Rewanui - a 1940 West Coast NZ layout in 1:64

[Lawrence@NZFinescale]

As I'm new here I thought I'd outline the magnum opus.

Rewanui was a branchline terminus serving local coal mines.  The area had no road access and was in an extremely tight valley.  The line was so steep that a Fell centre rail was used to assist in braking descending trains, and the trains were short.  New Zealand railways are 3'6" gauge and the model is 1:64 on 16.5mm gauge track.

I've been working on this, to a greater or lesser extent, for around 30 years.  However I take the view that, much like cooking dinner, I want everything to be ready at the same time.  So many things have been developed to a proving stage, completed but not painted, kits produced but not assembled and so on.  Fortunately the very first model still stacks up well with the current stuff. As I've developed kits for all the locomotives, rolling stock and castings/etchings/prints for almost every detail including figures, plants, track, couplers, wheels, building details and so on, it has taken a lot of time.  Very little on the finished layout will have been purchased as a model or 3rd party kit.  It's all scratchbuilt or assembled from my own kits. This DIY approach is partly driven by a lack of available models (to the standard that I want), but mainly by personal preference.  On the plus side original drawings for almost everything are available. 

The layout is extremely modest, being just the station yard, a short length of scenic track and the backshunt bridge.  This is indicated by the red rectangles on the attached drawing. The idea is that empty trains will arrive at the station and be shunted to the off stage coal bins.  Full wagons will be shunted back, marshalled and full trains will descend.   A fiddle yard allows the full and empty rakes to be exchanged out of view. Despite the small size there is a lot of detail around the station buildings that should be full of character.  And it was a busy place, often with 2 locomotives present and many short trains arriving and departing on a typical day.

I try to model to a high standard of realism, but I'm cognisant that a) models need to be robust on a model railway and b) individual models are part of the greater whole.  Therefore I'm not above leaving things out that are realistically never going to be seen.

Right now the layout is mainly bare boards with only one section being actively sceniked.  The current thrust is to get the 10 or so passenger cars and a couple of vans (cabooses) completed. The first of these is attached, but as it is a kit 2 similar (though visually different) cars are nearly done too.

The line (but not the terminus) features in a period news reel here: https://www.nzonscreen.com/title/coal-from-westland-1943 (4:48 - 6:00)

If there is interest I'll try to post some catch up on the state of play on the various elements.  After that is done I guess I'll slow down to the occasional update.

[finescalerr]

Please "post some catch up on the state of play." -- Russ

[Ray Dunakin]

Please "post some catch up on the state of play." -- Russ

Yes, please do!

I enjoyed the old newsreel. Looks like a really great subject to model!

[Lawrence@NZFinescale]

Since the guv'nor requires...

The start of my journey was We377. 

NZR's B class 4-8-0 tender locos were built in their own workshops and by Sharp Stewart in Scotland.  The 4 Scottish examples had a long coupled wheelbase that was problematic on early colonial track.  This problem was partly solved by converting two of the class to We class 4-6-4 tank locos in 1902 (and a third in 1943!).  One of the tender bogies was modified to become the trailing truck. In 1940 (the setting for my railway) both We locomotives exclusively worked on the Rewanui incline. So I needed a model of at least one, and for such a restricted class there was no kit in sight.

I'm not sure when I started this loco, but I was a student, so that makes it at least 30 years ago. By that stage I'd done plastic kits and a bit of scratch-building in other scales, but I was pretty new to modelling NZ railways, brass and 1:64.  Fortunately I had a few mentors. The model remains as in the photo attached.  It is largely finished and I remain very happy with it.  I've made a conscious decision to finish it without upgrading it to modern standards for the most part.  The brass here is cut by hand from sheet or turned from bar.  The tanks are brass shim with a solid core of that brown electrical composite that goes by various names. There are some commercial castings for standard parts, and a few of my very first castings from hand made patterns.  The tank fillers are from 3d prints as I never really got these sorted back in the day.  Wheels are the third or fourth set fitted.  The first set were castings I had from another scratchbuilder that didn't really satisfy. One set were hand made, but using info for another member of the class that it turns out has different counterweights.  At some stage I've adopted scale wheels over RP25, and thus the current set was printed and cast.  Rods were milled.  The front truck never worked and it took me years to discover the axles were not 2mm as I'd thought, but 2.03mm.  As the photo shows, this has been rebuilt.  There are plenty of mistakes in this loco that I'm well aware of but others less familiar with the prototype will overlook. In many ways the little errors, and my attempts to fix them, are what makes this loco so special to me.  Bits of it span my whole serious modelling career, so, while I could etch and 3d print something better it wouldn't be the same.  Of course I could still do We376 that way.

So as with much of my modelling, 377 still needs to be finished (It does have a cab roof and interior, but these must have been elsewhere when I took the pic).  It's all part of a larger plan since, had I finished the loco years ago, fitting DCC and bringing it up to standard would have been problematic.  In it's current state it will be relatively easy to finish it to blend in with everything else.

I'm not sure who took the image of the prototype in 1937, but it is labelled Stan Rockcliff collection.

[Barney]

Welcome - Very nice stuff with excellent detail and interesting Website
Barney

[Chuck Doan]

Very nice work so far!

[Ray Dunakin]

Looks great!

[1-32]

Hi Lawrence.
Great to have another modeller from across the ditch along.
cheers.

[Hydrostat]

This is going to be a quite inzeresting thread!

[Lawrence@NZFinescale]

Thanks to all for the encouragement and comments.

