30 May 2009

Solving the branch line conundrum

My layout plan provides for a branch line: it has an approach to the station, and an access point to the storage sidings. But until today I had no clear idea how these two ends were going to join up. At first I thought the branch line should cross over the main line on a bridge. This would look good and add interest to the model; however, the track plan for this required too many over-steep gradients (maybe even a helix, which is not really practical in the available space). So the branch line has instead to cross under the main line and somehow weave between the network of feeder tracks on its way from the station to the storage sidings.

Here is my solution:

The photo shows boards F3 and F4 (as well as my new Hornby 2009 wagon, purchased today!). All of the branch line track in the above photo will be covered by scenery: the line will emerge from a tunnel portal just to the left of the photo, and climb up and round to the station. At the top right there will also be a short section of uncovered branch track - just before it dives under the main station boards - to provide a bit of "watch for the train" interest!

The curves in the "buried" section are tight - all 1st radius - which will limit the rolling stock that can use this line. But for light freight and DMU movements it will be fine. I have cut the track bed but have not fixed it down yet. It winds its way through the main line supports and I will only need to move one support to provide enough clearance.

I am relieved to have solved this. More construction to follow: the last bits of track bed are about to go down.

Completing the circuit

A major milestone today: on a sweltering hot afternoon in the loft, I completed the inner and outer loops. It should now be possible to run a train all the way round the main circuit (although I've not tested this yet).

The final linking sections took some work, as I had to get the levels and alignments exactly right, but on the whole it looks pretty good. My work rate today was slower than usual because of the heat (far warmer than it was outside - the loft absorbs and retains heat rather well). But I have reached the point I wanted to reach by the end of today.

Here are some close-ups of the two linking sections:

The long gap in the above photo will be the bridge mentioned in a previous post. I have removed some of the framework to allow me to put in a lowered section when the time comes to add scenery.

Note the double-slip crossover in the above. The hanging track to the left is the start of the branch line.

The time for laying track is fast approaching, but there are a few bits of tidying up and levelling to be done. I hope to avoid these dragging on as I really want to get started on the railway rather than the construction. And I also have some news about the branch line: more on that in the next post.

26 May 2009

An optoelectronics novice

Because the weekend has been spent almost entirely in the garden, there has been no time for any work on the railway. So I have been thinking and planning, but not actually doing.

The current thinking topic is electronics and control in the storage sidings. These tracks will go in first and I want to put in the wiring at the same time. Because the storage sidings are less accessible than the top tracks, not to mention invisible in normal operation, I want to be able to (a) switch all the lower level points remotely, and (b) know where my trains are so I can set the correct route for a given train.

(a) is relatively easy, if not necessarily cheap. I can motorise each of the points, and if I use the standard Peco PL-10 point motor I can also attach an accessory switch that is operated by the motor itself. This switch can be used to light LEDs depending on which way the point is set. So when the point is changed, the appropriate LED lights up to confirm that the motor has actually fired. Combine this with a probe-and-stud panel showing the track plan, and you get a handy control centre that shows the available routes. Doing this is probably within my electronics ability, if I think about it for long enough.

(b) is rather harder. In the real world, low-voltage track circuits are used to determine where trains are. This is not easily possible in the confined lengths of a model railway - or at least, not without electronics knowledge I don't have. MERG do all sorts of stuff in this area using comparators, but it's not for the faint-hearted.

So I might use an optoelectronic approach. Maplin sell an inexpensive infrared detector kit consisting of one IR LED and one IR phototransistor, both mounted in "side looking packages". With appropriate circuitry, a lamp can be made to light up if the beam is broken. My idea is to have a bunch of these positioned along each of the storage tracks: LEDs opposite phototransistors, acting like security beams. When a train passes along the track it cuts through the beams and causes indicator lights to illuminate on the control panel. With enough such units along each track, you get a signalman's-eye view of the sidings and can see, to whatever accuracy you like, where the trains are.

The theory is simple, but execution requires a lot of components and a lot of extra wiring. I will see how my ideas develop. But before anything else, I have to complete the trackbed construction. Next weekend includes the promise of some time to work on this.

18 May 2009

Layout test #2

The purpose of the second layout test is to determine the final alignment of the track beds. No track has been fixed yet but the track is more or less in final position.

The photos below are a selection of views from the test.

Above: views of the station area from the left-hand and right-hand approaches respectively.

Above: a train waiting to depart from platform 3. The alignment of this line presently has a switchback curve (under the loco in the above photo) because the chimney breast turns out to be larger than I allowed for on plan (...doh). It means that the available length for platforms 2 and 3 is a bit shorter than I'd like. I am thinking about ways to move the entire station alignment forward, which would allow more space for the platform structure. However, doing this puts pressure on the curves on F5i which are already as tight as I want them to be. I feel some playing in XTrk coming on...

Above: the station throat on B5u. I have added the new board section at the front of the board, but not fixed it yet.

Above: a view of the entire layout as it stands now.

04 May 2009

Second incline revisited

A few tweaks later, and the second incline is a bit better than it was.

I moved the start of the incline back to F1, and adjusted the height of the supports all the way along the trackbed. With a longer incline I have been able to soften the gradient and (so I hope) make it easier for trains to get up the slope.

In the course of making these changes I decided to test the slope with real trains. (Only afterwards did I realise that this was the first time I had run a train in the loft!) The revised slope is fine for locos that are running well: I recently acquired an Airfix class 25 from a friend and this pulls a train up the slope quite happily. But an older Lima class 33 struggles; the lesson is that I need my locos to be running as smoothly as possible.

03 May 2009

The second incline

Change of plan again. Rather than work on the outer loop (F2o-F1o) I have instead made progress today with the inner loop from B1l round to F4i.

I decided that I really needed to finish off the lower levels, so I know where I can fix the supports to carry the upper-level baseboards. This issue was stopping me from finalising the station approaches on both B1 and B5. With the lower-level track beds all in place I can now finish off the top tracks. Fingers crossed that everything will link up without overly tight curves on F1o/B1u and F5i/B5u.

The "spur to nowhere" visible at the back of B2 will eventually be the feeder for the cassette loader mentioned in a previous post. I won't do any more work on that for the moment as it would just get in the way of more important things.

The trackbed joins at B1l-F1i and F1i-F2i are interesting because here the track is not running perpendicular to the baseboard breaks. If I had made a straight trackbed join at these points, I would have had to make a diagonal cut across the tracks, which would make the alignment less stable. My solution is as follows:

The trackbed join now runs perpendicular to the track for each line, so there is a sawtooth pattern where the boards slot together. This pattern should help the boards to marry properly when being assembled. I will have to do the same on F4i-F5i as here the track will be starting to swing round to meet up with the station approach on B5u.

The "sharp" bits of the sawteeth overhang the boards they adjoin, resting on the frame of the next board. So there's now a rigid sequence to assembly and disassembly of the layout, and all future overhangs have to be consistent with that. Board F1 will now always be the first board to come out, followed by F2...F5 and then the back boards. In assembly, this sequence is reversed.

I am pleased with all of this work except for one thing: the gradient of the inner loop is much steeper than I wanted. Whereas I had planned for the incline to start on B1l as soon as the lower trackbed emerges from under the upper tracks, it actually doesn't start until halfway across F2i. So there is less distance to gain the required height, and the resulting gradient is only marginally shallower than the "standard" 6% Hornby gradient with which my trains have struggled in the past. I will look at that on my next loft visit; it may be possible to make the incline start sooner and so soften the gradient.