Building the West Hickory bridge

[randgust]

This is another major project I've been looking at, well, forever.   At least since 1975 when I dreamed up the idea of a modular, portable, N scale layout that could go to shows, and built the original 18x36 inch Hickory Valley.

The prototype was basically a logging railroad in Northwest Pennsylvania that incorporated the mill-to-interchange portion of the track as a common carrier, the Hickory Valley Railroad.  That portion of the line was only about three miles long, had one locomotive, and ran between 1887 and 1938.  See my web page at http://www.randgust.com/wd1.htm

The interchange with the PRR (multiple predecessors)  was on the west side of the Allegheny River at West Hickory, PA.   To get there, the lumber company built its own private bridge across the river, originally a Howe wood truss, and then a steel Pratt pin-connected truss in 1896.  It was a toll bridge for vehicles, decked over, and had the railroad track right down the center.   There's a 1925 end shot in a local history book that also shows a small hip-roofed tollhouse, with a Model T straddling the tracks halfway down the bridge.   My great uncle was the toll tender.   When the railroad was abandoned in 1938, the Pennsylvania Department of Transportation took over the bridge, tore the track and wood deck off, and made it a paved one-lane highway bridge across the river - as it was only 14' wide.  That bridge lasted until 2007, and I drove over it many times.   I had one my and only locomotive-car collision (when I was 19) right at the end of the bridge on that PRR crossing, in 1975, and I was in the locomotive.  That incident ended my interest in an in-the-cab railroad career.

Here's the 1920's ICC valuation map extract that shows the layout of the bridge, West Hickory, and the rather odd joint bridge/diamond/PRR interchange:  http://www.randgust.com/West%20Hickory%20Valmap%20Extract.jpg

This was actually a pretty big bridge for a logging operation - 641 feet and four spans.  By 2000 it was 100 years old and getting a lot of historic preservation attention although the handwriting was on wall for a one-lane 100-year old bridge, even if it didn't have weight limits on it.  As far as I can tell, it was built for at least 50 ton.  It was torn down and replaced with a prestressed concrete span in 2006-7 and only the endplate was saved for display.

Here's the historic preservation page on the bridge:  http://historicbridges.org/bridges/browser/?bridgebrowser=truss/hickory/

So you can see all the challenges to try to model this.   First problem was West Hickory itself - I looked at an Ntrak module for years and finally decided I couldn't make it work with the track layout and curvatures.    T-Trak to the rescue, I could fit what I needed on a Triple, built that last year and had it at the Altoona show.   I've always had a problem with the concept of the Hickory Valley in that it needs to be a branch off of a main line / module, but whether a peninsula or a other method, I need separation.   I need that bridge.  Anyway, it's just fun, I've never built anything like this before.

Here's the one shot I do own that my mother took, this is taken from the depot area looking northeast (south side of the bridge).  Year is uncertain (I'm guessing early 1930's), but two facts are clear, you can see the white guardrail on what's a black bridge.   It was always painted aluminum once PADOT took it.

.

That's the PRR main in the foreground, today completely and totally gone- that ran up the river from Pittsburgh, Oil City, West Hickory, Warren, Olean, connecting to Buffalo line there.   Severed by the Kinzua dam in 1965, West Hickory ran as a branch until 1976.

[randgust]

The conflict between an historic model and a practical module is significant, and it's straddling the line here to do both.    After I got through the mental block of West Hickory, with the depot area, and future tannery, and PRR interchange, this is the next obstacle. 

Two big issues here, if I were to make it full size it would be a good four feet long.   I built my "Ross Run" logging module 42" long, and that's just a beast to move around, compared to 36" on the HVRR module box.   The 36" module size for West Hickory as a triple worked very well, easy to handle.  So 641 feet was just too darn long.  Not sure about that one.

The other major problem was the bridge.   I took a lot of close up photos of the bridge itself before demolition, trying to get a handle on how it was built as a pin-connected truss.   Wow.  It was really complicated.   If it was to be made from brass, it would have to be etched.  The other thing was that in many ways, the highway deck obliterated a lot of the coolest detail that you'd make pin-connected model.  Worth it?



Looking at kits, and parts, nothing much scored 100%, and some of the brass bridges that 'might have been' were no longer available.  But one thing did intrigue me - the Central Valley kits.    They had an older-looking Pratt truss that while not pin-connected, had most of the same general features, was the right span length at 150 feet, had some light-sections and wire bracing, was affordable, and looked like it could be kitbashed for both some more detail and and a full vehicle deck.   So I bought one to test the idea.    Lots of studying later, it's possible.   It's a go.   So, may as well start the construction thread, here we go.

