Wrack and Ruin

Rust never sleeps

Here's a tip: test your track by running a train both ways round it, with the engine facing in either direction. My circuit is pretty reliable if run clockwise with the engine facing backwards. If it hadn't been for my junior navvie and her preference for forward facing in the other direction, I wouldn't have discovered two bad rail joints. Which led to the first serious bout of track maintenance.

If run any other way than clockwise and backwards, the results are ... well, let's say "undefined", but typically involve derailments or wheel slip. The latter seems to be caused by the trailing wagons pulling the loco back slightly and causing the front end to rise, which then fails to grip on some of the gradients. I tried gluing a short length of steel bar under the front axle for added adhesion, but it has only been a partial success. In retrospect, I wish I'd built the chassis and underframe first, allowing me to align the motor better and insert some extra weight under the bonnet.

The derailments were due to a misaligned rail joint, causing the wheel to knock, and another joint where the rail had slipped out of the joiner, creating a sudden wheel drop that generally caught my lightest plastic wagon. I underdid the adjacent plate and replaced the joiner. I also gave the misaligned joint a twist with the pliers to ease the disconnect. Should have used more bricks on the curves. One at each end and one in the middle aren't enough to keep the track solidly in place; I'd have been better advised to place a paver under each joint midway.

While I had the tools out, I also replaced some of the mending plates that had rusted badly over winter. They're zinc-plated, but in the worst throes of a Welsh winter, zinc apparently means zip. Worse, the screws had corroded too, in some cases so badly that it was tricky to undo them. I thought they were brass screws, but on reading the packet, it turns out that "brass" also means "zinc-plated" in this context. Being unable to extract the screws would have been quite bad, and the only solutions would have been to either cut the sleepers or wait further until the mending plate disintegrated altogether. However, greater love hath no man than that for his favourite tool (ahem), and I am currently betrothed to my Kamasa ratchet screwdriver, which thankfully did the job. A tip for removing chewed-up crosshead screws: use a slot screw bit of the closest size for extra grip. On removing the screw, the very worst thing that can happen at this point is for a small piece of ballast to drop into the hole - necessitating a frantic minute of poking around with a needle to clear the obstruction.

The mending plates have fared variably with no obvious pattern to the rusting. I used both yellow and silver zinc plates, but I'm fairly sure both types have corroded in places and survived in others. Some plates that were fitted first have lasted, while others that were used later have failed - and vice-versa. I suspect several inches of snow and wet leaves may have accelerated the process on the curves. Now I need to revisit B amp; Q to find something more durable. I've seen some heavy duty plastic-coated ones, which may be worth a try. Although in my experience, once the water finds a gap in the plastic then the plate starts rusting on the inside until the coating bulges and splits, and the resulting appearance is twice as ugly. Perhaps a coat of Hammerite would be better.

Bridge

I'm weak. I couldn't resist. I had to have one. Coming on top of a new garden bench, and ahead of the imminent delivery of a tree seat and a playhouse, meaning that the garden will soon resemble a crowded woodyard, this small ornamental bridge is a little extravagant - another good month for my flexible fiend.

On the other hand, certain small people much prefer trip-trapping over the bridge to stamp-stomping over the track in their size sixes (you'd be amazed how much pull the average welly tread can exert). And even if it's massively overscale, it looks lovely - this is supposed to be a garden as well as a railway.

I had a minor worry about the clearance underneath, but a quick Google shows that Accucraft's Edrig loco is six inches high and the tape measure says we've got seven under here, so we should be golden. (Of course, for the price of the bench and bridge, I could have had Edrig but still, another day...)

Driving the last spike

Well, the railway per se isn't finished - there still remains fettling and tidying and stations and borders and rolling stock, oh my - but the track is. The last yard fitted into the last gaps with nary a millimetre to spare, which you could see as a tribute to my Brunellian engineering but which is actually more luck than judgement. We screwed down the last plate, checked the line for debris, pushed the test wagon round once (probably the most back-breaking part) and then ran the green diesel round for a few circuits. And then we all went inside, because the wind was a bit chilly.

