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

New project - Patin KS12b-1 Jägersteuerung
Old 07-09-2018, 09:04 AM   #1
Funksammler
Member
 
Funksammler is offline
Join Date: Jan 2009
Location: Normandy
Posts: 2,943
Default New project - Patin KS12b-1 Jägersteuerung

For many years now I have slowly been collecting parts for another autopilot system. The system in question is a late war development of the PKS11 system: the Patin Jägersteuerung KS 12b-1.

As the name suggests this was an autopilot developed for fighter aircraft and it was applied in late war Bf109 and Fw190 (Ta152) "Schlechtwetterjäger". Before going into much technical detail it is worth re-reading my work on the PKS 11 system, as rudder servo system is common to both the KS 11 and 12 systems: https://www.dropbox.com/s/25w6yppx2e...pilot.pdf?dl=0

The Patin KS-11 system obtained the signal for the angular deviation from either a "Kurskreisel" or from the "Kurszentrale" gyrocompass. The use of a gyrocompass however introduces an error when using high roll angles, one of the reasons why the application of previous autopilots to fighter aircraft proved problematic.

Instead of measuring the angular deviation with a separate gyroscope, it can alternatively be calculated by integrating the angular velocity signal so it was this principle that was applied in the Luftwaffe's "Jägersteuerungen".

It is worth remembering the fact that the earlier Patin KS-11 system used a "Dämpfungskreisel" that generated a mixed signal from the angular velocity and angular acceleration. This mixture is not suitable for integration, so a new "Dämpfungskriesel" was developed that produced purely a angular velocity signal. This simplification also reduced the manufacturing times for the "Dampfungskreisel" significantly, a factor become increasingly important towards the end of the war. The missing angular acceleration signal could easily be regenerated by another calculating trick: differentiating the angular velocity signal.

The new system with the simplified "Dämpfungskreisel" was called the Patin K12. A unit called the "Lage Integrations Gerät" or LI-Gerät was developed to calculate the angular deviation. The LI-Gerät replaced the gyrocompass to create the Patin KS 12 b-1 "Jägersteuerung".

Schematically the system looks like this:



The "calculations" in the system are done by two Patin relays. These were specially developed for the PKS 12 b-1 system and were called the SR5 and SR6. Each of these relays has five input windings (as opposed to four in the SR2 used in the PKS-11).

I was extremely fortunate to spot two SR6 relays in a recent Ebay auction.



Both relays proved to be in serviceable condition (although one has been messed with and needs a bit of work...) meaning that the restoration of a working PKS12 b-1 system using original components is now a step closer.

regards,

Funksammler
  Reply With Quote

Thank you
Old 07-09-2018, 02:33 PM   #2
LA6NCA
Member
 
LA6NCA's Avatar
 
LA6NCA is offline
Join Date: Mar 2013
Location: Larvik, Norway
Posts: 152
Default Thank you

I'm impressed with what you get to work.
I am looking forward to knowing more about this.
Very interesting.
LA6NCA
  Reply With Quote

Old 07-10-2018, 02:00 PM   #3
rv12p2000
Member
 
rv12p2000 is offline
Join Date: Jan 2012
Location: france
Posts: 383
Default

Hello funksammler

Bravo for this new project

Could you restore this link : https://www.dropbox.com/s/25w6yppx2e...pilot.pdf?dl=0

because it ends with the error message 404

Thank

Regards
RV12P2000
  Reply With Quote

Old 07-10-2018, 02:58 PM   #4
Funksammler
Member
 
Funksammler is offline
Join Date: Jan 2009
Location: Normandy
Posts: 2,943
Default

Quote:
Originally Posted by rv12p2000 View Post
Hello funksammler

Bravo for this new project

Could you restore this link : https://www.dropbox.com/s/25w6yppx2e...pilot.pdf?dl=0

because it ends with the error message 404

Thank

Regards
RV12P2000
What does the link refer to? I can't work it out from the code...

regards,

Funksammler
  Reply With Quote

Old 07-10-2018, 04:11 PM   #5
Gabbertje
Association Member
 
Gabbertje is offline
Join Date: Jul 2008
Location: Germany - Near Frankfurt a.M.
Posts: 111
Default

Looks and sounds like a great project! :-)

I'm looking forward to see this next project grow!


