being decoupled and when transmission stops the receiver being connected electrically onto the antenna system. A T/R stage should provide optimal signal transfer- and system decoupling, combined with minimal loss.



On  23 September 2015


We have encountered the word Wellenschieber or compensator, but how does it look like?


The Siemens Wellenschieber or Kompensator (W.Sch. 402)

(Frequenz, Bnd  9 / 1955  Nr. 10 pp 351-354)

Please look carefully at this photo; with some imagination you can recognise that in the rear of this frame we encounter an identical system.

Please, notice also the small circular disks left of the coax cables and connectors. These were, according AIR 20/1702, glass covered windows for monitoring the rotation of a helix.


Hence, this module constitutes a dual compensator system for the right- and left-hand antenna side.

The box on top, contains the remote controlled driving system and Selsyn repeaters (Drehfeldgeber)

Regarding the photo: the left-hand side coaxial cables being connected onto lower antenna array section and, consequently, the right-hand side being connected onto the upper array section.

The cables fit somewhere at the centre are the compensator tapping connections. Which being fed onto a combiner circuit first.


I guess, that their cables lengths do obey to n x ½ λ. In such case the mismatch, which parallel wired coaxial cable impedances would constitute is not causing serious signal reflection.  Henk did correctly suggest that more appropriate would have been implementing ¼ λ  lines, thus  3, 5, 7 ... times. Even times would implement ½ λ lines, which we just have considered being inadequate.

Similar technique was also accomplished within the Mammut system. The implications of interconnecting coaxial cables straight away together will be dealt with later.

Quoting from AIR 20/1702:

The tapping leads, which are all of equal electrical length, consist of flexible cables and terminate in two plugs for connecting to the helix tapping point and to the distributor. Their impedance is matched to half the radiation resistance of each dipole group, and is 35Ω.

Please remember where the left- and right-hand cable connection being connected onto:


There hardly is a way around that some drawings have to be repeated, especially when we deal with complicated systems. I trust - that both this line drawing and the foregoing photo will gain better understanding

(Frequenz 1955/10)

Just popping up: we will later notice - that the stacked antenna groups where fed with different ratio changes. Please notice the lines going out from points A and B in this drawing. Their tapping points do correlate quite well to the time delay necessary to match onto the building up of the antenna beam pattern.

Please be aware, expressing it differently - that the Wellenschieber or call it compensator, did have the same function as does have different lengths of coaxial antenna cables; both the Wellenschieber as well as the application of different antenna cable lengths will resulting in a vertical tilt of the antenna radiation pattern (regarding that both left and right-hand side will feed each section with equal cable length too).

Quoting from AIR 20/1702 based on a translation of a German document:

The purpose of the phase shifter is, therefore, to produce the differences in length in the feeders to the array necessary for electrical elevation shift.


Until now we have heavily relied on Zetzmann's paper, but we will go more in detail and therefore considering a wartime British report on the Wassermann Wellenschieber apparatus


AIR 20/1702   

Date: January 1945

(Crown copyright with courtesy Phil Judkins)

Air Scientific Intelligence Technical Translation No. 7 Wassermann M II and M IV broadband phase-shifter


Some information can be found at several places, but a line drawing explaining how the Wellenschieber was being built up?

I believe that we first should look at the Siemens designed 'core' of the Wellenschieber


A photo of the Siemens Wellenschieber innovation

(Frequenz, Bnd  9 / 1955  Nr. 10 pp 351-354)

Not well visible, the ribbons are actually the groves of a helical line, it constitutes a long-stretched solenoid forming a delay-line.

The tooth-wheel is connected onto the: Abgriffleitung or tapping lead. The tapping leads are connected onto the (black?) coaxial cables in the centre of the previous Wellenschieber photo.

Let us not forget, that finally there existed Wassermann systems operating with up to 100 kW transmission pulses. Consequently, the electrical reliability must have demanded great care. According AIR 20/1702: the internal mechanic of the Wellenschieber was screened off from environment (dust and moisture).  

Siemens was very proud of this construction, which they claimed could be operated over 4 octaves (= frequency ratio 1 : 16). A precondition for broadband operation (covering a wide frequency spectrum). This meant actually, that it could be employed for all (future) Wassermann derivates; there existed no radar system able to cope with the same antenna array with a frequency ratio of 1 : 16.

I might become clear what the central helix below is about.

For better understanding, I would like to reproduce a line drawing which I once have used for the Mammut Wassermann web page.


Please notice, that the sliding arm is in this case the tapping-wheel in the foregoing photo, and the transport movement being caused by rotating the helical body (or, was the cog wheel driven?)


The British file AIR 20/1702 quotes: The requisite length of the "Shifter Lines"(Schwenkleitungen, AOB) depends on the height of the aerial frame and the amount of elevation required. With radar sets FuMG 402D and E for example, an angle of elevation of 20° requires a shifter line of 8 m long. Only half of this line is used for shifting the phase as only upwards shift is carried out. In order to keep the dimensions of the phase shifter as small as possible, the part of the line used for phase shifting is made in the form of a helical line, and the part not used formed by an auxiliary length of the aerial feeder.

Maybe in this context conclusive, quoting from the same source: When shifting the elevation electrically, the lengths of these perpendiculars can be considered as the distances which must be covered by the transmitted radiation from the separate aerial groups in order to achieve the required change of direction in the radiation wavefront.

Omitting mathematical considerations, in my brief wording: This latter paragraph explains, that beam forming is the result of all the antennae elements involved. The final radiation pattern is the accumulation of all electric field components present at all points concerned. Let us consider that the antenna field originating from the most outside dipole (radiator) should add to the central generated EM field - their mutual EM field-phase is crucial. When a signal arrives in an opposite phase (say, - instead of +) than the local existing field is undergoing subtraction. We have to deal here with two components, the time it does take for an EM wave radiated by, for example a dipole,  to arrive at a certain point in free space - and with the signal phase of any other dipole element being supplied by means of their antenna cables. The ultimate antenna pattern is than the sum of all the EM fields involved. It is evident, that great care have to be taken when a narrow beam pattern of a high gain is to be formed (generated).

Final beam forming takes place some wave lengths in front of an antenna array. We encounter first the so-called 'near field', farther away from the aerial system will be finally creating a 'far field' radiation pattern.

When a group of radiators (like dipoles) being grouped together as to create a particular radiation pattern, this will always result in a main beam as well as null-field sectors (gaps); sectors where the field summation resulted in a signal fading out. On the other hand, positive summation always resulting in unwanted side lobes too. However, careful antenna design can have some influence on the field gaps as well as side lobes. Though, total elimination is hardly possible.      

The crux is to control, with great care, the time delay caused by the parameters of coaxial cables and the building up of the EM field components in free space in front of an antenna system (array). 



An example of near- and broad-band antenna feeding, but what is interesting is an example of main antenna lobe as well as side-lobes; by the way, there can be many more of them! When a radar system would like to hide himself - side lobes have to be prevented as much as possible

(ZWB- Berichte, look for Zinke's contribution)

May be enlightening are the two sorts of antenna array feeding. The upper situation concerns the narrow band- the lower wideband operation.


After this interlude let us continue with the mechanical construction of the Wellenschieber. We will only consider a single subunit; because the Wellenschieber was build up from  2 x 6 equal sections.


Not well visible, is the way the helical body being rotated (motor driven)

(AIR 20/1702 with courtesy Phil Judkins)


Please bear in mind first the photo of the left-right antenna switch with the by-passing delay line (phasing loop = Umwegleitung) and let us follow Zetzmann's genuine German text first:

Um die Abmessung der Umwegleitung klein zu halten, wurde die Innenleitung (the helical cylinder) als Wendel (helical) ausgeführt, wodurch sich eine geometrische Verkürzung bis zu einem Betrag von max. 1 : 6.5 erzielen ließ.  ... die Wendel (helical conductor) der koaxialen Leitung mit Z = 35 ohm.

I guess, that he means - that the impedance at the (tapping) cog-wheel is 35 Ω. Otherwise the cable impedance would become 35 : 2 = 17.5 Ω. 

(Frequenz, Bnd  9 / 1955  Nr. 10 pp 351-354)


Quoting from AIR 20/1702:

The 12 coaxial slider-provision (Abgriff) these cables being 35 Ω.


For better understanding I would like to remember what the Unwegleitung was about

(Frequenz, Bnd  9 / 1955  Nr. 10 pp 351-354)


However, Zetzmann's main point is that the by-pass delay-line e (phase loop  = Umwegleitung) should have been (too) long; by means of the application of a helical line (constituting an inductance and due to its capacitance against the surrounding metal cylinder getting also a capacitive loading and allowing it to act as a (delay-line) the actual mechanical length of line e was reduced by a factor 1 : 6.5. Please consider also the schematic of the antenna array and focus particularly on the coaxial switch g.


Quoting from AIR 20/1702

Very significant is understanding the slider controlled phase shifter.

We always should bear in mind, that the Wellenschieber cables connected onto the 6 upper antennae groups and the six lower antenna sections. Neglecting the other side of the vertical array.

The generated wave fronts from each section should arrive in free space as well as supplied via coaxial cables at a determined moment. Antennae more to the outside does have longer coaxial cables as well as need for beam forming a longer transition time (in free space) than does need an aerial section just boarding the antenna centre. Keeping both in predicted good order, needs careful design.

Quoting:    The purpose of gearing is to turn each set of six helices of the two halves of the phase shifter with six different speeds with a ratio 1 : 3 : 5 : 7 : 9 : 11. A shaft carries six worms with pitches in the ratio  1 : 3 : 5 : 7 : 9 : 11. Worm-wheels are geared to the worms; to the shaft of each worm-wheel is fitted with a further gearwheel. This is geared to the cogs driving the helices of the the two helical tubes are fitted onto the frame with the deflection speed ratios increasing from bottom to top. The topmost pair of helical tubes (feeding the two most extreme groups of dipoles, AOB) are therefore driven at greatest speed.

Quite clear, because all governors should arrive at the same instant at the virtual antenna centre (equilibrium); no beam tilting.  


Please consider the two coaxial systems left and right of the helix section.


This drawing constitutes a HF transformer. It is found on either side of each of the twelve delay lines. This drawing differs from the one showing the helix provision

(AIR 20/1702 courtesy Phil Judkins)

Please compare both this drawing with the forgoing.

Its design is typically meant for impedance matching. In HF coaxial systems the impedance is determined by the ratio between D : d (considering air dielectric)

On the left-hand side connecting onto an aerial section; on the right-hand side the helix (distributor) being connected onto.

I guess, that its complexity might have been influenced by the fact that this radar system was providing broad band operation.

I would like to explain the Wellenschieber mechanical construction by means of an additional compilation



The by me modified drawing provided in AIR 20/1702 is not entirely conclusive - essentials for better understanding have been skipped

(Crown copyright courtesy Phil Judkins, modified AOB)

Let us first consider the helix which should function as a delay-line; but a delay-line does need a capacitance against ground.

The helix actually constitutes the coil (solenoid); its windings being loaded capacitive by the helix covering tube.

Also explained quite fuzzy, how does function the tapping of the helix driven cog wheel and its connection onto the coaxial tapping leads?


 Let us consider the Wellenschieber photo again

(Frequenz, Bnd  9 / 1955  Nr. 10 pp 351-354)


Do we see evidence that the helix tapping does move together with the Helix gear wheel (cog wheel)?

I must say no!

In that case there must have existed a slit along the tube (helix length); which we don't see.

A slit would (also) contravene with the provision of screening its mechanism off from environment.

Hence, the moving helix gear wheel (cog wheel) must have been interconnected on a different way.

How might they have solved this problem?

I presume, the helix gear wheel mounting is sliding along the, what I designate: Helix covering tube.

All within such central Wellenschieber tube?

I don't know, because the impedance might have become quite low (bear in mind the ratio of D : d). We know that the tapping provision should have had an impedance of 35Ω*. When the diameter of the helix was relative small, it might have been possible.

* This might indicate that the arriving two impedances (from left and right) having 70Ω ( 70 : 2 = 35). Wherefore were then the matching provisions?

Just popping up - the helix might have been driven by a shaft running parallel onto the helix covering tube. Might this also have functioned as an HF support?


Who can help us?  



On 23 September 2015

A few hours after part 2 was put on the web.

I did send an e-mail to Phil Judkins and Mike Dean in which I told them that a new Wassermann page being published, and got nearly instantly a reply with an attached photo. This photo was once taken in the SM depot at Wroughton in the early 1990s.

Mike advised me to apply for a better photo, but I prefer this one, as it is genuine and don't need copyright permission, with all the nuisance.


The question stays still open, how does the centre tube look like inside? The white box in front left is the cover to the motor drive

(Photo courtesy Mike Dean)

Left-hand text: Nicht betreten or do not stand on it

Right-hand text: Nicht an den Rohren heben or do not lift it on the cylinders (tubes)





On 23 - 25 September 2015


 The Wellenschieber had been remote controlled; for it a Selsyn (Drehfeldgeber) did repeat the position of the helices mounted in a cabin in the rear of the antenna frame, and was indicated to the operator. We hereafter will understand that several operators must have been engaged, because for a single person there are too many screens and controls to be monitored (watching). (please notice further down)


Controls 2 and 3 indicate that virtual beam tilting could be accomplished motor driven (wobbling elevation up and down on occasion)

(AIR 20/1702 courtesy Phil Judkins)

Working currently on a Post Mortem document, they speak about a meter presentation, but in my perception, the repeating selsyn or Drehfeldgeber should have feed a rotating scale instead, because a meter would become far too inaccurate.


In this stage of our Survey we cannot come further due to lacking technical information on the Wellenschieber device.


However, we still have some information on the radar equipment involved.

Generally speaking - Wassermann, like Mammut, did rely on GEMA technology; though Wassermann was often accompanied with a mixture of GEMA and Siemens gear.


Let us consider how the operator desk once did look like, according instruction papers


The operators working space; I presume that two - three or even four men were handling the various CRT screens and controls

(Courtesy Fritz Trenkle)

Let us in succession deal with the drawn modules

* Strange, we don't encounter the IFF receiver codename Gemse


In 2011 we received from someone two photos with a query - what is this display about?


My first impression was - it might concern a Mammut display unit

(Courtesy Colin?)

My second thought, correctly, that it might concern a wide range display.

In the Mammut Wassermann survey I dealt with it more extensively.

Significant are, the two transparent range scales in front of both CRT screens.

Please look closely, at the lower CRT screen. The figures being a bit blurred, but we can recognise that the figures consist of two digits. I therefore estimate that it did cover a range of 0 - 100 km.

The upper CRT scale does show three digit figures; I guess, covering 100 to 300 km. This was the maximum capture range possible* because the Z-Gerät generated 500 Hz signals. 500 Hz PRF constitutes a wave length of 600 km. But, within a radar system, the radar signal has to travel towards a target as well as returning to radar station again. Therefore range have to be divided by 2 (600 : 2 = 300 km).

*  This might not entirely be correct, as sometimes over sea the capture range could be even up to 380 km. For it you have to monitor weak unsteady signals at relatively short range; these can only originate from far off targets (as long as these stay quasi locked).

Monitoring a full range of 300 km from a CRT screen with a diameter of 10 cm, would have been very inefficient.

I therefore tend to the opinion that this module once belonged to the P-Gerät. The GEMA CRT frames (units) were all more or less of similar construction.

The just discussed P-Gerät display must have been mounted on the left-hand apparatus side. What module was kept on the right-hand side I don't know. Maybe, the two time bases; because the regular time base range was fit for covering 0 - 150 km which isn't according the two scale divisions likely.


Just last evening I received, like was during the course of the previous Mammut-Wassermann Survey, a bunch of documents from Mike Dean. He most generously provided copies of photos and documents he collected and traced in the US and elsewhere.

As consequence of the previous Mammut-Wassermann Survey, a lesson is learned - that we should keep the train rolling and implementing new materials in the succession of arrival.


A "Deux ex Machina" just in time Mike Dean provided the following photo


The operator desk to a smaller Wassermann system, I presume; because elevation steering is not provided

(111 SC 269075 - "US National Archives" courtesy Mike Dean)

Photo taken during Operation Post Mortem, shown the outfit of the Wassermann radar site at Vorupor* (DK)

* In Mike's photo series there might have slipped in some contradictions; I therefore believe, that some photo might have been taken elsewhere in Denmark. In wartime days, Denmark was the home of many German radar sites. Wassermann was not a rare system, it was the backbone of the GAF long range surveillance.


What can be learned from this, in my perception, unique photo?



   Let us reconsider the Mammut operators desk, qua instrumentation some similarity is encountered.


In some respect the similarity is striking

(111 SC 269017 - "US National Archives" courtesy Mike Dean)

Albeit, that the Wassermann operators work desk looks a bit less "shabby".

The medium and high voltage R-Gerät is standing in the Wassermann operator room on the left-hand side.

In both cases of some interest is that the official room drawing differs slightly from the practical system layout, where the Z-Gerät module being exchanged from left to the right-hand side of the O-Gerät. I guess, for operational reasons.

On 15 and 16 January 2016 I was engaged with copying an R.A.E. file RAD No. 204, dealing with the NE110 Einsatz. This module is the display unit of the so-called N-Gerät comprising of the receiver as well as display scopes.

Since I have been forced to approach this subject again, because I might have made incorrect assumptions. The question concerns a means of 'magnified' range presentation.

The question, without going here in to particular details, down the CRT presentation window on the far left-hand side, there should be a control knob Pos. 301, which cannot be accessed because the Beam-steering terminal is blocking its access.

