Winter 2016 Challenge - Jim Botten - Breakheart Trestle

[jimbotten]

Though perhaps not strictly needed for the "Challenge", I feel a little background and context may be in order! My first-ever railroad-under-construction is intended to fit in an L-shaped space around the top of the stairway up into our attic sitting-room. The layout will be some 10'6" by 5'3", on boards about 30" wide, in N-scale and set in the early steam period (1890s to 1915?) on an imaginary railroad which draws its "history" from the 1975 film "Breakheart Pass" (starring Charles Bronson), which used the Camas Prairie Railroad for its location-shots and dubbed its fictional version the "Wasatch & Nevada".

Due to the nature of the site I have been working on completing as far as possible the short leg of the "L" out in the main room, prior to pushing the whole unit into place above the stairs. This section is effectively a diorama, designed around a trestle-bridge inspired by the CPRs "Half-moon Trestle".

DSC01728 by Jim Botten, on Flickr

Above is the view down the staircase. The "storage-bench" to the right will be the long leg of the "L", and the area straight ahead with a jumble of paint-tins etc. will be the short leg. Just visible top right is the unit which will (I hope!) slide across to the left.

thumb_DSC01729_1024 by Jim Botten, on Flickr

Above is a closer view of the short-leg unit as it exists today. I thought I needed to complete the landscaping as far as possible before working on the trestle itself - due to the need to work around and behind where the trestle will be. The track across the canyon is supported by the base foam that was cut out from beneath the track in making the valley, held in place by blue electrician's tape. (Note the slanted ends, making it possible to rest those ends on a firm base at either end of the gap.)

DSC01730 by Jim Botten, on Flickr

Above an even closer view of the crucial area. Note the short length of trestle-base lying on top of the foam which is (so far) the only casualty in my process. (The base was built and glued to the bottom of the track before the supporting foam was strapped in place.) The black line you can see along the top of the foam is just that, the marking for the track-centre as originally traced onto the foam base when it was whole (i.e. this is just one piece of foam, not two side-by-side). The foam-supported track has taken quite a few knocks in the landscaping process, so I'm grateful to have just the one casualty to glue back in place in due course.

Progress on the layout so far has, I admit, been slow... (I told my wife it was a ten-year project, and she keeps asking when that started from.) Which is why I have leapt at the chance to use this "Challenge" as a spur to get things moving. I've just about convinced myself that there is no more tweaking to do to the surrounding landscape (there may be more trees to add in front of the trestle once it's built), so the building of the trestle itself is the focus for this log. As emphasised in my second sentence I am a first-timer at all of this, so any helpful comments and tips as I go along will be very much welcomed. Here goes...

[jimbotten]

Just a brief note on the research and development that's already taken place before the starting-gun on December 1st. Apart from general reading up on trestles and how to model them I managed to locate two good overall views of the prototype from the internet, one looking in each direction along the Half-moon Trestle, and I've also been freeze-framing through the DVD of "Breakheart Pass" and making notes and sketches. The internet images show that the two ends of the structure adopted different types of "footing". The "right-hand" end uses timber-walled caissons, while the "left hand" end uses more rough-hewn, cut-into-the-earth steps - probably due to differences in slope which are not mirrored on my model. This leaves me in a quandary as to whether I want to have the timber-lined caissons on both slopes, or... (Any advice or comments appreciated!) One certain difference between prototype and model is that the Half-moon is (as its name suggests) an arc, while the model has to take an S-shaped curve (the taking on of which may yet prove to be my downfall).

halfmoontrestle by Jim Botten, on Flickr

halfmooncamas by Jim Botten, on Flickr

It's almost two years now (!) since I bought the JV Models kit for a Curved Trestle Bridge, and supplemented it with a lot more scale-timber of appropriate cross-sections from Northeastern Scale Lumber. The freeze-frames from the DVD have given me more close-up views of the actual construction of the trestle and its footings, and shown that at some points I will need some round-section "log"-like timbers - which I have now sourced as lime dowel from Cornwall Boat Models. I have also laid in a supply of terracotta-coloured Jovi modelling clay in the hope that it will be of use in modelling the caissons/steps.

