Kramer X-4 Air-To-Air Missile

[HoHun]

Hi T. A.,

I'll agree with the second, two axis control, but not with the first. You can't deflect a solenoid half way. It's on or it's off. That doesn't preclude rapidly banging it on and off, but you can't get partial deflection because solenoids don't work that way.

Ah, thanks for the clarification! Now I understand what you're getting at.

Partial deflection does in fact happen during the time the control moves between the two extremes, but I'd say in practice this transition is rapid enough that we can neglect it for now. On the X-4, the two extremes aren't even neutral and full, but "opposite" and full.

The effective control force is however determined by the duty cycle. If the duty cycle is 60% "left", then the control will be deflected left 60% of the time and right 40% of the time.

Ignoring the rotation for the moment, if the missile were to follow the control input instantaneously, it would zigzag along on a sort of sawtooth curve, which leads the missile on a wide left turn, though 40% of the time it would be actually turning to the right. The individual segments of the zigzagging turn would be tight minimum radius turns.

Increase the duty cycle to 100%, and the missile would make a smooth and tight turn to the left. So even with the asuumption that the missile follows the controls without delay, you get different turn radii though the missiles zigzags around the desired path.

If we look at the missile in greater detail, it doesn't actually react to control inputs instantaneously. What actually happens is that the deflection of the control introduces a force that creates an angular acceleration of the missile. As the missile begins to increase its angle of attack, its main wings begin to ramp up a lift force that accelerate the missile in the commanded direction. The full (centripetal) lift force will only be reached after a delay, because it takes a while until the missile has rotated to the point where the equilibrium between control-induced force and stabilizing tailfin-induced force is reached.

This is basically a more extensive description of what Kramer itself described very briefly:

"When the spoiler follows the varying pulses at sufficient frequency so that the inertia of the missile only permits the pulses to take effect according to their average, a simple yes-no control with full deflection has become, in the final result, a virtually proportional control without requiring any more effort."

So if you have a 60% "left" duty cycle, the missile will probably oscillate from something like 50 % "left" to 70 % "left" angle of attack, and fly in a wide S-ing turn to the left, instead of the segemented tight-turn path from the "instantaneous" reaction model.

With the missile actually rotating about two times a second, all if this gets a bit more complicated as it happens in 3D, but the principles stay the same

I hope my explanation makes some sense at least? It's sort of difficult to describe without sketches, and I'm not very good at sketching, I'm afraid!

Regards,

Henning (HoHun)

 

[Dilandu]

It is something in between of digital and analog. The trigger timing is fixed, the voltage has no influence on the control which is typically digital, but the pulse lenght can vary continously, which is analog.

It is fully analogue. Just more complicated method of encoding the signal.

 

[Nicknick]

It is a method which enables a signal transfer with very little distrortion compared to a purly analog system because it combmines features from digital and analog signal transfering.

By the way, it is also the most simples method of archieving this, another alternative would be a control by a variable current (often used for transfering measering signals). Here the effort of controling an actuator would need much more effort. Using a variable voltage instead of un variable current would lead to many signal dilution and distortion.

 

[T. A. Gardner]

What the Germans used here is a variant of what is called a "stepper motor."

Everything You Need To Know About Stepper Motors | RS

In our comprehensive guide we look at what stepper motors are, what they’re used for, as well as the different types that are available on the market.

uk.rs-online.com

A stepper motor is a type of brushless synchronous DC motor that, unlike many other standard types of electric motors, doesn’t just rotate continuously for an arbitrary number of spins until the DC voltage passing to it is shut off.

Instead, stepper motors are a type of digital input-output device for precision starting and stopping. They’re constructed so that the current passing through it hits a series of coils arranged in phases, which can be powered on and off in quick sequence. This allows the motor to turn through a fraction of a rotation at a time - and these individual predetermined phases as what we refer to as ‘steps’.

(It helps having nearly 40 years of experience with industrial motors and controls...)

 

[Nicknick]

No, the flaps were controlled by simple solenoids. So more current flows through them, so higher the electromagnetic force and so higher the steering angles of the flaps. Unlike in a stepper motor, you cant block the engine/flap at any point stepwise by sending sufficient current through the cables. It is always a balance of power between the electromagnetic forces and the aerodynamic forces.

BTW, Telex machines, or at least their predecessors also used such a system, were a solenoid acted against a spring. The types were defined by the level of current, which determined the position of the solenoid.

 

[T. A. Gardner]

No, the flaps were controlled by simple solenoids. So more current flows through them, so higher the electromagnetic force and so higher the steering angles of the flaps. Unlike in a stepper motor, you cant block the engine/flap at any point stepwise by sending sufficient current through the cables. It is always a balance of power between the electromagnetic forces and the aerodynamic forces.