For more details consider Y11A  

From H.J. Zetzmann's paper in Frequenz 1955/10 we know, that the earlier Wassermann systems have been fit with Freya like transmitters, providing the regular 8 kW transmission pulses.

It is, however, evident that increasing transmission power was necessary.

The early low power state of affairs:


Operating regular Freya TX outfit

Please notice the symmetric antenna connection on the top right-hand side

(source TME 11-219)


This situation was succeeded by a Siemens upgrade. On the left-hand side we notice the T/R switching unit - which was similar operated within the Jagdschloss radar system. Strange, this apparatus was designed to be tuned over some spectrum, but the transmitter does not show apparent tuning facilities. Definitely not yet the ultimate state of the art.

(IIISC269034 "US National Archives" courtesy Mike Dean)

You cannot imagine how delighted I am to encounter this particular photo.

Because it shows clearly - that Siemens techniques used in conjunction to Jagdschloss, was also adopted for the Wassermann system; to what extent stays open yet.

Also of interest, is the construction of the transmitter module on the right-hand side.

The black box above hand written marked 1235 - is also found on next the photo.

Please compare both - this photo with the previous photo and consider the next one: 


Jagdschloss equipment; the TX on the right-hand side is having band tuning, which is lacking in the previous photo. Please consider also the next photo:


Transmitter code-name Eibsee (to Jagdschloss)    Please look carefully at the top horizontal loop with its two sliding contacts. This version constituted the ultimate state of the art

Both consider a modified technique, where instead of coupling out the transmitter energy at the top side of the frame, the antenna energy being provided on the housing right-hand side.

Thus, this particular module provided quick frequency change, not yet provided at the forgoing IIISC269034 photo.

Though, this mediate stage of development operated already a pair of TS 41, whilst the regular Freya transmitter being fit this the low power TS 4 valves.

From the Jagdschloss A manual we know, that provisions were made for 100 kW operation.


Therefore special 30 kV provision being adopted

The high tension provision is self explaining.

For it, I would like to remember briefly, that 30 kV anode voltage was not provided by the R-power supply apparatus, but generated by means of a power modulator. Where the secondary windings of the output transformer providing 30 kV pulses. This is a far more efficient means of keying a radar TX PA than using the odd grid blocking method.

Isn't it interesting, to follow the process of coming into being of (high) power radar gear?



Another interesting photo, showing a low power transmitter* combined with on the left-hand side, in the grey box, the so-called: Simultangerät or, in modern terms T/R switching unit; typical Freya technology

(IIISC269079 "US National Archives" courtesy Mike Dean)

* Because the antenna being connected on top instead on the right-hand side.

In my perception a bit crude, the upper symmetric lines interconnect onto the antenna array. Thus when following the route from the antenna downwards: antenna - transmitter - T/R unit and passing through the wall the receiver connected onto the N-Gerät.

This consequently means - that the TX anode circuit is all the time loading the antenna system. Although, when it is in its off mode, the coupling loop and accompanied anode circuit might not hinder proper energy transport onto the receiver.

Maybe the U-shape section  between T/R box and the transmitter feeder might having ½ λ length. Roughly, this could have been valid. According a recent report provided by Mike Dean RAE Farnborough Technical Note No 380 Wassermann M II and M IV both operated in the spectrum of 120 - 158 MHz; thus in the genuine lower Freya spectrum; I myself consider generally being 125 MHz λ ≈ 2.4 m. 

I believe that inside the T/R switch they operated so-called Nullodes, glass envelopes filled with low pressure 'water vapour' (sometimes traces of some radio active substance being added as to speed up ignition as well as switching off performance). It has, nevertheless, to be noticed that in those days often SD 6 diodes were the core of the Siemens T/R units. On the other hand, in the translated German periodical Funkmeßnachrichten 19 issued on 25 February 1945 (Radar News 19), is noticed that these valves, owing to failing reliability, should be replaced by Nullodes.     SD 6 were sold in the late 1970s in quite big quantities, I then wondered where these might originate from?


When you look carefully again on the upper left-hand wall you just can recognise the two symmetrical antenna wires feeding onto the receiver unit N

(111 SC 269075 - "US National Archives" courtesy Mike Dean)

In my perception, therefore the N-Gerät being placed next to the wall, providing the shortest possible antenna connection.

The high power version did need a sufficient power supply; which the regular Freya like R-Gerät could not supply.

In my perception, a pragmatic solution was - adding an additional R-Gerät.


This photo shows how this once had been accomplished

(111 SC 269078 - "US National Archives" courtesy Mike Dean)

When we compare both this photo and the foregoing picture, we might notice that the latter did not represent a high power station. The contradiction is, that on Mike Dean's photos both times .. Wassermann Vorupor being noticed (and not only there).

What other option might bring both consideration in line? We don't know yet.

Maybe, the operator room R-Gerät was for this occasion only used for supplying the medium power voltages, and the photo above is representing the HT for the transmitter. The painted symbols on both (blue) glass windows, might underline this thought.

Nevertheless, I have seen also photos where a dual R-Gerät being used in conjunction to a Jagdschloss radar system.

On the other hand, I get the impression that although Vorupor being provided as photo-title, that we may have to deal with at least two different Wassermann stations sites.*

* Mike Dean had to rely upon the photo captions found in the US National Archive; this omission cannot be corrected as all systems have vanished.

Please look carefully at the open symmetric line on the right-hand side of the photo; I have used the genuine photo which's quality is better and consequently data is a bit too big for the web. The text down the symmetric open line entering the wall:

Bei AN-Peilung ...

This clearly points at the application of what I have described previously on A/N Peilung, an historical use of a term originating from navigational systems, where in Morse A = .- and N = -. being exploited. Like in the Lorenz Blind-Landing system, the correct DF is indicated when A and N overlapping one another; then a continuous signal being indicated. We have noticed before that split-beam provided also two signals, only when both were found equalling the correct bearing is accomplished.

Considering the foregoing, we may estimate that at least split-beam operation was accomplished at that station. Whether the dark box on the far right of the Vorupor operators desk has a link to AN operation - I don't know. A wastebin is unlikely.

However, according to AIR 20/1702, as well as the genuine drawing of an operators desk:


Both point in the application of a remote elevation-control-box (Peilgeber Höhe)

(Courtesy Fritz Trenkle)

Just popping up in my thoughts: split beam operations might have been operated not only within the M II and M IV system type, but in many more Wassermann systems (maybe all).

The next photo provides a different situation



Again: the remote control unit for elevation being not existent

 (111 SC 269075 - "US National Archives" courtesy Mike Dean)

We know from the Frequenz 1955/10 reference (Zetzmann's paper), that elevation (upwards) was accomplished over 20 degrees.

Do we see any indicator having a scale of 20 degrees?

The only matter I can recognise, is the compass like rose; in my perception more pointing on to azimuth search.

The two heavy switches on its front penal might constitute a kind of main switch.

Don't think of a simple motor support. The driving power should have been considerable.


An example of a Wassermann azimuth driving system.

(Frequenz, Bnd  9 / 1955  Nr. 10 pp 351-354)

After considering this photo again, I discovered that the driving motor pulley might be fit with a break system. Maybe operated for fine azimuth search or as a safety provision against heavy wind and inertia. Most Wassermann systems had been erected in coastal areas (poor weather).

We have, however, to think of a not too big Wassermann system.

Considering the metal side walls - I estimate it might have been for driving a Wassermann L type apparatus.


Comparing both this photo and the previous one, we get the impression that "standardisation" wasn't a great deal

 (111 SC 269082 "US National Archives" courtesy Mike Dean)

I believe, the two die cast modules constitute gear boxes. The one most on the right-hand side might have been a 90 degree gear box; and the solid black shaft in the rear the azimuth driving output. Considering the rear wall this was all kept in a bunker.

My estimation of a 90 degrees gear box is because electric driving motors are, to my knowledge, looking differently. 


Chapter 3 has become heavier loaded then was foreseen, but Mike Dean's materials are too attractive as to keep them for a future occasion.

This is just the right moment!



On 28 September 2015


Among the materials kindly provided by Mike Dean, we find an interesting report made during "Operation Post Mortem".

Although, the pdf reference curiously gives: Wassermann - Unterofficer Rieke, in modern days the Germans would say: 'new German'.

However, its genuine title is:

Technical Description of Wassermann (data chiefly from Unterofficer Rieke - Wassermann Mechanic)

Apparently, most data being provided by Obergefreiter Kennel

My first thought, simply putting the PDF file on the web, but its content is then not really accessible in a digital manner. I therefore take the burden typing it in full length over; though larding it with details where these can enhance understanding.

In the course of this contribution, we will also meet facts which we previously have dealt with. Some can enlighten unsolved contradictions in regard to our foregoing determination of Wassermann technology.

The author*, sometimes uses words like Ferraris motor, where he might have meant selsyn instead; somewhere he draws (or was Uffz Rieke or Kennel?) clearly a selsyn system (Drehfeldgeber) being meant.

* On a query put forward during a vivid exchange of e-mails between Michaël Svejgaard and Mike Deam and myself, I discovered who was the person named Kennel (see later); Mike did supply additional information:

This is obviously a rapidly developing correspondence - from your later message I think you have worked out that Kennel was an Obergefreiter. (O/G)

 For most of the reports, it is likely that some of the sketches were done by the British scientists, the reports appear to have been typed up in Denmark.

 The initials appearing before the date in some of the file titles eg MR-BRP denote the principal author and his associate. MR = Martin Ryle.


As to complete this chapter, the genuine PDF file will be added as well, and can be accessed down this report. But I would like to advice: read my comments (in red) and genuine text first.


1     The operating frequency is 136 Mc/s. and is not to be altered except on orders from German Ministry of Air. To alter it would involve moving two shorts on Lecher bars and tuning the output circuit.

        Receiver tuning controls are set by the mechanic on a P.E. and are not altered during operations.

2    Presentation

(a)    The range operator has an 80 km. trace with lens viewing. (a frame carrying optical lenses, AOB)

    A phasing control moves the time-base along until the blip is on the cursor line (designated Lichtmarke, AOB)on the tube face. (looking the screen of an OB-Gerät, AOB) Range is then read from a Veeder counter.

            I.F.F. responses appear on a lower trace on the same tube. (they used for all CRT scopes dual trace tubes AEG type HR2/100/1,5. Each CRT (trace) system was fully a copy of the other, and independent operating, AOB)

(b)    The supervisor has to two tubes, the first with 200 km horizontal trace, and the other with 40 km. trace.  (we might think of such kind of CRT scale, AOB) A phasing control is available to adjust the position of the 40 km. within the 200 km. range. (I guess, he operates the P-Gerät, which will be described in more details later, AOB)

(c)        D.F. operator has two tubes also. The upper has a 200 km. vertical trace, which doubles when split (operation, AOB) is switched on to give left/right indications. This is the coarse D.F. tube.  The lower tube is for fine D.F. and has a 200 km. horizontal trace, with downward deflection. Above the trace is a strobe trace about one cm. long on which the blip which is on the range operators cursor line appears. (I guess, that they mean the signal fed also onto the OB-display, AOB) When split is applied, a second 1 cm. trace appears  beside the first and the left and right echoes appear side by side on those two traces. (In my perception, the upper cathode ray tube section being equal to what is being displayed on the OB-screen originating from the fine range unit, AOB) (It might concern the HP-Gerät, which is to be dealt with soon, AOB)

        There is a knob to set the blip on the strobe when it is on the operators cursor line. This is only adjusted by the mechanic. With no split they are probably within ± 1° on their azimuth readings. (hand written, AOB)  (The author might not fully understood what the implication were. The display might have been equal to the OB-Gerät display; and the cursor is, from inside behind the CRT screen, illuminated by a projection system creating a vertical lighted line. For more details notice Mammut (B) , AOB)


        Horizontal beam width at transmitter half-power points (3 db, AOB) whether split or not. Each side (left or right) consists of 12 groups of eight half wave dipoles.


Each group is fed separately; the feeder coming down to split switch.  Here all one side are joined, since no height finding is in use. (My guess, the group cables joined somewhere. Each antenna side left or right, after combination, being fed onto a 'split switch relay', AOB

(e)    Split


moves the lobe about ± 2½° from physical centre. The whole array is thus used on split, and when no split is used the beam is 2½° off centre so; this correction is added to each bearing by the D.F. operator. (quite logic, because the antenna lobe being switched by relay; which's movement being interrupted. But, how do they know into what position, left or right, it just stopped?  Or, it might have been that the relay when ac driving power fails it stays in a determined switch position, AOB). Lobe 2½° off target gives 20 - 15% less volts out. (consequently, when no split beam is operated, only a (vertical) half of the huge antenna array being active, quite a waste, AOB)

        The aerial can use auto-scan to and fro (but not continuous rotation) but this is seldom used. Can scan almost 360° - but their usual arc of search for manual is from 230° - 320° - i.e. 90° sweep.  (owing to the fact that cables prevent this, AOB) (These bearing figures may given by the location of this particular radar site versus the to be expected aircraft routes, AOB

    (i)    Turning gear - Servo system

                A Ferraris motor (the author might error, it should have been selsyn instead, AOB) on the mast gives a feedback voltage into a valve amplifier which controls the field of the generator which provides the D.C. field for the turning motor. (known as Ward-Leonard motor, AOB)

3    Further Aerial Array Details

    Feeding the groups:-


    The back netting (which is not soldered) is 25 m. tall and the bottom of it is 5 m above (ground? AOB) (considering the measures it might concern the small Wassermann type L, AOB)

4    Measurements of Main Lobe (On P.E.)

    Echo amplitude down to half of maximum at ± 9½° off centre. This should be about transmitter half power point. Hence transmitter half power bandwidth is 19° approximately.

5        Receiver

        S + N = 2N at signal of 10 micro-volts about (O/G Kennel) (O/G might solve one of my later queries - who was Kennel, was he German or British - O/G might have stood for Obergefreiter, AOB) Noise itself is 7 - 8 micro-volts.

(a)        One R.F. stage using Philips 4673 "built into" tuned circuit.    Is not a vari-µ valve.

(b)        Local oscillator on 111 Mc/s. is a RL12T1 given from the 4673 mixer a first I.F. of 25 Mc/s. (two AF 110s here).

(c)        Second I.F. is 5 Mc/s. (with Local Oscillator on 20 Mc/s.). Two AF 100's with a bandwidth of 500 Kc/s. - (as the first I.F.). Gain maximum is 2 x 106. Usually work with about ½ cm. of noise.

(d)        Output valve is a P.10. (video amplifier type RL12P10, AOB)

(e)        Measurement of width of echo from Sylt is 1.2 kms. wide at half amplitude. (what is meant here I don't know. Do they mean that the strong Sylt signal, as we know now operated at the same frequency provided a 1.2 km broad pulse at the CRT screen? AOB)

6        Modulator and Transmitter

        The 500 c/s. waveform (sine wave, AOB) goes into the modulator were it is squared. The square wave switches on a pair of LS 50's which have ringing coil in their anodes. This produces a 2.4 kV., 2 micro-second pulses - all but the first half cycle being damped by a diode.

        The transmitter has 2 x TS 41 valves which are normally biased off at - 2.4 kV. on their grids. The incoming pulse switches them on. Peak power is 40 kW. (this is most interesting that were might know now what the previously discussed modified Freya TX is giving for transmission power, AOB) (this can be an important information, for grid-keying you don't need high energy, the only way to increase transmission power is to step-up the anode voltage. We have dealt with the dual R-Gerät set up, were the High Tensions wired in series? Somewhere in this report was mention anode keying with 22.5 kV; I estimate this voltage can be supplied by two R-power units in series, AOB)

        Has lecher lines in anode and grid circuit coupling  circuit is tuneable by a condensor. (the latter tuning capacitor was existent in the regular Freya TX, but cannot be seen within the modified module, maybe there is one, but not inside the modified Freya TX, AOB)   

7        Common T and R Unit

        Is not wide band, but can be set up on any frequency from 125 - 145 Mc/s. 

The TR switch in this equipment is a Nullode, i.e. it has no electrodes within the bulb, being simply a concentric glass tube containing a mixture of neon and hydrogen. The R.F. is fed to it by concentric tubes, one fitting inside the bulb and the other outside.  

        Its resistance when conducting is low, presumably far less than 70 ohms, since it is across a 70 ohm line. (there is happening something different. The glass bulb is having a stub inside the glass bulb and a cylinder around it, due to the transmission - electric field between these two electrical conductors, the gas causes a kind of short circuit and detunes the (stub) circuit. In the FuG 220 there was in series to the central stub a coil, when the Nullode ignites - the circuit is loaded via the coil - conducting (capacitive) onto earth. This will be rapidly repeating, because when the Nullode ignites the HF being detuned and the Nullode will stop conducting. But instantly igniting again. Both foregoing photo examples were taken from our operational Würzburg system. AOB)    The maximum voltage rating is 7 kV. (this is not possible, because at 8 kW the voltage at the coaxial line is P = U x I → U : R = P → P x R = U2 →   8,000 x 70 = U2  → U = 748 V. Even when they should have dealt with 40 kW, what wasn't the case, we should get: 1673 V, AOB) The life is estimated at several thousand hours - this one has done 3 - 400 so far.

