wehrmacht awards


Go Back   Wehrmacht-Awards.com Militaria Forums > Wehrmacht Uniforms and Equipment > Communications Equipment

Communications Equipment Radio, telephony and radar equipment

Reply
 
Thread Tools Display Modes

Another Luftwaffe autopilot comes to life
Old 11-26-2017, 12:31 PM   #1
Funksammler
Member
 
Funksammler is offline
Join Date: Jan 2009
Location: Normandy
Posts: 2,959
Default Another Luftwaffe autopilot comes to life

I have been collecting bits and bobs of an early type of autopilot for a while, recently I started testing and restoring some of the components. The Autopilot (more accurately a "Kurssteuerung" or yaw control, with the roll and pitch still being controlled by the pilot) was called the Lz12 and was manufactured by Askania:



As the above Luftwaffe date sheet shows, the Lz 12 was used on early versions of the He-111 and the Do-17. Aircraft of these types were used during the Spanish civil war, by the start of the war the Lz12 was already outdated and being replaced by the Siemens K4ü, but some Do-17P still played a role during the Blitzkrieg campaign and the early He-111 were converted to transporters and played a role in supplying the Demjansk and Stalingrad pockets.

The Lz12 was one of the last in a line of pneumatic autopilots developed by Askania in the 1930's, so it represents the pinnacle of pneumatic engineering of the time. As the "Nicht for neue Muster" on the Luftwaffe datasheet suggests, by 1938 more modern electro-hydraulic autopilots were taking over. This did away with one important disadvantage of pneumatic systems: their reliance on air pressure limited their operational height.

The technology was not completely redundant yet though, in 1943 Askania developed the autopilot for the V1 based on components used in these early autopilots.

The Lz12 is the first pneumatic autopilot I have worked on, and I was not sure if I could find a power source small enough to use on a rotating model. Not having had previous experience I guess the best way to find out is to try and I managed to find a relatively compact and light pneumatic pump that could provide both compressed air and a vacuum:



With a power source available, I could start playing with some of the components. First was the pneumatic Master compass:



Obviously it is not possible to convey the absolute compass direction with a pneumatic signal, but what is possible is to convey the relative direction to a given setpoint. If the compass points left of the setpoint, the first of a pair of pressure outputs receives a vacuum pressure, if it points right the other output receives the vacuum. These two vacuum outlets can be connected to an instrument, which will indicate left or right. Another instrument will be needed to change the setpoint of the master compass and relate this setpoint to a compass direction:



On the left of the panel the left/right indicator (marked with "Kompaß") is fitted above the compass setpoint controller. By turning the handle the indicated compass direction will change and via a flexible drive cable the setpoint of the master compass is adjusted accordingly.

Testing the compass revealed one of the problems in these pneumatic systems, rubber hoses inside the master compass had perished so there was no vacuum output. The compass had to be dismantled so that the rubber could be replaced. With this done and vacuum applied to the master compass, the system operated surprisingly well, the indicator instrument swinging left and right with the compass.

It is interesting that this indicator instrument was retained in later types of autopilots, the "Kurszeiger" used with the Siemens and Patin autopilots was an exact electrical version of this display instrument.

So that was the first part of the autopilot working again....

regards,

Funksammler

Last edited by Funksammler; 11-26-2017 at 01:26 PM.
  Reply With Quote

Old 11-26-2017, 12:48 PM   #2
IK5QLO
New Member
 
IK5QLO's Avatar
 
IK5QLO is offline
Join Date: Nov 2017
Location: Lucca - Italy
Posts: 35
Default

Excellent work, congratulations! pse keep us informed about your progress.
  Reply With Quote

Old 11-28-2017, 05:05 AM   #3
Funksammler
Member
 
Funksammler is offline
Join Date: Jan 2009
Location: Normandy
Posts: 2,959
Default

The last picture in my first post shows more instruments in the panel; on the right hand side sits a "Fernkurskreisel" or gyrocompass. Since magnetic compasses are not stable enough to provide an accurate directional signal for an autopilot, a gyroscopically stabilised compass is required. The gyrocompass provides stability but it will slowly wander from the true compass if left alone, so some means of keeping the gyroscope locked to the magnetic compass is required. This is done by the "Umwandler" in the system:



The Umwandler is fitted in a round instrument casing, has two pneumatic and an electrical connection, it is fitted on the bottom right on the backside of the panel. It is connected pneumatically in parallel with the left/right indicator instrument, either one of two contacts is closed in the "Umwandler" dependent if the magnetic compass indicates left or right. This electrical signal is fed to a pair of precession coils in the gyrocompass and will make tiny adjustments to the gyroscope. Over time these tiny adjustments will average out all the swings of the magnetic compass and keep the aircraft pointing in the set compass direction. The electrical switch on the front panel serves to switch this electrical signal on or off. The small round indicator on the gyrocompass instrument indicates white (as is shown on the picture) when the system is on.