What I'm doing for the bridge part is a pretty standard project that a lot of people can adapt, but I'm throwing in some wrinkles that I hope will be useful for others.  Most layouts have problems with spanning openings, aisles, and I have to develop solutions for all that.   Another painfully obvious thing I ran into was track design, and trying to figure out how I was ever going to keep the track clean as well as make the entire mess highly portable and durable.   I want the entire thing modular as I have no permanent place to set up the logging line as one intact layout.

[randgust]

Since West Hickory was already T-track standard, and this would end up coming off the diamonds at the front, it occurred to me that I should 'probably' make this as T-trak compatible as I could; i.e. it still should comply with length and vertical and horizontal standards even if the bridge was only one track, and even if one-track standards didn't actually exist in T-trak.    The other thing I really liked was the proven ability of the Kato Unitrack connectors to hold up in the field for rapid setup and teardown.

The basic geometry of a three-span Central Valley bridge unit would fit - I have 36 5/8 inches length for a T-trak triple, and 33 1/2 worth of bridge if I went three span instead of four.    My West Hickory module ending at the riverfront would need an adapter built about 4" long to complete the curve and get to the tangent on the bridge from the 60-degree diamonds, that was nonstandard for sure, but if I kept the bridge module straight and made that a separate little curve adapter module with the 9 3/4 radius it would still work.  This was fully prototype, and I needed a spot to put that cool white hip-roof tollhouse on anyway.   Another thing I liked about the CV bridges was that even if they were heavier and more modern, the general appearance and span distance was right - if I even wanted to replace them with an actual 100% design-accurate brass etched bridge, it wouldn't be impossible as long as I didn't make it equally impossible to remove the spans from the abutments and piers.  I wanted the entire bridge to be able to be removed for inevitable repairs, cleaning, and possible replacement.

The design of my West Hickory module basically left me with an 8" width for the bridge module, doable, and wide enough to protect the model.

I had roughly 2 3/4 vertical elevation to work with, so a deck design for the module with plywood on the bottom would be used.    T-trak had already pretty much established how a bridge module should be built and that was a solid concept with plywood end plates and adjusting screws.

All that being said, the geometry on making all this fit, line up, and actually work is pretty much field engineering.   Everything comes from 'top of rail' and works down to make it fit precisely.  Step one was to design and make the adapter from the West Hickory module to the bridge module, using 2 1/2 x 8 end plates from 1/4" plywood, cut to embankment shapes, edge-shaped with a dremel, and fitted with cork roadbed for the road/track section.   Object is to precisely establish 'top of rail' on the adjacent bridge module, as fitting all this up is a significant chore.



I realized that both the horizontal and vertical set on the bridge to do it this was a lot harder than I imagined, as things like abutment position and pier heights had to be calculated backwards based on railhead down, not table-up.   So you have to establish railhead first, and at least partially build the bridges to figure out where the bridges ended up in relationship to track dimensions.

The 8" x 2 3/4" endplates were made as pairs, clamped together, and drilled for 1/4" wood dowels to cross-pin with each other.   My 'nonstandard' adapter will have the male end so that the 'standard' endplate on the bridge module will stay flat.  Pinning dowels like this makes it pretty much of a press-fit that can be backed up with a screw or two later in the field.

[randgust]

One headache I ran into that I doubt anyone else would was the dual-mode use of this bridge.   I needed a full-width deck, and the one end shot with the track made it clear it was a wood deck, level with top of rail, wood stringer outside edges only, and a wood guardrail.   

I went ahead and assembled only the bottom stringers and cross-stringers of the bridge kit to see what it was going to do here.

Here's the CV bridge, stock, from their web page:  http://www.cvmw.com/imagecvmw/1800/1810-bridge-1.jpg

Here's a great end shot where you can really appreciate just how spidery the prototype was for a truss:  http://historicbridges.org/truss/hickory/p1010052.jpg
This bridge was exhaustively documented prior to destruction, that's just a sample, but a real boon to modeling it.

General lines are pretty much identical, major departures on the bottom chord is a lattice rather than pins and bars, and note that the first vertical truss beam on West Hickory is actually a wire member rather than the model steel beam.  I'll modify that, maybe, after I get a sense of how strong the model is if I mess with it structurally.  It looks like the way to do this is to glue the basic bridge up and then surgically remove and replace anything I want to modify in a lighter cross-section or with tension rods/bars.

The CV bridge, as it comes, has a plastic bridge tie insert that custom-fits the cross-stringers that you are supposed to put your own rail on.

Right from the start I could see that probably wasn't going to work.   I got a piece of Micro-Engineering bridge tie flex track just to see if that might work (code 70 rail) and was reasonably impressed with it to at least solve the rail problem although I was right back where I started on the deck.    I also ordered a walkway section to see if that would work, and that turned out to pretty much be a bust, not even sure any usable parts came off of that idea.

