domain of time
Page setup on 19 November 2012
Status: 26 November 2012
In the Nachtfee 'things done' webpage I dealt in section 18a/b/c with new findings as to how Nachtfee should have been, and still is, aligned upon the, via the IFF transponder, returning EGON and Nachtfee 'order' signals. The EGON signal is, however, not a Nachtfee parameter. Although, EGON definitely is part of the Nachtfee guiding system, both using the same signal carrier, but using complete independent PRFs.
This principle schematic of the Nachtfee ground console is quite conclusive though, the interactions between the three phase controlling goniometers may need some special explanation
Please double click on this drawing as to open it in PDF
Using the same colours as in the previous drawing (like is done in all drawings of this page). The red-line circuit is handling the sine wave derived from the quartz controlled channels. The blue lines are constituting the coherent signal chain belonging to both the 'order' data chain as well to the controlling feedback time-base section. The yellow section is handling the returning feedback signals
Please double click on this drawing as to open it in PDF
Let us resume what the three phase goniometers is about.
Control A is the 'Phase' control which is lacking any form of scale. It is thus not possible to control what it actually is causing. That is why there is no way around the fact that, in some way or another, a comparing signal has to be considered. The only non-coherent signal constitute the time-base-phase-reference of the aircraft 'order' display. Please bear in mind, that the Nachtfee system can only work properly, as long as both the internal Nachtfee clock and the virtual aircraft clock are running synchronously. A great matter of concern, as in previous experiments I proofed, that we have to cope with accuracies of at least 10-7. A dimension which hardly could have been realised under military operational conditions in the 1940s! No doubt, Nachtfee itself cannot provide this accuracy! That it, nevertheless, could have been operated may only when a the human factor becomes involved. He must manually compensate 'time domain' errors. To what extent I cannot judge.
Control B is the 'order' or call it 'command giving phase-goniometer', which create a 0° to 360° phase shift in accordance to the setting of the 'order pointer' (please notice the photo below)
Control C is a very significant control, that I called 'Range off-set', which compensate for both the distance between the Freya-EGON/Nachtfee ground station and the airborne aircraft under guidance, and, the electrical delays owing to the various electronic circuitries involved.
The main control panel of Nachtfee
The big compass like scale is showing a variety of 'orders', which actual Commands have been dealt with previously. (Bladwijzer 33) It constitutes goniometer B of previous drawing. The black 'Phase' control knob (down on the left-hand side) is representing goniometer A. The CRT in the upper section is the 'control Tube' LB2. The illuminated scale pointing in between 0 and 280 km is the, what I have designated 'Range off-set control'.
This drawing shows why it is necessary to compensate for distance off-set
Please double click on this drawing as to open it in PDF
In the dotted box most to the right-hand side represent what Nachtfee-EGON related electronics is involved. The dotted TB signal is providing a most significant system parameter, but, there are a few manners to let it enter the closed loop system. I have experimented by injecting its signal at the same point where the Nachtfee data signal is provided (pin 9 of the test-connector). I once modified the FuG25a such, that it is possible to inject the TB signal in a pre-stage of the IFF transmitter. The latter manner is by-passing the recognition-keys. As no one can solve this query, it has to be left open. Nevertheless, repeating, in some way or another, the aircraft time-base-phase signal has to be compared to on the Nachtfee control CRT screens! As otherwise 'Phase' control on the Nachtfee front panel does not make any sense! Even when one would use atomic clocks on both side of the chain, at least one time, after every new starting up, phase-comparison should be accomplished!