Apart from this small electrical system, the gyrocompass is pneumatic. The gyroscope is powered by vacuum and -like the compass- it produces vacuum on either of two pneumatic outputs dependant on a deviation left or right from the set course.

The set course is shown on the top scale of the gyrocompass. It can be changed with the little handle fitted to the instrument which is coupled to the setpoint controller of the compass.

The gyrocompass suffered from the same problem as the master compass, a number of internal rubber connections had perished, so the instrument did not produce output pressures. When I got the instrument, the input air filter was also blocked, meaning that insufficient air could be sucked in by the vacuum to power the gyroscope. After cleaning and repair the instrument worked fine. The contacts in the "Umwandler" also needed carefully adjustment to ensure that the switchover occurs exactly at zero differential pressure.

The testing of the vacuum instruments proved that the compressor was able to provide ample airflow to power these instruments, so I will fit a vacuum control valve (as was fitted to the original system) to ensure adequate airflow on the suction side of the compressor.

regards,

Funksammler
  Reply With Quote

Old 11-28-2017, 11:17 PM   #4
Quatsch
Member
 
Quatsch is offline
Join Date: Oct 2015
Location: USA
Posts: 127
Default

After the funkwagen I expect a full me262 or do117 restoration.
  Reply With Quote

Old 12-01-2017, 06:10 AM   #5
Funksammler
Member
 
Funksammler is offline
Join Date: Jan 2009
Location: Normandy
Posts: 2,959
Default

Quote:
Originally Posted by Quatsch View Post
After the funkwagen I expect a full me262 or do117 restoration.
On my wishlist, haha!

Failing an airplane, a bit more on the autopilot... In order to stabilise an aircraft in flight, at least a 2nd order control system is required, so the Luftwaffe used a damping gyroscope to measure the yaw velocity in the Lz12 system. This gyroscope is built into the "Kurs Steuergerät". Apart from containing the gyro, the box also adds up all the different signals (from the gyrocompass, the damping gyro, the feedback from the rudder servo and a curve signal) and produces a power output signal for the rudder servo:



The above picture clearly shows the damping gyroscope. The crescent shaped cutouts cause the gyroscope to spin when air is blown on them, you can see the feed line to the air nozzle running above.

You will also note that the box is quite dirty on the inside. Rather than vacuum, the "Kurs Steuerkasten" works on compressed air, typically an air compressor was fitted to one or both engines of the aircraft. These fane-type compressors require lubrication, an air cleaner is required downstream of the compressor to clean the air from the lubrication oil. It is clear that the aircraft mechanic responsible for this particular "Kurs Steuerkasten" had forgotten to empty the oil catchment vessel of the filter system which resulted in oil carryover into the pneumatic equipment. I suspect, rather than a laborious cleaning, they chose to replace the "Kurs Steuerkasten" with a new one. I suspect the aircraft mechanic got a bit of a telling off! The oil film did help to preserve the interior of the "Kurs Steuerkasten", but the oil had thickened and hardened over time so it did require a bit of a cleaning up.

After cleaning and finding pneumatic hose connections that fitted, I hooked up the "Kurs Steuerkasten" to the gyrocompass and the compressed air supply:



Much to my satisfaction, my compact compressor worked a treat, the gyroscope in the "Kurs Steuerkasten" spins up to tremendous speed. I did find that the output valve of the box needed a bit more cleaning, but in the end I could feel a differential pressure on the output connections.

The final major component in the system is the rudder servo:



This is powered by the compressed air coming from the "Kurs steuerkasten". Again pneumatic air connections of the right size were found and it was hooked up to the system. To my even greater satisfaction the rudder servo responded to movement of the "Kurs Steuerkasten" and the gyrocompass. The system did need some adjustment though as it was strongly biassed to one side. The "Kurs Steuerkasten" has a number of regulator screws that allow you to adjust the sensitivity of the various inputs and the bias balance. On the picture that shows the damping gyroscope you can actually just see this balance control: an axis with a worm wheel drives a small gear which causes a mounting to slide left or right. These adjustment also needed to be cleaned of the old oil before it moved freely, but after that the bias could be balanced accurately.