best regards,

Stefan
  Reply With Quote

Old 07-10-2018, 04:29 PM   #6
rv12p2000
Member
 
rv12p2000 is offline
Join Date: Jan 2012
Location: france
Posts: 383
Default

Hello

Funksammler

When I click on this link : https://www.dropbox.com/s/25w6yppx2e...pilot.pdf?dl=0

The following error message is displayed in French at this url :

https://www.dropbox.com/s/25w6yppx2e...pilot.pdf?dl=0

404 Nous ne trouvons pas l'objet de votre recherche (404 We do not find the object of your search)

regards

RV12P2000
  Reply With Quote

Old 07-11-2018, 01:23 AM   #7
Funksammler
Member
 
Funksammler is offline
Join Date: Jan 2009
Location: Normandy
Posts: 2,943
Default

Ah OK, it is the PKS11 link. I thought I already repaired it....

https://www.dropbox.com/s/25w6yppx2e...pilot.pdf?dl=0

regards,

Funksammler
  Reply With Quote

Old 07-11-2018, 04:14 AM   #8
rv12p2000
Member
 
rv12p2000 is offline
Join Date: Jan 2012
Location: france
Posts: 383
Default

Hello Funksammler

Many thank

regards
RV12P2000
  Reply With Quote

Old 07-11-2018, 02:04 PM   #9
Funksammler
Member
 
Funksammler is offline
Join Date: Jan 2009
Location: Normandy
Posts: 2,943
Default

Another key element is the "Dämpfungskreisel" of the PKS 12 system:



It is a fairly simple instrument, essentially a turn indicator but instead of driving an indicator needle, the gyroscope drives an arm with two potentiometer runners attached. The length of the arm ensures a very sensitive measurement of the turning velocity.

Note the relatively crude finish on the die cast frame, it is designed and fabricated with the minimal amount of parts, a far cry from the "gold-plated" engineering at the start of the war.

regards,

Funksammler
  Reply With Quote

Old 07-13-2018, 07:27 AM   #10
Funksammler
Member
 
Funksammler is offline
Join Date: Jan 2009
Location: Normandy
Posts: 2,943
Default

The Patin PKS 12 b-1 was developed and manufactured towards the end of the war and was used from the end of 1944 in "Schlechtwetterjäger". The main purpose was to help relatively untrained pilots to fly blind through clouds and darkness. It was applied in the Messerschmitt Bf 109 G10/R6 version which came in operation in late 1944. One source states only some 180 aircraft of this type were built. It was also used in some Fw 190 A8's, some of these aircraft were even used as night fighters where an autopilot would definitely have helped the pilots navigate the dark skies. How many FW190's with autopilot were built is not stated, but I assume similar numbers as the Bf109 versions. So overall only a couple of hundred aircraft were ever equipped with the PKS 12 b-1 making surviving components pretty rare.

I was extremely fortunate to pick up a number of new old stock spare components, I was even more fortunate to find a wreck of the "Steuerkasten" that held them:



The SK15 "Steuerkasten holds the damping gyroscope, the SR5 computing relay, the wobble motor and the differentiating capacitor (to generate the angular acceleration signal). The box I obtained had been flown and crashed and spend several decades in the ground so not quite "mint" but worthy of restoration:



From the last picture you can see that the box was twisted on impact. Somebody had started on an attempted restoration removing a number of components and cutting out wiring. Unfortunately that makes it a lot more difficult to work out the exact circuitry and components. Since no circuit diagrams of the SK15 survive I had to re-engineer the circuitry and wiring based on the earlier PKS 11 box and the high level descriptions and drawings in the operating manual.

The serial number on the box is 1683 while "gzy" is the three letter code for Patin:



It is not known if this indicates the true size of the production run, they might have started at 1000 to hide the true production figures. In any case it makes the SK15 "Steuerkasten" and extremely rare survivor; I personally only know on one other example in existence in the famous Beck collection.

After a slight polish it turned out like this:



Now old stock "Dämpfungskreisel", "Steuerrelais" and "Wobbelmotor" are now fitted to the box.

The "Wobbelmotor" generates a low frequency AC voltage which is fed to the "Steuerrelais". This increases the sensitivity of the "Steuerrelais" but does not influence the signal itself.