BTW, Telex machines, or at least their predecessors also used such a system, were a solenoid acted against a spring. The types were defined by the level of current, which determined the position of the solenoid.

The control handle / knob / stick worked like a stepper motor. As for solenoids, they deflect to the full distance they can go. With DC applied, they move that distance then remain there until the power is turned off.

 

[Nicknick]

Not really, in a stepper motor you have, to little suprise, steps and an incremented movement. Here, the stick procuces positve and negative pulses with an andjustable ratio of pulse lenght. These impulses are not directly linked to the position of the flaps. An approximated standstill of the flap can only be produced with the right ratio of positive and negative pulse lenght. In a stepper motor, the positon can only be fixed in given increments. For a fixed positon in a stepper motor, you need DC current in one of the three circuits.

 

[HoHun]

Hi T. A.,

The control handle / knob / stick worked like a stepper motor.

Schöer pointed out that there were three types of control handles, the Gb 203A with the rotating drum encoder featured in the videos linked above, the Gb 203B using the mechanical components of the Gb 203A but with potentiometer pickups and a different (electronic, it seems implied) encoder to replace the rotating drums, and the miniaturized "Knirps" with potentiometer pickups and (I guess) the same encoder as the Gb 203B.

I don't think I've seen a schematic of the electronic encoder yet, though I might have browsed past it ... would that be the one you'd compare to a stepper motor with regard to its working principle?

Regards,

Henning (HoHun)

 

[robunos]

I think this may be of interest . . .

https://rclibrary.co.uk/files_titles/1053/Galloping_Ghost_booklet.pdf

cheers,
Robin.

 

[Nicknick]

Still, proportional control of guided bombs, idenpendantly which ever came first (surly not Felix), might have been one of the first application.

I think the first were the fire control systems on warships, probably.

I don’t think so. For the fire control system, you needed an optical range finder which was operated manually and of course the information about the position and speed of your own ship and that of the enemy (probably estimated with the range finder). All those information were fed into an analog computer to do the calculation. These results were transmitted by phone to the turrets which manually adjusted the guns to the target. The pulse width systems like we’ve seen it in the X-4 were not really well suited for very high accuracy like it is needed for long range ship cannons.

 

[T. A. Gardner]

Still, proportional control of guided bombs, idenpendantly which ever came first (surly not Felix), might have been one of the first application.

I think the first were the fire control systems on warships, probably.

I don’t think so. For the fire control system, you needed an optical range finder which was operated manually and of course the information about the position and speed of your own ship and that of the enemy (probably estimated with the range finder). All those information were fed into an analog computer to do the calculation. These results were transmitted by phone to the turrets which manually adjusted the guns to the target. The pulse width systems like we’ve seen it in the X-4 were not really well suited for very high accuracy like it is needed for long range ship cannons.

That is how all the early German SAM systems were to guided. A set of optical fire controls, as used with flak guns, would input the range, altitude, bearing, etc., of the target to an analog computer. This was then run through an analog computer that calculated the position of the target and the missile upon firing. The missile would be displayed as a moving dot on a screen where the target was at the center (in the crosshairs if you will). The operator would then use a joystick to move the missile towards the center of the screen, with commands sent to it by a transmitter via polarized antenna pointed at the missile.

The Germans proposed to switch out the optical system for radar with one tracking the target, the other the missile, using a similar or identical control system, but these never got to a point where they were set up and tested before the war ended.

Such systems were found unworkable post war--the French and Russians both tried them while the US and Britain abandoned similar ones they had independently developed by late 1945.

 

[Nicknick]

Interesting, but my interpretation of Dilandus posting is, that pulse width control was used in naval fire control systems before it was used in rockets (ballistic guns). I don’t see the point, where a pulse width modulation could have been very useful in this regard.

 

[T. A. Gardner]

Interesting, but my interpretation of Dilandus posting is, that pulse width control was used in naval fire control systems before it was used in rockets (ballistic guns). I don’t see the point, where a pulse width modulation could have been very useful in this regard.

I suggest Naval Ordnance and Gunnery Navpers 16116-B (the 1950 version) which doesn't seem to be available online. It has a chapter (Chapter 10) on 'automatic fire controls' where servos and syncros are explained in detail. These are the same system the Germans are using. Their technology is not unique, but rather application of an extant control method to a new purpose.

 

[Dilandu]

Interesting, but my interpretation of Dilandus posting is, that pulse width control was used in naval fire control systems before it was used in rockets (ballistic guns). I don’t see the point, where a pulse width modulation could have been very useful in this regard.