thumb_DSC01816_1024 by Jim Botten, on Flickr

As seen above, some of the JV kit-parts have already been stained, in tandem with the long-ago building and installing of the timber road-bed under the track. The template on the right is that included in the JV instructions, while on the left is an expanded version (photocopied and extended by hand with the aid of a ruler) to cope with the much greater maximum height of the intended model. Another copy of the expanded version has been mounted on board and provided with some glued-on guides of scrap-wood at crucial points (though it is yet to be tested...) So, onwards and upwards.

[deanobeano]

Hey Jim,

Research or otherwise that's one large volume of timber in that there trestle - good luck and looking forward to the story of the build.

Regards Dean

[jimbotten]

Hi Dean

Ah well, the model is nothing like the length-to-scale of the original (but still a lot of pokey little sticks to juggle, true). Fingers crossed.

Jim

[jimbotten]

Ok, the 1st December starting-gun has been fired, and there has been some progress. I began by painting/staining a good supply of scale lumber, using two shades of brown tempera powder-paint and a grey made up from black and white acrylics, mixed-up with plenty of water. I used a brush to slop this on to the lumber in various combinations and sequences. Once the timber was dry I made up some “footings” for the bents. By stop-framing through the “Breakheart Pass” DVD I have got a pretty good idea what these distinctive (though I doubt unique?) footings looked like, and how they were made, though the sizing is only my best-guess. I glued four lengths of “post” side-by-side together and divided the length up into 5-scale-foot segments to each represent a pad. Prising some pads in half supplied the timber to be glued onto each pad as the stem of the inverted T-shape (much more easily seen in pictures, below, than explained!)

thumb_DSC01825_1024 by Jim Botten, on Flickr

thumb_DSC01827_1024 by Jim Botten, on Flickr

Using a stiff paper template I measured the needed height for the longest bent, gradually trimming from the bottom until I had the best vertical fit.

thumb_DSC01822_1024 by Jim Botten, on Flickr

(Finding my limit seems to be 3 photos per post.... continued in next post!)

[jimbotten]

I began assembly of the first bent using the prepared jig (as per the instructions in the kit). You can see I am using a hypodermic to apply the glue, and each time I cut a horizontal sash I duplicated it, ready for the other side. (I found the scissors good for snipping through the thin "sash" lumber, better than a knife.)

thumb_DSC01833_1024 by Jim Botten, on Flickr

Once the glue had dried with the sashes all done for one side I (gingerly) freed the assembly from the jig, trimmed its length in accordance with the paper template, turned it over and fixed the sashes for the other side.

thumb_DSC01834_1024 by Jim Botten, on Flickr

I repeated the process for the sway-braces (measure and cut two for each length, glue one of each on one side, then turn over and glue the other side likewise). I then used small dots of glue from the hypodermic to try to represent bolt-heads. (There are plastic bolts supplied with the kit, but they look over-sized and terribly fiddly to fit, as well as nowhere near enough to model every joint as needed. Looking at the prototype photos the bolt-heads are just visible in close-ups but are far from emphatic, and colour-wise pretty well blend in with the weathered timbers – rather than being distinctive iron-and-rust, so I’m hoping the glue-dot approach will suffice.)

thumb_DSC01836_1024 by Jim Botten, on Flickr

[jimbotten]

The various images (photos and freeze-frames) I have of the prototype from different periods show a variety of sashes in use at any one time; some solid through-beams and some pairs of parallel boards to either side of the posts. At least for this first, central, bent I opted for all side-boards, to maximise its strength (the posts thereby being continuous). However for the base-sash (or “sill”?) I filled-in the gap between the boards with wood-filler, to make it a solid through-beam.

thumb_DSC01837_1024 by Jim Botten, on Flickr

The footings were then glued in place under the sill. Note the miniature spirit-level which was used to keep the bent assembly as vertical as possible when it was installed...

thumb_DSC01838_1024 by Jim Botten, on Flickr

...and finally glued in place, using wood-glue at the top to attach it to the already-in-place deck and No-Nails under the footing pads to attach it to the landscape and also achieve the correct height.

thumb_DSC01839_1024 by Jim Botten, on Flickr

So, as of this date we have the single largest bent installed, seemingly reasonably strong and looking the part (enough for me, anyway). Just twenty-odd more to go!

