Tuesday, 31 December 2013

The Year Ahead and Behind

It is that time of year when Blogger shows you a full 12 months of posts down the right hand side of the page and you click on January to see what has changed since then.

From this

To this

Quite incredible progress, I'm sure you'll agree.

Of course in reality there is a big difference between a RTR 16.5mm turnout and a hand built 18mm gauge one, although I still think that the Tillig track has much to recommend it. All the same if I have a regret now it is that I didn't build Apa in EM.

However, I did build Apa, and it is more or less finished and time to mover on to the next project. With careful choice of camera angle and some post processing I'm quite pleased with how Apa turned out.

So what did I learn in 2013?

A lot was about observation and layers of seeing. There are elements of the overall mix that make up a convincing layout that perhaps I hadn't realised were important before, perhaps because I hadn't consciously observed them when looking at the real world.  I'm amazed how I can still find things in photos of the TVR that are blindingly obvious once you've noticed them.

One result of this is that I'm much more aware of track work, point rodding signal cables and other things of that ilk. In the past I've know they were there, but only in a vague sense that they were part of the picture.

Colour is something else I think I understand a  little better than I did before but I still struggle to produce the colours I want first time around, as evidenced by  the multiple repaints of the goods yard and the stone bridge. One major investment I'm making this year, to benefit my photography as well as my modelling, is to shift to the consistent use of 6500k lighting in the study.

Looking Forward to 2014

I've also got  a much better idea of the skills  I need to develop before building a proper layout, which helps set the agenda for the next intermediate project that I'm planning for the next eighteen months or so.

Top of that list is to shift the emphasis towards track and  a few key items of rolling stock. the scenics can sort of take care of themselves although I do want to build new versions of all the buildings.

Saturday, 28 December 2013

Playing Around in Templot

Now how did that happen?

Printed out and with the TVR goods shed  placed on the plan it still seems to work, though the narrow clearance in the loop doesn't look right.

Which leads on to this iteration. The turnouts are B6 except fir the one to the goods shed that is an A5 but still gives a minimum radius of  42" that I hope will just about be acceptable for shunting with a 14XX.

To make it believable will mean some scenic tricks and obviously the small TVR buildings are going to help. Having just measured up the  load space in the new car* I guess  I can go up to 18" baseboard width if I can keep the weight down.

My mind tells me the best approach is to flip it over so the goods yard is at the front and to place the platform on the loop, but my heart wants to keep it this way and keep to the "bitsa" station idea I used on Apa, bearing in mind in theory there shouldn't be any passenger services except the odd brakevan special.

*That is if she doesn't crash it again - a week old and it is already in the body shop.

Friday, 27 December 2013

The Lessons of the Point - Part 1

So much has been written about how to build points that you would think there was nothing left to find out for yourself. As a beginner though that is part of the problem. Such is the volume of what has been written that it is hard to filter out the essentials you need during the nerve wracking exercise of building that first point. Much that is written is also influenced by the unconscious competence of those writing it.

In my professional life this is something that is exercising much of my time at the moment. How do you distill the wisdom of experts into a form that allows the less experienced to follow their advice and get short term results that are satisficory whilst also providing a foundation that doesn't constrain future development.

So here are my initial notes on building that first EM gauge point using the C+L system . I will hopefully update it in the light of the next five I have to build in the coming months. Mostly it is litany of the mistakes I made, and needs to be read in conjunction with much of that good advice which is out there.

Choice of construction

I'd built a few copperclad turnouts in OO9 a long time ago and was tempted to make a start using that method in EM gauge, using the SMP kits. I might still buy one to try since they are so cheap. To be honest if I need to build a single slip, a feature of so many Iain Rice trackplans, I think I would go that way until I've built up an awful lot more confidence. The other appealing aspects of copperclad are speed of construction, which can also equate to speed of learning, and the ease with which mistakes can be rectified with the touch of a soldering iron.

So why did I go for the C+L system? Well I already had a stash of EMGS ply sleepers so I thought the TimberTracks base would be a good visual fit, I wanted to use steel track, and I know that on the finished layout I wanted to use the correct pattern of chairs and retrofitting them on to a copperclad point sounded like hard work. In retrospect I don't think it would have been since the C+L chairs are very easy to cut in half. There is also this alternative from  PH Designs that could be worth investigating.