A new differentiation capacitor block fills the rear component. The four screws on top of the capacitor block serve to connect or disconnect four individual capacitors in parallel, providing a simple method to "tune" the capacitance to the required value. I ordered special unpolarised electrolytic capacitors as the voltage will be positive as well as negative dependent on which way the aircraft turns. The small cover under the "Dämpfungskreisel" on the front of the box hides a number of tuning resistors, allowing the strength of the various signals to be varied and tuned.

The box has since been fully fitted, rewired and tested and is now fully functional again. Not bad for something that crashed and burned to the ground nearly 75 years ago!

regards,

Funksammler
  Reply With Quote

Old 07-13-2018, 07:45 AM   #11
rv12p2000
Member
 
rv12p2000 is offline
Join Date: Jan 2012
Location: france
Posts: 383
Default

Hello Funksammler

I found on the net a very interesting document of which you are the author, this document was translated into Czech

Please do you have the original in English?

link : http://www.luftschraube.g6.cz/?tag=pks-12

Regards
RV12P2000
  Reply With Quote

Old 07-13-2018, 08:45 AM   #12
Funksammler
Member
 
Funksammler is offline
Join Date: Jan 2009
Location: Normandy
Posts: 2,943
Default

Ah yes, I did write on the PKS12 before, I forgot about that one. I guess I will update the article at some point with the current experiences gained during the restoration of the equipment....

https://www.dropbox.com/s/zs7gcbzdm8...erung.pdf?dl=0

regards,

Funksammler
  Reply With Quote

Old 07-15-2018, 04:51 PM   #13
rv12p2000
Member
 
rv12p2000 is offline
Join Date: Jan 2012
Location: france
Posts: 383
Default

Hello Funksammler

Many thank for this document

Regards

RV12P2000
  Reply With Quote

Old 07-17-2018, 11:29 AM   #14
Funksammler
Member
 
Funksammler is offline
Join Date: Jan 2009
Location: Normandy
Posts: 2,943
Default

As discussed in my first posting, the "Kurs" signal is generated by a "LI-Gerät" instead of by a classical directional gyroscope. I was very fortunate to pick up an "LI-Gerät" over 10 years ago with the faint hope of completing the system somewhere in the future:



When I got my "LI-Gerät", it it was fitted with a SR5 relay, but I know believe it was originally fitted with a SR6 relay instead while the SR5 was used in the "Steuerkasten". I have not been able to find much differences between the relays, the input impedances of all the windings and the output impedance are all equal. The only difference I spotted was that the SR5 has quadruple runners on the output potentiometer instead of the double runners on the SR6, so it seems that the SR5 is designed to deliver a bit more current than the SR6.

The "LI-Gerät" stands for "Lage Intergrations Gerät" as the unit integrates the turn velocity signal to calculate the directional deviation. So an important part of the unit is an integrator, which uses the Patin control relay and a capacitor:



This picture shows the capacitor more clearly; three bipolar 70µF block capacitors in parallel giving a total capacitance of 210 µF:



One element was missing, a small potentiometer. In the schematic this is drawn as the "Nullpunktsabgleich", it took be quite a while to figure out how this was supposed to work.

The first job was to create an experimental cable tree to be able to link the various units together. This is always the most nerve wracking part of a restoration as one little mistake can have disastrous consequences. The Patin control relays are full of extremely delicate contacts and wiring and they can not withstand short circuiting. The SR5 and SR6 relay even have input windings that are too fragile to take the full supply voltage so great care needs to be taken. Since no original wiring schematics exist, a fair bit of research and engineering needs to be conducted before a working autopilot is recreated. So here is the partial system rigged up on my window ledge:



Apart from the "Steuerkasten" and the "LI-Gerät" there is the three phase "Umformer" to drive the gyroscope. I have added a "Koppelschalter" switch. My latest addition proved to be crucial, a resistor network that mimics the function of the PFK/f3 "Führertochterkompaß".

The PFK/f3 contains an output circuit that generates a signal that indicates if the compass is left or right of the setpoint. This is done by a Wheatstone bridge consisting of two fixed resistors and a potentiometer. I discovered that the "Nullpunktsableich" potentiometer is part of a second Wheatstone bridge using the same fixed resistors in the PFK/f3. So the "LI-gerät" can only properly be adjusted if it is connected to a PFK/f3. I suspect that this was the reason why the resistor was removed from my "Li-gerät", it was probably tested postwar in an aircraft without a Patin compass system and they thought they could make it work without the adjustment. My initial tests were also made without the adjustment circuit in place as I thought it would make no difference.