I ended up putting on individual .020 styrene plates across the full width between the cross stringers, on top of the main stringers, between them and the bridge ties.   That raises the track by .020 but that's not much of an issue.   I then secured the flextrack to that with a light touch of Goo, removing some ties in the middle to clear the cross stringers like the original CV tie strip.   Now I had to build up beside the flextrack tie section to make a flat section, so I scored off multiple strips of 3'3" wide .060 styrene.  I didn't glue any of this to the bridge stringers themselves; the track with the .020 plates glued to it lifted right out. So far, removable track deck is feasible.

Now, here's a tip.   If you are laminating up styrene layers, don't use styrene cement.   I'm not sure why, but no matter what you do, or how thick material you use, it will eventually warp when layered.    What I discovered is that using ACC doesn't warp, and is just as durable if not more so.  So the .060 filler strip beside the ties was ACC'd to the .020 deck plates.  That will allow putting a real stripwood deck down full width here.   At this point I spray-painted the entire deck flat black.  If you look at the shot above you'll see the first bridge span at this stage.

[randgust]

I've gone ahead and pretty much assembled the first span, trying to figure out what I can change to more closely resemble the actual bridge and what I can't.   The model bridge is a later, heavier span than West Hickory, but the basic Pratt truss reinforced with bars and rods for diagonals is still there.   The plastic is easy to work with, glues up well with solvent cement.

Found a nice exploded view of the kit online, this shows how it is designed:  http://www.blwnscale.com/blwphotos/cv-1810a.jpg

If you look at the actual span, here's the most significant departures from the Central Valley model:

* the first vertical beam/member on the span from either end is a pair of rods, on the model it is a solid beam.  I'm planning on cutting this out and replacing it with a .020 wire pair AFTER I get everything else reinforced up.   The actual bridge is rather odd that it is only a tension member.  I've never seen a Pratt truss, or any truss, with that first vertical member being rods only.   Take a look - notice the rods on the vertical support and a pair of bars on the diagonal, and also the overhead detail:  http://historicbridges.org/truss/hickory/img_0038.jpg

* The horizontal bottom chord on the span is a heavy pair of pin-connected iron bars, on the model it is a lattice beam.   From the way the model is built, if I leave off the lattices on the top and cut off the bottom  and put some pin detail at the joints with .030 wire, it will be 'close enough'.

*  The top chord diagonal braces are rather light bars on the real bridge, on the model they are lattice castings.   Those can be trimmed back to just a bar portion, first test of that idea worked rather well.

This will definitely make the model 'lighter looking', but it has to be done carefully as the model isn't exactly rigid until everything is glued up. 

Now that I'm really studying this bridge, it's rather interesting that it's built early enough to still use major pin connections instead of riveted gussets, but late enough that there's a lot of riveting going on with the top chords and lattices.   It's really a rather odd bridge.

The instructions on this thing are rather lacking.   There's lots of ways to put it together wrong, some of which I discovered already.   Nothing critical, but with no part numbers on the sprues and with left/right/up/down not particularly clear either, stuff happens.  Also, thread is furnished for the rods on the diagonal bracing, I'll probably use .020 wire, as sooner or later, thread goes slack and looks terrible.

As I assemble the second and third spans I can document this better in photos.

I got some Chooch cut stone abutments and piers that are most certainly close enough, and from the looks of things, just a hair tall.

[Ed Kapuscinski]

There may not be any replies yet, but I'll make this one just to cheer you on.

Go Randy!

[peteski]

Nice to see your work Randy.

[randgust]

OK,well, this should help explain this better.    I have the basic noT-track 'triple' base built, the adapter plugged into the one end, the one abutment propped up, and the semi-finished first span to the point where I can at least prop it up on one pier.   I'll get to more detail on what I'm doing on the bridge, but for now, here's the module concept that will eventually have all three spans set up on it -   450' worth of N scale bridge.

This is sitting on the storage box for West Hickory - standard 'base construction' for a T-trak bridge module triple length, just single track and 8" wide.  The 4" adapter curve on the first shot is plugged in the other end of the bridge module, with a matching embankment profile end plate.



Side view of the semi-finished Central Valley 1810 span, still putting it together but you get the idea.  For anybody freaking out over the C80, remember the only visible part of the finished product will be the railhead, the rest is decked over level for the roadway deck on the bridge and the approach adapter.



The idea is to fit a 1" chunk of Unitrak on the bridge end so it can mate with a standard T-trak connector, for lack of a better idea, and then trim the abutments down to get the approaches and everything to line up to railhead height.   The sub-deck for the roadway is in there, just hard to see.



So that should help show the idea here....