It is clear that after the 'order' phase-shifting goniometer caused a particular change of signal phase, that the signal had to pass through quite some stages in the Nachtfee apparatus, followed by some stages in the Freya-EGON system. Every stage in the electrical chain is causing an amount of signal delay. We count normally in free space that velocity 'c' of EM waves is 3. 108 ms. Whatever the value of system delay may be, v < c. Fritz Trenkle gave, for instance, that EGON used in combination with Freya radar, that on the radar screen an EGON signal was legging 400 m. We know that 1 µs is equal to 300 m EM wave displacement. I guess that the main parts of this delay is being caused by both the electrical circuitry of the FuG25a IFF transponder- as well as within the Gemse receiver system. The same apparatus configuration we still are using in the reconstruction of our hypothetical Nachtfee system. Thus at least 1.333 µs delay is to be counted with. However, Nachtfee should be capable to operate over a distance of 250 km (theoretically 300 km). Hence 400 m versus 250,000 meters is a to be neglected figure, as 400/250,000 = 0.00016. The only parameter that matters is thus 'Range off-set'.
However, our experiments are showing a much higher system delay than may be expected from the above brief calculations. What may causing this? We operate over a distance off-set of 2 x 2.5 meters only! The simulated airborne system is, however, using additional pulse forming and amplifying modules.
As is already noticed, our used pulse-forming and amplifying transistor stages may causing a considerable time delay, thus phase shift (neglecting signal level and supply voltage). We also must take into account, that when a stage (circuit) is inversing a signal, thus rotating it over 180°, that this may give the effect of a distance of ½ λ. The circuits I use are mainly constructed in a manner that in- and output are in phase. However, it is found that when transistors are brought in saturation a phase-shift between in- and output will occur. This phenomenon may not have counted much in the valve era. This may be the reason for their quite low system delay of only 1.333 µs in the wartime system. It may be worth changing the transistor modules for MOS devices. Though, we do not have to fight a war, our only aim is finding out how Nachtfee works and how it most likely was being operated. This alone is already a huge challenge!
Nachtfee and EGON as well operate with say 500 Hz PRF. λ · f = c → λ = c / f = 600 km. From the previous drawing we can see that the signal range has to pass the operational range twice. Thus 600 km has to be divided by 2, providing a system limit of 300 km! But, there exist a great deal of system or range symmetry. Both, the upwards going and downwards facing waves have to bridge equal distance. That is why the Nachtfee 'Range off-set' scale is calibrated in 0 - 300 km.
We also know, that the Nachtfee controlling CRT screens can only show signals that entre through the feedback input (Impulsamplitude) (pin 1 of the List-Stecker). The objective of using EGON is to measure the time delay (in casu distance) between the instant that the transmitter sends a pulse and its appearing at the radar display. EGON is also called a secondary radar system and its range measurement is less accurate than when primary radar signals are involved, thus signal bouncing at a metal surface of an object (like an aircraft).
Let us close the system loop.
Presenting the Nachtfee and EGON system loop differently (the used colours are equal in all drawings on this webpage)
Please double click on this drawing as to open it in PDF
Range constitute the operational distance, which is symmetrical for both the upwards- and downwards travelling signals.
My next contribution should deal with the implications of 'caused phase shifts' and its appearing on a controlling CRT screen in the closed Nachtfee loop.
On 21 November 2012
As to get a feeling for what is happening within the Nachtfee CRT deflection system we consider first the next drawing.
Shown is a section of the first drawing on this webpage (The LB2 deflection stage is for simplicity omitted)
Please remember, that this drawing section originate from the begin of this year; when I did not yet understand the function what was then designated 'Number or degree scale'. We know now that its purpose is to compensate for a 'Range off-set'. The goniometer designated 'C' in the second drawing on this page, is changing the signal phase at will. As soon as this has been accomplished, the signal is as it has been before. Imagine, viewing a film strip where a single frame has been removed or is being added. You will hardly notice that it once has happened, because there does not exist a frame counter to compare with. Just this is what is happening in Nachtfee! The deflection or time-base system in Nachtfee (only the one to the dual trace CRT is drawn) is being fed with the regular sine wave signal. The LB2 is having its virtual spot starting point at due North. In previous explanations I spoke about rotating the time-base, which is not fully in accordance to what actually is happening! Let us assume, that we are viewing two fully synchronously projected films simultaneously. The viewers will see entirely equal events. Imagine, in one film strip we add or remove some frames, from now on synchronism is no longer existing, though the viewer of one of the films is not aware of it at all. Let us consider the un-touched film being an external reference clock. Watching the difference between both film strips is only possible by comparing their mutual states of affairs. The un-touched film strip may be considered the aircraft time-base, and shifting phase (manipulation) only taking place within the Nachtfee console.