As far as I know, was used to automatically feed data from rangefinders to calculators and to send the calculated angles to guns. By late 1930s, most fire control systems tried to get rid of human involvement as much as possible (sure, humans still were needed to keep rangefinders on target, but not for data transmission). The 1930s fire control systems were complicated; they calculated multiple parameters, including wind force and direction, atmospheric pressure, temperature of guns and shells, Coriolis force, ect.

 

[Nicknick]

The fire control systems are largely a forgotten antique high-tech and I don’t really know a lot about how they works, but I for believe signal transmitting a voltage system or current would have been the preferred method. To my knowledge, PWM signal are not common for precise measuring signals.

 

[Nicknick]

Interesting, but my interpretation of Dilandus posting is, that pulse width control was used in naval fire control systems before it was used in rockets (ballistic guns). I don’t see the point, where a pulse width modulation could have been very useful in this regard.

I suggest Naval Ordnance and Gunnery Navpers 16116-B (the 1950 version) which doesn't seem to be available online. It has a chapter (Chapter 10) on 'automatic fire controls' where servos and syncros are explained in detail. These are the same system the Germans are using. Their technology is not unique, but rather application of an extant control method to a new purpose.

You might be right or not, it depends whether you fully understood the pulse width modulation of the rocket controls or if you still confuse it with a stepper motor.

 

[T. A. Gardner]

Interesting, but my interpretation of Dilandus posting is, that pulse width control was used in naval fire control systems before it was used in rockets (ballistic guns). I don’t see the point, where a pulse width modulation could have been very useful in this regard.

I suggest Naval Ordnance and Gunnery Navpers 16116-B (the 1950 version) which doesn't seem to be available online. It has a chapter (Chapter 10) on 'automatic fire controls' where servos and syncros are explained in detail. These are the same system the Germans are using. Their technology is not unique, but rather application of an extant control method to a new purpose.

You might be right or not, it depends whether you fully understood the pulse width modulation of the rocket controls or if you still confuse it with a stepper motor.

Servos and syncros are a variant of stepper motors with a specialist application. The technology is largely obsolete (servo / syncro) as found in the 1950's. The same is true of magnetic amplifiers, another version of that same technology.

 

[HoHun]

Hi T. A.,

Servos and syncros are a variant of stepper motors with a specialist application. The technology is largely obsolete (servo / syncro) as found in the 1950's. The same is true of magnetic amplifiers, another version of that same technology.

None of these technologies were used on the X-4, were they?

Regards,

Henning (HoHun)

 

[T. A. Gardner]

Hi T. A.,

Servos and syncros are a variant of stepper motors with a specialist application. The technology is largely obsolete (servo / syncro) as found in the 1950's. The same is true of magnetic amplifiers, another version of that same technology.

None of these technologies were used on the X-4, were they?

Regards,

Henning (HoHun)

Not that I know of. The X-4 used a simple 200 VDC +/- system where the input was negative or positive and fed by one of two wires to the missile. These translated into up / down or left / right in effect. The missile and the launch aircraft had limited space for the system to be placed, so it was made as simple as possible.

 

[Nicknick]

@T. A. Gardner :
For you, everything electric is the same. If the system of the X-4 would have been used to move a hughe gun with electric motors, you could have never achieved any precision. The whole gear train would have vibrated and rattled all the time. In the X-4 the the controls didn't directly define a position, but only a force on the flaps. The air resistance acted as a couter force (there might have been springs involved too). Pushing a 50 t gun against a spring to archieve any pressision would be insane. I’m sure, this was not the case and it worked entirely different.

 

[HoHun]

Hi T. A.,

The X-4 used a simple 200 VDC +/- system where the input was negative or positive and fed by one of two wires to the missile.

Actually, the X-4 used a two-channel encoded PWM system in which all signals (which were analogue representations of the joystick position) were transferred to the missile over the same single circuit consisting of both of the wires.

We're not talking about telegraph landlines here, there's no common earth fighter and missile could use to eliminate the return wire.

Regards,

Henning (HoHun)

 

[Nicknick]

So, the X-4 and Fritz-X could stil have been the first application of PWM (pulse width modulation). Any other suggestions which might have been first?

 

[T. A. Gardner]

Hi T. A.,

The X-4 used a simple 200 VDC +/- system where the input was negative or positive and fed by one of two wires to the missile.

Actually, the X-4 used a two-channel encoded PWM system in which all signals (which were analogue representations of the joystick position) were transferred to the missile over the same single circuit consisting of both of the wires.

We're not talking about telegraph landlines here, there's no common earth fighter and missile could use to eliminate the return wire.

