Saturday, September 28, 2013

The Austrian station block: An animated explanation - the internals

Deutsche Version

This second posting about Austrian interlockings of type 5007 has the goal to present some internal mechanical details. Most prominently, it explains the inner workings of the Siemens block instruments, whose fundamental function I explained at the beginning of the previous posting.

The Siemens block instrument was invented by Carl Frischen, a co-worker of Werner Siemens at Berlin, in 1871. Although its initial application was with block working over the line—where the secure communication need was most pressing—, it was soon used to secure the dependencies in the area of a single station with multiple signal boxes, which were under the lead of a single traffic director. This use of block instruments became later known as "Bahnhofsblock," which I translate as "station block."

While I intend to present the internal workings of a command frame and a lever frame in sufficient detail to show the mechanical and electrical dependencies, I have abstracted away many (many many!) parts that are necessary in the real world, but are only distractions for that general understanding. I dropped some parts completely, e.g., all electrical circuits, or the double locking of route bars—maybe I find time to explain these parts later.

To see what is going on inside these frames, I go once more over that sequence of six steps from the previous posting, where the traffic director, the pointsman and the train work together to get the train into the station.

4.1. Traffic bureau: Transmit order

Again, the traffic director tilts the small route lever and blocks the Ba instrument—which at the same time unblocks the Be instrument at the signal box— and then triggers the track indicator:

We now take a closer look at the blocking and unblocking of the two block instruments.
Let me stress a last time that block instruments consist of many more parts than shown in my explanations. However, their basic function is exactly as shown.
A block instruments consists of a
  • toothed rack, whose teeth engage with the two teeth of an
  • anchor, which therefore locks the rack in place.
When, however, the anchor is moved back and forth by two electro-magnets, the rack can wiggle downwards or upwards tooth by tooth. When the block instrument is moved to the blocking position (it is "blocked"), the rack moves only by gravity. However, there is an additional obstacle: On the rack, there is a pin, which is held up by the rack guide of the locking rod. The locking rod is held up by a spring, so altogether the rack is held in its upper position. To free the way for the rack, this rod must be pushed down; the respective operator (in our case, the traffic director) does this indirectly by pushing down the knob (or handle) atop the handle rod which, in addition, also connects the magnets to the inductor.

Therefore, by pressing the knob and turning the inductor crank, the block instrument can be blocked. At the end of this process, the rack has moved down, turning about its axle by about 60 degrees. The stopped anchor again locks the rack in place, which now continues to press down the locking rod even if the handle rod is released. This allows the block instrument to lock whatever it is supposed to lock in the frame below.
Actually, the Siemens block instruments were so predominant in Central and Eastern Europe that all frame designs could be combined with them for locking purposes—down to quite simple devices only used for temporary interlocking installations like the Austrian "Trommelschlüsselwerk" or the German "Schlüsselwerk" (two types of lock-and-key matrices).
In the command frame, the locking rod locks the route bar. Here is an animation of this blocking action:

Blocking the Ba block instrument unblocks the corresponding Be instrument at the signal box. Also there, the locking rod is pulled up by a spring, but because of the blocked-down rack, this locking rod is currently in its lower locking position. When the instrument is unblocked, however, the oscillating anchor allows the rack to move upwards tooth by tooth, pushed at the rack pin by the rack guide, which is lifted by the locking rod's spring. The lock is thus released:
By the way, although springs are used in these devices at crucial places, they are designed in such a way that a broken spring will never compromise security. What might happen is that a rod drops at a place where it should be in its upper position—but this will always lead to secure situations, even if all signalling might come to a grinding halt. Actually, this requirement is not always easy to fulfill—there are, as far as I know, places where this rule requires some additional locking elements.

The animations above show how two corresponding block instrument can exchange a locking requirement between two remote locations. But why use such a complicated mechanism at all? Wouldn't simple direct connection, as e.g. used in English line block instruments, do the same job? I did not find any positive and explicit reasons and reasonings for this decision. However, from the quite interesting book "The Application of Electricity to Railway Working", written in 1877 by William Edward Langdon, and the entry "Blockeinrichtungen" in Röll's railway encyclopedia from 1912, it seems to me that the story could have run about like this:
  • In the 1840s and 1850s, short intervals of current supplied by a battery ("single DC pulses") and weak magnets, like magnetized needles, were a customary design for railway circuits—including long distance applications like block working which would change their state due to short, i.e. non-continuous application of current.
  • During the 1860s, it became clear that such designs had inherent risks: "Atmospheric electricity" could create a sufficient voltage to induce a current capable of activating sensitive magnetic devices; and lightning strokes could create arbitrary voltages and currents—depending on their distance from electric lines they might destroy or merely disturb the instruments. Because of these problems, alternative solutions were actively sought, among them: (a) Continuous currents for danger-carrying states, like "line is free." (b) Strong magnetic fields, by using large magnets (and small soft-iron parts). (c) Alternating currents, which allowed to keep the principle of non-continuous operation for state changes.
  • Additional factors had to be considered: DC could only be provided from batteries, which however had to be changed at intervals—either by replacing the whole battery, or by replacing its depleting components, i.e., the electrodes and vitriolic solution (I have not found any reference what was common practice in those initial years). AC could only be provided by hand-cranked generators, which however did not need any supplies.
  • The number of wires needed for the whole system was an important factor. Therefore, up to the end of the 19h century, an earth link was considered acceptable; electric traction at last put an end to this practice.
  • By around 1875, all the design questions had essentially been settled, and—so it seems to me—electricity research and development for railways went into new areas, e.g. into power applications for traction during the 1880s and points during the 1890s. Thus, the experimental era of block working and communications via electrical currents was over, and the respective solution concepts (battery currents in England, alternating currents in Germany) were never again left, rather the chosen solutions were enhanced and completed for large-scale and standardized application.
If someone could shed more light on the history and reasons of this "division," I'd be happy about a corresponding comment or message!