So now with the system working (probably for the first time in 75 years) I must say I am pleasantly surprised at the accuracy of the pneumatic control. Turning the gyrocompass a fraction of a degree out of alignment results in a movement of the rudder servo.

For now the system is only rigged up for testing:



This provisional rig-up has proved that my compact compressor can adequately supply the autopilot of both vacuum and compressed air and it has proved that all the components appear to be working adequately. Above all, it proves that everything is there to build a working demonstration model of the Lz12.

The next stage will be to build the autopilot components onto a rotation platform like my other autopilots. Hopefully I will be able to make progress with that this winter....

So are there any other autopilot collectors out there....?

regards,

Funksammler
  Reply With Quote

Old 12-13-2017, 07:20 AM   #6
Funksammler
Member
 
Funksammler is offline
Join Date: Jan 2009
Location: Normandy
Posts: 2,959
Default

After the initial "proof of concept" I started building the autopilot into a working model. The first challenge is to fit all the elements onto a compact space, balancing the weight. An additional challenge for the Lz12 model is to keep the master compass as far away as possible from any iron, electric currents etc. My main worry in that respect was the pneumatic pump, some initial tests showed I had to keep that at least half a meter away from the compass.

So I ended up with the instrument cluster and pump on one end, the compass on the other, and the rudder servo and control box somewhere in between:



I wanted to hide the modern pneumatic pump, so this is actually suspended underneath the frame, so from above you will only see the original autopilot components.

The other side shows the master compass as far to the back as possible:



At the moment the frame is not yet able to rotate freely and is supported on two stands. First of all I want to mount all the components (including the drive mechanism for the rotating platform). This will allow me to find the exact centre of gravity without the need to have to fit any balancing weights to the frame. The frame is getting heavy enough as it is!

This picture shows how the pump is slung underneath the frame:



Fitted to the pump is an original "Sogregler" or vacuum regulator valve. This ensures that the vacuum pressure does not exceed the maximum in case the airfilters on the compass and gyroscope clog up.

The vacuum supply runs to an original "Sogverteiler" or vacuum manifold fitted next to the gyro:



This manifold distributes the vacuum pressure to the different vacuum instruments. Additional instruments such as a turn indicator could be added if required. Each of the four outputs from the header has a different hole size, getting smaller towards the top. So the gyroscope, fitted to the bottom connection gets most air. Originally the compass system would have been fitted to the top (smallest hole) connection, but I found I got better sensitivity of the compass system giving it a bit more air.

I particularly like this bit:



Spagetti junction! I made some wooden clamps that hold the pneumatic lines neatly in place. The metal line near the bottom holds the flexible driveshaft connected to the master compass.

During the next phase of the construction I will fit the drive system for the rotating table and a "rudder" to show how the aircraft would behave. I am still looking for an old bicycle to cut out the steering bearing/front fork assembly to use as a bearing for the frame. So if anybody has an old bicycle to sacrifice (must be steel so I can weld it), let me know!

regards,

Funksammler
  Reply With Quote

Old 12-13-2017, 11:49 PM   #7
Yuri D.
Association Member
 
Yuri D.'s Avatar
 
Yuri D. is offline
Join Date: Apr 2001
Location: Western Hemisphere
Posts: 6,172
Default

My God! Where do you get all this stuff ?
  Reply With Quote

Old 12-14-2017, 12:12 PM   #8
Funksammler
Member
 
Funksammler is offline
Join Date: Jan 2009
Location: Normandy
Posts: 2,959
Default

Quote:
Originally Posted by Yuri D. View Post
My God! Where do you get all this stuff ?
I ask myself that question as well! Every few years or so, an interesting piece turns up. So very slowly, bits and bobs come together until there is enough of it to attempt to get it all to work. The key is to know what you are looking for. The autopilot is far from complete, I have a relic main operating switch which will need to be completely rebuilt, but at least I have one... I would love the find the correct "Kursmotor" and "Richtungsgeber" for the Lz12 somewhere in the next decade or so!

regards,

Funksammler
  Reply With Quote

Old 03-07-2018, 05:59 AM   #9
Funksammler
Member
 
Funksammler is offline
Join Date: Jan 2009
Location: Normandy
Posts: 2,959
Default

Another few months have passed and slow progress has been made on the Lz12 autopilot. After a long search, an old bicycle frame was found and the front fork assembly recovered. This allowed me to build up the turntable for the autopilot model.

For the model to work a "rudder" had to be incorporated, so that the movement from the rudder servo results in the rotation of the turntable in the same way as an aircraft would do. Since the model can not depend on airflow to create rotation, a special electrical drive had to be build in order to achieve this.