It works! Begger me, but it actually works. The loco made quite a few runs round the complete track, always managing to overcome small obstacles such as leaves and twigs after a brief pause (I suppose this means it is a Really Useful Engine). The track isn't perfectly laid, but it's not bad given the pains to which I didn't go. I think I'd describe this as a low-effort railway (which is not the same as "no effort", or even "effortless"). Turning the engine round and running anticlockwise, it became apparent that there is a moderately stiff gradient on the final corner (nearest the camera, above), but nothing too problematic. I think I waved a spirit level over the track once or twice during construction, more out of curiosity than diligence, so this isn't a bad result. The bricks make it quite easy to pull and lift the track in small incremements, with the gravel quickly resettling to hold everything firm.

Of immediate priority is stabilising the soil around the outside of the permanent way. The cutting side has already collapsed once in the winter rain, and the compost bin is slowly sliding downhill, threatening to end up on the trackbed. I still need to buy some roofing slates to shore things up, and think about planting as the warmer weather appears. After that, we need a station site to give it a focal point, and then I have to build the kits received at xmas so there is something to run. Lurking in the background remains the quest for steam traction, the line's raison d'etre. If I can resolve that, we can think about a more formal opening. Until then, here's a quick trip round the garden (notice that the camera wobbles more than the loco).

2:15pm Sunday afternoon

"Evie, guess what?!"
"What?!"
"We've finished the railway!"
"I want to go on the swing, Daddy!"
"sigh... OK, Evie."

Mind the gaps

Amazing, it's just past mid-February and spring is here already. (But then, winter was a month early.) Perfect weather for getting back into the garden with some track and a few bricks. Just for once, the DIY indoors can go hang.

Today, I laid the final curve and the last whole straight length. These are the two gaps remaining in the circuit, and it looks like the only remaining yard should be sufficient to plug them with a bit of careful sawing which I hope to carry out tomorrow. There are a few other outstanding jobs as well, such as using the turf removed so far to fill in some of the bare patches on the lawn created during construction, and to clear a small mudslide in the cutting. Regrets? I have a few, the chief one being that I didn't curve the lawn section the other way, inwards to the middle instead of outwards away from the last curve. It's going to make for some rather wibbly trackwork to join the ends up. But still, it's prototypical so long as the trains stay on the track (and if they don't, it's even more prototypical - and just as infuriating).

The more pleasant surprise has been that, despite barely thinking about garden railways through the winter, my enthusiasm is brimming nicely as I finally return to the task. I was worried that the thrill had worn off at the first delay, but with the finished line coming within approx. 30 inches and one more day's work of existence, I find I'm spurred on to see my rackety little diesel wobble round the garden. Cross your digits.

Curve no. 3

This curve represents the sum total of progress in the past two weeks, all of which occurred in the last hour of daylight yesterday and despite several squalls of rain. The weekend before, of course, was ideal weather for tracklaying outside - mild and not too wet. Needless to say, I was inside fixing light fittings when I wasn't out shopping. Even Saturday's weather was better than yesterday, and indeed I was out in the garden - filling eight bags with fallen leaves.

Now I did say that the previous length of line might well be the last track laid until spring, so one can't complain too much about even the least amount of progress. But c'mon! It's immensely frustrating to waste the opportunity of what little favourable weather turns up. Added to which, my plan to spend the winter indoors kitbuilding has been temporarily stymied by the Cooper Craft slate wagon kit I ordered turning up without wheels.

You'll note that this curve is not actually connected to the rest of the line yet. This is because I have to lay this bit, plus another three yards and the final curve, before I can see how big the gaps are going to be and whether the last yard of track will be sufficient to fill them in. This last piece will be the only one that requires cutting, and it needs to be absolutely precise, so I wanted to have the rest in place first. (A wiser man would probably have started here, rather than finished.)

I've trialled a slightly different trench technique on this stretch - only the locations of the bricks have been fully dug out, the rest of the trackbed merely has the turf removed. So instead of requiring a whole bag of gravel per yard, this section only required, er...two halves of a bag. Shome mishtake poshibly. At least I didn't dig as much.

A word about track layout: doubtless you'll be thinking, "This must be the most boring garden railway in the world. It's all flat. No water features. No bridges. Not even some gentle curves. And no sidings!" Well no, there's none of that. It isn't entirely flat, as I'm sure we'll discover when we try to run some proper locos on it. But the garden itself is mostly flat, and I lack the imagination to alter the landscape at this stage. As for sidings, I can't see the point (arf) - yet. There's only going to be one engine in steam because, well, there's only me. It's a ground level line, I don't foresee buying a radio-controlled loco and thus I'm unlikely to want to carry out any shunting. Finally, SM-32 points are expensive and complicate the trackwork, thus violating the first principle round here - Let's Not Go To Any Trouble. One train, round and round - that's it. Maybe a little halt too, but absolutely nothing else is planned at this stage. That is my dream, and it's an incredibly modest one as well. (Seriously, I'm fixated on the getting to the point where all the track joins up and it's possible to run a loco without needing to reverse it.)