Around the time I started We377 I also scatchbuilt a 30' guard's van (caboose) from styrene with balsa roof, and commercial cast bogies.  This model has been a bit damaged over the 30 or so years since and I need at least two, so when I revisited the subject last year I decided to do it as a kitset using printing, etching and casting.  The model here is the pilot, and some minor improvements have been made since.

I'm in two minds about printed models.  On the one hand they are detailed and quick to make.  On the other there are compromises I'm not entirely happy about.  That's true of any medium, but I still like metal best. A really big plus is that I can build stock so much faster than before that the enjoment factor and probability of completion has increased immeasurably. Not that I'm impatient, but spending weeks on one wagon is fine. Getting motivated to repeat the process for a dozen siblings is more difficult, remembering that these individual models are just small components of a larger whole.

The photos are a bit cruel, especially showing some gaps at roofline and between the body/undeframe.  The model screws together and can be disassembled for detailing, which is why I have not filled these gaps.  In addition they are far less obvious at real size 1:64.  I'm learning a lot about aging from the forum, and my weathering could use some work.  However these vehicles would have been grubby, but otherwise maintained quite well, so extensive peeling paint is not the finish needed here.  As I noted in the title post my plan is not to finish components one at a time but to get everything to come together at the end.  I'll revisit weathering of all the stock to make sure there is a degree of homogeneity and as my skills improve.

Using the printer for this introduces some interesting conundrums and the design of parts to minimise print artifacts is key.  There is also a need to accept some compromise as printing allows fanastic detail that I probably could not do otherwise, but some large flat areas do show some layering.

The van body is 1 piece.  Side doors are separate and can be opened.  The numberplates are etched, but other plates (load/postal) are printed in place.  Bogies are all resin prints and the spring detail is nice and open.  I've since revised these with an etched subframe for a bit more strength.  The footsteps here are styrene, but it tends to move a lot with temperature change so I'll replace it with metal or wood at some point and for future models.  As I always do, windows are designed with slots and recesses to take the glazing.  The end doors have slots.  The lookouts have a recess that allows the one piece glazing to be scored, folded and clipped into place.  Protective bars over the windows are etched brass.  The roof is in two pieces dowelled together, glued and filled.  This enables optimal print orientation so that the verandah can be detailed underneath.  The various ventilators are added separately to allow the roof to be sanded smooth.


There is a fuller description of the van on my blog: https://nzfinescale.com/the-back-end/

[finescalerr]

Lawrence, I think the reason may be that you linked photos from another website instead of uploading individual original jpegs. When I open your images in a new tab they show up there so apparently the our two websites are having a little digital spat.

I've written this a dozen times and most of the guys now understand why: You must upload individual jpegs of not more than 230 kilobytes, even if that means using a separate post for each. No post can contain more than 230 kb. The problem with linked images is what you've encountered. Eventually, in nearly every instance, the links break and the entire thread then becomes useless. Our software is old and relatively crude, we can't replace it without losing everything, and we all suffer from its limitations. It is such a problem that a sticky exists to explain how to resize images to deal with the software's exasperating limitations.

Now aren't you even more delighted than ever that you asked to join us?

Russ

[Lawrence@NZFinescale]

Yep, web site spat sums it up.  I can embed Volker's images, but not mine even though they show up fine in the debugging window.

The advantage of embedding images from my site is that I can post here in 10 minutes, rather than sitting down to rework all of the images.  While I get the underlying issues that doesn't change the fact this barrier is going to reduce the number and extent of posts I can find time to make.

For those with Photoshop (Elements, and probably other software too), there is a 'process multiple file' tool that allows you to resize, adjust res to 72dpi, adjust compression etc., saving copies of the modified files to a location you choose.  In other words prep your images for this forum in a single step for all images.  Not ideal, but it does reduce the work needed a LOT.

Lawrence

[Ray Dunakin]

Beautiful!

[Lawrence@NZFinescale]

One of the challenges to the finescale modeller in smaller scales is undoubtedly weathered timber.

Chuck Doan et al. set a high standard here, which had left me a little dissatisfied with some of my own attempts at old wood.

Over the last few days I've been playing with a solution.  The illustration here shows progress.  The effect was achieved in just a few minutes and there is certainly a lot more I can and will do in terms of distressed paint colour variation and so on.  But that's just finishing, the hard bit is sorted.  It is part of a larger model that I can now progress.

The full story is at: https://nzfinescale.com/going-with-the-grain/ . As I work up the model I'll post updates.

Lawrence

[WP Rayner]

That is very effective for 3D printed wood and an interesting approach that I wondered if anyone had experimented with in the smaller scales. You mention on your site that you are using Blender which can be an excellent choice, primarily because it's open source and free, thus not requiring a huge financial investment for the software, though as with any 3d modelling application, the more powerful your multi-core CPU and GPU the better, which requires an often significant investment in hardware. Concerning the photographic textures you used, did you prepare your own normal maps? Size and resolution of the photographic image is also important as you want to avoid tiling, an important consideration when searching for photos on the internet. Generating and flattening the UVs required so the texture maps properly on the model can also be something of a challenge. Am I correct in understanding that Blender has the ability to do that in a fairly simplified process, at least in comparison to 3dsMax, Maya, Modo, Rizom, etc? Though not impossible, it is important to note that this process is much more complex to accomplish with a model created by a 3D CAD application. The CAD model has to be imported into one of the 3d modelling applications, then retopologized (generally into quads) in order to be able to generate the correct UVs so the texture will map correctly on the model. All in all, it's an interesting approach and I'm looking forward to seeing what more you can do with it. Also, check our Megascans... they have a lot of excellent textured surfaces, though, not having any experience with Blender, I don't know if there are any compatibility issues.