Tuesday 18 November 2014

Investigating the Quarry

1st October 2014

Today I had enough of the track modules for the quarry sidings complete and ready for a live operational test. There are two removable modules; the "black plank" and the "quarry turnouts". The "black plank" is effectively a bridge allowing space below it for the tree roots to grow. It contains the main line, single track at this point, the shunter refuge siding, and a turnout switchable between the main line and the quarry sidings. The "quarry turnouts" module contains two turnouts toe-to-toe, fed from the quarry turnout on the "black plank", and feeding back to the refuge siding and forward to the exchange siding or the quarry floor.

The turnouts are switched using Peco point motors, which are hidden under appropriate-looking building. The building on the "black plank" is a ground-frame cabin, based broadly on the one at Midford North sidings near Bath. It was purpose-built from "plasticard". The other building covers two motors, and I found something suitable in Hornby's range, a weighbridge office I think it was described as. I decided that in real life, these turnouts would have been operated by the train crew using a special token to unlock the ground frame. In model form, therefore, I use the same principle; I have a ground frame in a box, which is plugged in to the track modules via a length of cable, and used to operate the points. (If you are an expert in these matters on real railways, please let me know whether the terminology in my description is correct.)

The shunter has brought two loaded wagons from the quarry floor. It will now propel them to join others in the exchange siding. The refuge siding allows a complete train of loaded wagons to be assembled, while keeping the main line free for other trains
The quarry has been disused for some time. Nature is taking over again.

The shunter descends the final short 1 n 5 gradient to the quarry floor.
The way it was; the quarry in October 2011, with very little greenery. A shunter nears the top of the 1 in 5 gradient from the quarry floor.

Loco Testing

5th September 2014

Anyone running a 00 garden railway in Britain will very soon discover that keeping it clean can take up a lot of time. And it's not like cleaning the car, or the house, which can always be put off until tomorrow. If your rails ain't clean, the trains won't run.

Something our railway owner might also discover is that running trains regularly helps keep the track clean, especially if assisted by an occasional run with a railhead polishing and vacuum cleaning train. If done once every few days, this seems to be adequate; if left for a week, a manual clean with a track-cleaning rubber will be necessary.

It's not just the rails that need cleaning. The tunnel attracts a wide variety of wildlife, from insects, woodlice, spiders that build webs across the tunnel bore, slugs and snails that derail your train, to mice that eat nuts and acorns and leave the shells on the track. I've even found a large frog jammed in the tunnel.

Today's short running session doubled as a running test of two locos recently acquired second-hand, and needing quite a bit of attention to minor but annoying faults. The first was a Bachmann "Hall", originally supplied with DCC decoder fitted. In its first outdoor trial, it had suddenly lost the changes I had made to CVs (mainly just changing the address away from 3) and reverted to the default (address 3). So I changed its address in the handset, and continued the trial. For its next trick, it waited until I was bringing it to a stop, and when the selected speed was down to 1 or 2, it suddenly shot forward for about a second at high speed. Not good! I opened it up, and the only thing I could see that might have caused this sort of problem was the two capacitors and inductors fitted in the feed to the motor. I thought I'd start with the capacitors as they are easy to snip off, which I did. Today's trial was to see if the symptoms had gone away. I did several runs at different speeds up the 1 in 50 with 6 coaches (all it can manage), and it ran sweetly. It will take a few more hours of trouble-free running before I am confident the fault is fixed, but it looks promising.

The other loco was a Bachmann BR standard 4MT 4-6-0, to which I had fitted a decoder. It ran up and down the 1 in 50 with trains of between 7 and 11 coaches, double heading with a class 4 2-6-0. The only observation was that the tender seemed to be dragging on one side, and at intervals would emit a creaking, groaning noise. On later examination, the cause of this proved to be a large dob of carpet fluff and hair wrapped tightly around one of the tender axles. When I had first received the model, the loco's axles had been similarly encumbered to the extent that it would barely move at all, and I had spent a couple of hours with a pair of tweezers removing it. But I had missed one of the tender axles, an omission I have now rectified, and I hope that there will be no groaning at the next running session.

Attached are a couple of pictures of the double-headed train. If you are a regular follower of these reports, you might notice that I have put the GWR Centenary coaches away in their box, and found a rake in the BR maroon that is more appropriate to most of my locos.