But when it's taking you two weeks to lay three bits of curved track, there isn't much point shooting for the moon.

Crossing the lawn

For years, the lawn would prove a formidable barrier to the coming of the railway. In 1854, John Evans of Glyncoch raised a share capital of five thousand pounds and set out with an army of 15,000 navvies to construct a standard gauge line running east to west, directly across the grass. Every man, and all their equipment, disappeared into the mist and was never seen again, although rumours of their eventual fate were rife for decades afterwards...

At the turn of the century, further attempts were made to cross the lawn, including at least two aborted proposals by the GWR. All foundered due to lack of investors, poor planning, inadequate lawn-mowing, bad weather or plain bad luck. It looked as if the railway would continue to go "the long way" (via the flower beds).

...Until yesterday, because the eldest JRA and I successfully began to lay track across the barren, muddy waste of the lawn! Screwing our courage to the sticking place (or something?), we cut into the turf and dug out a dozen sods to form the usual trench, filling it with bricks and gravel as is our wont. I was reminded of Stephenson throwing moss and heather fruitlessly into the bottomless bog of Chat Moss, or Maddocks emptying wagon after wagon of stone into the Glaslyn estuary as he attempted to build the Cob embankment across it; for however much gravel we pour into the trench, it never quite seems to rise to the level of the track. That'll be some more backbreaking, suspension-bending trips to B&Q then. (Btw, my JRA turns out to be a dab hand when it comes to brushing ballast around the track.) I'm sure I ought to be able to get away with a few deep holes for the bricks and a much shallower level of gravel inbetween (say a couple of inches). This was originally part of the point of using the bricks as anchorage. Must try harder.

The track has a pleasing curve to it, although perhaps I should have gone the other way as the proportion of lawn I've eaten into is somewhat embarrassing for a "low-impact" line. (The GRA is currently looking at bridges so that small wobbly people may cross safely.) I tried to raise it on a subtle gradient to meet the lawn and create less of a dip. Fortunately, the lawn also falls away slightly as we progress across it, so the two might actually come together on a level by the third curve.

Laying a curve and another two yards of track has been good progress, given the myriad distractions this week. Although if the weather continues in the same bitterly cold vein, it'll be the last tracklaying session until Spring - definitely a time for staying indoors with a glass of red and a volume of Boyd. To celebrate our work, we gave the ugly loco a quick blast round the full length of track and actually achieved one complete run without mishap before finishing. (The flanges on the loco are so wide that it bumps over the moulded chairs and this, combined with its breckneck hurtling, often conspires to derail it on the slightest pebble or twig.)

Incidentally, the leaf problem I knew about, but didn't realise I would have to contend with complete branches coming down as well.

Today I had another quick jaunt outside. No further digging, but I did lay out the remaining lengths on the grass to see if they're going to fit. It's going to be tight; there will be two gaps next to the remaining curves of approximately ten and twenty-six inches - the exact combined length of the last yard of track. Some precise cutting and bending will be required to make the ends match up. (Place your bets now on how many centimetres short they'll be.)

Progressive wrack

My week "off" has turned out much as I feared, if not worse. In a mocking echo of the sixties, when the motor car ran rampant and almost killed off the railways, there is not a day this week when I am not taking a (broken) car to a garage or fetching a (still broken) one back, leaving little time for railway construction. The eldest JRA and I have, however, been on a prototype train every day as a consequence. But as these were all soulless Arriva DMUs, they haven't been much compensation. (There are rumours of a more enchanting trip out on Friday...)

Today we found a spare hour after returning from the Big City Motor Dealers (home of the Whopping Big Service Bill) and doing our chores to get into the garden and lay the second curve. This one enters the previously virgin territory of the lawn; there is a now a dirty great trench making its way across the turf. I think we may have to introduce a shallow gradient, as the trackbed is presently in a cutting to stay on the level, which isn't very appealing and, more importantly, is bound to cause the youngest JRA to take a tumble.

