Of Interlockings And Other Machines ...: 2014

Monday, April 28, 2014

Test runs with the 12SA and 5007 lever frames at Grafenberg, 2014

Deutsche Version dieses Postings

My friend Werner has completed much of the work at his "Sammlung Grafenberg", so in the week before Easter, we did a few first test runs—which was quite much fun! Here are a few photos and videos from this meeting.

Werner has

completed "Steyrling" (which was the original station controlled by my 12SA lever frame),
hooked up the semaphore and two sets of points
and connected the telephones between the original "Absdorf" station and "Steyrling" ...

... and with that, we were up and running.

Not everything was working flawlessly at that time—the semaphore's lower arm was somewhat too low, the tracks were a little wobbly, and the telephone batteries had been depleted through the winter, and/or some contacts were a little corroded.

Since then, all that and more has been fixed:

There are now double on each side of the wagons and locos,
the locos have magnets to simulate track circuits,
the tracks have been screwed down at most places,
the semaphore's lower arm has been weighted,
and all the points at "Steyrling" have been hooked up to the frame.

Here are a few videos and photos from the "inauguration meeting"—where I discovered that my knowledge about Austrian train handling procedures has gotten quite rusty over the last 30 years. Thus, we skipped some of the necessary offer-and-accept procedures for single-line working (we convinced ourselves that the model railway's DC supply was a sufficient safety measure against collisions ;-) ).

Here is the 12SA frame, now located in an old work hut from the NÖVÖG, which is now "Steyrling"'s signal box:

12SA frame "Steyrling", Sammlung Grafenberg, 16.4.2014

Some levers got original chains which move the contacts to control the model railway. These chains thus do not lead to any outside devices, but are "short-circuited":

"Short-circuited" points chain, 12SA-frame "Steyrling", Sammlung Grafenberg, 16.4.2014

Here is a photo of one of the model railway's locos—more of them can be seen in some of the videos and on Werner's homepage:

Model railway, Sammlung Grafenberg, 16.4.2014

This is a shot of the old home signal A of Hetzmannsdorf-Wullersdorf, which is now used as home signal D of "Steyrling":

Home signal D of "Steyrling", Sammlung Grafenberg, 16.4.2014

After we met, Werner has also installed a model semaphore as starter C, which has also been hooked up (electrically) to the 12SA frame:

Starter C of "Steyrling", Sammlung Grafenberg, 23.4.2014

Besides "my" (but now Sammlung Grafenberg's) 12SA, one of the old lever frames that served at Absdorf-Hippersdorf has been moved to Grafenberg:

5007 frame "Absdorf", Sammlung Grafenberg, 16.4.2014

Frame and block instruments, 5007 frame "Absdorf", Sammlung Grafenberg, 16.4.2014

Track indiciator, 5007-Stellwerk "Absdorf", Sammlung Grafenberg, 16.4.2014

But now finally for a few impressions from our train runs. Here is my work place at "Steyrling":

Traffic director's (signalman's) workplace at "Steyrling", Sammlung Grafenberg, 16.4.2014

Track plan and instructions for handling of the 12SA frame at "Steyrling", Sammlung Grafenberg, 16.4.2014

Model railway transformers and track switches for "Steyrling", Sammlung Grafenberg, 16.4.2014

And now for some action: The first train is announced:

Here it is:

A contact drops the warning bell, and then there's some work to do:

Actually, I was a little late, as can be seen here. It took me some time to get used to the controls of the various track supplies, and the track at that day still had a small kink with consequences:

Still, the signal must be returned:

A little later, I had announced my train to "Absdorf"—however, I had forgotten to turn the engine on the its side. Ok—more work:

But Werner used the time slot to send me a light engine which should fetch a wagon from the loading track at "Steyrling" (there's a photo of that train somewhere above):

The two metal bars that can be seen here are necessary to reduce the wire travel for the semaphore from the signal lever's 500mm to the signal's 250mm. We took the raw design of this device from the Slovenian Railways, where old Austrian "Drahtnachzughebel" (wire pulling levers like this one at Greifenburg-Weißensee; the weights only have a function when one of the wires breaks) had been replaced with the simpler version at some places (an example can be seen in this posting in an Austrian railway forum). Our design is simpler than the original one and also matches the heights of chains and wires for the signal better. Unfortunately, there was still that kink in the track ...