[jimbotten]

With the central bent in place I decided to remove the foam support. Originally I’d thought to leave it in place a while longer, but that would just have made the installation of the other bents more awkward. To my pleased surprise the single bent seems to support the length of suspended track quite firmly!

thumb_DSC01840_1024 by Jim Botten, on Flickr

Removing the foam made it possible to repair the various bits of damage that the deck had sustained while the background landscape was being worked on. However, it also revealed the need for some other repairs. At the right-hand end the track and the deck had parted company horizontally (see photo below) presumably as the track’s curves have settled themselves over the year-plus it’s been in place. This was easily fixed by forcibly detaching the errant deck, adjusting it (with the aid of a craft-knife) and re-attaching it.

thumb_DSC01843_1024 by Jim Botten, on Flickr

More seriously, at the left-hand end the track itself was in trouble (see next photo). A lot of thinking and testing and some brave bodgery with a soldering-iron and files seem to have produced a track that gives a smooth run to my test-locomotive.

thumb_DSC01845_1024 by Jim Botten, on Flickr

So, as of this date all the deck-repairs are complete and I’m ready to start work on the next few central bents (at least that’s the plan at the moment). A more precise count than in my previous posting says “One down and seventeen to go”...

[PeterBowen]

Jim,

Well done for accepting the challenge. If you need more lumber I found a good source used by model airplane builders at http://www.slecuk.com/

As for staining the bents I found using shoe dye and rubbing alcohol worked very well to get the aged and weathered wood look.

Peter

[jimbotten]

(Thanks, Peter, for the tips.) However, I have hit something of a hiatus. I was happily measuring and building the second-longest bent whilst carrying on a conversation at the EarlyN Yahoo Group, starting with an innocent enquiry aimed at the appropriate placing of refuge-platforms (with their fire-barrels) on such a structure. What unspooled was a lot of information about trestle-building in general which has left me in several quandaries. Through following the JV kit's template, and my own innocent nose, I've ended up with bent-spacings too close together and a bent structure that's too distinctly different from the prototype (which has fairly even "storeys" all the way down, while my model currently has shorter storeys at the top, directly beneath the deck). So I've paused at a point where I can still do some re-thinking (and perhaps re-modelling).

[PeterBowen]

Jim,

This might help you:

Prototype Rio Grande Trestle details





Construction Methods
This sketch below shows the method used by the RGS to tie the bridge bents together. This style of construction has been observed at all of the major bridge sites on the railroad that contain enough material for examination. It is also evident in the water towers still standing.



A cap was attached to the posts underneath it by driving a round steel pin down through the cap and into the post. This was done for each post and for each story of every bent. Anywhere from one to three small bridge nails (9" long with 3/8" shaft) were used to attach the sway braces to the posts. These sway braces ran from the lower left to upper right. This unit or story was often assembled off to the side and then moved into position on the bridge.

Each storey was attached to the one below by toe nailing several full sized bridge nails through the post and into the cap of the story underneath. In the bottom story, the mud sill would probably be attached before the story was moved into place. As each storey was positioned and attached to the storey below, girts were added between the bents to stabilise them.

When finished these bridges were strong and stable. The bridge at Ames, (43A), which was built in a very precarious location, stood for almost thirty years with no maintenance whatsoever.