Engineering the Diversion

2nd September 2014

The first train ran through the diversion more than a week ago. Since then my railway time has been spent wiring, ballasting, weatherproofing, operator interface, and all the other jobs that go with a significant change to the system. This evening I put the diversion back into place for a few record photos.

The first picture shows the train entering the diversion through a short reverse curve including a left-hand turnout. The other arm of the turnout goes into the quarry. The ground-frame hut was purpose built to disguise and protect a Peco turnout motor and frog switch. The original main line is in the left foreground.

The second photo shows more of the diversion. The Oak tree that caused the problem is on the left. The main line runs out of the bottom right corner of the picture, and onto the Gate Viaduct.

The track had to be moved at the entry point of the viaduct. In the final photo, the ballast line of the old route can be seen curving off to the left of the new route. The viaduct is standing on a table well away from its normal position, to make ballasting easier.

Still  to be completed is the pointwork to the quarry, the point motors and the buildings to hide them in, and general landscaping and detailing.

Normal Service at Danes Wood

25th July 2014.

Services are running normally on the Danes Wood railway, apart from speed restriction past the oak tree. The first three pictures show the progress of a freight train of 36 loaded wagons, a GUV, and a brake van. It had set off from The Barn headed by a Stanier 8F, which failed before the whole train had left the yard. A West Country class pacific, normally a passenger train engine with a reputation for wheel-slip on gradient, was the only loco available to deputise for the failed 8F. A Jinty was standing by to assist as a banker if needed (which it was).

34041 "Wilton" deputising on a heavy mixed freight, crosses the Lawn viaduct below the 1 in 50 bank from The Barn.
The train pounds up the 1 in 50 reverse curves towards the Gate viaduct and the Oak Tree, assisted at the rear by an unidentified "Jinty" in ex-works condition.

The short level stretch above the Gate viaduct offers little respite as the train must be slowed to 5mph past the Oak Tree before the final 1 in 50 climb to the summit tunnel.

Shortly after Wilton has passed, the celebrity new-build A1 pacific "Tornado" clatters down the gradient from the summit with an evening luxury dining train.

30th July 2014

A couple more shots from yesterday to show that trains are still running despite the major construction work at Quarry Junction. The main reason I want to keep the circuit complete as long as possible is that most visitors (friends, relatives, neighbours) ask to see it running, whether or not they are normally interested in model railways, and we usually have a fair number of visitors at this time of year!

A short fitted freight train crawls tentatively through the diversion construction site.

Standard class 4 2-6-0 76011 rattles into a curve with a short fitted freight, as the last light of the sun is fading. The photographer has used several high-powered flash guns to make this picture, with the permission of the railway company and the full knowledge and agreement of the loco crew.

Friday 5 September 2014


21st July 2014

All is not well at Quarry sidings. The junction is close to the Oak tree, and it was always recognised that the tree's roots might one day cause problems. Over the past two years, the ground has lifted slightly, and one of the rail joints opened up by about 2mm. However, the situation is now changing rapidly. In the past month, the rail gap has opened up to 8mm, and there is a pronounced "hump" in the track. I suppose it's the opposite of subsidence. Whatever it's called, some speedy action is required.

A gap of 8mm where two sections of rail join.
There is a pronounced hump in the track.
24th July 2014

I now have a plan of action aimed at resolving the problem of track distortion due to tree root growth.

1. Move downhill end of rockery back by about 6" (further if practical).
2. Build a simple, "bridge" over the tree roots, along new route of main line (see below).
3. Re-arrange the three sets of points at Quarry junction, as shown in the first picture below.
4. Lay new main line to Gate viaduct.
5. Re-align track on quarry end of viaduct
6. Run test trains. If ok, open champagne. If not, go back to 1.
7. In 2019, run test trains again. If still ok, open another bottle. If not find another hobby.

Proposed new arrangement of points at Quarry junction.

 The first picture is a view looking down on the three sets of points at Quarry junction. Overlaid are three similar sets of points arranged differently so that they slew the main line onto a route further away from the tree. This scheme has the attraction that it doesn't require the purchase of new points; it uses the existing ones. The middle set is turned though 180 degrees, and the other two are swapped over.