... but it has been since repaired, tells me Werner!

And here, finally, the two track lists we filled out during our "game of trains". At least, we managed to end up with the same number of entries :-) ... but as I said before, I'll have to work a little at my knowledge of procedures!

Train register "Absdorf" of our running day, Sammlung Grafenberg, 16.4.2014

Train register "Steyrling" of our running day, Sammlung Grafenberg, 16.4.2014

Saturday, February 15, 2014

Richard Pike's signal boxes and some more explanation of tappet locking, 2013

Deutsche Version dieses Postings

A day before we left England, I got a short message from ... Richard Pike!—that "there were three more signal boxes nearby."

I wondered were they would be—no railway line there would still use old signal boxes. Then I found out: Two thirds of Richard's garden are sort of a signalling company's storage, with three fully working complex signal boxes, all connected to a relay simulator built from an old telephone exchange!

I made only a few pictures of the lever frames, as I had too much to ask and listen to during my visit. However, farther below are a number of photos from the locking bed of the tappet frame which (if I remember correctly) was brought to life again by Richard—I'll use them to explain a little bit more about tappet locking!

But let me start with some shots of Richard's working frames.

First, there is the mechanical frame of Ely North Junction:

Ely North Junction frame at Richard Pike's, 30.8.2013

Then, there is the electro-mechanical frame of Ely Dock Junction:

Ely Dock Junction frame at Richard Pike's, 30.8.2013

And finally ... unfortunately and for no reason I can remember, I did not take a picture of the third mechanical frame. However, part of Richard made it on the following shot:

Ely Station North frame at Richard Pike's, 30.8.2013

Outside, Richard has got quite a number of signalling equipment:

Shunt signal partly made of rubber (to reduce the risk of injury) at Richard Pike's, 30.8.2013

30.8.2013

Richard was, as remarked above, preparing a small ground frame. A few parts with fresh brown paint were drying:

And here are the final signalling-related pictures from our vacation in England, explaining the typical parts of a tappet frame. The frame is shown here from the back side, and the locking bed has already been opened. Some tools and some—well—other things are lying near and on the frame—so it goes ...

Here is a closer look at the locking bed. It is a quite simple locking bed—there are much trickier ones. However, it shows a number of typical patterns how levers lock each other which I'll shortly explain:

At first, we look at two neighbouring levers that lock each other like the ones in the ground frame at Wansford. Here, one can see the two tappets, with a pair of wedges between them. The left lever is locked in this situation—can you see why?

Here is the explanation:

In order to reverse the left lever, its tappet (the flat iron) would have to move (towards the top of the picture).
This would force the left wedge to the right, via the short tapered edge.
But this would also have to move the right half of the wedge pair (as the wedges touch each other at the top and the bottom), ...
... which is impossible as there is no cut corresponding to the wedge in the right tappet!

So the left lever is locked in place.

How can the left lever be unlocked? For that, the right lever must be reversed. Here we see the first fraction of it smovement—the cut on the right has moved (up, in the picture) a little bit (thereby, the wedge on the right wide has been pushed out—but we'll ignore this for the moment). The left lever, by the arguments above, is still locked:

When the right lever has been completely reversed, a cut in its tappet is now positioned opposite the wedge. This, of course, finally allows the wedge to move to the right if pushed by something! (Sorry for the hand in the photo ... it's probably mine, but I don't know why it's there ...):

And now the left lever can be moved! Here we see that its tappet has already been pulled (again upwards, in the picture) by its lever. The taper at the cut has forced the left half of the wedge pair to the right, which in turn has pushed the right half into the matching cut of the tappet of the right lever—which, therefore, is now itself locked into its place!

If the leftmost lever is moved back, the cut in its tappet will of course again be aligned with the wedge—but the wedge will remain outside the cut until the movement of another lever forces it back. Here is the situation after the first lever was moved back:

Four pictures up, you can see the situation where the wedge has been forced back into the cut in the left lever's tappet.

Up to now, we have only dealt with neighbouring levers. However, in many cases it is necessary to lock levers that are not neighbours of each other. This is where connecting rods become necessary. The next picture shows a simple example: There are three levers—left, middle, and right. The left lever should be locked against the right one, with no locking logic for the middle. The solution is easy:

The two wedges are placed against the corresponding tappets.
Below them, a connecting rod is placed with two noses that push the wedges.