Regards,

Henning (HoHun)

Um, no. The inputs were voltage high, voltage low, amperage high, amperage low. These were translated into right, left, up and down at the missile. The input device for Dusseldorf wasn't like the one for Kehl. The Dusseldorf joystick operated switches, not rotating oval cylinder and take off brushes like in Kehl. They were distinctly different systems.

 

[HoHun]

Hi T. A.,

Um, no. The inputs were voltage high, voltage low, amperage high, amperage low. These were translated into right, left, up and down at the missile. The input device for Dusseldorf wasn't like the one for Kehl. The Dusseldorf joystick operated switches, not rotating oval cylinder and take off brushes like in Kehl. They were distinctly different systems.

So are you're telling me there was no PWM encoding?

Regards,

Henning (HoHun)

 

[Nicknick]

@T. A. Gardner : The control was soley done by the pulse lenght, neither the voltage nor the amparage had any influence on the steering.

 

[T. A. Gardner]

Hi T. A.,

Um, no. The inputs were voltage high, voltage low, amperage high, amperage low. These were translated into right, left, up and down at the missile. The input device for Dusseldorf wasn't like the one for Kehl. The Dusseldorf joystick operated switches, not rotating oval cylinder and take off brushes like in Kehl. They were distinctly different systems.

So are you're telling me there was no PWM encoding?

Regards,

Henning (HoHun)

With Dusseldorf / Detmold the combo on the X-4, yes. no pulse width modulation. The missile used spoilers for control, and these were on-off. The video posted earlier even shows one at one point. This simplified system was installed to save weight and space in the diminutive X-4 as well as simplify and reduce the cost of the system. See about 5:48 in the video. It's on-off there is no half way.

It's 200 VDC (there is a 600 VDC system that's similar that was proposed for Hs 293 and the Fritz X. Higher voltage due to longer wire runout. That's shown in the video too). The signals are switched on-off and are high / low voltage (+ or -) and high / low amperage. That allows for differentiation between turn / roll and dive / climb. You move the control stick left-right or up-down to activate a switch and send a signal to the missile to turn or climb / dive. The gyro on the missile accounts for roll and orients the input towards a constant 'up.'

Pocock has electrical schematics of these included in his book.

It's a more sophisticated version of the one the British proposed for Artemis only using wire for the inputs rather than a radio signal.

 

[HoHun]

Hi T. A.,

With Dusseldorf / Detmold the combo on the X-4, yes. no pulse width modulation.

Well, that is in contradiction to all the other sources I have pointed out, including the US Navy intelligence report and the original German Baubeschreibung, which is a primary document.

Even if Pocock shows a different system, which I am not entirely ready to take for granted yet, I would still have to question why you favour one source over all the others it contradicts.

The missile used spoilers for control, and these were on-off.

The solenoids were sort-of on-off, but as has been explained ad nauseam in the thread above, that doesn't mean they weren't good for a very nice approximation of proportional control.

Here's the US Navy version of the explanation:

Well, I just found the following paragraph in OP 1666, German Explosive Ordnance, Volume 1, dated 11 June 1946:

GENERAL DESCRIPTION. The X-4 is a fin-stabilized guided missile with a proximity fuzed warhead developed specifically for use by fighter planes against enemy bomber formations. [...]

[...]

Steering is effected by means of rake spoilers located in the tail fins. These spoilers vibrate at a rate of 5 cycles per second, control being effected by making the period during which the spoiler projects from one side longer than that during which it projects from the other. When the two periods are equal, no control is applied. This method of steering has the disadvantages of appreciable drag and a certain amount of delay, but has the advantage of simplicity and instantaneous mechanical response.

Regards,

Henning (HoHun)

 

[T. A. Gardner]

Hi T. A.,

With Dusseldorf / Detmold the combo on the X-4, yes. no pulse width modulation.

Well, that is in contradiction to all the other sources I have pointed out, including the US Navy intelligence report and the original German Baubeschreibung, which is a primary document.

Even if Pocock shows a different system, which I am not entirely ready to take for granted yet, I would still have to question why you favour one source over all the others it contradicts.

The missile used spoilers for control, and these were on-off.

The solenoids were sort-of on-off, but as has been explained ad nauseam in the thread above, that doesn't mean they weren't good for a very nice approximation of proportional control.

Here's the US Navy version of the explanation:

Well, I just found the following paragraph in OP 1666, German Explosive Ordnance, Volume 1, dated 11 June 1946:

GENERAL DESCRIPTION. The X-4 is a fin-stabilized guided missile with a proximity fuzed warhead developed specifically for use by fighter planes against enemy bomber formations. [...]

[...]