8    I.F.F.

        The same type of transmitter is used for I.F.F., transmitting on 158 Mc/s. and receiving on 125 Mc/s. (this is confusing, because the FuG 25a IFF system in an aircraft, receives the Freya like radar signal at about 125 MHz and the transponder responds at about 158 MHZ. For it a separate receiver Gemse being engaged. However, most GEMA sets operated later at different frequency than 125 MHz, so did this just described system (136 MHz). In this occasion a separate transmitter KUH (Q-Gerät) being implemented, AOB)* The I.F.F. response is displayed on a lower trace on the range operator tube. (Displayed apparently on the screen of the OB-Gerät, all CRTs were full dual trace, operating the second tube system, AOB)

* It later was found that bigger systems like Wassermann and Jagdschloss had been fit with a Freya type transmitter module, but handling only I.F.F. signals. Hence, providing rather high power I.F.F. ground signals. The advantage being - that I.F.F. systems in German aircraft could be triggered at ranges over 200 km!

        In the waveform generator there are two phasing controls to adjust the phase of I.F.F. and the main transmitters to compensate for the relay in the aircraft I.F.F. set.

        Uses split if required. The (IFF, AOB) aerial array consists of 4 wide band vertical dipoles per left or right side. (the antenna section was always on top of an antenna array, but not every Wassermann being fit with IFF, AOB)

9        Extra Units

(a)      Extra range unit

                Switching this unit increases the range by 200 km.

            Can thus read up to (and actually beyond) the next transmitter pulse.

            Just add 200 kms. to the range reading - both range and D.F. operators traces being shifted the 200 kms.

(b)    Wasserflau (Wasserfloh, AOB

        Consists of a small C.W. oscillator with switch, tuning and output controls. (This MTI principle was also known as Laus and famous Würzlaus. Actually forcing the transmitter to start-up synchronous (coherence) with the 'Reguliersender', which often was tuned at a little bit higher frequency. It took me a long time myself before I understood what happens - the transmitter signal starts being locked onto the CW signal, but as soon as the transmitter is overruling the CW signal it jumps to its genuine frequency. Causing a signal with a coherent signal component (spectral lines).  A portion of the energy arrives also at the front-end of the receiver. Here it interferes with the returning signal. The interference being displayed at the CRT screen (causing an additional DC off-set). The coherent part of the radar signal will allow to distinguish between not moving (like chaff or window) and moving targets, like aircraft. The moving targets will causing a beat on the screen described as Laus, AOB)    Its output (of the CW generator, AOB) is fed into the transmitter box by a probe about 10 cm. long running parallel to the grid lines. The box probably has two valves, being oscillator and buffer stage. (In Würzburg they simply prevented from signal pulling by tuning the CW signal at ½ frequency, AOB) The C.W. gets into the receiver via the common T and R unit. (The CW source continuous and also present when the transmitter being blocked, AOB)

        The operators seldom use it, if they are interfered with they leave things to another station. (from Würzburg we know, that operating Laus = Wasserfloh was delicate and fine touch was demanded to adjust it properly, which the regular personal is apparently often lacking, AOB)

        On 24th June (first exercise) when Window appeared they said "Here comes that growing grass again", and more or less packed up in the Window sectors. With Wasserflau can sometimes see that there are aircraft present in the Window but still cannot count them, or follow a particular track.

(Later popping up in my mind, the Wasserfloh Reguliersender most likely was placed near to the T-Gerät, because it should be connected onto the transmitter; the receiver will picking up some of its signal content passing through the T/R switch. This is quite evident, because the T/X being straight away connected onto the in/out going antenna line. (at least valid for the low power version), AOB

(c)        Height Finding Equipment

            The height tube is already installed in this equipment. There is also a compensator in the cabin at the base of the array. (I doubt, because it should have been installed at the back of the antenna somewhere in the middle, but very down? AOB)

        The Wassermann mechanic (Uffz Rieke? AOB) says a compensator was tried but lost 40% (signal strength? AOB) (this type of compensator was based on techniques implemented in Mammut by NVK, but having here only 6 slotted lines of equal length, AOB)

        So they used a Wellenschieber (wave shifter) which consisted of lines (delay lines, AOB) with motor driven adjustments. (dealt with previously on this current web page, AOB)

        This is thought to lose only 10%

The above is installed and working on Sylt.

The Sylt set is also wide band, 125 - 145 Mc/s.

According to Kennel (Obergefreiter Kennel, AOB) they use a compensator just like the Mammut only smaller.

Can theoretically shift the lobe 0 - 90°, but will probably only use about 15°.

He says he can set up the compensator to lose only 20%.

Says cannot use split and height finding at once, but one after other.

10        Wassermann Turning Gear

            Bayley of RAE says the amplifier is all A.C. fed, the output being the field supply of  a DC. generator. The output of this generator supplies the (field?) of the turning motor. (Ward-Leonard technique, AOB)

    Diagram Kennel

The amplifier has two thyratrons, one of which strikes when the aerial goes clockwise, the other when it goes anti-clockwise.



1         Wassermann

(a)        Frequency range 119½ to 156½ Mc/s. They usually work at high frequency end of band.

(b)        I.F.F. trace is on the D/F operators upper tube - (this one covers the entire range, AOB)  being another trace on the right. (I cannot understand what this means, AOB

(c)        Height lobe may be swung 0 to 20°. There is a hand tuning control which makes the motor on the Wellenschieber, and its rotation is repeated back with Ferraris motor (should be correctly selsyn or Drehfeldgeber, AOB) to a meter in the operation room. (this is not in accordance to AIR 20/1702, a selsyn or Drehfeldgeber should have rotated a scale disk instead, AOB)     No split no height, just work maximum. At the moment the hand turning is u/s. and they just have two way switch for up and down and wait till the beam gets there.    On 1st July followed an aircraft into 25 kms, where it was at 18,000 feet. Sometimes they tilt the beam to try to reduce jamming. Did not while we were there on 4th July exercise.

(d)        For bearing can use auto-rotation at 3 r.p.m. (I believe, following further this text that he meant wobbling the virtual height-beam up and down, AOB) There is the usual bearing meter by the handle, but there is also a Ferraris repeater (should have been selsyn or Drehfeldgeber, AOB) from the aerial to the usual pair of bearing dials.

(e)        Transmitter uses 2 TS 41's but has high power anode modulation, with 22 kV. pulse. Has one knob ganged turning control with calibrated dial but you have to open the transmitter front panel to get it. (please notice, that the hyperlink photo might not provide a high power modulator, but the TX is according the text, AOB)  Also the D.F. operator (Feldwebel) who does the frequency changing has to walk round the back of the gear to do it. (I would say, of course, AOB)

(f)        Receiver has one knob tuning - by Operator who is on height. (let us consider the drawing of the operators desk. (The D.F. operator, might sitting in front of the HP-Gerät on the far left; all is quite fuzzy described, AOB)

(g)        Frequency changing procedure in:

        (i)    D.F. operator goes round to transmitter and (reads, AOB) from a calibration chart, sets on a suitable frequency (what he thinks fit).

        (ii)        Also from chart, sets two controls on the common T and R unit. (notice the transmitter tuning on the right the two T/R tuning controls, AOB)

        (iii)        Returns to front and tunes the receiver on a P.E. This has now taken 80 seconds.

        (Iv)        Then he trims the T and R on shorts from another operator, taking about two minutes to do it.

        (h)        Has no Wasserflau (Wasserfloh, AOB)

(i)        Has a bandwidth switch which detunes one I.F. stage.

(j)        Wellenschieber


Other equipments

                                                Romo "Rose" Decimeter Link

(a)    Amplitude modulated magnetron driven 4 watt out.

(b)        Frequency band 600-625 Mc/s, with auto sweeping of T and R together.

(c)        Can handle 9 speech lines, being in the range 30 to 60 Kc/s. (of the side bands) (Träger-Frequenz, carrier telephony, the 10th channel is for service, AOB)

(d)        Each speech channel can be divided into three teleprinter channels.

(e)        Transmitter receiver separation should be more than 2½ Mc/s.



    P.P.I. Freya

    Normal frequency 146 Mc/s.

    Range 125 - 155 Mc/s.

Rotation at 6 r.p.m., maximum being 10 r.p.m.


SYLT CORFU    (Korfu, AOB)

Horn at 45° giving 45° polarisation. (this was often used by the Germans as it provide the possibility to receive both horizontal as well as vertical polarised signals. Albeit, with reduced signal level, AOB)


   The probe is a continuation of the feeder inner.

You can download this report in PDF format. 



On 1 October 2015

The CRT Presentations


Let us first consider virtually what the various presentations might have constituted.


A theoretical example of an operators desk

(courtesy Fritz Trenkle)

Operators: at least 3 men must have been involved at once.

Starting from the left-hand towards right.

D.F. operator: managing the HP-Gerät  -  He controlled both the HP displays as well as the elevation and likely azimuth remote controls as well.

Supervisor: the P-Gerät; maybe also the receiver of the N-Gerät

Range operator: managing the O-Gerät

Thus, 3 men were 'manning' the operators desk.

I guess, that at least a technician was on stand-by duty.

I believe - that most of you will not having an idea as to how the operational CRT screens might have looked.

The HP - as well as P-Gerät display presentation - is a reconstruction of what I imagine might have been once the case; sad - there do not exist conclusive documents.

The HP presentation is the most unknown aspect, because we have to rely just only on the few lines of brief explanations based on the foregoing Post Mortem Report.

The P-apparatus is less doubtful, but still we encounter differences to what the report text tries to express - and considering a current photo of a P-Gerät module.


 This is a photo of a Wassermann operators desk photographed in June or early July 1945

(111 SC 269075 - "US National Archives" courtesy Mike Dean)

Please notice again the undesignated frame with the three slanting lines (genuinely red colour). Comparing the foregoing drawing with the one above there can hardly be a doubt about the P-Gerät.

In this set-up we lack the HP apparatus, as well as the remote control unit for steering the virtual antenna beam elevation. Thus, we should consider this Wassermann site is lacking elevation beam steering (height finder).

  I am nearly sure, that when the desk would been fit with beam steering - that the HP-Gerät would have been marked with a series of red lines too. Because, both did constitute secret technologies; to what I remember having read once - in case of danger these devices should be destroyed first.

Please remember: that the drawings do show fantasy pulses and other means; its only purpose is to let you see - and understand - what it once might have been about. Neglecting a general noise floor of about ½ cm, according to foregoing report.

Starting from the most left-hand side


The HP-Gerät*

* All other modules carrying code letters, of which for some the first character designate to a module; like OB-Gerät part of the O-Gerät, and NB-Gerät which was occasionally part of the N-Gerät.

In my perception HP stood for: Höhen-Peil Gerät.


The next virtual presentation should have been monitored by what the foregoing report designate: D.F. Operator.

The upper CRT might have been fit with a range scale, but I have no example to refer onto, therefore it has been omitted.


HP-Gerät dual screen presentations, yet no split beam operation for fine D.F. (DF); beam is 2.5 degrees out of the vertical antenna axis


All screen presentation examples are drawn freely without reference to other screen contents.

The yellow cursor line being projected from behind the phosphorous CRT screen.

The upper trace painted on the lower CRT screen, equals the OB-range operator display.

My difficulty, is, to translate the fuzzy explanations into a sense making instrument.


HP-Gerät dual screen presentation, the main difference on the lower CRT is the upper trace line showing side-by-side the split-operation D.F pulses.     I cannot say why- and how- this technically was accomplished. The lower trace is equal to the previous example


Viewing the upper CRT presentation for what it eventually might have looked having scale divisions coarse D.F. tube. The scale design has been derived from two examples - first the CRT of the P-Gerät module and from our museum sample.

However, the lower CRT trace is providing entire range capture, but being polarised downwards.

I must admit, that I have no idea how these two left- and right signals being combined at a single trace, nor why.

This section has been derived from the foregoing report 2 (c)




More towards the centre of the operators desk we encounter the N-Gerät


The N-Gerät: on the right the receiver module NE; on the left a dual CRT presentation known as NB apparatus

(derived from the Fu.M.G. (See takt) 40 G (gB) manual)


Let us view what kind of presentations we will meet.

Working on the Wassermann chapter Y11A I discovered that my perception of as to how the former NB 110 screen presentation was apparently incorrect; and needs instant correction!



The lower CRT screen presentation repeats the OB range-measuring screen


The upper screen displays the entire range capture at a determined azimuth. The trace with signals pointing downwards provide the according I.F.F. signals matching onto targets.


On 17 January 2016 I have redrawn a screen simulation, in the line of my perception.

Although, I have to admit that I do not understand the mechanical implications of the gated video fully.


Please compare this screen presentation reconstruction with the foregoing drawing

First, the screens being interchanged. Second, the upper - what the British investigators designated 'slow' time-base - constituted a variable chosen range section of approx. 40 km.

How it really must have once been accomplished I don't know yet. Because all sources available do not provided accurate or conclusive information. Even not the text with a genuine Seetakt manual.

Please consider for the cross out text and smaller than usual drawings my additional contribution Y11A


When their are no I.F.F. signals accordingly visible - than it is possible that it concerns an enemy target. Of course, radar signals with according I.F.F. response is regarded being friend.

According a foregoing report, there existed a provision to add an additional 200 km range; one only has to add 200 km to the read-off of the veeder counter (Zählwerk). The range operator on the far right-hand side will adjust a target against a projected cursor line (slit illuminated from behind the CRT envelope)

The supervisor should instruct the range measuring operator (sitting right of him) at what range (target) he should focus upon.


The P-Gerät  is rather fuzzy described in the foregoing report 2 (b)


The supervisor's CRT presentations 


The P-Gerät


Let us first consider how such display module genuinely looks.


The foregoing report stated that the lower screen:

... and the other with a 40 km trace. A phasing control is available to adjust the position of the 40 km. within the 200 km range?

(Courtesy Colin?)

Do you understand where he is talking about? Frankly, I do not.

The upper CRT screen does not run from 0 to 200 km, because the (blurred) figures to the upper scale does show to consist of three digits. And the lower CRT clearly has a wider scale than 40 km only (and scale figures having two digits only).

I therefore would like to reconstruct, considering my imagination, what might have been displayed on the hand of the foregoing apparatus picture.



P-Gerät screen presentations (PB-Gerät??). Overlooking the entire Wassermann capture range divided 0 - 100 km and the upper screen displaying a range of 100 - 200 km


Recently doubt is risen whether my assumption that it concerns a P-Gerät. I cannot say it is - or it isn't.

The downwards facing pulses at the lower traces constitute the I.F.F. (correlating) signals.

In my perception, the station operator in charge needs an overall picture of all targets within the current capture range of his Wassermann system (at a certain bearing). Somewhere else in the foregoing report they refer to a horizontal beam aperture of 19°. To cover a wider arc azimuth bearing has to be altered.

The I.F.F. signals being visible on the lower trace lines. Recognition what particular friendly aircraft is concerned was provided by a repeating Morse code character (signals flickering in the rhythm of the Morse code). However, all targets correlating with a downwards facing I.F.F. signal being friendly; the rest might be enemy related.




O-Gerät frame, the range finder unit

(derived from the Fu.M.G. (See takt) 40 G (gB) manual)

On the right-hand side the delay-line module known as Messkette; range reading-off from a Veeder counter.

On the left the OB-Gerät, constituting the range display.

For more details I would like to advice you - to access: Exhibits-details 17 webpage

Its function is to delay the regular time base in such a manner - that the foot of an up-going pulse slope matches just on the right-hand side of an optical projected vertical white cursor line in the centre of the CRT screen. This light cursor* stays even when the beam is being entirely dimmed.

* For obvious reason, I have drawn it yellow.


The CRT scope being drawn at the lower end, because just this is what is visual when you look through the front window


The upper section being kept free, because in the NB chassis there we encounter the full range CRT implemented.

You might get the impression that the I.F.F. pulses are not well matching onto the according target pulses. In practice the I.F.F. signal pulses legged a bit behind the actual radar signals. According Fritz Trenkle, about 400 m (about 1.2 µs); on a 200 km scale this aspect can be neglected. But I would like to emphasise upon this phenomenon.

 Consequently, the OB- as well as the NB frames are about similar, the difference the OB-Gerät carries a single CRT, whilst the NB apparatus being fit with two CRTs. Both of the usual dual beam type HR2/100/1,5.



On 12 October 2012


As a result of quite lively discussions between:- Michaël Svejgaard - Mike Dean - Hugh Griffith and myself, Mike did sent us all a CD copy of material he possesses on Denmark.

A most intriguing series of photos and reports arrived. We have to notice first that the photo captions sometimes do contain  some reference errors. For example, from a radar site we possess so many different kinds of system photos that one or another must have been photographed elsewhere.



A typical mistake, I wondered first what is wrong? Because the antenna output never appeared on the left-hand side of a GEMA transmitter

It should show an I.F.F. (IFF) transmitter

(111 SC 269080 - "US National Archives" courtesy Mike Dean)

Nevertheless, intriguing is the air-cool-converter, consisting of a zigzag wound quite heavy tube; incorporating an electrical pump, similar to the one regularly used. 


Whereas it should be this way around (photo rotation 90° clockwise)

(111 SC 269080 - "US National Archives" courtesy Mike Dean, - 90° rotated AOB)

Please notice the grips for lifting the cover lids. All GEMA modules were being protected from humidity by cover-lid-seals, rubber seals protected the cable connectors too. Quite understandable when being off-shore, but also applied for shore systems. These kind of heavy protective provisions cannot be found within Telefunken radar systems even not within the Giant Würzburg (Riese Fu MG 65 or Fu SE 65).