My one concern with the C+L route was the need to get it right first time without that get out of jail free card of a dab of the soldering iron. In a post that is going to be littered with cliches I have to admit that turned out to be a red herring. Not only does the system allow some time for adjustment before the bond sets but it also proved quite easy to correct mistakes either by refloating the chair with more butanone, or by slicing it off and putting on a new chair, either by flexing it on to the rail or slicing it in two and applying the halves separately.

It appears there might be supply issues around the TimberTracks bases, and since most of the turnouts I need are going to be built on Templot templates this might have proven to be a bit of a cul de sac. Something else I hadn't realised, though I guess it was obvious, is that that the template provided with the C+L kit doesn't match the timbering of the TimberTracks base, which made life more difficult than it needed to have been when it came to using the template as a reference. Once I'd realised that I actually produced a rough template using TRAX 3, which allowed me to extend the rails beyond the length of the turnout, and then copied over the  relevant information. I made a mistake in doing this so I have one less slide chair on  one side than I should. .

The C+L kits are a good starting point (sorry) especially for those like me who struggle to find their way around the C+L product range and website. However they certainly aren't cheap.  I was also under whelmed by the instructions which seemed rather dense and in need of a lot more illustrations in place of words.

The common crossing must be a big part of the cost of these kits, since they are £16 to buy by themselves. If that wasn't incentive enough to learn how to make your own I'll also add that I manged to break the common crossing in my kit, which isn't surprising given the amount of handling it got as I played around with it, as any beginner is wont to do. Re-soldering it proved to be a trivial job and another confidence booster. So building my own in future is definitely a skill I want to develop.

My overall conclusion is I'm going to stick with C+L chairs on timber sleepers, but use Templot templates and in the longer term make my own common crossings.

Before Beginning

I think this is an area that many experts don't address from a tyro's perspective.

I cannot emphasize enough that building your first turnout to prove to yourself that you can do it is NOT the same as batch building turnouts for a layout. A workflow that works in the latter case can be full of pitfalls for someone in the first situation.

In particular in retrospect I wish I'd had a complete list of instructions to follow which also included notes on what was essential to get right at that stage and what would fall into place at a later stage.

Now I thought I did have a complete list of instructions. In fact I thought I had several complete lists. But I didn't. Not complete as in the old WW2 US Navy approach of designing everything so it could be operated by a farm boy from the Midwest.  Some things that were missing because the author obviously thought no builder could be stupid enough to not do it, others appeared out of construction order. I suspect much of what I write in the next few weeks will be on the subject of tiebars, but they are a prime example. In books and articles they are almost always referred to as an afterthought, but they aren't. In fact in many cases the author ends with a casual "Now fit the tiebar" . So I got to the end of building my point and realized that the inadequate provision I'd made for fitting a tie bar was going to be a major handicap.  OK, it was easy enough to fit something that worked, but not the sort of tiebar I'd been planning to fit.

A quick diversion here about one of the reasons why I'm building my own points. It was at Warley a few years ago and I was watching a really nice OO9 layout. I was really impressed until the CLUNK as a Peco point underwent the quantum change from being set from one road to the other. It wan't just that unrealistically sudden switching that ruined they illusion for me, it was that that it drew attention to that area of a RTR point that we all like to pretend doesn't exist.

Anyway a complete list of instructions is a must, so I'm going to create my own for next time, including a lot of copying and pasting so that things appear in the right order, and with the use of

large bold type

to remind me of things I'm likely to forget, like drilling holes fro the dropper wires or  not tinning the underside of rails before I've slid the sleepers on.

And I'm going to include tick boxes to make sure I've done what I should have done before proceeding to the next stage.

Incidentally I built an A4 turnout for my first attempt. This was influenced by the remote possibility that I might replace the track on Apa with EM. If I had done it would have  looked like this

Gauge conversion thanks to Photoshop.Once again I forgot to check the state of the fence, but trust me it has now popped back into shape thanks to the elastic "wire"
However I wouldn't recommend that choice. As is so often the case with modelling a smaller prototype is harder work and the tight radius makes this quite difficult to build accurately.  I think a B6 would probably be ideal.

Apart from compiling the list of instructions the other thing I should have done at this stage is to remove the chairs from their sprues and put the left and right handed ones in clearly distinguishable containers. This would have saved a lot of time later.

One other big takeaway from this stage, and it was the cause of some of my problems, is to decide whether you are building it in situ, or for installation elsewhere, and whether or not to ballast before laying the rails. I'll explain in my next post....