The initial tests showed some functionality, but the integrated signal returned back to zero after 5 -10 seconds, which is too quick for the autopilot to work properly. One of my problems is that I have not yet been able to model the integrating circuit using the PSR control relay and the capacitor and therefor I do not yet understand how it is supposed to work!

I did notice that the output signal of the drifted to one side and that it proved difficult to offset this drift. Part of the drift was caused by the damping gyro. Even in rest position this is giving a small output voltage. I tried to adjust the gyro springs to get the output as close to zero as possible, but it proved impossible to get below 0.1 Volt deviation. Over time, even this small deviation drives the integrator into saturation.

So I decided that I needed to recreate the "Nullpunktsabgleich" circuit to try and counter the deviation. This circuit essentially puts a small positive or negative voltage over another input winding of the control relay in the "LI-Gerät". Much to my surprise, this completely transformed the functioning of the "LI-Gerät", all of a sudden it started integrating properly!



The above trace from my oscilloscope shows the "Kurs" signal coming out of the "LI-Gerät". I turn the "Steuerkasten" to the left and the signal moves up to a new steady state until I turn it to the right in the number of steps. The next movement to the right does not change the steady state because I have reached a saturation point. I now move left again in a number of steps until I reach saturation on the other side so I move back to the right.

Each time I stop the movement, the signal holds steady. The integrating circuit "saturates" at about plus or minus 5 Volt, about a fifth of the supply voltage. .

So the "LI-Gerät" now works as I would expect it to do, but I have no clue why and why the activation of the "Nullpunktsabgleich" makes such a difference!

Anyway, I have much thinking to do to explain this theoretically, but from a practical point of view it allows me to move forward. One thing I can now do is to "tune" the signal from the damping gyro so that the saturation points appear at plus and minus 20 degrees from the set point (the output circuit from the PFK/f3 compass instrument only works in this range so it seems a logical limit to use; this limit is also used in the Patin "Kurszentrale").

One thing is sure, whoever tested the unit after removing the "Nullpunktsabgleich" system was probably disappointed as the "LI-Gerät" does not seem to integrate properly without it....

regards,

Funksammler

Last edited by Funksammler; 07-17-2018 at 11:37 AM.
  Reply With Quote

Old 07-18-2018, 06:50 AM   #15
Funksammler
Member
 
Funksammler is offline
Join Date: Jan 2009
Location: Normandy
Posts: 2,943
Default

Another day and a bit more progress and understanding... The euphoria of the "LI-Gerät" working properly was short lived. As I was trying the repeat the measurements in the evening, the unit was up to it's old tricks and no longer provided a stable output:



The right hand side of the screen shows the effect I was getting, there was quite a lot of noise and the voltage drifts slowly up after each steering input (the sharp dips). the noise is caused by an AC voltage and the only AC voltage in the system that could cause the noise is the wobble motor:



The WU10 wobble motor is a simple DC motor on one end which drives a permanent magnet which in turn generates a small AC voltage in the output winding.

The idea is that by supplying this small AC voltage to the SR5 and SR6 control relays, these relays vibrate ever so slightly, reducing the mechanical friction and increasing the sensitivity. The wobble signal is not supposed to influence the output of the control relays, which according to the measurements it clearly was (in fact the left side of the above oscilloscope trace is showing the output of the "LI-Gerät" with the wobble motor removed, so it clearly makes the difference). The wobble motor I was using was running too slow, getting within the operating frequency of the control relays (approximately <40 Hz).

This slight signal is enough to create a disturbance in the integrator causing it to drift. Fortunately I found another wobble motor which ran at the required speed and it transformed the behaviour of the "LI-Gerät":



The noise is gone from the signal and the output from the "LI-Gerät" stays stable after each turning motion of the damping gyro.

So my initial thought that the "Nullpunkstabgleich" made all the difference was wrong, it was pure coincidence, I must have accidentally knocked the wobble motor out of it's socket during my earlier test. At least I now have a repeatable result and I am slowly starting to make sense of how the integrator works.

regards,

Funksammler
  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