The horizontal U-shaped copper tubes (trombones) constitute a line matching facility; it is adjusted as to match the transmitter output impedance optimally onto the Sperrtopf (balun) visible on the far upper right-hand side (likely, as to obey to: n x ½ λ, n = 1, 2, 3...).

By the way, the first idea of an automatic triggered I.F.F. response was to employ the (in the beginning) regular Freya frequency of about 122 - 126 MHz radar signal. It was used to measure regularly range or distance as well as simultaneously triggering the FuG 25a aircraft transponder. During the course of history (jamming) they have been forced to widen the operational frequency band. Firstly, towards higher frequencies; we have to think of up to, say, 155 MHz. Later they went also down towards even 70 MHz and beyond. It is evident, that a system based upon a transponder input frequency spectrum of about 122 - 126 MHz could not cope with input signals beyond this 3 to 4 MHz wide I.F.F. band. The FuG 25a receiver front-end being wobbled continuously as to capture signals within this relatively narrow spectrum. Therefore an additional transmitter was implemented into the system, as to cope with the now different I.F.F. spectrum. Also known as KUH (Q-Gerät) maybe providing, I guess, 1 - 2 kW pulses; based on the same transmitter triggering pulse (both the radar TX as well as the I.F.F. transmitter did send simultaneously). Apparently, they also implemented quite high power transmitters (8 kW). Implementing high power facilities - allowed also another application - known as EGON (Erstling-Gemse-Ortung und Navigation)(Erstling was the codename of FuG 25a - Gemse meant the ground based I.F.F. receiver - Ortung is DFing  and range ...). In this case the I.F.F. transponder being employed as a guiding system; skipping Freya or equivalent radar pulses. This became known as secondary radar.

The FuG 25a transponder being followed beyond the system capture range; simply because the sensitive transponder picking up the upon them directed I.F.F signal (Freund-Feind-Kennung) and retransmitting it at about, say, 156 - 158 MHz. It is clear that this provision has great advances, because - the radar mode signal has to be bounced backwards onto the radar system. Never forget - that only those pulses can bounce successfully that arrive at the target skin under an angle of 90° (the rest being dispersed); this (tiny portion) only will have a chance being picked-up by the transmitting/receiving system  again. In contrast, the EGON amplified signal being about 800 W pulses instead of bouncing - mV or even µV signals. The downside being that range accuracy reduces and an additional system delay being introduced. According late Fritz Trenkle this constituted an about 400 m lag in range (+ ~ 1.2 µs). According another Post Mortem report reference:- the time-base generator (Z-module) provided a second variable pulse delay as to counter I.F.F. signal delay on the CRT screen presentation (drawn examples being described within chapter 5

We may estimate - that the amount of radar energy that returns at the receiver front-end is only about 10-16 or 10-17 th. Therefore, radar range was often less than the PRF would have allowed for. Freya and related systems operated a PRF of 500 Hz, allowing a radar range of 300 km (600 : 2); whilst it actually captured hardly 150 km distance (neglecting Mammut and Wassermann). Not because the signals were too weak for wider ranges, but because the amount that returned was too little as to surpass the receiver (system) noise floor.

However, the Germans exploited this technique quite often.

Another system - known as Freya Flamme - exploited the British I.F.F. Mk II. This phenomenon was discovered after a Freya received quite strong pulsed signals when pointing into particular direction (it operated at a slightly higher frequency about 128 MHz? than formerly used). It was soon concluded that the particular pulse modulation sequence originated from British I.F.F. This could work well at ranges of 250 km and beyond.

A new weapon was born; later British I.F.F. operated near- or over continental Europe could be lethal! Please remember, that transponder signals extended its capture range beyond the optical horizon. About September 1944 its application was strictly prohibited when approaching German controlled territories. Inevitable, however, within a bomber stream of several hundreds there always was at least one offending regulations. Also H2S was from then on restricted to be operated; only permitted when quite near a target (about 50 km). German Y-service proved to be operating better than was expected; this was particularly discovered (confirmed) during Operation Post Mortem.      



Another interesting example that one should critically digest photo captions:-


TU 100 transmitter, erroneously the lower section being designated power supply

(111 SC 269019 - "US National Archives" courtesy Mike Dean)

Whilst it concerns: - the modulator stage; on the left the box for the TX filament transformer.

This time the modulator power valves being a pair of RS 391 we will find sometimes LS 50s instead.

Please notice the cable connectors down on the right-hand side and compare it with the foregoing photo; it is evident that the latter photo represents the true situation. 


Compensator of the DK radar site of Hjaremeal

(111 SC 269035 - "US National Archives" courtesy Mike Dean)

By the way, it is likely that this compensator or Wellenschieber type was once designed by (KM) NVK. In some of the British Post Mortem reports a German technician quotes:- that these versions implemented an additional signal loss of 40 per cent. In another report quoting Obergefreiter Kennel - told his interrogators that with proper provision he could reduce this to a 20 per cent loss. The Siemens type Wellenschieber apparently caused an insertion loss of only 10 per cent. 

It is evident that it concerned the height-finding compensator. But we may assume that an equal ganged section existed too; given the fact that split beam operation necessitated exact beam-steering between the left- and right-hand (vertical) antenna systems.

I have taken time to look carefully at the slotted-line gap and discovered that indeed here they do not employ a linear coaxial line but a helical instead. The type of line erroneously referred onto in the Mammut report; in this context it matched the facts; though, not in respect to the regular Mammut compensator.

I guess, because the delay-line was constructed as to match onto the current Wassermann situation and operation at a higher frequency band up to 150 MHz. According a Post Mortem report we might consider 135 - 155 MHz. 


Viewing a similar-compensator (Wellenschieber)

(111 SC 269057 - "US National Archives" courtesy Mike Dean)

The difference appears when we compare it with the previous photo, it is clear the the latter is being fit with an additional 'Stichleitung' lacking in the previous one.


Magnified view into the slot, please notice the helical conductor construction

(111 SC 269057 - "US National Archives" courtesy Mike Dean, modified AOB)


Again, only those very familiar to German techniques and apparatus will notice that also this photo previously has been produced up-site down

(111 SC 269057 - "US National Archives" courtesy Mike Dean, modified AOB) 

There are two indications underlining my assessment: - German type number plates always carry the subject first; as well as the numbers painted on the compensator arm and cable combiner. Curious only is, that the arm pivot apparently is on the bottom side; or are we looking against the room or cabin ceiling?

Time and again:- what good looking at a photo can learn us.

A query, time and again we see a square grey cable junction box, what was their purpose? It might have been for interconnecting a servo provision, as the compensator-arm arc should have been repeated onto the radar operator. Although, it has to be said that we haven't seen a sign of such provision, yet.

A query to myself - why did have Mammut compensators real conductive (linear) sliding gear, whilst the Wassermann compensators did have, like the Wellenschieber, helical provisions?

I assume that the reason might have been - that the distance between the most outer antennae sections of Wassermann was far greater than it was within the Mammut antenna array. For it one need more transition delay; which is easier to achieve with a (coaxial) helical conductor than with a (linear) straight line conductor.



         On 15 October 2015


I would like to deal with a mystery:- how did the Wellenschieber apparatus actually worked; meant in technical/mechanical respect?

So far both Zetzmann (Frequenz 1955/10) and British AIR 20/1702 provide, to say it mildly - oquite fuzzy stuff. Photo or drawing are given, but nowhere is really described how it technically was done.

Let us reconsider what we already have noticed.


     I have two comments to make:- it is not well visible that the apparent sections are actually constituting a helix. More misleading is the that that the small sprocket or cog wheel is not used as a tapping or contact device, but constitutes a driving mechanism as to let the cylinder rotate. When the cog wheel being fixed at a position - the helix will start moving into a particular direction      

(Frequenz, Bnd  9 / 1955  Nr. 10 pp 351-354)

Giving even more confusion is a drawing found in AIR 20/1702


Am I mistaken, when we get in first instance the impression that the helix gearing moves along the helix?

(AIR 20/1702 with courtesy Phil Judkins)

This would have the consequence that the tapping contact should move along the screening cylinder.

But, considering a genuine photo of such device, we can clearly notice that this was not the case.


Here we encounter our dilemma

(Courtesy Mike Dean, photo taken in the early 1990s at the SM depot at Wroughton)

It is evident, whatever mystery is found inside, tapping (coaxial cables being already cut-off) is accomplished from a fixed position. On the left-hand side of the coax cable connections we notice small circular glass windows, according a reference meant for watching whether a particular helical rotates.

Having studied the introduction photo to this chapter and the foregoing lively discussions with Michaël Svejgaard he did sent me very kindly a series of photo copies which - after some considerations together with Phil Judkins - unrivalled most of the Wellenschieber 'secrets'.


Shown is a helix demounted from the cylinder or tube behind it

(Courtesy Michaël Svejgaard)

Please notice first the shadow thrown from the left-hand side door-post; and consider the shaft connected onto the helix cylinder:-

With some imagination you might recognise the right of the centring (transparent) triangle shaped disk - the shaft coming out of the (left-hand) helix passes into a slightly thicker tube leading to the left-hand side. Please bear this in mind too. 

A key finding is the quite flat plate with a tapered hole toward the inside of the helix housing tube; how it might have worked I have no idea yet. But it should in someway or another enable the sprocket- or cog-wheel to drive an inside helix.

Please notice additionally the group of cables passing through the cabin floor. Considering all together we may assume that we deal here with a cabin somewhere at the back-side of the giant Wassermann antenna array.

Luckily Mike Dean provided to us a report made by British investigations during Operation Post Mortem at a visit to the Wassermann station at Sylt.

Obergefreiter Cove*, a technician, provided crucial information. It has to be said, that significant information mainly came from interrogated German technical personnel; valid for quite many reports. They knew where they were talking about; whilst the interrogators often being hampered by their own understanding of the matter originating from their different background.

* Cove is not a regular German family name; nor is:- Kennel. Neither is another example:- Fl-Oberingeneur K. von Gregor interrogated about the same time on Klein-Heidelberg; he wrote a key description on 31 May 1945, when he was in captivity at Grauel camp, also in the area where Operation Post Mortem should take place soon.


For Phil and me a tantalising finding

(Post Mortem Visit to Sylt on 4.7.1945, courtesy Mike Dean)

For the first time it becomes clear to us how the helix did fit into the Wellenschieber apparatus.

It was driven by means of a fixed cog- or call it a sprocket-wheel. Consequently the helix is being forced to move (rotate).

Together with the foregoing photo we can understand how this was technically commenced.

The open queries being how was this - magic - mechanism driven. Also intriguing to know how the motor on top of the Wellenschieber apparatus and the previously given information that the 6 delay-lines where driven in a ratio of: 1 : 3 : 5 : 7 : 9 : 11. Consequently, my understanding is:- that the upper system (nearest to the electric motor) will be driven at the most high speed, thus being linked onto most outside antenna array groups. Although, careful looking might unravel some additional details.



This text section quotes:-

.... The power is fed in via a fixed spring-loaded silver brush at the centre of the main tube. This system has great advantage that the system can be matched throughout, (i.e. high-impedance phasing line is not used) and thus wide-band operation becomes possible.

Does this imply - that the NVK type compensator did constitute a high-impedance line?

However, we know now how signals were tapped from the Siemens type helical line; moving from left-to-right v.v..

In my perception it might make sense to transcribe the five pages of this report, as to allow Google searching this document to your comfort. The integral copy will be made available at the end of this transcription.

For practical reasons, the illustrations being skipped but are included in the genuine PDF document version; and also implemented in my following explications.


Quoting - and transcribing from the genuine report:-

Visit to Sylt (4.7.45)

  1    Interrogation of Obergefreiter Cove

        (1)    Wassermann

        Wassermann on Sylt is M II. Serial No:- 1266.    The Wassermann at Römö is a M I - having a fixed frequency band.

        The M III & M IV are later versions, and he believes give more accurate height finding.

        The cylindrical mast type is a very early version and is obsolete.

        The M II comprises a wide-band system fitted with improved height-finding system, (Wellenschieber), which does not introduce the loss of the original Compensator used in the M I. (apparently, a NVK type compensator, AOB)

        Frequency coverage of the equipment Sylt equipment is 119 - 156 Mc/s although they normally work on 152 - 153 Mc/s, only changing from this frequency if jamming is experienced (on their initiative). Over this frequency range operation appears perfectly satisfactory, although a column showing the normal amplitude of a standard P.E. (I don't know where this abbreviation stood for, AOB) is added to the columns of dial readings against frequency. (This figure varies from 8 to 42 over the range 120 - 150 Mc/s).    This may merely be frequency sensitivity of the P.E. and does probably not indicate variation of sensitivity of the equipment.

        When changing frequency it is merely necessary to reset four controls, (one TX, one on RX and two on Simulton) (apparently using the German expression to T/R also known as:-Simultan Betrieb, AOB), to the readings shown on the table.    Markers are fitted to the scales so that 3 or 4 spot frequencies can be selected rapidly without reference to the table. (From this description we might understand that they already were being fit with the modern version of the regular NE receiver module, called Kreuzeck, AOB)    Frequency can be changed within about 1 minute.    Change frequency, results in the necessity for correcting the measured elevation, and a height conversion chart is used to find true height when operating different frequencies. (I believe a quite interesting detail, AOB)  Normally a height accuracy of not better than 500 m can be obtained.

        Aerial System

        The aerial system is of similar size to the Wassermann I and comprises 24 H.P. dipoles high by 8 wide.    3" diameter radiators are used.    The array is divided into two halves vertically - fed from an azimuth split motor.    The main feeders to the two halves are taken to two "Wellenschiebers" (ganged mechanically) (actually being housed next to one another within the same mounting frame, AOB).

        These comprise phase-shifters consisting of a spiralled conductor mounted on an insulated shaft which can be rotated and screwed along, end connections being made by means of sliding concentric connections to the 70W aerial feeders. (he meant 70 Ω impedance coaxial line and cable impedance; however, the central tapping line impedance should then become 70 / 2 = 35 Ω, AOB)    The power is fed in via a fixed spring-loaded silver brush at the centre of the main tube (in my perception the silver-brush should have been quite small in the horizontal plane, because it otherwise might short-cut the left- or right-hand helix winding, AOB).    This system has the great advantage that the system can be matched throughout, (i.e. high-impedance phasing line is not used) and thus wide-band operation becomes possible.

drawing skipped

        Unlike the Mammut, the phasing is done only in 12 bays (two sets of dipoles being in phase), but this should be quite adequate for the small degree of phasing necessary for elevation.

        The general arrangement of the Wellenschiebe is seen below, showing the arrangement of the repeating mechanism which is described later.    This, together with the split switch is mounted in the base of the aerial mast, thus involving only one RF connection to the tower (apart from I.F.F.)

drawing skipped

Each bank is presumably driven at a progressive slower speed. (1 : 3 : 5 : 7 : 9 : 11, AOB)

drawing  skipped   

            The use of wide-sound aerial elements and careful impedance transformation, together with the matched phasing lines, has resulted in a system which requires no adjustment over the range of 119 - 156 Mc/s

        Simultan Gerät

        The Wassermann Simultan originally used diodes (type SD 6, AOB), but these gave trouble when they got old, as the impedance did not remain sufficiently low, with the result that receiver protection became poor.

        With the development of the Nullode, the protection was just as good, and the life more or less indefinite.    On this equipment the Nullode systems only became available in March 1945. (notice for it Radar News 19 notice replacement at page 14!, AOB)

        The equipment was obviously modified, with the original diode holders in position and with a meter for the diode filaments.

        The exact layout of the Simultan circuit was not familiar to him and it was not possible to examine it without taking the station off the air, but details of the Nullode itself are seen in the diagram below:-

drawing skipped

He did not know what gas was used, he thought neon, but the discharge was blue.

        The device only needs two tuning controls, and these are fitted with easily read scales for quick frequency change.


        The transmitter is contained in the standard Freya container (meant housing, AOB), but the chassis is entirely different.    The same valves are used, (TS41) but in a circuit capable of single-knob tuning over the band.    Grid, anode and aerial circuits are made identical and ganged.    Fixed cathode chokes are used.

        Anode modulation is used, instead of the usual grid modulation and the pulse (22,000 volts, instead of 8,000) is obtained from three paralled TS41s feeding a pulse transformer. (8000v HT).    The grids of the three TS41s are driven via another pulse transformer, from a fourth TS41 which is driven from the original Freya modulator.     The system therefore consists of the original Freya modulator/transmitter assembly, with another box containing the additional modulator stages (to be dealt with later, AOB).    He did not know the peak transmitter power, except that it was greater than 40 kW.


        According to Crove, although a separate I.F.F. receiver is fitted. (fed from the normal I.F.F. aerial at the top of the tower), the main transmitter has to be used for I.F.F., by returning 125 Mc/s. (this could not be checked, but it seems most unlikely). (we just previously have dealt with the photo which was in first instance rotated anti-clockwise, this apparatus constituted an I.F.F. transmission system. Of course, it was housed as all Freya and Seetakt like apparatus in an equivalent mounting, AOB)


        In spite of the facility of rapid frequency changing he stated that electronic jamming was very much worse than window;    no particular part of the band had been found to be better from the jamming point of view, but they usually moved from the normal frequency of 152 Mc/s down to about 120 Mc/s.    In general charging frequency did not help very much as the jamming used to come back again.

        During part of exercise 5, the equipment was observed and the jamming was not at all effective, being generally only 2/3 X noise.    (Note:- This is presumably due to working frequency being above the normal Mandrel barrage, while it may well have been overlooked by Mandrel III operators).