I think it is clear that this part of the apparatus behaves exactly like the one above, only now for two levers some distance apart:

One can see that two noses project from the connecting rod that push or are pushed by the wedges. However, the connecting rod is not centered, but somewhat offset. This makes room for another connection rod which can connect other wedges in the frame. In the locking bed shown here, the trough where the connecting rods run can contain two connecting rods; in other designs three or even five rods could be placed in one trough. But of course, more than one trough might be necessary:

Either because more connection rods are necessary than would fit into one trough;
or because a tappet has to be locked with more than two wedges (only two wedges per tappet per trough are possible, one on each side).

In this frame, the center tappet engages with three wedges, therefore at least two troughs are necessary:

Of course, a single lever can lock more than one other lever. Typical examples of this are signal levers which lock all the points levers (or the facing point lock levers) that are guarded by the signal. In this frame, there is an example where a single lever locks three other levers: If the leftmost tappet in this frame is moved, its wedge (marked red)

... will directly push over the wedge marked white;
... will push the connecting rod with the nose seen in the red area, which in turn ...
... will push (with two more noses) two wedges against the tappets of the remaining two levers in the picture.

So the leftmost lever here locks all other three levers if it is reversed:

The next picture shows the situation when the lever has been reversed:

The tappet (1) ...
... has forced the wedge (2) ...
... to the right, so that it pushes (with the protrusions at its top and bottom) the next wedge (3) into the neighbouring tappet's cut and locks it into place.
Moreover, the wedge (2) pushes the nose (4) to the right, which shifts the connecting rod (5).
With its nose (6), the connecting rode pushes the wedge (7) into the cut of the tappet (8), thereby locking this tappet.
Also, the nose (9) shifts the wedge .......

... and here I ran out of numbers. I think you get the pattern:

The final pictures show that the rightmost lever is locked twice:

Here is picture of the full frame in this situation ...

... and here is a picture of the two dependency chains that start at one tappet and force wedges and connecting rods to the right so that the rightmost tappet is doubly locked:

This is by no means the high end of tappet locking—much more elaborate and even tricky locking beds are possible and necessary for large signal boxes. If you are interested in the intricacies of large interlockings, you might want to read pages 270 to 300 of Railway Signal Engineering (Mechanical), by Leonard P. Lewis, published in 1912. Be warned: It might take you a week or two if you want to understand each line of text.

And that's all from our 2013 "signal box" vacation in England!

Ketton signal box, 2013

Deutsche Version dieses Postings

Here are a few photos from an operational signal box with a traditional lever frame. Ketton signal box is on the Birmingham to Peterborough line, next to a level crossing, which it controls together with a crossover and a track into a cement factory:

Ketton signal box, August 2013

The most interesting feature of the Ketton signal box is a lower quadrant semaphore in a tight curve. Because of that curve, it is placed on the wrong side of the track, namely next to the right track, although of course it controls train movements on the left track!

Semaphore, Ketton, August 2013

Here is a train coming from the west. One can see that the signal is high enough so that it can be seen from a train on the left track even if there is another train on the right track:

Semaphore, Ketton, August 2013

Ketton signal box, August 2013

A short time later, another train is going in the opposite direction—unfortunately, I did not catch it at the signal:

Semaphore, Ketton, August 2013

Level crossing, Ketton, August 2013

Ketton, August 2013

And here are a few photos of the frame in the signal box. Almost all operational levers (but not the white spares) have shortened handles to indicate that they work electrically controlled devices. The one exception is the lever for the semaphore which can be seen right after the row of spares:

Ketton signal box, August 2013

This is, as far as I know, the control stand for the barriers:

Ketton signal box, August 2013

The blue lever that is reversed here unlocks a ground frame for points leading into the large Ketton cement works—obviously, a train is shunting there, but I had forgotten to take a photo when it had rumbled by a few minutes earlier:

Ketton signal box, August 2013

Here is the block shelf with the track diagram:

Ketton signal box, August 2013

And here is a final shot of the tranquil surroundings of the level crossing:

Ketton level crossing, August 2013