Trains.com

Subscriber & Member Login

Login, or register today to interact in our online community, comment on articles, receive our newsletter, manage your account online and more!

Bridge for a curve

4300 views
11 replies
1 rating 2 rating 3 rating 4 rating 5 rating
  • Member since
    July 2006
  • 9 posts
Bridge for a curve
Posted by Espeefan on Monday, December 28, 2020 1:52 PM

Hello everyone!  I’m building a model railroad in HO scale and I need to figure out how to build a bridge on a 24” radius curve.  I understand the basic idea, I think, of needing to miter multiple individual bridge sections to match the curve, but how do I go about that exactly?  How do I calculate the miter angle, and how much to remove?  Also, what do I do about the joints, where each section of bridge meets on a peir?  

To give a little more detail to my question, I’m looking at using multiple Micro Engineering open girder bridge kits to have a multi-span bridge, supported by stone or concrete peirs.  I would suppose multiple shorter spans would be better for a 24” radius curve?  30 foot bridges, or 50 foot?  

After cutting the girders to a miter, and shortening the girder on the inside of the curve I assume I might have some issues with the cross bracing no longer matching up to the ends of the girders because the inside girders are now shorter.  How do I deal with this issue?  Center most of the crossbracing between the girders and leave some of the crossbracing off at the ends, as some of it may no longer fit flush?  What would a prototype practice be?  

Also, are gaps between the girders, when they meet at the mitered angle prototypical, or would they butt up tight to the next girder in line, at the pier?  

One last question - are the track ties allowed to over hang the girders on the edges, while the rail must stay within the girder footprint? 

Thank you for your patience and advice!

 

 

  • Member since
    February 2005
  • From: Vancouver Island, BC
  • 23,330 posts
Posted by selector on Monday, December 28, 2020 2:44 PM

You draw your centreline and track profile on a sheet of graph paper, and then draw a series of segments wide enough for 'gauge loading' for the scale of rolling stock you are using.  So, you draw your curve track, then use some measure to place distance between the railings, if there are to be railings, or a refuge station if it's a trestle type bridge, and draw lengths to wherever you will need pylons or an abutment.

 

  • Member since
    March 2017
  • 8,173 posts
Posted by Track fiddler on Monday, December 28, 2020 2:52 PM

How ya doing Espee?

I ran across this similar situation on a 22 or 23 inch radius curve.  There is quite a few lines running underneath so the situation was a bit complex and a girder plate Bridge was the only solution for this situation.

I had cut a stepped curved piece of Masonite and used 2 Kato Bridges and dissected them to cut the girders to the right length to get the angles for the piers, bents to fit between the lines below.

In your situation I would think you could put a 24 inch radius curve on poster board and set your ME girder plate sections on that curve to fit nice and trace a template to cut out for reference.  ME also sells just single flat girder plates that you could cut to fill the ends of the slight triangles on the outside radius of the bridge in between your pre-made sections.

I don't think I'd worry too much if you're ties hangover the deck girders a little bit.  Sometimes custom stuff in model railroading has a bit of a prototypical trade-off.  I personally don't worry about it too much as long as it isn't too overwhelming.

Here's an example how Duluth's curved girder plate deck Bridges going to the ore docks look underneath.  If you examine closely you'll notice they have rectangular sections with slightly angled shorter sections in between.  If you could find similar supports, putting two together at an angle or scratch build a tripod with X-suports in between.  Perhaps you could do something like that.

 

Hope that can help.

 

Nice Bridge Selector! and here comes the double stacks!Stick out tongue

 

 

TF

 

 

 

  • Member since
    September 2003
  • 21,669 posts
Posted by Overmod on Monday, December 28, 2020 3:08 PM

I think I'd consider making the piers heavy enough to have a complete arch springing on each 'facet' to which a span goes, with both the vertical (at the arch springing) and horizontal (at the deck) thrust carried through the masonry.  Ballast the deck and lay the track on a ballasted continuous prism.

You can do the same with a girder bridge, as supporting the sections either straight or 'skewed' is a simple matter of loading.  Just be certain the spans are wide enough to keep the rails well inboard of the supporting rockers.  You could also tie them together, but I'd stay clear of modeling an indeterminate truss over multiple piers on a curve -- the Japanese can and have done that, but for esthetics, not economics or sound engineering...

I would NOT lay bridge ties as the only support between the rails and the trusses or girder structure.  Let ballast handle the lateral loading.

You could build a concrete arch a la Nicholson curved -- in fact I think the Lackawanna built those.  Again make it well wider than the tie ends require, if the span is long.

What's possible on a single span is a bit tighter -- see the track on the bridge at Dupont that tripped up Amtrak 501 for the poster child of curved track on a straight bridge...

  • Member since
    July 2006
  • 9 posts
Posted by Espeefan on Monday, December 28, 2020 11:41 PM

Lastspikemike

You could make a drawing full scale. 24" is your centreline radius. Draw two more arcs equal to your roadbed width. To do that you split the difference and add half to 24" and deduct the other half to get the limits of your desired bridge deck width. Draw three arcs. The approximate radii will be a bit less than 22", exactly 24" and a bit less than 26". 

The number of segments you will  need depends on the length of each segment you choose. There is no set length. Shorter segments more closely match the true arc. The closer you want to be to a true arc the more segments you need and the narrower the widest part of the bridge deck becomes. In theory, an arc is an infinite number of infinitely short segments.  