But now let me continue with the internals of those type-5007 frames!

4.2. Signal box: Set up route and clear signal

After the command from the traffic director has arrived (i.e., the Be instrument has been unblocked), the pointsman sets the points (and reverses the FPL levers where necessary). Then, he locks the points levers of the route mechanically by tilting the route lever, which moves the route bar in the locking bed so that all catch handles of points required for the route (directly or for flank protection) are locked in place. Next, the route bar itself is locked electrically by blocking the Ff instrument (which, at the same time, unblocks the Fa instrument at the traffic bureau). Finally, this enables the pointsman to reverse the signal lever of the corresponding signal:

Let us take a closer look at the reversal of a points lever. The following animation shows that the catch handle releases the lever. After reversing the lever, the catch handle is released, which locks the lever in place. But pulling and letting go of the catch handle each moves the lever leading into the locking bed. Because the points lever is reversed by about 180°, both the movement caused by pulling and the one by letting go of the catch handle add up to a combined movement that raises or lowers a hook in the locking bed used for locking the lever:

One can also see that the catch handle is not connected directly to the levers leading to the locking bed, but rather via a spring. The following animation shows the reason for this design: Locking the points is accomplished by moving the route bar to a position where it prevents the locking hook to be lowered. Therefore, the releasing pin on the lever does not leave the corresponding slot on the frame, and hence the lever cannot be reversed. Still, the catch handle can be fully pulled—but this will only expand the spring without unlocking the lever. If the catch handle were connected directly to the levers leading into the locking bed, a forceful pointsman could possibly warp parts of the delicate mechanics or even overcome the lock and force the lever to the opposite position, both of which must of course be prevented.
Actually, this crucial spring is hidden somewhere inside a lever. A signal engineer and I searched for it at a lever for some time, but we did not find it ... The springs visible at the outside of real levers pull the catch back in normal operation—something I have omitted from my animations.

After the points have been set and the route has been locked mechanically, the Ff instrument is blocked. Here is a short video showing the inductor used for blocking, which generates alternating current of about 70 to 90 volts at around 12Hz. Of course, the traffic director had used a similar inductor in the first step when blocking the Ba instrument:

Here is another video, this time showing the blocking of an Ff instrument. The handle has already been depressed at the start of the video:

For specialists: This is, actually, a DC-AC instrument of a central interlocking—a video of a pure AC instrument is still missing from my collection.

4.3. Train passes by

The train releases the button lock. This is accomplished by a short insulated rail which, when shortcut by the train's axles, energizes a relay which in turn sends a DC current from a battery into a separate electro-magnet (therefore, these special instruments are called DC-AC instruments, in contrast to the AC instruments using alternating current for both blocking and unblocking):

Here is a picture of such an instrument removed from the frame. At the left, there is the rack (without the black-and-white faceplate), at the back are the magnets moving the anchor (the picture has been turned by 90°—actually, the instrument was lying on a table):

The following picture shows the back of the instrument, with the anchor's AC magnets at the top (one is behind the anchor shaft) and the DC release magnets at the bottom:

4.4. Signal box: Return signal to stop

At the released button lock, the handle rod can now move downwards. Therefore, the pointsman can now unblock both the Be instrument and the Ts instrument via the connected handles to their locking positions. At the same time, the Ba instrument at the traffic bureau is unblocked and releases the route bar:

4.5. Traffic bureau: Release route

The traffic director blocks the Fa instrument, which unblocks the Ff instrument at the signal box and releases the route bar in that locking bed. At the traffic bureau, the route lever is returned to its normal position so that other routes can be selected and corresponding commands can be sent to the signal boxes:

4.6. Signal box: Return points to normal position

The electric lock at the signal box has been released when the traffic director unblocked the Ff instrument, and therefore the pointsman can also return the route lever to its normal position. This, at last, releases all the catch handles of the levers that have been locked until now (see animation under 4.2), so that the FPLs and points can be returned to their normal position:

This concludes my animation sequence of the principles of type-5007 command and lever frames. Of course, one could now expand these explanations in quite a few directions:
  • How where the various locks accomplished that prevented unacceptable operations, e.g. clearing the signal as long as no command had been transmitted?
  • How does one deal with failures in the apparatus, and with operational changes like changed order of trains after a command has been given?
  • Which other typical installations are there (e.g., many stations only have a single frame in the middle)?
  • What about more technical details—especially the locking bed with its route bars deserves a more complete explanation, but also levers for points, electrical circuits and the provisions for torn wires might be interesting topics.
  • How does line blocking work, and how does it interoperate with station blocking?
If someone is interested in one of these topics, he or she can leave a comment here—maybe I find time to create additional texts or animations!

No comments:

Post a Comment