Rio Grande Southern Railroad had 142 bridges scattered along it's 162 miles of track . The most spectacular bridge was 9A. This wooden trestle was 836 feet long and had a 202 foot Howe deck truss in the centre. It was 134 feet high, prompting the early crews to call it the high bridge. The original 9A strained the capabilities of wooden bridge building to their limits and, as a result, it was never really successful. The crews didn't like it or trust it, and neither did management. Unsurprisingly it was among the first of the bridges to be replaced. 9A was removed and replaced with an earthen fill and box culvert in 1903. When this culvert washed away in 1908, it was replaced with the second 9A, which was a more typical curved open deck trestle.

Like 9A, almost all of the original bridges were rebuilt or replaced at one time or another. Most of the Howe trusses were replaced with simple open deck trestles, and many of the smaller bridges were replaced with earthen fills and culverts. By the end of operations, 111 bridges were left on the railroad; virtually none were original. The reasons for replacing a bridge were many, but for the most part, the arrival of heavier motive power or damage caused by weather and high water resulted in replacement.

At a glance, most of the bridges appear to be alike, a closer look however, shows that all differ in detail. The RGS built their bridges following accepted and proven engineering practices. Thus, all of the bridges look similar in overall design.



A Robert W. Richardson view of the Butterfly, Colorado trestle Bridge 44A from November 7, 1952

Detail differences in the bridges occurred due to unusual circumstances in placement or geography. Changes also occurred when the bridges were repaired. Thus, over time, all of the bridges slowly changed.



I know that it can be frustrating not having a clear plan for construction so I did my own drawings which might help you:



I hope this helps you, let me know,

Peter

[jimbotten]

Huge appreciation from this beginner for the all the help that's been forthcoming here and via the EarlyN Yahoo Group. I have now located and downloaded "A Treatise on Wooden Trestle Bridges According to the Present Practice of American Railroads" (1897!) which tells more than anyone could possibly want to know about the design and construction of trestles way back when. It can be downloaded free from Forgotten Books so long as you’re prepared to put up with every few pages being obscured by adverts, or else you can get the whole uninterrupted thing by subscribing for a month (or more, if you get carried away). Lots for me to read and think about, but two things have immediately become clear - firstly my 16' spacing of bents is just about right (by some miracle), and secondly the scale 10 x 10 posts supplied by JV kits need beefing-up to 12 x 12s. Fortunately I'm not too far committed yet not to be able to go back and remake/rejig. Just need now to absolutely sure what I'm aiming for before I commit for a second time. All this may well mean I over-run the 3 month deadline for the Challenge, but as long as I get a finished model I won't mind!

[jimbotten]

There’s been a natural hiatus in my progress due to the realities of the season, however there have also been on-going and fruitful discussions via the EarlyN Yahoo Group (which is highly recommended for those with a period interest). Charles Sloane in particular has been indefatigable and ingenious in analysing the available photographs and coming-up with detailed construction diagrams and pdf templates (a process which hopefully continues). Anyone wanting to see his analysis of the Halfmoon Trestle really needs to sign up with EarlyN. So I’ve been doing some heavy thinking, balancing what I think of as achievable for me with the most accurate reflection of the prototype. As I’ve said to my EarlyN pals, I’m more of an “impressionist” than a “rivet-counter”, and my model can only be “inspired by” the Halfmoon, rather than being a model of it (for a start my model is 120’ high compared with the originals 140’, and is only half the length, plus being on an S-curve rather than an arc; but apart from that...)

I have adjusted to the idea of using scale 12 x 12s in place of the existing 10 x 10s, and the fact (unearthed by Charles) that the central bents of the original are 6-post, while those towards the ends are 5-post. I shall also use the bent-dimensions Charles decoded, with the bent-towers being built in 24’ storeys (bents) atop another, with sashes at 12’ (put very simply; there’s a lot of variation as the bent-towers cope with different overall heights). My biggest compromise, though, is over the “batter” (the outward slop of the outermost posts). The real-world engineers recommended a slope of 3” outwards for every 12” down; for the Halfmoon, Charles has calculated it is 2.5” in 12; while the JV kit template is a skimpy 1.5” in 12. The latter is just about within the realms of real-world usage (though not on the Halfmoon) but I have decided to stick with it – mainly because the underlying landscape was built with the JV design in mind, and the wider, more realistic, base resulting from the prototype batter would, I think, consequently look wrong (even though it would be more correct!)