Naturally, it started spotting with rain shortly after we stepped outside, but we ignored it long enough to get filthy-dirty. Nothing must stop the railway! (Except teatime. And the laundry. And broken cars. And the rest of life.)

Here's a song to be going on with, which I'm posting here for no better reason than that I like it and it's tangentially about railways. This is Trains by Porcupine Tree (2.8MB, MP3). It's only 64Kbps quality, so if you want high fidelity, you'll have to download it from iTunes or, better yet, buy the In Absentia album. I particularly like it when the main riff comes crashing back in after the folky interlude, heralding a tour-de-force outro from Gavin Harrison on drums.

"A sixty ton angel falls to the earth,
A pile of old metal, a radiant blur...

Driving the first spike

(Well, the first 50mm screw actually.)

I have actually been a supporting member of the Welsh Highland Railway Society for several years, but my contributions in rebuilding that line have been purely financial rather than participatory. This was for several reasons: a) the GRA and I had a new house to refurbish first, so we couldn't spare the time; b) if you'd seen the bodged mess we'd made of the house, you really wouldn't want to let us near your railway. To this, I can now add, or at least confirm, reason c) it's too much like hard work.

Somehow, the weather mostly held off this weekend and I found sufficient free time to put down not just one length of track but all the track required for the first phase of the plan: six yards. (It helps when you ignore all the DIY elsewhere in the house.) Although at one point, I had to distract the navvies with some toy wagons to prevent them redistributing the ballast all over the garden. It's a shame, because small people who are closer to the ground are much better suited to this type of work than old men with bad backs and knees.

As you can see, my experimental/theoretical method of anchoring the track worked out in the end. Had to abandon the washer idea due to clearance, but I finally discovered that the part I originally wanted to use but couldn't locate is called a "75mm mending plate" (the Screwfix website is handy for identifying odd bits of ironmongery). B&Q sell them, in bags of twenty. As used, they allow some lateral and longitudinal movement for expansion, although there is a minor danger of sufficient movement to either foul the flangeway or disconnect the rail joints. We'll see. They're a bit ugly, but ballast can hide a multitude of sins (e.g. train suicides).

More by luck than judgement, I even got the curve laid successfully, despite not starting at that point (somehow, it happened to join up with the existing lengths correctly). Next time: lay the curves first, then you can cut the straight lengths to meet up with them.

The curve has actually turned to be my favourite feature so far. As you can see, to maintain the level it ended up in a cutting, which looks quite attractive. Haven't done anything to stabilise the soil yet though; some relatively thin stone slabs (slate?) might be nice. (It's at this point that I remember with regret the large stack of spare roofing slates we left behind at our last house.)

The track has turned out not so wiggly as I would have preferred, but then given that I slapped it down fairly quickly, it isn't entirely straight either. It's more or less close to level; a bit of brick-wiggling sorts out most gradients. Quite tricky to find the level between the soil, the edges of the trench, the brick and the gravel though. In places, the soil threatens to spill down over the tracks; in others, the gravel spreads out over the soil. One thing that this brought home to me: the larger scales are much more appealing. SM-32 is satisfyingly chunky to look at and feels more like "real" track, especially laid outdoors. I never felt this way about N gauge (doctor).

After completion, I was able to successfully push a couple of lightweight plastic wagons around the track. There were a few problems due to leaves (which are obviously going to be a nuisance, and we haven't even had the major fall yet) and bits of 10mm gravel in the flangeway, but the state of the track itself did not cause any derailments, nuclear spills or passenger carnage. This counts as a roaring success by the standards of a Rixon project.

For an encore, I even figured out what was wrong with my cheapie battery-powered loco from eBay. (Loose gear wheel on an axle; needs some glue.)

Now to find some rolling stock...

A bit more of a start

I was beginning to worry that the two weekends work so far would be all I would accomplish until Spring 2009, given that summer now appears to have briefly come and gone and we have settled back into watching the rain sheeting across the valley. Plus, much time has been spent recently battening down the hatches for a long, cold winter with high heating bills.

However, we did have an hour spare at the end of yesterday where the rain held off and, urging the navvies fiercely on in the finest spirit of Stephenson, I was able to fill in the other bit of trench.