The Rockery re-aligned.
The second picture shows the site after the section of rockery opposite the tree has been moved back by about 6". It also shows where the earth has been dug out under the concrete trackbed to check the clearance between the tree roots and the track.

Double Heading

14th July 2014

Here are some of the results from my latest operating and photography session. The operating side was interesting, as I was controlling the two locos independently from separate throttles on the same handset, and balancing their individual throttle settings by watching the tension in the coupling between them. The technique seems to work well. Since my normal loco policy is to have a cosmetic hook & screw coupling on the front of the loco, I keep a few 3-lonk chains to couple to the locos in the event of double-heading, banking, or hauling wagons only fitted with hook-&-chain couplings.

And now for the photos. They are aimed at finding some new viewpoints on the railway, using an 11-coach train headed by Hornby Bulleid pacific 34067 "Tangmere", and piloted by Midland 3F 0-6-0 43218. This is a combination that could have happened in real life in the 1950s or early 1960s on the Somerset and Dorset line between Bath and Bournemouth, where freight engines were regularly used to assist the pacifics over the 1 in 50 gradients of the Mendip hills. A train of 11 coaches would certainly have required assistance, and the same is true on the 1 in 50 gradients of my railway, where Tangmere would not have been able to move 11 coaches without the pilot engine.



Sunday 3 August 2014

Ready for the Masses.

23rd June 2014

In a week's time, our garden will be open to fifty or more friends and neighbours for our annual garden party. There has been a concerted effort over the past months to complete all of the track-laying, and that has been achieved.
Last year, the railway provided a bit of an attraction in the sidelines, but now there is a viaduct right across the middle of the lawn, on the route from the house to the bar! I have done a few things to avoid the need for stepping over the viaduct:
  • The removable section will only be fitted at quiet times and for a demonstration run including two trains crossing on the viaduct.
  • At the end of the viaduct, where the ground rises, I have raised it a little further to provide a "step-over" crossing.
  • A couple of well-placed signs will send people in the right direction.
The first picture shows a train with its rear three coaches on the crossing. The grass on either side is new turf laid on top of a pile of soil from elsewhere in the garden. The brown pole above the engine is support for one end of a handrail. The 00 scale fencing is intended to discourage stepping on the rails!
A train at the "step-over" crossing.

Since all track has now been laid, I have been running various types of train and loco to find and eliminate any problem, and provide a reliable railway. The next pictures show two of the first complete circuits of a loose-coupled (hook and chain) freight train.

We do freight, too.

A freight train hauled by an S&D 2-8-0
Incidentally, you may wonder why I have paired a set of GWR Centenary coaches with a BR standard 4-6-0 in the first picture. Well, the coaches belonged to my father, and I wanted to give them a run. However, I have no GWR express passenger loco capable of running on the garden railway, so I chose one with at least a semblance of a GWR livery.

The garden party went very well, we had about 70 people there, and there was a continuous demand for "driving experience", especially among the ladies (of all ages from 4 to 88). We had trains going in opposite direction around the circuit, each with its own driver, and passing at speed. Any conflict for occupation of the single-track section were resolved amicably without any input from me. And the locos, stock and control equipment all operated reliably throughout the afternoon.

When most people had left, the "lads" took over, and their driving was gradually turning into a speed trial. I was just considering whether to leave them to it or be a spoil-sport when one of the trains left the track at quarry junction. It may have been the uneven track at this point due to the oak tree, or it may have been one of the many large wood-ants found there, but Pacific loco 34041 Wilton" and 4 of its 7 coaches ended up on their sides in the bottom of the quarry.

Later analysis of the lap times indicated that the train had been averaging between 90 and 100 scale MPH, though I am assured it was going much more slowly at the time odf the derailment. Surprisingly, I could find no sign of damage to the loco or coaches. The cut weed trimmings I had forgotten to remove from the quarry may have helped. The train was soon back on the track, but now running at a lower speed.

To avoid any future risks to my Bulleid pacific, into which I had just installed YouChoos sound, I dug James's old Lima IC125 out of the loft and fitted a decoder so that it could run in the garden. Its running is a bit rough but it's improving.

And finally, another freight train, this one hauled by a rather mucky Stanier 8F.