Wednesday, 25 December 2013

Making My Point

So, unbelievably, here it is. My very first EM gauge point. The last point I made was in OO9 with copperclad sleepers and was a million miles away from this. Oh yes, and my eyesight was better in those days.

OK it looks untidy and I know there are a host of faults with it and a few things that I need to finish of but overall I'm quite proud of myself.

Underneath there somewhere is a C+L kit on a TimberTracks base, though built on a template from TRAX

I've learned an awful lot in building it, and discovered that some things I thought were going to be really difficult just fell into place. The real difficulties I faced have all got solutions. The one big question mark I have is around the tiebars. I didn't use the C+L versions  because they aren't right for the TVR and at the moment I just have a piece of copperclad doing the job. I have a practical answer to how to make a functional below baseboard tiebar involving the usual mix of dressmakers pin, tubing and fibreglass strip but it would be nice to have an option to build a surface mounted version that approximated to the GWR pattern

Since I'm writing this on Xmas day whilst the rest of the family watch Downton the finishing touches and the next round of thinking will have to wait for another time. I promise I will spill the beans (or superglue) on some of the things that I got wrong.

Big thanks are due  to Geoff for giving me the encouragement and practical advice to get started.

Tuesday, 24 December 2013

Serious Learning

I know I haven't yet revealed what went wrong with my point building exercise. I'm waiting until it is finished and I can report on all my mistakes en masse.

But that, after all, was the point of the exercise - to make mistakes and learn from them, and, to a lesser degree, to learn what worked and use that to build confidence.

Looking back on Apa Valley it was primarily an exercise in getting my confidence back,getting my modelling back to where it was in my OO9 days. As it happened I tried out a few things that were new to me, but that was never the main purpose.

I'm very aware though that the next project, and the things I'm trying out on my test board as well, is about moving into new territory. That means I'm going to have to get used to making more mistakes as I get out of my comfort zone. Some of those will be mistakes of execution, some of design, and some the result of failing to observe and research the prototype properly.   That probably means progress will be slow even by my usual snail's pace, and also thinking the design through so that I can replace anything that doesn't work.  If at the end anything is reusable then that will be a bonus.

In total contrast to Apa the focus is going to have to be at track level and on operation. That means getting to grips with Templot and settling on a workflow for turnout construction that suits me. I'm suffering from a surfeit of advice and techniques at the moment and whilst all of it is very useful I need to settle on what works both for my simple mind and in the context of how I'm building the layout. I seem to have spent an inordinate amount of time peering at photographs looking for things that I'm sure would jump out at other people. Examples: The interlacing of sleepers, the joggling, the use of  concrete blocks and tie bars, the point rodding runs....all the things that if I build one way I know will be proven wrong by a photo I've had all along. That though is the whole point.

Then there is the operational side, the regauging of stock to EM, the building of a least one decent chassis and the changeover to S&W couplings. Not to mention settling on a reliable way of operating finescale turnouts and a an appropriate approach to building the baseboards

All of which should keep me busy in 2014

Tuesday, 17 December 2013


By now you might be expecting to see a photo of my first attempt at building an Em turnout.

Draw your own conclusions from the absence of a photo.

To be continued...

Friday, 6 December 2013

Two Plans

Llansilin Road

So these are the latest versions of the two plans.

First the modified version of Llanrhaiadr Mochnant based on the valuable  input from Richard Ough.

Perhaps the most obvious visual change is the move of the catch point at the up end of the loop. I've also dropped any hint of the second platform.

Then the modified version of TaOC, which isn't any more, because thanks to a suggestion from Kane the plan has shifted to being based rather more on Easingwold East but still with the station at the entrance to the scenic section and the goods shed at the down end.

Of course now the penny has dropped that since building this new compact design, provisionally called Llansilin is going to take the next eighteen months to two years then by the time I get to build Llanrhaiadr I might have enough space available to build it to fit a larger space so I can revert to a more spacious and authentic version. After all the kids are bound to leave home some time, aren't they?

Wednesday, 4 December 2013

Of Trees and Track

I'm always fearful that my posts about the design of the next layout might give the impression that I'm not actually doing any real modelling.

So here are the meager fruits of this weekends work.

I'm not sure but I think this is my first ever completely scratchbuilt tree. The wire was from the scrap box and rather thicker than that normally advocated. You can see the bare armature, coated with hot glue, in the current header photo of the blog. I sprayed it with Plastikote Manhattan Mist stone paint, which is one of those handy products for adding a bit of texture in all sorts of situations.

The tree is intended to disguise the sharp bend in the backscene behind the station building, and seems to work for that purpose.