Steering is effected by means of rake spoilers located in the tail fins. These spoilers vibrate at a rate of 5 cycles per second, control being effected by making the period during which the spoiler projects from one side longer than that during which it projects from the other. When the two periods are equal, no control is applied. This method of steering has the disadvantages of appreciable drag and a certain amount of delay, but has the advantage of simplicity and instantaneous mechanical response.

Regards,

Henning (HoHun)

And all that is in line with the description I've given. The spoiler is either up or down based on the voltage applied to the solenoid. Solenoids are all-or-nothing, they can't go halfway. That makes use of a variable signal worthless. Instead, the application of a DC voltage + or - would determine the direction of movement.
The description above is partially wrong. If, again, you watch the video where one is shown, the spoiler unenergized simply can flop up or down between the two energizing coils and would as the missile rotated. Application of a signal--closing the switch and applying voltage--would lock the spoiler in one of the two positions available (up / down or left / right).

I agree with the assessment that there would be a delay and creation of drag. I would say that those same problems would have been present with the Artemis rocket design, and even worse given only one spoiler being used. This means there is going to be a big issue related to this response delay and the accuracy of the missile. To overcome some of that, orders from the control aircraft could only be received when the missile was at an interval of 45 degrees to the zero axis as determined by the gyro. Thus, a slight delay could be imposed so the command was less likely to oversteer / correct the missile's course.

Again, the video makes the spoiler operation clear. On / off. Which spoiler(s) are activated depends on how you push the control lever and the switch closed as a result.

I'm not faulting the Germans here. It was, at least in theory, a workable system. In practice however against a fast-moving bomber, even one on a steady course, with parallax errors and the pilot's (or operator's) hand-eye coordination just wasn't going to cut it. For a proof look at ATGM's of the late 1950's through the 70's that used MCLOS. These against even a slow-moving tank have success rates on the order of 30% with operators who have practiced hundreds or more simulated launches.



All of that makes sense in a system that is operated on a DC voltage.

 

[HoHun]

Hi T. A.,

And all that is in line with the description I've given.

To address the first elephant in the room: Your thinking stops at the spoiler position.

If you look at the missile as a whole, spoiler position only determines angular acceleration. The integral of angular accelerations determines angular velocity, and the integral of angular velocity determines angle of attack, which determines turning radius.

With the same missile at the same mass and velocity, you can fly a wide range of different turning radii even if you only have a spoiler that always is in either of the two end positions.

In the word of the missile's designer Kramer:

"When the spoiler follows the varying pulses at sufficient frequency so that the inertia of the missile only permits the pulses to take effect according to their average, a simple yes-no control with full deflection has become, in the final result, a virtually proportional control without requiring any more effort."

Do want to disagree with that or not?

Regards,

Henning (HoHun)

 

[Nicknick]

"Solenoids are all-or-nothing, they can't go halfway"

This is false, Solenoids need a certain time to go from neutral to the End position. If the actuating frequency is shorter than this timespan, the spoilers/flapps will wobble around the desired position. With 25 Hz, this is certainly the case.

 

[T. A. Gardner]

Hi T. A.,

And all that is in line with the description I've given.

To address the first elephant in the room: Your thinking stops at the spoiler position.

If you look at the missile as a whole, spoiler position only determines angular acceleration. The integral of angular accelerations determines angular velocity, and the integral of angular velocity determines angle of attack, which determines turning radius.

With the same missile at the same mass and velocity, you can fly a wide range of different turning radii even if you only have a spoiler that always is in either of the two end positions.

In the word of the missile's designer Kramer:

"When the spoiler follows the varying pulses at sufficient frequency so that the inertia of the missile only permits the pulses to take effect according to their average, a simple yes-no control with full deflection has become, in the final result, a virtually proportional control without requiring any more effort."

Do want to disagree with that or not?

Regards,

Henning (HoHun)

Kramer is correct to a degree. In theory, he's absolutely correct. In practice, without some more advanced computer control, and just hand-eye coordination using a simple on-off joystick controller course corrections will be rough and abrupt.

What Kramer, given the Dusseldorf Detmold system, is saying is the person controlling the missile can flick the lever in one direction or another for a varying amount of time--that is they could tap it over for less than a second, or hold it over for much longer--varying the 'up' time of the spoiler and its effect on course.

I can't see anyway what Kramer describes for Dusseldorf Detmold to work that way on the circuitry provided alone. It is entirely possible he was being overly optimistic of his system of spoilers given that it wasn't copied anywhere post war for use other than in a few low speed ATGM's that proved less than successful. For example, this system was copied, more or less, in the French SS 10 ATGM but abandoned and replaced with a thrust vectoring system in SS 11.