Hence, when I spoke about rotation of the circular LB2 time base line, I actually meant that its starting point at due North (0°) is delayed (manipulated) in time such that a pulse will arrive at due North; whereas it otherwise would have been painted at another circular base-line vector. When we consider that the circular LB2 time base is derived from a sine-wave, and that 0° and 360° are both just at due North - the course of the sine-wave equals thus the rotation over 360°. Due South represents thus 180°. When the EM wave has travelled 300 km (500 Hz equals λ = 600 km) the LB2 spot is just pointing at due south or 180°. As to rotate the vector to point at due North, the 'Range off-set' control should be set at 300 km, thus adding half a cycle. However, distance has to be bridged twice, and the next returning 300 km (out of 600 km) is added in signal travelling time. 'Range off-set' has to be adjusted at 300 or 0 degrees. Just where a new cycle is due to start.
Nachtfee is sending via Freya-EGON a pulse towards an aircraft under guidance, its 'order' pulse is passing through the transponder system and being afterwards received and made visible on the LB2 control CRT screen. In this case the 'Range off-set' control is to be adjusted such that the returning signal will point exactly at the same vector as when it was transmitted. We have already noticed, that the most effective vector is zero degrees or due North. The 'Range off-set' control is now to be adjusted at the effective off-set of range. We have also noticed that electrical system delays may be in most cases having a minor value and will ultimately be incorporated within the compensating 'Range off-set' value. (Bladwijzer189)
Please bear in mind: when a Nachtfee pulse leaves the Freya-EGON antenna array, it will constantly travel with the speed of light. The pulses are derived from sine-wave signals. Though, both keeping a tight relation. When the pulse bridged 300 km - the sine-wave originating from has rotated its vector over 180°. Although, what is noticed are still only pulses, but we compare these pulses against sine-waves of origin. The pulses are just derived from the instant where the sine-wave crosses 0° in a positive sense. The steepness of a sine-wave signal is just there having its highest value; most favourable for first order pulse differentiation.
Also in the aircraft display the time-base is constituted by a sine wave generator. According information we have, the CRT in the aircraft was similar to the LB2 in the Nachtfee console. Magnetically deflection is normally derived from a sine-wave.
It is evident, that the time-base signal phase should be running exactly equal to the one in the Nachtfee ground console. This is impossible. Though, Nachtfee is having a 'Phase control' goniometer (A), which is changing the signal phase in the domain of time. The blue line connected system will not notice what manipulation once took place. As the entire blue line system is derived from a mutual point. What happened before this point cannot be controlled directly within the Nachtfee controlling CRTs. Now a crucial factor (aspect) is entering the arena, the state of affairs (imagine the frame numbers) of the film strip in the aircraft. Of course, I mean the aircraft time-base-signal-phase should be taking into account! This is only possible when the actual aircraft time-base-signal-phase (TB) is also combined within the Nachtfee feedback signals. The TB pulse is incorporating information of the actual state of affairs of the TB phase in the aircraft (again in the domain of time). The Nachtfee feedback signal burst and the time-base pulse will bridge equal distance and time as to be made visible on both the LB2 as well as dual trace HRP2/100/1,5(A) CRTs. To my current understanding, the TB phase parameter is only to be adjusted upon at the LB2 CRT screen. Significant is, that also the TB pulse information will return in the aircraft after having bridged the 'Rang' twice; albeit, that the returning Nachtfee data pulse, which data-phase is manipulated by operating the 'Phase control' (A), is incorporating the human correcting factor. The Nachtfee data pulses are being changed in the domain of time in such a way that the 'order' or command signal will be made visible just where it should pop-up on the aircraft CRT screen. The Dual trace scope is only meant for adjusting accurately the Nachtfee feedback pulse onto the actual signal phase of the LB2 circular deflection. Which, as you know, is being adjusted upon in combination with 'Range off-set'. The vertical pulses being the feedback signal and the time-base is adjusted in such a manner that the feedback pulses will correlate just in the centre of the dual beam CRT. When this is done correctly, the Nachtfee pulse is pointing due North at the LB2 CRT. Of course, only valid as long as the 'order' pointer is also pointing at due North. Please consider carefully: (Bladwijzer189) and (Bladwijzer190!)!!