This is certainly the first hand made track I've tackled since my OO9 days. In those days I used copperclad sleepers. For this I've gone back to riveted ply because that's what my disastrous attempts at EM track building in my teenage years used, and I wanted to lay that ghost to rest.  I'm not actually planning to use this technique on the layout itself, but I wanted to make sure I could do it if I decide I need to use rivets rather than C+L chairs to strengthen the points. Oh, and I had some FB rail I wanted to make use of.

Monday, 2 December 2013

Compromising Design

Few classic layout designs can have been both so inspiring and frustrating as Roy Link's "The Art of Compromise" in the October 1978 Railway Modeller.

I know I keep coming back to it on this blog, and I know others do as well.

I think, from a personal perspective, it is time for me to take a critical look back at that original article

The inspiration comes from the possibility that in a 6ft by 1ft space you can build a model that is "a subtle blend of the diorama... and a working layout" that is a "moderate operational unit itself" and that "incorporat(es) details not normally found on the average model railway"

Remember this was back in '78.  Lichen is king.

Roy's coloured illustration  accompanying the article is  itself inspirational, but so is the fact that the design could, in those days, be built using the readily available Prototype Models card kits, incorporated just about all the features of a typical branch line station except an engine shed, and had an interesting track plan that separated the loop from the platform. It is no coincidence that the same issue included an article on South Leigh station, again with a  track plan illustrated by Roy. South Leigh of the few stations designed on those lines. A subtle twist Roy added was the gentle curving of the track with hardly any of it parallel to the front edge of the baseboard.

So far so good.

The frustration sets in when people realise that Roy designed the layout with that tool most of us have some where - the optimistic pencil. You know the one, it allows you to draw that turnout in to fit a ridiculously small space and connect up two pieces of track without the need to check that the tangents align. The same pencil that convinces you that because there is clearance between a lineside feature and the centreline of the the track that you've drawn then there will still be clearance when the track is actually laid.

On top of that there is a presumption on the part of the reader, though never mentioned in the article, that a layout in the RM designed to use kit built buildings is probably one also designed to be built using Peco track.

I think that is probably the first pitfall people come across. I've spent hours in the past trying to come up with a workable plan using code 75 Streamline and this is still the closest I can get.

Actually perhaps it isn't that bad. I think it needs that one curved turnout to maintain the flow of the loop. However it certainly doesn't allow as much room for scenic development as Roy drew on the plan.

Roy also suggested the model would be suitable for operation by the then new Lima 45XX as well as a Metro Tank or a 14XX. That was certainly part of the appeal when I first saw the plan, even if it took me another thirty five years to buy the 45XX.

In reality though the use of anything bigger than a 14XX is a step too far to be practical.

I suspect as well many who have been attracted to this design because it has so many elements of a "proper" branch line terminus have been tempted to push the plan a bit too far  themselves by adding other features like an engine shed or a proper signal box. They then either decide it can't be done, or expand the design to fit in  a larger space.

Reading the article from today's perspective it is also interesting to see that the suggested baseboard construction was extremely traditional, including a Sundeala top.

So if the design is only marginally feasible in OO why am I still pursuing it as an option for an EM gauge layout?

Well first of all the 6ft by 1ft size has a very specific value in that it means it could fit across the length of my desk in the study.  That also means I could work on the layout in decent light. The possibility of using several small lightweight baseboards also means I'm going to be able to turn them around and over to get at things. Apa Valley has already taught me how important and useful that is at this stage of my learning curve.

Modelling the Tanat Valley brings another key advantage. The station buildings on the TVR were all quite compact, and the goods sheds road loading bay is at the end not the side, so you don't need to allow for as much space alongside the goods shed. That alone possibly saves a couple of inches of width.

There is the small issue that this isn't a track plan the TVR would have used, but it does contain elements lifted from individual stations. For instance the combination of the overbridge, platform, signal box and start of the loop is pure Blodwell Junction

As for the use of EM, well, we've already established that this is a track plan that needs hand built track, and I think, using Trax, that it just about works.

The observant will already have spotted that this involves exactly the same amount of track building as Llanrhaiadr Mochnant, whilst probably demanding a higher degree of accuracy. My other concern is that building small areas of scenery is actually more challenging than building relatively large ones. There is still scope to ease some of the curves, though the way Trax draws things on screen paradoxically seems to make things less smooth the further you zoom out. However the final decision to build will depend on a detailed design in Templot.