        Remote indicator of Wellenschieber

        Details of the remote system used for height indication were also discussed in connection with Elephant and Wassermann azimuth indication and Askania^. The conclusion on these systems are dealt with elsewhere.  

^ The Germans designated the Seeburg Tisch often as:- Askania Gerät. Because, the Askania company made the facilities to the Seeburg Tisch convenient; as it allowed since remote control of most important operational parameters. In due course I would like to dedicate a separate webpage on the Seeburg Tisch subject, maybe even becoming a real survey.

The integral report in PDF:- VISIT TO SYLT (4.7.45)


Let us finally consider some of the other drawings incorporated in this report


The explanation as to how the Wellenschieber was implemented in the height finding provision of some newer Wassermann types

(Post Mortem Visit to Sylt on 4.7.1945, courtesy Mike Dean)

Considering both this drawing with the foregoing we get a quite good idea how the system worked.

The connections 1 and 12 where interconnected onto the most outside array sections and its ratio of (Wellenschieber) movement was 11 times faster than did move the antenna array helix linked onto the groups 6 and 7 - representing the most middle array sections.


This drawing shows how the most outer (top) array sections 1 was interconnected, and where 12 represents the most opposite section near to the ground

(Post Mortem Visit to Sylt on 4.7.1945, courtesy Mike Dean)

Please notice the vertical dotted line, which shows the virtual centre between the left- and right-hand antenna side (split-beam operation).

Height-finding (elevation) was accomplished by changing the phase between cables 1 and 12, followed by 2 and 11 and so on.

Azimuth (left-right) beam steering was arranged such that the observer had to find an equilibrium between the left- and right-hand antenna signals it has to be noticed that the latter display might have been presented by means of the HP Gerät, which was not everywhere installed, and that it expresses my vision on:- as to how it principally might have been displayed.


The Nullode* application; constituted a detuning device, which caused an impedance (loading) change during transmission power was present. Protecting the receiver front-end as well as preventing over-loading of the successive receiver stages - which could cause an unnecessary long cut-off cycle of the receiver (limiting its minimum detection range) 

(Post Mortem Visit to Sylt on 4.7.1945, courtesy Mike Dean)

* The shown Nullode operated within our Würzburg apparatus; notice also the blue discharge.

Shown is the way the Wassermann Simultan or T/R unit being modified with a Nullode and the post war Nullode patent version explicit extended onto symmetrical-wire-application. According to this Sylt report this modification was accomplished in March 1945; very well in accordance to the information stated in Radar News 19 (notice page 14). A translation of the German radar periodical:- Funkmeßnachrichten 19, issued on 25 February 1945; containing interesting information on the state of their affairs.

I wonder, whether they simply used the same Nullode type used within the Liechtenstein SN 2 and/or FuG 200?



On 18/19 - 20 October 2015


Time might be right to concentrate on some of the unsolved details.

Like the one on the magic black box on top of the power version of the Wassermann- as well as Jagdschloss-radar-transmitter.


What was the function of the next black box?


Please view carefully inside the transmitter stage and in particular to the upper right-hand cabin side (right of the 'interlock')

We see some indication of an up going (brass) line; most likely the transmitter output.  Wherefore the just visible screened cable fed onto the right-hand apparatus side was meant for I don't know. Apparently being interconnected onto it (about the level of the right-hand tuning scale) I have no idea. Not very likely that this constitutes the modulator pulse input (being between 22 - 30 kV).

By the way, at the end of chapter 8 we will draw the conclusion that the lower compartment housing incorporating the regular modulator stage - could 'drive' Siemens' high power (100 kW) modulator module, which provided between 22 - 30 kV anode pulses onto the transmitter stage shown here and the one next.


The top u-shaped horizontal loop constitutes the out-coupling circuit. Its sliding contact connection (at the far right-hand side) does well fit into the previous photo

Nevertheless, I still don't know wherefore the black top box being meant for. Maybe in includes a balun and/or matching provision.


Another subject is the T/R or Simultan apparatus, but we have just looked at it from outside only.

Why not taking a look what is inside?


The quite big (bulky) diodes type SD 6 are well visible, whether operating in push-pull or other means I cannot answer, yet

(Please consider also the additional description further down this page)

The soundness of the quite elaborate construction is evident.

Considering the two vertical tubes, one might get the impression that some kind of a symmetric system-tuning is dealt with. Whether this is valid I don't know.

Viewing this photo carefully again, you might recognise that at the lower side of the top cabin section, two shafts coming out of it. It might be that the lower tuning scale being interconnected onto this tuning provision.

This latter might imply, that the lower the lower (vertical) scale pointer becomes the lower the according operational frequency was.

We notice three outlets on top of the Simultan housing. I assume the most right-hand one constitutes the common antenna in/output interconnection. The tube on its left might be for cooling and the one facing left is feeding the RX; the TX input is not visible from this perspective.

Please look at the high tension 'blitz' like red sign on the far left-hand side. This might indicate that real high tension being involved. Would this imply HF? I don't think so purely, my first thought goes in to the direction of a high voltage used for saturating the two diode valves SD 6 as to constitute a very low impedance - necessary for energy cut-off protecting the receiver front-end. The moving coil meter is calibrated in amps and guessing numbers quite some current involved. Knowing these valves physically - quite understandable because a series of parallel tungsten wires constituting the filament (cathode).  

Another side effect of saturated diode valves, is, that these can generate considerable amounts of noise, when matters go worth it might even adding onto the receiving signal. In this respect a Nullode is a favourable switching device.


On this photo we find no indication for such cable provision. But the black coupling box is good visible

The top cabin contains the (wide band) transmitter module - I assume, the lower section being the regular modulator; but, its signal is not de-blocking the transmitter stage though is feeding the driving-power-stage onto the main (high) power-modulator (dealt with next).

Maybe the unusual cabling in the middle on the right hand side might be originating from the additional high power modulator stage.


So far we have dealt in foregoing British reports with a 40 kW anode modulated power stage (and some beyond) - where the transmitter-anode-voltage being provided from the modulator output transformer. As well as the fact that several times is quoted:- ... 22,000 V (22 kV) anode voltage being supplied onto a stage. When this is in conjunction to an anode-modulated power stage, we may consider that the modulator itself being supplied with 22 kV anode voltage. Quite well in accordance to the application of two R-Geräte (medium and high power power supplies) interconnected in series.


During the course of this Survey we have to confront ourselves with Siemens' high power (HP) transmitter version. I did encounter so much confusion that we have to dig into this module in quite great depth.

Some of you might ask - is this worth it?

I would like to reply - everybody dealing with the history of technology needs to ask himself, from time to time, does it really make sense what I am doing?

Most will correctly say no, it does not!

Nevertheless, having quite unique materials at hand, and having the knowledge of understanding what the subject is about - at least we should invest some time in the following matter.

In Holland we say:- Het bloed kruipt waar het niet gaan kan.



On 6 November 2015

Please re-read this chapter, because someone turned my attention onto the Jagdschloss manual again, which after all gives a completely different - and far better understanding of what it all is about.


On 28 October, I received an e-mail from Thomas Van de Velde from Belgium, who referred first to the open TS 41 queries, then he mentioned the Jagdschloss manual. This aspect I had neglected, why, I don’t know.


Therefore I have to enter this subject again.


An example of a 100 kW modulator stage. Operated for Jagdschloss- as well as some Wassermann systems (the latter maybe experimental)

You can have a PDF version of this photo in 600 dpi, just click at it

The construction is a bit cramp, but apparently it has to fit inside a regular GEMA cabinet, about similar as was used for the combined transmitter/modulator stages. 

It might seem that two quite heavy wires leaving the bulky transformer on the left-hand side being interconnected onto one of the high voltage ceramic capacitors, but careful viewing shows that this it isn't the case. This module uses 4 x TS 41. Each valve consuming 10.5 A at a filament voltage of 10.2 V. Considering the outgoing wires from both sides of the transformer, I assume that each transformer side supplies a pair of TS 41s thus providing 22 A.

Considering the same photo but combined with my vision on what the module is about.


Again, all is the result of very careful looking for tiny details, often just visible, but eventually conclusive

The input transformer is not valid, but the cylinder most likely is a high tension input capacitor, described later!

Choosing photos is always delicate, as one always would like to know what is another perspective offering us.


Viewing it from another perspective

You can have a PDF version of this photo in 600 dpi, just click at it

The transformer is quite bulky, it does not look like a pulse-transformer for 500 Hz but more being a filament transformer device.

However, what might be the case, is, that the sealed-off metal box in front incorporates the actual pulse transformer.

Considering the grip on the left-hand side - we may assume that this section is facing outwards when the module being inserted into its cabinet. 


This perspective causes quite some confusions

Pre-driver output transformer is not correct, details dealt with later.

What are we looking at?

How does interact the various components?

The final text version had at least to be rewritten four times; why?

Because it really took time to unravel all the components and wires we are currently looking at. 


Great confusion brought the fact that we are looking at modules that underwent change in between both genuine photos had been taken

Pre-driver output transformer is not correct, details dealt with later

Visible is the fact that the on the first two photos all three PA valves being linked together and only two modulator PA valves were operational on the other.

We may thus draw the conclusion - that what is shown is a module in an experimental stage.

When I may draw a conclusion, my comment is that it might have provided between 40 - 100 kW pulses onto the Jagdschloss and Wassermann transmitters; but its concept was quite crude, as we hardly meet provisions for blocking off hum and ripple. On the other hand it provided every 2 ms a single pulse of only, say, 2 µs duration; in between each cycle the transmitter (thus the modulator) was kept in its off mode as to allow reception of returning target signals (2000 µs - 2 µs = 1998 µs pulse pause).


Continuing after Thomas Van de Velde's hint from

28 October 2015,

with a schematic derived from the Jagdschloss A manual.



Comparing both my perceptions isn't too bad! But, there are differences as well


Ü 2 being my transformer at the most right-hand side

The main difference lays in the input circuit of the driver stage. Input circuit drawing faulty. Being corrected hereafter.

But, also the information on the anode voltage used: 8 kV, whilst it should have been at least  ≥ 22 kV according other sources of information.

About the quotations within British reports, these rather often deal with anode voltages of 22 kV.

I first considered when 22 kV would be also supplied onto the HP modulator the output transformer ration might have been 1 : 1.

But, apparently in the Jagdschloss A manual they deal with a 8 kV modulator HT; in that case they necessitate a voltage step-up in a ratio of 1 : 2.5 (Ü 2)

The driver transformer Ü1 had a transformation ratio of 2.5 : 1. Consequently, the conveyed anode-pulse-voltage is being reduced by a factor 2.5. For example, 6 kV pulse being reduced to 2.4 kV. 


However, comparing both the Siemens photo with the modulator schematic, the common grid resistor (W 6) is not interconnected onto W 2.

Another matter is the components constituting the driver input circuit. After consideration, I have to revise my previous assumption - that the circular device was a choke of transformer.

 The power modulator schematic (Hochtast Gerät) tells us that the dark cylinder is a high voltage (ceramic?) capacitor of 10,000 pF.


In that case it obeys to the modulator schematic


Do you see an extra wire?

Please click on the above photo as to get it reproduced in 600 dpi.

The only strange matter I can see, is the black-ring like provision on the tap of the pulse driver transformer and the lead from RC combination on the right-hand side.

Maybe this query will be never answered entirely.


The schematic of the pre-modulator driving stage


Continuing on the line set out in the Jagdschloss manual


On 30 October 2015

I decided on this day:- that we should consider the under-laying theory of the high power modulator (Hochtast-Gerät)


Quoting from page 55

    Das Hochtastgerät

Aufgabe:    Das Gerät befindet sich in einem besonderes Gehäuse in der Nähe des Senders. Der Einsatz (Hochtastgerät or high power modulator module, AOB) is einschiebbar. Die Verriegelungs-Schleife wird, ähnlich wie bei den anderen Geräten, über Deckelkontacte (interlocks, AOB) geführt, die sich oben am Deckel befinden. Die Spannungszuführung erfolgt über Messer (Kontakte, AOB). Der Impuls wird über ein abgeschirmtes Kabel geführt. Eine Kühlung ist wie bei den anderen Geräten vorhanden.

Wirkungsweise:    Um zu der für die Anodentastung erforderlichen Leistungsabgabe zu kommen, verwendet man eine Gleichtaktverstärkung. Nach einer Vorstufe (TS 41) werden 3 TS 41 parallel geschaltet. Die Anodenspannung von 8 kV kommt aus dem R-Gerät (external power supply R, AOB). In einem RC-Glied wird durch den Gitterstrom der positiven Impulse automatisch eine negative Gitterspannung erzeugt (dealt with later, AOB), die die negativen Impulse unterdrückt (Gittervorspannung bei einer Anodenspannung von 10 kV etwa - 1,8 kV). Der positive Impuls, vom Steuerteil kommend, erzeugt an der Primärwicklung des ersten Impulsübertragerers (Ü 1, AOB) einen negativen Impuls. Anodenspannung bricht zusammen)

Es soll nun die Übertragung vom Impulsen über Transformatoren etwas näher betrachtet werden. Dabei wird vom Ersatzschaltbild des Übertragers ausgegangen. Die Sekundärseite des Transformators soll mit dem Gitterkreis der folgenden Röhre belastet werden (Kapazität Ce) CP stellt die Kapazität der Primärwicklung dar, CS die der Sekundärwicklung. RP, RS stellen die ohmschen Widerstände der Wicklungen dar (Bild 1)    Bei weiterer Vereinfachung ergibt sich Bild 2. Dabei ist:

R = RP + (W1/W2)2 RS

C = C1 + Ce'

C1 = (W2/W1)2 Ce

Ce' = (W2/W1)2 Ce

Die Kapazität der Primärwicklung CP soll gegenüber C1 und CE'vernachlässigt werden. CE und Cp sind etwa von der gleichen Grössenordnung, so dass CE' > CP wird. Ferner ist wegen der grösseren Windungszahl CS > CP. Bei der Übertragung von Impulsen muss ein Frequenzband übertragen werden, das um so breiter sein muss, je grösser die Formtreue der Impulsübertragung sein soll. Die Begrenzung erfolgt durch die untere und obere Frequenz.  Bei tiefen Frequenzen geht der Widerstand ???? (I cannot read the rest of the text line, AOB)


Continuing with page 56

Quoting further:

Man kann daher für die untere Grenzfrequenz schreiben: ωu = (Ri + R) / L. Die obere Grenzfrequenz ist bedingt durch die Resonanzerscheinung zwischen C und L0  

ω = 1 /√Lo C .

Für eine formgerechte Übertragung muss das Verältnis für ω0 / ωU sehr gross werden (breites Band). Dadurch Einsetzen erhält man:

ω0 / ωU = [ (1/ √lo) C ]  [ L1 / (Ri + R) ].

Das Verhältnis wird umso grösser, je kleiner die Streuung (meant is mutual coupling; all magnetic field-lines in de primary coil section should be transferred into the secondary coil section. But, there are field lines that do not contribute to the secundary coil output, these being lost,  AOB) wird. Mit Speziallegierungen gelingt es, L0 / L1 auf Werte von 10-4 zu bringen. Bei besonderen Wicklungsanordnungen erreicht man sogar Werte von 10-5 - 10-6. Ferner soll Ri + R klein sein, d.h. man wird mit Vorteil Trioden verwenden. L1 soll nicht zu gross gemacht werden, da eine Steigerung von L1 auch R vergrössert. Ferner soll noch C klein sein. (consider the foregoing Bild 1 and 2, AOB)

Einen Impuls kann man sich aus zwei Vorgängen zusammengesetzt denken.

1    Aus einem Spannungsstoss von -1 nach +1 zur Zeit t = 0 und

2    Mit einem Spannungstoss von +1 nach -1 zur Zeit t = τ . 

Die Summe der beiden Spannungsstösse ergibt den Impuls.

Es soll nun die Impulsform in Abhänigkeit von ω0 / ωU betrachtet werden. Für den Ersten Stoss ergeben sich auf der Sekundärseite die in Bild 3 und 4 gezeichneten Kurven. Man sieht dass der Impuls allmählich ansteigt. Die Laufzeit wird umso grösser, je kleiner         ω0 / ωU ist. Für den zweiten Spannungsstoss ergibt sich dasselbe Bild mit entgegengesetzten Vorzeichen um die Impulsdauer nach rechts geschoben. Die Überlagerung der beiden Spannungsstösse ergibt die Impulsform auf der Sekundärseite (Bild 5). Impuls gelangt an die Gitter der drei parallel geschalteten TS 51. An der Sekundärseite des Ausgangsübertragers entsteht ein positiver Impuls von etwa 20 kV und 300 kW. Die im Anodenkreis eingebauten Induktivitäten L1 - L3 und die Kapazitäten C3, C4 dienen dienen zur Unterdrückung von Eigenschwingungen. (preventing self oscillations, AOB)

Stromversorgung:    Die 4 Röhren TS 41 sind mit ihrer Heizwicklung parallel geschaltet und werden von einem im Hochtastgerät eingebauten Heiztransformator gespeist. (what I can see on one of the Siemens photos - is that most likely they used two filament supplies, AOB)

 So far the additional genuine German circuit description, implemented on 30 October 2015.







Viewing the g1 and anode pulses of the final power stage, consisting of three parallel TS 41 valves; apparently valves 2, 3, 4

For better understanding we have to follow some lines of the German description first.