Oh, one more thing.   To understand my semi-emotional link to this spot in my own life and family history (both sides).... this shot was taken the same day as my West Hickory backdrop, shooting up the Allegheny (north) at this location, and other than the wire I have to photoshop out, shows the amazing unspoiled beauty of the upper Allegheny.  This very much could still be 1925, and this will be the backdrop for this scene, cropped and printed.   For the rail-oriented, the PRR line would have been on the left bank.

[SSW7771]

Very cool! I will watch this build with interest.

[randgust]

As I start researching detail on this I'll make the observation that if you're going to build a bridge, it sure doesn't hurt having it nominated on the National Historic Register, even if it has been demolished - because the level of detailed, high-quality photo documentation on this thing is simply astounding.  The only thing I'm surprised about is how fuzzy they are on the original railroad relationship.   I may yet scan that book photo just to have a thread link to show the rather amazing shot of the tracks with the wood planking and Model T on it.

I also came up with this bonehead idea of making a couple adapters to standard T-trak - probably two doubles - that would take the front main alignment from 1 1/2 inches in and swing it over to my single track centerline of 4" on the bridge, and take the second main track and swing it BEHIND the backdrop divider on a 5" x 36 5/8 'sneak track'.    That would make this whole bridge module adaptable to standard T-trak.  I think its at least possible, but I'll have to check out some geometry.   I only use the Kato unitrack for adapters on and off the modules, everything in the middle has turned out to be just about everything but.

I also have to give some credit to Daryl Kruse on his Mississippi bridge crossing concept:  https://i.ytimg.com/vi/HKzj8yX3owY/maxresdefault.jpg particularly in the idea of a narrow, long shelf with a real photo backdrop on it to convince me this is worth it.      I also got to see Jerry Britton's magnificent Duncannon module firsthand with the photo backdrop and that convinced me of the long and narrow bridge module concept validity  http://kc.pennsyrr.com/photos/01501/01598_m.jpg
That's the impact of final results I'm aiming for here.

[Ed Kapuscinski]

Ok, the one thing I will say. It pains me to see such fine modeling saddled with the code 80 "fencetrack".

For stuff that's not mainline, is it too late to consider something more finescale and befitting your talents?

[randgust]

Ed - not paying attention !      That 9 3/4 curve on the adapter will be completely buried in both deck planks and dirt up to the railhead as well as the little chunk of Unitrak, and the code 70 on the bridge will be planked in the center and on both sides with only the railhead visible.   No visible anything.

Good view on that end before it got demolished: http://historicbridges.org/truss/hickory/img_0007.jpg
The HV track was in the center of the bridge, and as soon as it got off the span, curved sharper to the inside of the road curve here.   Note the concrete retaining wall right near the approach.  1925 shot has it there with  wood railings on it.   Track was planked over in the center, and outside of the ties, dirt outside of that flush to top of rail.  The PRR diamond crossing was just barely off the pavement to the right in that shot, about where the speed limit 25 sign is.  The tollhouse was on the left about where that utility pole is.  That's a neat shot because you get a sense of fear of what it was like to drive over the darn thing - a complete game of Chicken.   Other than the fact it was wood planks instead of asphalt, no other changes on the bridge itself.

The other (east) end (if you look at the ICC valuation map link at the top) had it's own wonderful feature, a 90-degree curved wooden pile trestle cutting across the diagonal on the road "T" intersection   If I need to make a sharp 90-degree curve module for some reason, that's what was there.  Not a trace today, one good photo of it I've found with a train on it.

One of the few recognizable traces of the HVRR today is the railroad grade just north of that intersection, it crosses Route 62 and is very noticeable, mild oblique angle with a visible roadbed on the east side. 

One thing I have learned is that if you're going across a modular table joint on a curve, you're a lot better off to use a piece of sectional and 'tweak' it if necessary, any curve sharper than 20" is just way easier to build and hold to a consistent curve that way.   Flex will always, always kink at the joint.

[PRRATSF]

Its a shame the old bridge was torn down. There's a deck truss RR bridge just east of Tyrone Pa., and I'm using the CV kit to try and reproduce a reasonable facsimile. Watching your progress closely.
Sam

[Ed Kapuscinski]

Ed - not paying attention !      That 9 3/4 curve on the adapter will be completely buried in both deck planks and dirt up to the railhead as well as the little chunk of Unitrak, and the code 70 on the bridge will be planked in the center and on both sides with only the railhead visible.   No visible anything.

Ah, nice. I didn't realize the approach track would be all hidden.

Consider my concern withdrawn. 

[randgust]

There's a deck truss RR bridge just east of Tyrone Pa., and I'm using the CV kit to try and reproduce a reasonable facsimile. Watching tour progress closely.
Sam

The mystery bridge beside Rt. 453 and Plummers Hollow Rd?   Yeah, I know it.   That could work.   Where the heck did that line ever go?   Up into that valley someplace and just dead-end?  That was an expensive bridge for a dead-end branch.