It is evident that at some instant the TB reference pulse arrives in the Nachtfee controlling circuitry. It may have any phase difference compared to the actual Nachtfee internal signal phase. Shown is, that the first step of adjustment should be setting of the correct 'Range off-set'. This guarantees that the distance the Nachtfee signal phase is equal to when it left the Nachtfee console. The only factor with uncertainty is, what is the actual TB phase inside the aircraft display system. This latter time-base is entirely independent and cannot be manipulated from the ground.
The only means possible is the introduction of a deliberate, but controlled, phase shift. For it the only existing means is the 'Phase' control goniometer (A). It is capable of changing the overall Nachtfee signal phase (in the domain of time). It is not at all changing the 'Range off-set' adjustment, because this is a quasi coherent system-loop.
I have to confess, that this is my presumption: it is known that the parameters of every stage of the Nachtfee system is known. It is thus likely that the TB pulses should get a particular off-set. Whether this coincide with the due North feedback pulse or at another LB2 screen vector I don't know. However, I have tested this method with great success. Being informed with a TB off-set parameter it is possible keeping the Nachtfee 'order' signal blip on the (simulated) aircraft display exactly in line with the one in the Nachtfee console.
This already elsewhere used photo shows what dual alignment is about
The due North pulse constitute the Nachtfee feedback pulse, which is appropriately aligned by means of 'Range off-set'. The pulse or blip at about 45° represent the correctly adjusted pre-setting of the TB pulse by means of the 'Phase' control. Just at this vector our simulated system is tuned exactly in accordance to the actual aircraft phase arriving via the downwards feedback chain. Now the Nachtfee 'order' signal (blip) is pointing correctly due North at the simulated aircraft display.
It may be, nevertheless, necessary to check whether the actual aircraft TB vector off-set stays constant for several cold start-ups. It worked fine during previous experiments, but it has to be confirmed by a series of cold synthesiser (PM5193) starting ups!
I checked today whether it does matter what actual signal phase the PM5193 is providing in the domain of time. So far, the correction factor stays equal. My conclusion is that its actual signal phase does not matter. The reason may (also) be, that the painted Lissajous figure, constituting a circle, is also derived from this signal source. Under all circumstances, due North stays due North and just at this reference vector we compare and align all system parameters.
However, how the TB pulses were incorporated within the feedback signals I do not know. It may have been done on demand or having a lower sequence; one thing is certain - in some way or another its signal should return towards the ground station frequently. For it the negative impact of a useless 'Phase' control on the Nachtfee front panel would have been too disastrous. Even a slightly touching of its control is virtually rotating the Nachtfee 'order' signal against the painted aircraft time-base-line. Hence, its then painted 'order' vector is becoming unreliable and faulty.
On 26 November 2012
Please consider my new approach: explaining it a bit differently, by means of a PowerPoint presentation converted into PDF
Please consider also new findings via Wikipedia: Was Nachtfee ultimately a failing system concept?
To be continued in due course
By Arthur O. Bauer
Please return to, or continue with: Nachtfee things done page