Let us transcribe what is being told

Überprüfung der Arbeitsweise des Geräts

    Die vom Steuerteil ankommenden Impulse (positive und negative) kommen an das Gitter der Rühre 1. An der säkundären Seite des Übertragers Ü 1 ergibt sich am Punkt P 1 der in Bild 1 dargestellte Spannungsverlauf. An den Anoden der 3 parallel geschalteten Röhren zeigt sich am Oszillografen das Bild 2. An Punkt P 5 ergeben sich die in Bild 3 gezeigten Impulse, die in Bild 4 mit gedehnter Zeitachse nochmals zu sehen sind. Die Messungen erfolgen mit Philips Oszillographen, der an die Meßstellen kapazitiv lose angekoppelt wird (2 kV Vorsicht). And very much higher voltages!

For me still an open question - how are the real potential levels? According the foregoing explication, measurements had been accomplished by means of a probe in the vicinity of  wires concerned. We have therefore not yet an idea how the actual potential levels were.


Anyway, all foregoing considerations forces me to modify my schematic of the high power modulator stage

One of my major concerns was: - what prevents the power stage to be kept blocked in the pause interval.

Please remember: PRF was 500 Hz or 2000 µs. Let us estimate that the output pulse was lasting, say, 2 µs. Hence, pause time becomes 1998 µs. 


By means of this principle a kind of 'automatic' negative bias had been created

Grid current flows from grid to cathode (filament) and then entering the grid resistor from ground. It is regarded - that the point where a current flows into a system, that this point being positive. The other end of the resistor constitutes a negative potential against ground. This negative voltage is essential for our proper operation, as well as protecting the stages during pulse-pause-duration.

The RC combination is (also) constituting a time-delay circuit.

τ = RC = 106 ● 10-7 = 0,1 s

I my perception quite long in respect to the 2 ms (0,002 s) lasting PRF cycle.

However, capacitor C2 did have another fundamental function, it should act as a low impedance as to close the circuit between cathode (filament) and the driving secondary winding of the driver transformer. Because, driving energy, or call it voltage, should be applied between grid and cathode. What counts here, is, that they dealt with quite sharp pulses constituting by it very nature a range of harmonics (signal bandwidth).  


The Jagdschloss A manual also provides an interesting instruction on how to check the power modulator final stage. (Hochtast-Gerät)

Quoting the main text section:


    Das Gerät wird mit sein Gehäuse eingeschoben, die Netzspannung und der Impuls angeschaltet. Die Netzspannung muss wieder genau 220 Volt betragen. Sie darf auch bei voller Belastung durch das R-Gerät (main power supply, AOB) nicht absinken. Der Ausgang des Übertragers Ü 2 (pulse output transformer, AOB) wird mit zwei parallel geschalteten Widerständen von je 4 kOhm und 100 Watt Belastbarkeit abgeschlossen. Es sind dazu Schicht- und keine Drahtwiderstände zu benützen. Die an den Widerständen auftredebden Impulsspannung wird einen Spitzenspannungsmessgerät (Bereich 20 kV) gemessen. Der Strommesser  im R-Gerät (Bereich 45 mA eingerichtet sein, die drei Widerstände von 20 kOhm im T-Gehäuse (pre-modulator stage + TX, AOB) sind kurzzuschliessen. Die Zusammenschaltung von R-, T-, Z- und Hochtastgerät (high power modulator stage currently here discussed, AOB)  ist aus beiliegendem Verkablungsplan ersichtlich.

    Beim Hochfahren der Anodenspannung auf 10 kV ergeben sich folgende durchschnittliche Betriebswerte:

        Stromaufnahme                                    ca. 30 mA

        Gittervorspannung                                ca. 1,8 kV negativ gegen Erde (this we are looking for, AOB)

        Impulsspitzenspannung                      ca. 20 kV (pulse output voltage, AOB)


Die Gitterpannung wird am Verbindungspunkt PA (Ü1 - W 2 - W3 (should be W 6, AOB) C2 gegen Erde mit einem statischen Voltmeter gemessen und ist abhänig von der Höhe der Anodenspannung. (Hence, the grid bias being measured by means of an electrostatic volt meter, as to prevent loading of this quite high resistive circuit, AOB)

    Bei der Messung der Impulsspannung ist zu beachten, dass diese u.U. (= unter Umstände, AOB) erst nach einer Betriebszeit von etwa einer halben Stunde auf ihre volle Höhe steigt. (Also z.B. von anfänglichen 18 kV auf 20 kV). Es hat diese seine Ursache in der Röhrenstreuung und der während der Einbrennzeit noch stattfindenden Aktivierung der Kathode. (hence, the brand new valves do increase performance during the first hours of operation, AOB)   

    Die Prüfung des Ausgangsübertragers Ü 2 auf Überspannungsfestigkeit erfolgt im Leerlaufbetrieb (Belastungswiderstände abgeschaltet). Da in diesem Ausgang von 30 bis 40 kV auftritt, ist diese Prüfung nur stichprobeweise und kurzzeitig (etwa 5 sec.) vorzunehmen. Die Serienprüfung erfolgt mit reduzierter Anodenspannung (8 kV).

    Behelfsmässige Prüfung

    Bei Zusammenschaltung des Hochtastgerätes mit einem Eibsee (Bereich I) (meant is the broadband TX version for the regular lower spectrum 119 - 126 MHz, AOB) muss letzterer mit mittleren Bereich etwa 100 kW Hochfrequenzleisung (HF output power, AOB) abgeben. Diese behilfsmässige Prüfung ist dann vorzunehmen, wenn die entsprechenden Vorrichtungen nicht zur Hand sind. Eine Untersuchung auf grobe Fehler hat dann selbstverständlich vorauszugehen.


    We may draw the following, simplified, value of the accumulated grid currents through W 6.

Given is: 1 MΩ and 1.8 kV is applied across it:

U = I x R →  I = U / R → 1800 / 1000,000 = 1.8 mA

about 450 µA per valve (when we consider 4 valves roughly)

According a previously quoted description, is the grid current the result of the driving signal amplitude passing the actual grid-voltage-space; also known as Class C, its efficiency might be up to, say, 70%.


Let us close this chapter 8c with a TS 41 data sheet. AEG was the company who delivered to Siemens the valves incorporated in the high power modulator dealt with.


TS 41 Data sheet

Interesting for us might be:

Emission 5 A at 800 V

- 2kV grid voltage.



Going through the genuine Jagdschloss trainings manual another interesting aspect is found.

Let us first remember how the Simultant Gerät, or T/R switch did look like.


I chose the open version, as you may recognise some similarities between this photo and the next technical drawing

Please notice that what is visible above is (partly) just turned upside down in the schematic below. Quite easy to understand at what end of the system the two SD 6 diodes being shown.


The schematic of the Reißersee I  Simultan-Gerät or T/R switch (low band spectrum)

(Jagdschloss A trainings manual)

We may consider, that the tuneable section K is representing the the upper section of the foregoing Reißersee photo. And, the Posaune (trombone) is constituting the lower apparatus section.

It is clear that the two diodes are of type SD 6

By the way, Ant. means the common connection onto the outside antenna; N is connected onto the receiver front-end; and T is the transmitter.

A curious detail, the receiver NE did all use symmetric front-end input, implying an additional balun provision.


Electrically the circuit can be brought back to the next circuit.


The two diodes are interconnected at 'b'.

(Jagdschloss A training manual)

T = is the TX and N represents the receiver front-end.


Please notice the trombone named Pos.


This might lift the secret about the reason of the 'magic black box' between the TX output and the Simultan-Gerät (T/R switch) input.


According this information the TX output impedance was (symmetric) 850 Ω

(Jagdschloss A trainings manual)

The symmetric line having an line impedance of 250 Ω

The balun differed from the standard Sperrtopf type (1 : 1), as it transforms in a ratio of 1 : 4.

The meaning of the text is: that when the TX is in its off-mode its impedance is ca 4000 Ω after the balun is not found 70 Ω but say 300 - 350 Ω .


Isn't it plausible that this provision is kept within the magic black box?

Apparently, the old Freya type transmitter did have a lower impedance, but symmetric as well.


Another aspect


Dealing currently with high power provisions in respect to the Wassermann system, was its actual HF concept really allowing high power transmission of up to 100 kW?

I would like to look critically at some of the (high power) bottlenecks.

Heavier power stages have been manufactured by Siemens, but I strongly doubt whether the (Wassermann) auxiliary antenna components could really cope with the according high tensions.

Let us remind that:-    P x R = U2  → 100,000 x 70 = 7,000,000 → U = 2650 V (ac)


Do you honestly believe - that the shown centre connector can cope reliably with 2650 V ac often within a quite humid environment?*

* Revision

Henk Peek discussed my doubt whether this connector could stand say 2600 V ac?

He rightly pointed - that it is ionisation that causes sparking and ionisation needs time for building it up. He suggested that 2 µs is a too short interval for it. I pointed, that a humid environment is causing a rising level of ionisation. Whatever the truth might have been, a weak point it still was.

40 kW might have constituted the maximum, which according a foregoing brief calculation gave 1673 V, but 1000 volts ac more?

I once have been taught briefly that:- one have to calculate for 1 mm air spacing 1 kV, when no additional dielectric is incorporated.

Nonetheless, the real - quite heavy coaxial connectors on either side (left and right) of the split beam relay, could doubtless cope with such high voltages.

When the critical sections would once have been kept pressurised it would not have caused nuisances; but it definitely was not!

One might reply: - they took a risk but the circumstances forced them to apply for heavier beaming power.

Another option might have been, to replace critical HF -  HT components.



We have gone through the power modulator (Hochtastgerät) and dealt with query whether 100 kW transmission power was feasible.



On 8 November 2015

Mike Dean did send me a Danish language report copy on German communication and radar systems captured in their country.


Kommissionen til Besigtelse

af de her i Landet oprettede

tyske Radiostationer.


Kort Rapport

over tyske Radiopejlanlæg.

ad 654/1946

The title as being given


I  was intrigued by some drawings on Wassermann antenna and related techniques.


Wassermann - (a part of) a broadband antenna group and Sperrtopf

(post war Danish report, courtesy Mike Dean)

"Balanceringsfirpol" might have stood for: 'balancing four pole'. Also known a balun. The Germans called such device Sperrtopf.

One aspect is bothering me:

I do wonder however, apparently dealing with a wide band system between 115 and 160 MHz; and the balun or Sperrtopf is having a ¼ λ system incorporated. However, ¼ λ  means a quarter wave length; though - at what wave length?

I assume, that the equilibrium being chosen somewhere in the middle of both maximum deviations; thus at about 137.5 MHz.


Wassermann was designated by the Allies as: Chimney; which quite well correlate with its tall construction.

Basically it consisted of two vertical antenna groups.


The two vertical groups each consisting of 12 dipole groups

(post war Danish report, courtesy Mike Dean)

Shown is an example of a broad-band system.

Group 1 -12 looking 2.5 degrees to the right and antenna groups 13 - 24 facing 2.5 degrees to the left (viewing it from an operational perspective)

The earlier models used, like most other radar systems, narrow band antennae.

During the course of the war, systems were improved and provided with height finding facilities. For it each vertical group being split into two sections; according the following scheme. Please rotate virtually this drawing 90 degrees clockwise; then becomes group 1 the top section and 12 is the antenna group nearest to the surface.


Each antenna side constituting two antenna sections each consisting of 6 groups (from ground to the top: 2 x 6 = 12 groups)

(post war Danish report, courtesy Mike Den)

Let us consider how the grouping have been accomplished.

Antennae 1 and 12 constituted the (vertical) most far end groups; thus the most outside systems.

Hence, group numbers 6 - 7 were adjacent to one another just up and down the virtual (horizontal) middle.

When the antenna pattern had to 'look' without 'beam tilt' the phase shifter or Wellenschieber should deliver equal signal phase onto all groups; hence, being set just in a middle or centre position. The delay-line helix having equal phase shift towards either antenna side (up and down).

It is also evident, that when the phase shifter is set into an extreme position (maximum phase shift) that the phase shifter connected onto 1 or 12 is prevailing at one of the two groups. These antenna groups being mounted several wave lengths out of the virtual antenna centre. It should therefore being fed retarded accordingly.

Bear in mind: how beam tilting had been accomplished.


It is understandable, that the antenna groups 6 and 7 will be also differing in signal phase but - of a far less value.

When the most outside antenna group has to be brought in line with the middle groups, the most outside needs a faster delay-time change - then do need the centre antenna groups 6 and 7.

To cope with it, each delay-line was driven with a different speed reduction ratio:                 1 : 3 : 5 : 7 : 9 : 11.

The (reduction) succession is to be understood being gearing ratios. Thus, the most outside groups were geared with 1 (thus no speed reduction) and adjacent group viewing it in the direction of the antenna centre - were driven with a gearing ratio of 1 : 3; and the helix of the group adjacent to the centre moved most retarded with a ratio of 1 : 11.

Hence, when the most outside groups reached centre the most inside groups should reach centre as well.


We hereafter will see how the Wellenschieber delay line had been accomplished. But, please bear always in mind - that the helix in each delay-line section (or bay) rotated (and thus moved along) with a different speed.


Let us close this section with a reminder, a drawing used manifold on this webpage.


It can clearly be recognised that the outside delay-line group (1 and 12) has to rotate faster than the group attached onto the virtual antenna-centre  

(Frequenz 1955/10)

Not yet discussed is the top section in designated Fig 3 Danish post war drawing.

Antenna groups A and B

Sende-modlageantenna for Feind-Freund Kennung (German wartime expression), know as I.F.F.

The foregoing Danish drawing shows it with:

T Stykke

This provision might have implied a left-right or split-beam operation.


The designated: Relae og Forsinkelse til Straaleskifting, samt T-Stykke;

meant is the split-beam switch previously dealt with (then - whether it really could have stand 2685 V ac?) Henk Peek suggested, that ionisation needs to be build up, and 2 µs might have been too short, but it was repeated 500 times per second, and how fast was existing ionisation vanishing?


The next drawing is in some respect of interest, as it provides measures to the Wellenschieber construction.

"Endlich" (Finally) technical facts; all drawings encountered did fail in this respect.


Thanks to Mike Dean's commitments and a post war Danish drawing we can calculate roughly what the line impedance have been

(post war Danish report, courtesy Mike Dean)

Because the driving cog-wheel or sprocket being kept fixed, rotation will causing a rotating as well as moving helix (like a screw moving in or out). We know from previous explanation, that at the centre of the helix housing tube that there is a cable connection (junction); the contacts consisting of a pure silver brush.

My personal fear, is the brush really kept just touching onto the (deposited) conductive silver layer only; or is it now and then short-circuiting adjacent windings?


   Another critical point of my concern - is the space between the conductive helix-layer and the inner surface on the screening tube.

66 - 60 = 6 mm


6 / 2 = 3 mm.

Is this a sufficient (operational) space - when at 100 kW transmission power at some places they will encounter, say, 2.6 kV a.c.; in a sometimes hostile environment?



Let us do some brief calculations on the measures provided.

 I would like to rely on the equation (formula) used in Rothammel's antenna book, where he gives for a concentric line (equations can be found elsewhere too):

Z = 138 lg D/d

(Antennenbuch Karl Rothammel, 8th edition, Militärverlag, DDR, 1974, pages 73-74)

The shaft on which the helix was mounted is given for 20 mm and the tube in which it forms a coaxial system being 66 mm

we get now: Z = 138 lg 66/20 = 71.5 Ω, quite near to 70 Ω


However, how did they once calculated the impedance of the helix within the same tube diameter?

Using the foregoing equation:

 Z = 138 lg 66/60 = 5.7 Ω

What goes wrong here?

How does influence the helix the coaxial line impedance?

What may play a crucial role, is the fact that the helix system is not entirely coaxial or concentric; implying that both conductors having (not) the same centre (or point) of radii.

It just popped up in my mind - capacitance against ground is, on the other hand, also necessary for proper delay-line function.

We might encounter contravening physical phenomena.

I myself admit - that I do not have the necessary expertise.

In this case - a real expert should answer: - as to how it all obeyed to physics theory.



On 16 November 2015


The seemingly never ending Wassermann Saga is chasing me day and night.


First the existence of the next photo: 


The so-called Wellenschieber or phase-shifter device

Genuine text: Wendelleiter einer Wellenschieber-Umwegleitung*

(Frequenz 1955/10)

* Where stood Umwegleitung for? Umwegleitung is often a provision by which means two systems being compared and a phase shifting line (Umwegleitung) being introduced as to create split signal provision.

I cannot see any sign of such provision, in conjunction to the Siemens Wellenschieber apparatus and Wassermann.     Umwegleitung accurately must have been an additional route or trajectory, but the Wellenschieber did constitute a variable route. My suggestion:- Wellenschieber-Kompensator or route-compensator would have been, in my perception, more appropriate.


There is something curious with this photo:- Is this device a real one or a dummy?

Does it give the impression the windings are real windings or constituting only circular sections?

I cannot find signs of real windings as then the 'groves' should run under an angle (pitch) and not being (seemingly) perpendicular to the cylindrical axis. Also the narrowness of the 'winding separations' do intervene with proper signal pick up without shortening between windings (due to the size of the tapping silver-brush). The cog- or sprocket-wheel is not what I have thought for a long time - a contact- but only a provision for driving the rotation of this helical device.



let us now consider how such device finally looked.