 

[Nicknick]

"What Kramer, given the Dusseldorf Detmold system, is saying is the person controlling the missile can flick the lever in one direction or another for a varying amount of time--that is they could tap it over for less than a second, or hold it over for much longer--varying the 'up' time of the spoiler and its effect on course"

That is false! Please try to understand the working principle! There is even this nice movie on youtube (see aboove) where you can actually see how it works. If you change the lever position, the ratio of positive and negative pulse length will vary 25 times per second). You dont have to push it for 1/25 th of a second!

Please inform yourself on how a PWM system works, we are moving in circules!

 

[HoHun]

Hi T. A.,

Kramer is correct to a degree. In theory, he's absolutely correct. In practice, without some more advanced computer control, and just hand-eye coordination using a simple on-off joystick controller course corrections will be rough and abrupt.

So it seems we're in agreement now that a simple on-off control surface deflection system is no hindrance for smooth control. Great, we're making progress.

With regard to your assertion that the joystick controller is "on-off", that doesn't agree with the German sources.

From the original documentation reproduced in the Waffen-Revue:

"III. Transmission

For transmitting the command from the carrier [aircraft] to 'X-4', the wire connection and the direct current pulse transmission analoguously to the technique developed for the 'Fritz X' are used."

In other words, the X-4 uses pulse width modulation just like the Fritz-X.

With regard to the joystick used, the same Beschreibung notes:

"Due to the low wire resistance of the 'X-4', it's possible to operate with a very low transmitter voltage and thus use the small joystick 'Knirps' ['Midget'], currently under development for the 'Fritz-X', directly as transmission device, by which means the transmitter and receiver are reduced to the simplest possible technologies."

This confirms the initial statement that the control system of the X-4 is principally the same as that of the Fritz-X, and a joystick used with the PWM-controlled Fritz-X obviously could not be a "switch stick".

So, smooth input, smooth output.

Regards,

Henning (HoHun)

 

[HoHun]

Hi again,

This confirms the initial statement that the control system of the X-4 is principally the same as that of the Fritz-X, and a joystick used with the PWM-controlled Fritz-X obviously could not be a "switch stick".

So, smooth input, smooth output.

Well, I've managed to get my hands on a copy of Pocock's "German Guided Missiles" now.

Here is his description of the X-4 (my highlights):

"The X-4 was a small cigar-shaped missile with four swept wings and four tail fins. It was propelled by a liquid-fuelled rocket motor and controlled in flight by electric pulses transmitted from the parent aircraft across two fine steel wires which unwound from bobbins on the missile wing tips. The control orders were given by the fighter pilot on a joystick in the same manner as the orders for the radio guided bombs using the Kehl-Strassburg apparatus. This Ruhrstahl missile was the prototype of not only the modern air-to-air missile but also of most modern anti-tank missiles used by infatrymen, which combine the characteristics of rocket propulsion, wire guidance and spoiler control."

On the Kehl-Strassburg system:

"When no order was transmitted, the control relays remained for an equal time at each end of their travel. With an order transmitted, for example, an order to the SD 1400 X to 'steer left', the relay remained for an increased time at the 'left' extreme and a reduced time at the 'right' extreme, so that the total switching time remained unaltered.

Switching fo the modulating frequencies was performed by an 'order distributor unit' at the transmitter, this unit consisting of a rotating cylinder divided into two segments and a sliding contact on the surface of the cylinder which reproduced the movement of the bomb aimer's joystick. With the stick held in its mid-position, each segment was connected to the contact for exactly half of the cylinder. A movement of the stick would move the contact towards one end of the cylinder, where the segments were shaped so that one was of larger area than the other. Thus one segment would be connected to the contact for a longer time, and the other for a shorter time, but the total time of revolution was unaltered."

Clearly, this is a description of a pulse-width modulation system.

Then there is the Düren-Detmold system:

"This system could only transmit two orders - a 'left/right' movement and an 'up/down' movement [...] These orders were given in exactly the same way as in the Kehl-Strassburg transmitter, the switching contacts being operated by the bomb aimer's joystick."

Obviously, since Kehl-Strassburg was a PWM system, so was Düren-Detmold.

Pocock continues:

"As the SD 1400 X went out of service, from the spring of 1944, production of the 'Düren-Detmold' also stopped, but existing apparatus was modified into 'Düsseldorf-Detmold' sets for the Ruhrstahl X-4 air-to-air missile. This system was almost identical with the 'Düren-Detmold', but with a lower working voltage."

So, Düsseldorf-Detmold clearly was a PWM system just like Düren-Detmold.