Here's the trench, complete with bricks - no thanks to the navvies, who seemed keener on using their new trowels to scatter gravel over the soil.

One of the bricks also has a plug and screw in it, ready for track-fixing. This was purely to test whether our ancient electric drill was up to the task, as I once nearly burnt it out trying to drill into a solid Victorian housebrick. But you could view it as Sculptural Art too - the screw is Peace, intruding into the Space represented by the brick as Marxism, with the gravel flowing around it to show Clubcard Points. Or sommat.

In the course of doing all this, the GRA's real interest in the railway slipped out - she sees it purely as a means towards gaining the garden feature of her dreams, namely a railway sleeper. I tried to break it gently to her that the whole "16mm scale" thing means that the sleepers are about the size of her little finger, but it turns out she wants a full size eight foot sleeper to form a flower bed above the track (usefully holding back the soil on the slope). There's a salvage yard in Newport that nicks sells them, apparently (probably fresh off the Cardiff-London main line). I guess I'd better add this to the spreadsheet, since it's clearly non-negotiable.

So that's two sections of trackbed completed, barring the corner joining them (for which I need the curved track to mark out the course). This has highlighted a gentle gradient running up and towards the east of the lawn. Again, not sure how to tackle this:

• Put everything at the lowest level, which will effectively mean the line running in a cutting below the lawn on two sides. This doesn't sound very appealing.
• Put everything at the highest level, which will require some kind of embankment across the lawn. This sounds like a recipe for accidents and mudslides on the line, mostly involving Tilbo The Destroyer.
• Muddle through, roughly following the ground contours and hoping the locos can handle it, and then rejigging things if they can't. The line of least resistance giving the Line of most (rolling) resistance.

I think at this point, I need a proper plan. More on this anon.

(Btw: Barry, if you're reading this, stop at once 'cos Pamela says you're supposed to be building an extension instead.)

Anchorage

Having second thoughts about my clever "cheap fleximount" idea for attaching the track to the brick. What if the track moves around the washer until the washer obstructs the flangeway? Derailments. Calamity. Public inquiries, media outrage...no, getting carried away there. It'll be the peaked cap next. But not good anyway.

I guess I could chicken out and put a screw through the sleeper like everyone else does. But I don't want that degree of rigidity since it will doubtless cause gaps when the track expands or contracts.

I really, really need some track to test all this. Where's the eBay fairy when you need her?

A further start

Well, I now have a short trench lined with weed control fabric and filled with 10mm gravel plus a brick.

Behold, I am the natural heir to Capability Brown.

Of course, my gang of crack navvies immediately commenced shovelling the gravel into small buckets and standing on the side of the trench to see if it would collapse (it did).

The materials came from that well-known emporium of disappointment and specialist non-suppliers of the exact thing you want, B&Q. In a horrendous misjudgement sure to be corrected next time they restock, they accidentally had all the items I wanted. At least, if I went to my local store and the one near work. The bricks in particular were a steal at 4p off the RRP, mainly because to the poor plebs on the tills, one unmarked brick looks pretty much like another when you're trying to find it in the price list. Bargain! I should have tried the builders merchants down the road instead of supporting another generally useless conglomerate, but I couldn't face the "where's the site the lorry is delivering the pallets to? what do you mean 'a small railway'?? how small?" conversation when I was only after single numbers.

• If I dig the trench this deep, it will require approx. 1 bag of gravel per 4 ft of trench, or 13 bags all round. Not extortionate, but a lot of lifting. I'm sure I can get away with a shallower trench.
• It's easy to get the brick level. It's also easy to knock it out of true again.
• I don't yet have a good solution for the edge of the fabric. A minor cosmetic detail (dear)!

Not sure how much further I can go with this without some track to lay, as it will help to keep the course straight, place the bricks at the right intervals and demonstrate how stable the trackbed is likely to prove. But I hesitate; a couple of bricks and bags of gravel are a minor flutter that can be written off later as an unfortunate error of judgement, but a dozen lengths of SM-32 represent a Significant Investment from which there can be No Going Back. And there are all these bills this month too...

A start

The last two weekends, I cleared two stretches of border around the lawn, removing all the weeds and plants that might as well be weeds so far as we're concerned. Then I dug a couple of short lengths of trench around the top edges, just because...

P'shaw! You don't dig trenches without some idea of what to put in them. Next on the list: weed control fabric, gravel and bricks.