There exists quite some differences between this- and the foregoing 'idealised' Wellenschieber or phase shifter

(Courtesy Michaël Svejgaard)

Considering the white colour of the cylinder - this most likely is consisting of a ceramic carrier or body. The silver conductor is also good visible (not yet much oxide). Seemingly, spacing between windings is also wider. At least providing a better chance of proper tapping without short circuiting phase-shifter windings.

Please bear in mind:- that the junction or tapping-brush consisted of pure silver. Such a device cannot having a too small diameter, as power current bursts must be dealt with.

P = UI  →  I = P / U

100,000 / 2645* = 37.8 A

* Value have been calculated previously.

It has to be noticed though, that this current was split into 6 equal trajectories for each Wellenschieber section. Hence, current load remained only 37.8 / 6 = 6.3 A

This might seemingly being a moderate value, but when the (tapping) silver-brush does not make proper contact, quite some problems may occur. Although, this current burst flows only for approx. 2 µs and 500 times per second.

I estimate - a manageable value.


Another point of discussion, is, what was the actual impedance of the phase-shifter?

Someone quoted what Zetzmann did say in this (Frequenz paper of 1955)

Quoting from page 353:

... die Wendel der koaxialen Leitung von 35 Ohm.

Is this really true?


  The Danish authorities luckily investigated the Wellenschieber in quite great detail

(post war Danish report, courtesy Mike Dean)

We have already calculated the in- and/or outgoing line impedance at the ends of the helix (Wendel).

Let us repeat the calculation:

Z = 138 lg D/d = 138 lg 3.3 = 71.55 Ω

When the helix (Wendel) impedance would have been 35 Ω there would, consequently, exist an SWR of 1 : 2. In my vision unacceptable - because there would be a power loss of roughly:- 11.1 % or say a return loss of about 9 dB.

In my perception impossible for such advanced system.

However, what is more likely is the fact that Zetzmann meant the impedance at the junction or tapping point. We may consider that at the junction or tapping point two coaxial systems of 70 Ω being wired in parallel and hence becoming: 70 / 2 = 35 Ω. 

There simply is no way around this circumstance; all as long as the SWR stays with neglectable proportions. (say 1 : 1.1 or that like). On the other hand, cable length does also 'smooth mismatch'; hence, the resulting SWR

Another consideration - the impedance of the phase-shifter device (Wendel or Helix).

Why not considering being: a helical strip-line?

From this latter drawing we know some details, nowhere else being given:

Helix or Wendel length 510 mm; Helix or Wendel body diameter 60 mm; delay-line coil having 33.5 turns, of which - I guess - that about 30 windings being used. The helix or Wendel rotated within a tubular construction having a diameter of 66 mm, leaving a surrounding free-space of 3 mm. 

Hence, when the Wellenschieber being set in its central position (delay line equilibrium) that, say, 15 coil-turns left and right being operated.


The shortcoming in my previous considerations - as to how we can calculate its impedance constituting a coaxial line? My first premise was apparently faulty.

What I have neglected - was that the inner section of the helix device is (quite) open against ground.

Hans Jucker suggested that Gerhard Megla's book - he used during his study at the ETH:

Dezimeterwellentechnik, Theorie und Technik der Dezimeterschaltungen - might solve the queries.

He did send me a single page section (later followed by a complete paragraph). I estimate that this might not solve our problems yet, because this chapter deals with shortening (tuned) Lecher lines by means of the implementation of coil-wound provisions. And Zetzmann quoted that the Wellenschieber bandwidth covered four octaves; thus equations where λ being involved can hardly be used.

I instantly ordered a book copy via AbeBooks.de

What I desperately would like to study in great depth.

It definitely has the potential of providing a solution.

One lesson has been learned - don't trust without some deal of scepticism so-called expert papers. Especially not when more than a decade have been passed since - his commitments and writing a paper on it - have passed. Some details might have been remembered - but just the crux being mixed-up. Like occurred in our case.

Zetzmann certainly was an expert; but be honest - do you really check all your details when using it on another occasion? I must admit - I might fail either.


Before we can give it a trial - calculating the impedance of, for example, a virtual strip-line we need to have an approximation of the measures involved.

I did import the foregoing photo into my CorleDraw program and compared the measures we know from the Danish 1945 report and the measures derived from the photo; it was possible to calculate a certain multiplication factor, in our case being: 7.08.

A nuisance, with photos of reduced resolution, is the point where digital pixels do become blurred. The art is just trying to find an equilibrium (sharpness versus realistic dimensions).

My first attempt was, to take the helix diameter measured on the picture and multiplying it with the multiplication factor of 7.08. This provided 60.11 mm which must be regarded quite good as it should have been 60 mm.

Measuring the broadness of the silver conductor (layer) quite some uncertainty being involved. We may, nevertheless, assume given the circumstances, that it must have been between 10 and 12.5 mm (1 - 1.25 cm); a value which might have been realistic.

Finally checking the helix drum length. In the CorelDraw picture I measured 71.624 mm, multiplying this with 7.08 we get 507.09792. Given is 510 mm - don't you agree that this value is quite well proving the exactness of our way of deducing other measures (sizes) as well? 

Repeating the same way of measuring, but now referring to the Wendel or Helix shown in Zetzmann's paper; I got a multiplying factor of 2.58, when referring to the Wendel or helix drum length.

Accessing the broadness of the conductor it gives 12.78 mm. I would tend to believe that its actual coil-conductor broadness must once have been between 10 and 12.5 mm. Quite well in accordance to my foregoing calculation.  



On 5/6 December 2015

I would like to give it a new attempt to calculate the impedance of the Wassermann Wellenschieber delay-line system.

Using the book: Dezimeterwellentechnik by Gerhard Megla, fifth edition, issued 1961 DDR. (Issue 1962 was no longer published in the DDR but in Western Germany; the reason - was that Megla escaped to the US in 1960, where he became professor as well. In these cases, a person became in the DDR: persona non grata. All his beings and intellectual existence was to be ignored since) 

It is apparent, that Megla dedicates only limited attention onto the aspect of the impedance of so-called Wendelleitungen.

Before you start digesting the following contribution, I would like to add what otherwise might cause some confusion:

'a' is the distance between the centres of a single winding; I calculated: 1.5 cm. However, in the equation is not dealt with 'a' but n - the number of windings over a unit of length in casu 1 cm (values being expressed in CGS measures). I did short-cut a and n and regard a = n; what is not entirely correct, but is where the equation is dealing with.



My first aim is to rely upon equation 4.36

(Gerhard Megla page 109)

Actually all three equations are equal to one another, albeit, in some respect notation  do differ.

Do they?

No, not really!

I therefore would like to use 60 /√εr instead of Z0 / 2 π√εr

Both originating from the impedance of free space which is calculated for: 120 π ≈ 377 Ω.

The first point of consideration - what should we do with the value of: √εr ?

In my perception, εr is constituting the dielectric between the two conductors involved.

Let us consider what system parameters we have to deal with: 


  Homogene Verzögerungsleitung (mit gewendeltem Innenleiter)

(Gerhard Megla page 108)

There is nowhere an indication where ε comes into account.

We may, nevertheless, assume that ε should be considered being a dielectric in between both conductors (the coil or helix and the inside of the conducting cylinder surrounding it).

For simplicity we have to neglect it and putting ε = 1, as the only dielectric medium is air. Maybe not entirely correct, but we have also to take into account, that this book still relies upon CGS measures. Hence, we should calculate in cm and (gram) instead of metres and kilogram or Newton.


Let us also reconsider the measures of the Wellenschieber delay-line system.


Luckily we can rely on real measures and no estimations

(post war Danish report, courtesy Mike Dean)


An important measure is 'a' or the spacing between the centre of a winding (per unit of length in cm).

According to the foregoing drawing the delay-line counted 33½ windings.

When we leave out the ½ conductor turn we hold 33 windings.

Previous considerations, based on estimation from scaled up photos, gave a conductor width of 1 - 1.25 cm.

Let us now assume that the conductor width is 1 cm, the surface of the carrying ceramic drum is partially covered with 33 x 1 = 33 cm. We know that the drum length is 510 mm or 51 cm.

We can very simply estimate what the space between each winding is left:

51 - 33 = 18 cm - leaving 18 / 33 = 0.545 cm space between each winding.

We have already decided to deal with a conductor width of 1 cm; distance between coil windings is measured at its centre. Knowing this, we get - a spacing factor a of:

0.5 (half width of the conductor) + 0.5 (spacing between the conductors) + 0.5 (half width of the conductor) = 1.5 cm

My fault first was that I did enter for a = 1.5.

During my proceeding sleep I waked up and realised that my assumption was incorrect, it should have been 1 / 1.5!

a = 1.5 cm it should have been less than a single winding per cma = 1 / 1.5 = 0.666

b = 6 cm

d = 6.6 cm

cot ψ = n π b     (n is the number of turns per unit length in this case 1/1.5)

For simplicity we use: n π b = 1.5 3,14 6 = 28,274 but:  12.55 → therefore

(n π b)2    = 799.419    = 157.598        

b / d = 0.909 → (b / d)2 = 0.8262

d / b = 1.1  →   ln d / b = 0.0953    and: 2 ln d / b = 0.1906  

Knowing these small terms, new calculation generated a completely different outcome:

We get ZL = 223.516 Ω   ZL = 68,78 Ω (neglecting round-off errors)

After having encountered an incorrect assumption first, we finally have reached a value that matches onto the known coaxial line impedance of approx. 70 Ω.

To be sure that I did not make calculation errors again, the entire procedure have been repeated several times.

I must admit, that yesterday a discouraging impedance of about 223 Ω was calculated. My mistake was - that I did enter for a = 1.5 instead of 1 / 1.5 = 0.666!


I honestly confess, that when I started some time ago with finding a solution for the Wellenschieber system impedance, I had quite some doubts about its outcome.


As to let you see what the implication of the actual Wellenschieber helix concept was, I would like to refer onto a graph used in Megla's 1952 book (carrying the same book title as the improved 1961 issue).


Wellenwiderstand eines Wellenleiters b = 1 cm; d/b = 3,6; Z = 70 Ω

(Gerhard Megla's book issued 1952 page 41)

The ratio d / b of 3.6 is about the ratio value of regular coaxial lines with air spacing for 70Ω. What astonishes me is the difference in equation between his 1952 book version and the 1962 issue apparently does have so much difference.

However, I assume that some pressure from the practical and/or theoretical field caused a more comprehensive approach; with favourable result.


Have we reached today a temporary conclusion?

I prefer quoting James Bond: Never say never again.

After a good night sleep, I did remember that the Danish report also dealt with the construction of the 'combiner'. Where the 6 coaxial cables join - interconnecting the Wellenschieber with the radar system.



On 8 December 2015


On the fore-last page of the Danish report we find the next drawing:


'Impedance transformer'

It constitutes a provision by which means 6 coaxial cables join at a single point together, will be matched onto a single coaxial cable (v.v.), which device I would like to designate being a "combiner".

(post war Danish report, courtesy Mike Dean)

An instant comment - I personally doubt whether the dielectric ends that much abrupt; I do expect, however, it should slowly widen, so that impedance jumps being prevented.

A second observation - please view carefully at the drawing and particularly at the outer screens of both - the system on the right-hand side as well as the one on the left. You might recognise, that the screening outer conductors do couple together by means of two ¼ λ tubular stubs, but are not directly earthed (connected) together. I guess, that on the other hand, earth potential was mutual through the Wellenschieber frame as well as through the split-beam switch. But this might constitute long lines with considerable inductance?     

As nearly always in this Survey, another query has been engendered.

Where came these 70 Ω cables from (I mean their apparent impedance)?

Does this imply that the interconnecting cables between the "combiner" and the Wellenschieber tapping points, were standard 70 ohm cables? Or, used they 35 Ω coaxial cables instead?

The latter would have been appropriate, as the, what we previously have calculated constituted a helical line impedance of approx.  70 Ω, where at the tapping point also constituting a point where two 70 Ω line-sections join together - the impedance would become 70 / 2 = 35 Ω.

However, we know also that there are various means transforming impedances. Like, for example,  n x ½ λ lines; these do behave like 1 : 1 transformers, where the line impedance is not really of importance.

But, we would have got to deal with 6 x 35 Ω impedances at the left-hand side of the impedance transformer device.

Where might it have gone wrong?

Thinking that the coaxial cable type was actually 70 Ω, neglecting the fact that the cable impedance should have been 35 Ω?

Or, did they use 70 Ω cables and chose for n x ¼ λ (n = 1, 3, 5 ..) cable length?

Or, does this configuration show a provision operated when the systems did not possess a Wellenschieber? Linking the antenna groups up and down the virtual (horizontal) antenna-centre together?

This latter aspect we have not yet dealt with, but likely was more common than systems fit with height finding facilities through a Wellenschieber. 


Let us give it an endeavour getting at least some measures out of a genuine Wellenschieber apparatus photo:



Please bear in mind, that this photo shows some kind of parallax error

(Frequenz 1955, modified by AOB)

Relying on the measures provided in the Danish 1945/46 report, shown may times on this webpage.

However, on our CorelDraw program I calculated a scaling factor of 10.099, thus, say, 10.

However, we must consider that such method always includes some percentage of error; this time maybe a bit less, because the screen objects are relatively large to one another. But errors must exist and - in some way - have to be counted with.

The second step was taking a thin thread and laying it over the computer screen covering several visible coax cables. The outcome was about 14.1 - 14.3 cm. Hence, multiplied by roughly 10 we get  141 and 143 cm estimated cable length. We have, however, to take into account the velocity factor of the cable in use then. Now a guess: let us estimate, like still today is quite regular for solid dielectrics in coax cables: v = 0.6. Hence, electrical length becomes then: 1.42 x 1/0.6 = 2.36 m  We know that some modified Wassermann systems, in contrast to Mammut, could operate between 119 - 156 MHz ( 2.52 m - 1.92 m); I estimate, that these systems once have been calculated (optimised) for 135 MHz = 2.2 m*. But, on the other hand, we should bore in mind that the way the actual cable length have been determined incorporates some inevitable errors. Let us say +/- 10 %; thus 2.36 - 0.236 = 2.124 m or 2.36 + 0.236 = 2.596 m. Whether we may estimate that a deviation + 10% causes exceeding the band-spectrum and - 10% is more in accordance to the wave spectrum stays open. We might tend to believe - that what was measured was a tiny bit too much optimistic in length (λ becoming a bit longer).

However, the aim of this exercise was to determine roughly whether the interconnecting cables between Wellenschieber and the combiner stood in a relation to n x ½ λ or n x ¼ λ (the latter n = 1, 3, 5... ).     

* We know, that on many occasions transformation relied on ¼ λ sections. But  ¼ λ is only valid for a single wave length. I therefore tend to believe, that they could have chosen the equilibrium of the to be expected wave spectrum, in casu, 135 MHz.

Comparing both - the rough cable length and an estimated cable velocity factor of 0.6 giving 2.36 m electrical response, which is well within the covered wave spectrum. Hence, they most likely operated electrically full-wave cable lengths.      



The cable lengths between the tapping points in the centre of the helical tubes and the what I designate combiner, most likely was having a full (electrical) wave length within the regular Wassermann frequency spectrum.

It has to be noticed though, that Zetzmann in his 1955 Frequenz paper, quotes that the Wellenschieber device was capable of coping within a spectrum covering 4 octaves.

Noticing my forgoing arguments, this latter can be true, but only when divided into smaller band-sections; for various reasons. Because the matching stubs - and baluns (Sperrtopf) and other means relied upon ¼ λ  and/or ½ λ techniques, which never allow direct very broad-band coverage! Four octaves implies a spectrum coverage of 1 : 16! For example - take: 30 - 480 MHz = 10 m to 62.5 cm!

Impossible in this context!



On  15 December 2015


Browsing through the 1945 Danish report again, I came across some detailed information on the Wassermann type IV serial number 1301

Although, the introduction page isn't teaching us matters we haven't dealt with before, the second page is, nevertheless, giving us an inside vision on the apparatus outfit.



We notice that the covered frequency spectrum had been widened even up to 160 MHz (=1.875 m)

(Danish post war report, courtesy Mike Dean)

On top the box in the centre is, what might be expected the Wellschieber as to allow height finding.


For us is the summing-up of significance

(Danish post war report, courtesy Mike Dean)

Some confusion is causing Eibsee Simultangerät and Riessersee (should have been written Reißersee I. I assume, that a mistake is involved, because Eibsee definitely concerned a wide-band transmitter and  'Reißersee I'  being the Simultangerät (T/R switch)

Kreuzeck the receiver we have dealt with on other occasion before

P-Gerät for Grob and Fine DFing    (We might discover that something is curious about it)

HP-Gerät Höhenpeiling (Height D/Fing)

Freund-Feind Kennung (I.F.F.)

R-Gerät power supply to the TS-Gerät, pointing straight away onto the application of a high power I.F.F. transmitter. Relying upon the regular Freya like transmitter providing aboout 8 kW pulses.

T-Gerät, being integrated within the TS housing. Combined with its according pulse-modulator-stage; like in regular Freya.

Gemse-X is not clear to me. Gemse constituted normally the ground I.F.F. receiver.

About the P-Gerät I still have my great doubts.

Please keep in mind, that this unit is only a display facility, always accompanied to an adjacent module carrying the rest of the electronic circuitry. 


Please notice the two CRT scales

(Courtesy Colin?)

Do we see really fine and coarse scales?

I highly doubt this!

Please confirm: the lower scale does have two digits, whereas the upper scale does possess three digit scale number. Do you honesty believe that this once provided a fine and coarse (grob) scale?