In the context of the "Kogge" and "Karte"/"Pol" systems, Pocock also mentions a new type of joystick:

"A new order distribution unit unit was also included in this transmitter. Instead of the rotating cylinders, the joystick was now used to control the constants of a multivibrator circuit. This new order distributor was given the name 'Karte' or 'Pol' according to wether cartesian control (as in Wasserfall) or polarised control (as in Schmetterling) were used."

Pocock doesn't mention whether this joystick (at least the cartesian version) was the same "Knirps" mentioned in the context of the X-4, but I wouldn't be surprise if at least the circuitry were the same.

And yes, there is a (highly simplified) diagram of the X-4 "basic circuit" in the book that shows encoder switches on the transmitter side, and a floating comment "coupled to joystick". However, it's pretty clear that these are the encoder relays that encode the signal arriving from the joystick, which is not shown but from the coherent description obviously provides the PWM-encoded input the encoder translates into signal that can be transmitted as 200 V DC PWM through a couple of kilometers of wire.

Regards,

Henning (HoHun)

 

[Dave_sic]

hey guys, im the creator of the Video shared earlier in this thread (thanks for watching) .
i hope i can clear things up on how the X4 was guided and throw my 1 or 2 cents in about the history of that project.

was the X4 cheap?
depending on the point of view...
it was made out of sheet aluminium, cast aluminum centerpiece, aluminium pipes ,steel warhead, wooden wings, steel combustion chambre and was fueled by byproducts of the german chemical industry the electronic parts were 2 identical relais and 4 solenoids.... those are relatively cheap materials. the parts are produced by inexpensive and quick manufacturing processes (bending on a master body, riveting, casting, etc). In a normal functioning economy this is very cheap,
however the german economy at the time was under attack, all kinds of materials (most of them alloying metals) were increasingly hard to get and slave labour as replacement for skilled workers killed/wounded at the frontline is terribly inefficient. in addition to that the government changed fiscal policys to artifically create money, blurring the image historians can get from documents and pricelists of that time. in a normal economy the x4 would have propably been cheap, in 1944/45 germanys economy it propably wasnt (like everything else).

the french X4
Kramer worked for the french after the war and france definetly copied the X4 (AA.10) but discontinued its production due to the increasing speed of jets and jet Bombers and its dangerous liquid fueled rocket engine. the oxidizer used in the X4 was Nitric acid that is unhealthy and very aggressive. it can dissolve metals like iron and ignites organic substances on contact. those arent fuel propertys you want in a weapon produced for long term storage (to be later used in the event of a soviet invasion) the german x4 had the exact same problem and was only fueled at the last moment before the carrying fighter took off.
a logical step was to replace the BMW liquid fuel Rocketmotor with a solid fuel rocket.
the germans wanted a Schmidding SFR but france never tried to replace the LFR and instead developed new missiles.

 

[Dave_sic]

the X4 was steered by 2 electrical circuts, the internal circut moving the solenoids and the guiding circut between the Aircraft and rocket.


the guiding circut carries an electric signal alternating in polarity (+/-) and current (high/low). the command information is encoded in those alternations (polarity change for left/right info and current change for up/down info) the frequency of both changes is the information on the stick position (how many percent the stick is moved off the center. for example 90% up and 35%left)
the frequency changes (pulse width modulation) are created on the triangular surfaces on the 2 rotating (20 rot/sec) drums in the "encoding devicve" in the plane.
for example in neutral position the contact detects that 50% of the drum has a positive surface and 50% of the drum is negative, creating a 50/50 +/- signal
if the contact on the drum is pushed slightly to the left 70% of the drum has a positive surface and 30% has a negative surface creating a 70% positive and 30% negative signal
if the contact is pushed to the far left 2 edges of the triangular surface touch and the contact is only touching the positive surface resulting in a 100% positive signal.
the other drum is wired in series and contacts low and high resistance surfaces to create the high/low current fluctuations

the wire spools of the rocket form one! circut with the control device (because as it was mentionedaircraft and missile have no common ground)


the internal circut is powered by a 9V battery. the battery is connected over the gyroscope wit the switches in the relais that "chose" in which direction the solenoid will move. the only job of the gyroscope is to assign the right command of the (rotating) guiding circut to the correct relay in the (rotating) missile so that the effect of the given command is "stable" (relative to the !not! rotating pilot)

by assuming the missile doesnt rotate we can completly forget about the gyro and assume a direct connection between battery,
relais-actuated switch and solenoid.

the internal circut can only move the solenoid to 100% left or 100% right position (or 100% up/down).