The Pertinax "rudder" was made long enough to reach the front panel. I will mount a little indicator on top so that you can observe the rudder movement while watching the panel.

The base shows the recycled bicycle bearing at the centre. It provides a strong support for the frame while being hollow so that electrical cables can be run through the centre:



Note that the vacuum tube from the pump the the vacuum regulator had to be routed around the base. Building a compact model like this is constant a juggle for space. Remember that the weight of all the components needs to be balanced and electrical and iron components need to be kept away as far as possible from the master compass.

The green "wire" running on the circular wheel is the drive cord; a small electric motor fixed to the turntable provides the drive.

With all this operational I could make the first attempt to test the autopilot in action. It quickly transpired that the autopilot could not be tuned to be stable, it kept swinging wildly from side to side. In itself this was useful to give the autopilot some exercise and clean up the various pneumatic and mechanical controls.

The reason for the oscillation in a control system is excessive gain in the feedback. In other words, the autopilot and the turntable drive were running too fast. The autopilot's speed can be controlled with a pneumatic feedback from the rudder servo, unfortunately the feedback cylinder was no longer present on my rudder servo. A suitable replacement cylinder had to be found and adapted to fit to the rudder servo:



This cylinder provides a small airflow which is fed back to the "Kurssteuergerät" to counter the movement of the output valve thus slowing down airflow to the rudder servo.

The slowing down of the turntable drive had to be done by reducing the torque of the drive motor which can be done by placing a resistor in series with the motor's rotor winding. By playing with the value of this resistor the right response of the turntable could be found. I also found that the motor torque was not equal in both directions, causing the turntable to run faster in one direction than the other. A resistor was added to one of the stator windings (this type of motor has two stator windings, one for turning in each direction) so that the torque could be equalised in both directions:



At the moment these resistors are just temporarily tied in; once the "tuning" of all the components is complete these resistors will be fitted permanently to the turntable model.

This result of the slowing down is a relatively smooth and stable operation of the autopilot. It is clear that the LZ12 is suitable for relatively large and heavy aircraft as it can only control movements that are relatively slow. In this respect it is outclassed by the later generations of autopilots.

Further work on the model will focus on refining and tuning. Once this has been achieved I will attempt to restore the autopilot's main control switch which works with Bowden cables. Another significant challenge ahead....

regards,

Funksammler
  Reply With Quote

Old 03-09-2018, 10:36 AM   #10
Funksammler
Member
 
Funksammler is offline
Join Date: Jan 2009
Location: Normandy
Posts: 2,959
Default

The last few days saw a lot of running in an attempt to tune the autopilot. A number of variables can be tuned on the autopilot: the gain of the directional signal; the gain of the angular velocity signal; the strength of the feedback system. Apart from these autopilot variables, the torque and speed of the drive system of the turntable can be varied.

It was quickly found that the model could only be stabilised with the gain of the directional signal set at minimum and the strength of the feedback system set near maximum, otherwise the signal from the damping gyroscope is simply not strong enough. This is probably due to the relatively low supply pressure to the "Steuergerät". In the aircraft the pressure was supplied by an engine driven compressor of 0.7 Horsepower, a bit stronger than my small compressor unit!

The lower pressure results in a slightly slower spinning of the gyroscope which means it produces a smaller signal, relatively weak in comparison to the directional signal which it is competing with.

Another reason for finding it difficult to tune is due to the dynamics of the autopilot. Unlike static feedback systems such as the Siemens K4 and K12, the Lz12 produces a dynamic feedback signal. In a static feedback system, the rudder angle is proportional to the deviation from the set course, so if the course deviation is held stable, the rudder will not move. With a dynamic feedback system the rudder speed is proportional with the deviation from the set course, so even if the course deviation is stable, the rudder angle will keep increasing.

I was somewhat surprised at the dynamic behaviour of the Lz12, as dynamic feedback autopilots like the Patin PKS11 were only "invented" much later. In any case, the dynamic feedback makes it much harder the tune the autopilot, it has probably taken me 10 running Hours to reach a setting I was happy with.

OK, enough talk for the technophiles! After the settings were found the various loose resistors and wiring were tidied and the autopilot was moved to its place in my autopilot rack:



Lying near the base are the original compressor and vacuum pumps. These took their power directly from the engine of he aircraft. The compressor and vacuum pumps are very similar, the compressor can be recognised by an extra sleeve and cooling ribs.