I don't!

In my perception: the solution, is, the lower scale provided 0 - 100 km and the upper one 100 - 200 km.

But, we know also, that there apparently existed a provision, that range could be doubled. Thus 200 km became 400; and 100 km should then have been read-off being 200 km.

Yesterday I did study our manual to the Freya LZ part 2. This volume was carrying the following header text:

An das

RLM NVW Abt. 4/1 C

On the right-hand side:

Bestellschein Nr. 0067 - 0073

Berlin - Köpenick den 13.9.1944

Beleg-Nr. 44/ 11 562

Anlage F 2212 LZ 768

However, I was rather astonished to find a PB display schematic. As well as an overall block schematic where the P-Gerät being integrated into a Freya 'LZ' system.

My first perception, was, that the PB Gerät (CRT module shown above) was only used in conjunction to the wide range systems; like was Wassermann. Please remember, that in such case the module consisted of a main-frame. On its right-hand side we find a chassis representing its main function. On the left-hand side we find a unit carrying a single or dual CRT display.

However, some doubt has come up recently, after studying the Freya schematic part of the manual we possess.


This small photo part of the only Wassermann operational gear we know; it shows a 'blow up' of what is visible through the CRT window of the so-called P-Gerät.

 (111 SC 269075 - "US National Archives" courtesy Mike Dean, part handled by AOB)

I have manipulated various picture-parameters as to get the best out of it. Please look carefully, you might see some figures at the upper top half of the main CRT screen.

A query: is the lower CRT screen comparable (in function) to the one in the foregoing  photo?


Viewing a block diagram to: D 1057 FuMG 501102 G01

(Gema manual to LZ 768)

I was really tantalised by seeing this Freya related block diagram, because it implies a P-Gerät module.

I have to apologise for its reproduction quality. Please notice that all schematics being a kind of 'blue print'; where filtering-out blue, likely, implies removing also the wanted cabling as well.

This genuine rather heavy manual consists only of schematics and a complete set of according component lists. All being of the same type of printing.

The schematics being recomposed as if it is a regular schematic again.

But, some schematics are more than 1 metre and some even nearly two metres long (wide). Scanning it - would destroy the paper.

My wife Karin has done here best to fit all together, but finally it did not work out what we have expected.

Therefore it is impossible to let you be part in determining the schematic details.


As to give you an example what it would become, please click at this hyperlink (shown is the schematic of the receiver module fit with a combined OB and NB scope)

I guess, you might agree that such reference is very inconvenient as to be used in a vivid discussion or explanation.

I therefore have to find a fair way in explaining it differently.

I would like, nevertheless, to continue, but you have to trust or rely on my observations.


For our convenience, let us reproduce the foregoing drawing again

Fu.M.G. 501102 G 01  latest date found 11.2.44

Let us focus our attention onto the two screened interconnections (Kb103 and Kb104); both also being interconnected onto the O-Geräte module - hence, the time-base and/or delayed time-base signal.

The N module provided the video output but also necessitated two time-base signals, one delayed by the range measuring unit OK as well as one  non-delayed for the full range display NB.

On the other hand, the O-Gerät also necessitated video for the OB display module.

A second, rather confusing fact found - the P-Gerät being supplied from the two beam-switches (split-beam operation). The transmitter apparently wasn't being operated in split-beam. I guess, because split-beam switches antenna signals, and doing this with high energies might causing failures.


Does this imply that the beam-split signals being displayed at the CRT screens?


Considering first: DP 1001G02

It apparently considered a module consisting of a receiver as well as a dual CRT section.

To me curious, the receiver was equipped with an input wobbling circuit. Why? This kind of technique was encountered in the I.F.F. Gemse receiver. Also is curious - that it consist only of a single IF conversion stage. Albeit it of 3 band-filter sections followed by a detector and two stages of LF amplification. However followed by a second LF amplifier section. Most signals being interconnected onto out-going cables.


Concentrating my attention onto schematic details my eyes were caught by printed information boxes inside the various module (schematic) sections.

Such as (all followed by 101 G 01):

DPB    its designation is in the regular line as B points in the direction of a CRT display module

DPV    a LF amplifier chain; having, however, its own power supply

DPE    being a receiver chain; which signal via a quite confusing route ending up in the CRT deflection circuit.     We might draw the following conclusion: The second frame designation like R, P, O, N ... being chosen randomly, as to confuse espionage. But, the third character designating quite logic symbols. Such as: B → Beobachtung; E → Empfänger; V → might have stood for Verstärker, S (like DTS) for Sender ...


Apparently the two CRTs being fed by a less sensitive (sophisticated) receiver chain.*

* Might this originating from the to be find-out signals originating from the two split-beam switches?


  Quite confusing is the extreme wiring running from far left to right and then to the far left again. For example line number 418 feeding one of the deflection chains. One have to start from amplifier input towards the frame input connection:

line number (lnb) 417 → via (C37) becomes (lnb) 417 → via a level dividing network and (C38) line number 55 → via connector Bu1 → Bu4 then running on the schematic towards the the left-hand side → Kl? (=Klemme) towards another connector; apparently interconnecting the main chassis frame with outside, but getting designation: cable number 9 apparently the number of the final outside connector Bu1. (here we encounter two different designations Bu1).

Let us discover where the second deflection amplifier signal comes from:

Valve Rö 13 gets its signal via line number 407 → (C32) becoming line number 408. Line 408 goes to potentiometer-arm of W72; fed from line 410. → feeding valve chain Rö 11 ... its anode output of Rö 9 is line number 428 → C40 → line number 401 → (C16) becoming line number 299 ending up CRT Rö 4 connected onto the Wehnelt-cylinder apparently for blanking purpose.

It is evident that two deflection amplifiers being fed in series.   

Another, in many respect confusing circuit is found around valve Rö 8.

We find a pentode acting as triode (anode and G2 and G3 being interconnected) and connected onto the high tension straight away. The cathode goes via (W88) and (D2) choke number 2 towards ground. However, the grid circuit being kept clamped by means of Rö7 (but also could have constituted a charging or loading circuit as to create a signal form of saw-tooth), a pentode wired as diode. It is connected with (C413) → line number 59 → (Bu19) pin number (r 16); by the way at the far right-hand side, apparently towards outside or being an incoming signal.    What was the function of the circuit around valve Rö8? An oscillator or even a saw-tooth or that like generator?    Working some paragraphs below, my attention felt upon CRT (Rö4) and I encounter here also line number 59! This time feeding both horizontal deflection systems (dual trace CRT).     This might have an interesting implication: my hypothesis - the circuitry of the oscillator with cathode-base driven oscillator might even have operated at a different deflection frequency. Was this meant for feeding the time-base of the double-range CRT display?    In that case, I would expect a link onto the original time-base signal originating from the Z- and/or O-Gerät. Then line number 59 might have constituted the locking signal; and the circuit around (Rö8) acting as a kind of regenerative divider!   


However, what is the purpose of the signal at the clamping vale Rö7, which also does have an interconnection onto line number 63. → via (Bu1) → (Bu4) to a kind of multi-pin circular connector pin-number 3 designated (Bu16).

This example being only documented - as to show you the complexity of the circuitry without any accompanied description. Please remember that the schematic actually is about 2 metres wide and about 28 cm broad (high)!

Digesting some findings, one get the impression that its receiving chain was, among its internal CRT displays, also fed onto other users.

It is rather sad, that you cannot share the real wiring with me, but without risking to destroy this unique document, its reproduction is out of the question.

What should we do next?

Do we have a complete schematic covering the entire inter-wiring?

We might, luckily possess such a schematic, designated: Zeichnung D 1042                    FuMG???ehstand last date found: 26.6.1944

Might have to be read: Fu.M.G. Drehstand

We therefore may estimate that it constituted GEMA's state of the art.  

 However, this should be our next encounter!



On 19 December 2015


Thinking over what we have just dealt with, I came to the conclusion that it does make sense to show you what the curious circuit is about.


All cable-screens being connected, as usually, onto ground potential

(AOB derived from the LZ 768 manual)

The schematic being redrawn for practical reason; but electrically all is according the matter of facts.

We encounter thus two lines ( 59 and 63) connected onto outside; signal direction not yet determined.

Reconsidering the DP 1001 G02 schematic again, I came across another curiosity. Rö 4 a dual beam CRT being fed from signal line 59, onto both, I assume, horizontal deflection systems; hence, line 59 apparently constituted an output signal line, whilst line 63 was an input line. Albeit, that plates were interchanged; I believe for the reason that one trace paints upwards and the second one downwards. But, now something unexpected found, the other deflection plates being connected onto the main deflection amplifier stage of Ü1 and valve Rö13; originating from the same transformer coil. But, this one possesses a tap connected onto ground. I therefore tend to believe the stage acts in push-pull to one another. But what is then the signal from valve stage Rö8/7 generating?

Curious is also that the second CRT Rö 5

Considering the same deflection system, nearest to the CRT screen, this system being connected onto cable lines 335 and 339.     This can be re-found on Bu3 → Bu6 → Bu? → Bu2 passing though a selector switch onto potentiometer supplied from a Stabilovolt Rö4 (multi-voltage neon stabiliser 0 - 70 - 140 - 210 - 280 V*, also licensed to England and the US, there even existed a CV? number!)

* Sometimes ground being connected onto the 70 volt tap, and 0 is then constituting a stabilised - 70 V

It becomes apparent: that each section did have its own valve numbering and likely components as well. The only exception were the line numbers; as long as these do not pass a selector switch or component.

However, line 339 becomes → line 54 being on the one hand fed via C1 → becoming now line number 469 and being fed on to G1 of Rö1 of the DPV section. Following a LF amplification chain. The same line 54 being interconnected onto Bu10 → Bu2 of module section DPE. This constitutes the output of the wobbled receiver chain. Clearly not as advanced as the receiver of the DN-Gerät also known as N-Gerät (the general Freya receiver).

Why was this receiver spectrum being wobbled?

However, the receiver input connector designated: Bu 3 → Bu 12 → Bu 11 (Außen) Hence, the outside antenna connector where we, among, should focus upon! 

Encountering such queries is symptomatic to 'reversed engineering'; it also constitute the main challenge!

Often the answer lays elsewhere.

Nowadays, we may consider that experts are no longer among us. The only one who I regard an expert is Harry von Kroge. But, I haven't spoken to him for more than a decade.

We have each time to start from scratch again.


The main system wiring schematic might even be 3 metres long!

Continuing: quite discouraging, the cable schematic only covers what once ran onto or from the so-called Schalttafel I and II. Which, with some difficulty, can be determined, is that output cable line from the P-Gerät came-in by means of Kb 3. However, the cable numbers change completely; and only three being visible.

It is no longer possible to determine what happens. For example, numbers are now related onto what contact number appears on the other side of a cable. Therefore, for example, the same wire is designated at one end (63) and on the other end of the same cable (9). The numbers now indicating opposite pin-connector-numbers instead.

The main problem currently confronted with, is the fact that inter module cabling is not being covered in the manual.

This really is frustrating.


I would like to notice, that I am no longer convinced that the display carrying the two CRTs fit with range-scales were once part of a so-called P-Gerät. The query may arise, what else could it (once) have been?

Honestly speaking, I have no idea!

The only option popping-up in my mind, the P-Gerät was once part of the EGON system.

Pointing into this direction is the GEMSE like receiver channel. On the other hand, EGON used the I.F.F. system and operated vertical polarised antennae, whilst the regular Freya systems, certainly the one delivered in 1944, operated for radar horizontal polarisations (wideband). And, the block schematic links the P-Gerät onto the radar gear.

Confusing is also, that the main wiring plan also shows a connection onto a so-called Q-Gerät (= KUH) or low power I.F.F. transmitter. To what we know from foregoing sources, the EGON stations might have operated with high power facilities. It might, nevertheless, have been a low power Q-gerät or KUH instead; because the regular Freya LZ and that like systems - lacked (space) facilities for an additional high power arrangement.

It was first my intension to close this survey with a wartime photo once reproduced in a British 1944 report on Wassermann radars.


Maybe not the most beautiful picture, but genuine reproduced. Recording date 21 August 1944. This Wassermann station stood in Bergen (NH)

(GB-NA AIR 20 1555 courtesy Phil Judkins)

Attached photo information:

(a)    540 Squadron P.R.-Mosquito XVI

(b)    21.8.44

(c)    100 ft

(d)    Chimney Radar of Gi?de? te??o at Bergen Aan Zee, Holland


What is the grey blanket at the lower part of this reproduction? Clouds or, quite often encountered on British wartime photos - censorship?  


Would this imply - that our Survey has reached a stand-still and have come to a virtual conclusion?

Apparently - Not!




On 17 January 2016


On late Cas Casper's funeral early this month, I met his son Remco again, we both are dedicated collectors of historical techniques.


He did send me, very kind, for copying a bunch of R.A.E Technical RAD notes, among it on GEMA related apparatus.

After having copied Note No. 204, and having digested what is dealt within, I realised that some of my previous assumptions may have been wrong - or at least being inaccurate. I therefore have to deal with some aspects again.




Viewing the N-Gerät opened front panel

[Fu.M.G. (See takt) 40G (gB)]


Let us follow briefly what occurred yesterday.


After having copied Technical Note No. Rad. 204  and considering the schematic and the drawing on its last page, I was really shocked, as I have never seen this before; most likely non of you either.


The point of of most interest is the dial designated: Range

(R.A.E. Tecnical RAD Note No. 204 last drawing)

Bothering me also is the tact that the functions of the above and lower CRT screen being alternated!

My first approach is considering the genuine GEMA manual on an EN 110 display module again.


Please bear in mind Pos. 301, which is the control designated by the Farnborough investigator Mr. Butlin:- Range

[Fu.M.G. (See takt) 40G (gB)]


Let us continue with considering the genuine German text:


I do not feel obliged to translate it

[Fu.M.G. (See takt) 40G (gB)]

However, the last sentence is intriguing:

Das Skalenrad (disk with attached scale) und das Fenster vor dem Bildschirm (a moving window in front of the upper CRT Pos 305) des Grobrohres (lower full range CRT screen) sind mechanisch miteinander gekuppelt und gemeinsam durch die Mitnehmerscheibe des Drehfeldschiebers (phase goniometer) Pos. 301 einstellbar (adjustable).


I must confess I have never seen any sign of such window like provision

[Fu.M.G. (See takt) 40G (gB)]

The circuitry around valve Rö 332 is a so-called transistron, an oscillator type providing two sharp defined signal stadia; like does a flip-flop.

This circuit is described well in the: Technical Note No. Rad. 204 report.


Mechanically there exists a discrepancy between the text and what is drawn on the last RAD 204 report page shown in the introduction of the current chapter. According the genuine German text - there must have existed a mechanical moved window, not a periscope provision. The reason might have been - that captured German gear (often) was in a poor state when it arrived at the Royal Establishment; whether caused by the Germans themselves, or by the capturing parties stays open; maybe by both.

One matter still bothering me - when in front of the full range CRT screen a window (Fenster) is being placed, how was it then possible to watch the entire radar range-capture? Was this window made of transparent perspex (Plexiglas)?



My perception as to how it basically once might have looked like. Above the magnified range-scale adjusted for 120 to 160 km

Any window like provision being ignored.


Maybe enlightening is the next drawing as to how screen presentation might have looked, through British eyes.



I must, nevertheless, admit that I do not fully understand the implications of the German text in respect to the RAE report explanations

(R.A.E. Tecnical RAD Note No. 204 last drawing) On


A question raised time and again -

is this the conclusion of our Wassermann survey?



On 13 May 2016

Last month, between 28 April and 1st of May, I joined the annual Funkfreunde Treffen, this time held in Gollhofen, Bavaria Germany.

On Saturday, usually in the later afternoon an exhibition being held by those who brought something along for display.

At a certain moment I set down next to JØrgen Fastner who told me straight away that I am not am a poor spy!

Don't you see something?

My reply, where should I look for?

Look carefully in the direction of Paul Reuvers.

Finally he brought my attention onto a ceramic tubular body, which I did not have the slightest idea for what its purpose might once have been. I thought it was the body of a rather heavy power resistor.


Do you know what it once was?


It is the tubular helix body of a Siemens Wellenschieber section!


Please notice the two dots which should indicate the maximum mechanical tuning deviation.

Before continuing our survey, it was possible to obtain this sample!


I only knew this photo of the Wellenschieber helix

Is this helix device equal to the previously shown device?

Yes it is!

One lesson to be learned - don't fall into the trap of scaling!


Comparing this diameter with another Danish drawing:


We notice that the ceramic tube diameter is given  for 6 cm


According JØrgen Fastner, the conductive layer helix is becoming, nowadays extremely brittle.

He told me also, that he owns one where still some of the conductive strip is still existing. I hope he will supply us with a conclusive photo.

He explained additionally, that the sprocket-holes being punched a bit into the (bronze like) conductor-layer.


2600 / 378 being the designations

My guess, serial number 378


Finally having a look at the broken-off shaft side

What is very good visible, is that the dielectric separation between the Wellenschieber coil-windings; once part of a theoretical discussion with Dick van den Berg, isn't that simple.

What I did not have noticed previously, is that the end-sections have been kit (cemented) onto (partly into) the main tubular helix body.

Time and again - one major matter being learned, never say never again; this was my foregoing query when chapter Y11A was brought to a semi-conclusion!


Maybe - to be continued in due course



By Arthur O. Bauer