decoding and steering is achieved by the missiles 2 internal relais. the relais can detect polarity and current changes in the guiding circut and switch the missiles internal circut, making the solenoids move in one or the other direction.
if we again assume that the missile isnt rotating, one relais is only reacting to polarity changes while the other is only reacting to current changes. if we know assume that for every second of "steering signal" the encoding drum in the aircraft rotates 5 times decoding my 3 examples becomes possible.

in neutral position the polarity switches 10 times per second [Edit : its 40 times because the encoder drum rotates with 20Hz , for the basic understanding its only necessary to know they are rotating so i wont change the following calculations.]
(5 drum rotations * 2 drum surfaces / 1 sec of command)
each position is held for one 10th of a second ( [1sec of command /5 drum rot ] * 50% positive contact per drum rot. )
as a result the solenoid is vibrating between the left and right position and in total projecting 5/10 of a second to the left and 5/10 of a second to the right.(5rotations * 1/10 holding time) the movement of the rocket is canceled out.

in slight left position the polarity still switches 10 times per second but the projection time changes.
[1sec of command /5 drum rot]*70% positive contact = 0.14 sec holding time for each left signal
5 rotations* 0.14 sec holding time = 0.7 sec total projecting time to the left.
[1sec of command /5 drum rot]*30% positive contact = 0.06 sec holding time for each right signal
5 rotations* 0.06 sec holding time = 0.3 sec total projecting time to the right.
as a result the rocket will turn to the left.

(extra example here 99% left)
in almost far left position the polarity still switches 10 times per second but the projection time changes again.
[1sec of command /5 drum rot]*99% positive contact = 0.198 sec holding time for each left signal
5 rotations* 0.198 sec holding time = 0.99 sec total projecting time to the left.
[1sec of command /5 drum rot]*1% positive contact = 0.002 sec holding time for each right signal
5 rotations* 0.002 sec holding time = 0.1 sec total projecting time to the right.
as a result the rocket will take a sharp turn to the left.

in the very far left position doesnt switch anymore , the constant positive polarity is interpreted by the decoder as constant projection to the left and the solenoid will stay in the left position.
the rocket does a very sharp turn left.


i hope this clarifys some open questions on the working principles, pw modulation and ammount/design of electrical circuts in the x4.
my information come from op1666 HoHun alredy mentioned and the book : Peenemünde-West: Die Erprobungsstelle der Luftwaffe für geheime Fernlenkwaffen und deren Entwicklungsgeschichte written by botho stüwe that was recommended by a friend of mine.

!My Video shows some Errors, the x4 didnt use rudders like i show them but little airbrakes and signal decoding happens after the gyro, my animation shows decoding before the gyro "horizontalification prozess". (improved video coming soon)

 

[Dilandu]

i hope this clarifys some open questions on the working principles, pw modulation and ammount/design of electrical circuts in the x4.

Thank you for the data and explanation!

 

[HoHun]

Hi Dave,

i hope this clarifys some open questions on the working principles, pw modulation and ammount/design of electrical circuts in the x4.
my information come from op1666 HoHun alredy mentioned and the book : Peenemünde-West: Die Erprobungsstelle der Luftwaffe für geheime Fernlenkwaffen und deren Entwicklungsgeschichte written by botho stüwe that was recommended by a friend of mine.

Great, thanks a lot for the additional details!

And it cracks me up that Stüwe has a good description of the technology ... I completely forgot I have had this book for a long time, and totally failed to check it even once! :-D

Regards,

Henning (HoHun)

 

[T. A. Gardner]

Additionally, the signals from the guiding aircraft are 600 VDC to cover line loss (resistance in the wire) while the wires are 22 ga spring steel from Sweden coated with varnish for insulation. There are two bobbins, one on the missile and one on the aircraft to pay out the wire.

The control surfaces on the missile are spoilers that are activated by solenoid. The spoilers 'vibrate' with the signal induced from the control system.

 

[Dave_sic]

Additionally, the signals from the guiding aircraft are 600 VDC to cover line loss (resistance in the wire) while the wires are 22 ga spring steel from Sweden coated with varnish for insulation. There are two bobbins, one on the missile and one on the aircraft to pay out the wire.

The control surfaces on the missile are spoilers that are activated by solenoid. The spoilers 'vibrate' with the signal induced from the control system.

thanks vor the 600V info, didnt know that but it makes sense due to the long wires.
i dont know about the second bobbin on the aircraft, ive read the wire ends of the x4 bobbins are simply plugged into the correct holes on the ETC 70 weaponcarrier and videos dont show that second bobbin/spool(maybe its internal and thus not visible).

the HS 293 Wireguided version (ausf A ?!) definetly had a spool in the wing tip and a second one in the control aircraft .