A few of the "modern" bits on the turntable showing the compressor, a 24V DC supply and the electronics for the drive system of the turntable:



With the loose wires and resistors gone, it all gives a tidier impression:



Especially the wring between the gyrocompass and the pressure switch needed some cleaning up:



The pressure switch translates the pneumatic signal from the magnetic compass into an electrical signal which automatically synchronises the gyrocompass with the magnetic compass. This was called the "Überwachung" of the gyrocompass. The next few photographs show the Überwachung in action during one of the running tests:



In this photograph you can see that the gyroscope course is aligned with the set course but the magnetic compass indicator is off to the right. The Überwaching will now very slowly (about 2 degrees per minute) precess the gyroscope, causing the aircraft to turn towards the magnetic compass course. After a few minutes, another picture was taken:



The magnetic and gyrocompass are now in alignment and will be kept that way as long as the Überwaching is switched on. You can check that it is switched on by the round indicator in the top left corner of the gyrocompass.

So even though the restoration is not yet 100% complete, there exists once again an operational Askania Lz12 autopilot!

regards,

Funksammler

Last edited by Funksammler; 03-09-2018 at 11:29 AM.
  Reply With Quote

Old 03-10-2018, 03:59 AM   #11
IK5QLO
New Member
 
IK5QLO's Avatar
 
IK5QLO is offline
Join Date: Nov 2017
Location: Lucca - Italy
Posts: 35
Default

Incredible Job, congratulations! Pse keep updates coming.

Regards
  Reply With Quote

Old 03-13-2018, 08:30 AM   #12
roger_m
Member
 
roger_m's Avatar
 
roger_m is offline
Join Date: Nov 2008
Location: Europe
Posts: 174
Default

Quote:
Originally Posted by Funksammler View Post
The autopilot is far from complete, I have a relic main operating switch which will need to be completely rebuilt, but at least I have one... I would love the find the correct "Kursmotor" and "Richtungsgeber" for the Lz12 somewhere in the next decade or so!
Very nice display. Pls post pics of what you are still looking for your unit.
  Reply With Quote

Old 03-13-2018, 03:31 PM   #13
Funksammler
Member
 
Funksammler is offline
Join Date: Jan 2009
Location: Normandy
Posts: 2,959
Default

Here are some drawings of the Kursmotor and Richtungsgeber used with these early autopilots. These parts were usually scrapped with the aircraft, so a bit like a needle in a haystack.

Here is the Kursmotor:



it has a connector on the top and the drive on the other end. I suspect it had a electrical clutch build into the motor housing. I am looking for the 24 Volt model, the Lfzm/4

The Richtungsgeber consists of two parts:





The base unit on top is called the Lfz/5 and contains the switch for the clutch in the Kursmotor, the actual direction switch, called the Lgri/3 was fitted on top of the Lfz/5

I rate my chances of finding these near zero, but you never know....

Anybody going to Speyer this spring?

regards,

Funksammler
  Reply With Quote

Old 03-14-2018, 03:09 AM   #14
roger_m
Member
 
roger_m's Avatar
 
roger_m is offline
Join Date: Nov 2008
Location: Europe
Posts: 174
Default

Quite rare bits indeed. But chances are that if they pop up will be very cheap. Very few people is looking for such prewar pneumatic stuff.
I never missed one Speyer in 25 years. Hope to see you there.
  Reply With Quote

Old 03-14-2018, 05:52 AM   #15
Funksammler
Member
 
Funksammler is offline
Join Date: Jan 2009
Location: Normandy
Posts: 2,959
Default

Unfortunately I can't make it this spring (I always have to choose between Speyer, Ciney and Militracks, I can only do one of them...) so I am looking for somebody who can keep an eye out for certain items (such as the ones below) in Speyer...

regards,

Funksammler

Quote:
Originally Posted by roger_m View Post
Quite rare bits indeed. But chances are that if they pop up will be very cheap. Very few people is looking for such prewar pneumatic stuff.
I never missed one Speyer in 25 years. Hope to see you there.
  Reply With Quote
Reply

Bookmarks


Currently Active Users Viewing This Thread: 1 (0 members and 1 guests)
 
Thread Tools
Display Modes

Posting Rules
You may not post new threads
You may not post replies
You may not post attachments
You may not edit your posts

BB code is On
Smilies are On
[IMG] code is On
HTML code is On

Forum Jump






vBulletin skins developed by: eXtremepixels
Powered by vBulletin® Version 3.8.11
Copyright ©2000 - 2018, vBulletin Solutions Inc.
Copyright Wehrmacht-Awards.com