However, I have not only been thinking. I have been shopping! Taking recommendations from Peter Bowen’s “Challenge” postings I have ordered some isopropanol and a “wood-chopper” (though not the NWSL one). Instead I have gone for a Proses T-105, partly to avoid the complexities & costs of importation, but also because it seems to offer complete flexibility on angle cutting (rather than being restricted to 30, 45, 60 and 90 degrees). While I wait for it to arrive I am drawing-up a “parts list” for my trestle, so that I can (I hope) mass-produce all the same-size items and thus speed the construction process – in due course. So, watch this space, and keep your fingers crossed for me.

[jimbotten]

As my thoughts gelled (more or less) and I postponed action until I was in possession of a wood-cutter, I did do some preparatory work. Over a year ago I had produced multiple copies of the JV template mounted on board (under the delusion that I would one day be doing mass-production!)

thumb_DSC01860_1024 by Jim Botten, on Flickr

These now came in useful as I reworked the design in accordance with Charles Sloane’s discoveries, using red ink.

thumb_DSC01868_1024 by Jim Botten, on Flickr

The (finally) successful (I hope!) reworking (note even here some crossings-out) was then converted into a jig by gluing on guides. For these I used some of the now very surplus scale 10 x 10 timbers, which should leave the 12 x 12 components standing slightly proud, making it possible to glue on sways and sashes without getting them stuck to the guides (a learning-point from my original effort). To align the guides I tacked 12 x 12s in place using Bostik “Glu Dots” and worked up against them, trying not to get too much glue on the posts as I did so.

thumb_DSC01869_1024 by Jim Botten, on Flickr

I have also now completed my initial parts-list, having worked out approximate numbers needed for each of the various lengths of sash, sway-brace and cill as they repeat from bent to bent, allowing for the fact that the number of 24’ storey units varies as the bents vary in height – so that more are needed of the shorter components. These (square-ended) components are I think unaffected by the changes from 5 to 6-post bents. My plan is to determine the lengths of the individual more vertical (angle-ended) parts in the process of building the first bent in the new style, and then to mass-produce a suitable quantity (bearing in mind the variations created by the changes in height). I hope that’s clear enough - I think I know what I mean, but if any further explanation is needed, please ask!

[jimbotten]

At least I see I’m not alone amongst the Challengers in having let time lapse since my last report. I have been doing some work on the trestle, though not as much as I ought, but there’s at least enough to be worth an up-date. So here goes... The first bent to use the “revised” design developed by Charles Sloane over at the EarlyN Yahoo Group (yes, I keep plugging it!) has been pieced together, inevitably taking much longer than the original over-simplified version, what with each full-length post (one piece in the original approach) now being made up of 5 or 6 separate segments with each end cemented to a through-cill. The result is on the left below, compared with the original on the right. To me it’s a huge difference! The end product was also surprisingly robust (I levered it out of the jig using a couple of artist’s palette-knives, with great trepidation).

thumb_DSC01877_1024 by Jim Botten, on Flickr

The next photo shows where a small set-square and a pen have been used to mark the level and extent of the base that will be needed to support the bent (which had been held square in place while the marking took place, calling for at least 3 hands).

thumb_DSC01879_1024 by Jim Botten, on Flickr

Three foam-nails were then driven into the area, with me hoping that they will help to provide a key to hold the modelling-clay in place. (Please remember that I am making this up as I go along and don’t really know if it will work at this stage.) The photo below (if you look hard) shows the foam-nail heads (and the markings for the base-area).

thumb_DSC01880_1024 by Jim Botten, on Flickr