The chosen lengths of your segments determine the angle of each cut needed to obtain your desired 24" on average arc. You place the two selected segments on your drawing and half the angle formed by the two segments make each of your adjoining end cuts. Bisecting the angle at each join.    

The minimum width of each segment is the governing dimension.

Protoype bridges exist that were assembled from a series of straight segments.

Using a full sized drawing as your pattern also allows you to include easements on the bridge deck if desired. 

The bridge decks I've seen are approximately twice the gauge. The long ties project beyond the main girder beams by a small amount. If you use through girder bridge sections then obviously the ties end within the sude girders  width. Bridge sections with the girders below would have the ties go out to the girders  and end there. There are such bridges with ties extending beyond the girders  underneath as for a trestle type.

Google "Kettle Valley West Fork Canyon Bridge" for a good prototype. This bridge not only combines a series of tangent girder bridges on very tall towers but the East end lands on a section of fill restrained by a retaining wall and then it also enters a tunnel with an exposed concrete entrance acting a slide shed and continues further into a second tunnel shortly afterwards. This prototype still stands, as part of a hiking trail, and has all the features a model railroader could possibly want. 

 

 

Hello Mike, thanks for sharing a straightforward explaination.  I will surely give it a try.  Drawing my curve fullsized and seeing how the girders line up in segments and at what angles makes it sound a bit less daunting.  I’ll have to experiment a bit and see what I can come up with.  

  • Member since
    July 2006
  • 9 posts
Posted by Espeefan on Monday, December 28, 2020 11:46 PM

selector

You draw your centreline and track profile on a sheet of graph paper, and then draw a series of segments wide enough for 'gauge loading' for the scale of rolling stock you are using.  So, you draw your curve track, then use some measure to place distance between the railings, if there are to be railings, or a refuge station if it's a trestle type bridge, and draw lengths to wherever you will need pylons or an abutment.

 

 

Selector, thanks for taking the time to type up an explanation for me.  Your method coincides with Mike’s up above.  Sounds easy enough.  I apreciate the reply.  

  • Member since
    July 2006
  • 9 posts
Posted by Espeefan on Tuesday, December 29, 2020 12:07 AM

Track fiddler, that’s a nice bridge you’ve put together.  Those bents look great.  I appreciate your input and photos.  Regarding the prototype bridge photo, I rememember reading about bridges on curves using shorter square spans and longer spans at angles, just as you typed.  Maybe even Micro Engineering’s website may have mentioned that in a pdf I downloaded for instructions on building their steel vaiaduct.  Something to keep in mind.  Thanks again!

  • Member since
    July 2006
  • 9 posts
Posted by Espeefan on Tuesday, December 29, 2020 12:15 AM

Overmod

I think I'd consider making the piers heavy enough to have a complete arch springing on each 'facet' to which a span goes, with both the vertical (at the arch springing) and horizontal (at the deck) thrust carried through the masonry.  Ballast the deck and lay the track on a ballasted continuous prism.

You can do the same with a girder bridge, as supporting the sections either straight or 'skewed' is a simple matter of loading.  Just be certain the spans are wide enough to keep the rails well inboard of the supporting rockers.  You could also tie them together, but I'd stay clear of modeling an indeterminate truss over multiple piers on a curve -- the Japanes can and have done that, but for esthetics, not economics or sound engineering...

I would NOT lay bridge ties as the only support between the rails and the trusses or girder structure.  Let ballast handle the lateral loading.

You could build a concrete arch a la Nicholson curved -- in fact I think the Lackawanna built those.  Again make it well wider than the tie ends require, if the span is long.

What's possible on a single span is a bit tighter -- see the track on the bridge at Dupont that tripped up Amtrak 501 for the poster child of curved track on a straight bridge...

 

Overmod, thank you for your reply.  Some of your nomenclature confused me a bit at first, but I think I get the gist of what you are saying!  I did consider trying my hand at building a concrete viaduc, based off one of the few prototypes I’ve seen.  There is a handful.  I may still consider that, as the construction of a bridge like that could be easier in some respects, and in others maybe not so much.   I could definitely see the curved subroadbed easily being disguised, but I’d have to research a bit on how to make the arches.  Extruded foam probably!  I picked up a Kalmbach book all about modeling bridges and it has an article that could hemp me out with that.  Oddly enough it never talked much about a curved girder bridge, or how to build one, so that brought me here.  

  • Member since
    October 2008
  • From: Calgary
  • 2,047 posts
Posted by cx500 on Tuesday, December 29, 2020 9:08 AM

Espeefan

  Also, what do I do about the joints, where each section of bridge meets on a pier? 

....... 

Also, are gaps between the girders, when they meet at the mitered angle prototypical, or would they butt up tight to the next girder in line, at the pier?  

Each bridge segment would normally be separated by a couple of inches to allow for expansion, rather like the gaps at joints in jointed rail.  In your case of very short spans, the bearings at each pier will be very simple to model since the amount of expansion will be minimal, merely a thin square of plastic under the girder end.  Add an anchor bolt casting if you want more detail.

John

  • Member since
    January 2004
  • From: Canada, eh?
  • 13,375 posts
Posted by doctorwayne on Tuesday, December 29, 2020 12:43 PM

These two bridges....

...along with this one (and a couple others)...

...originally existed as the 2" wide 3/4" plywood subroadbed that made up the mainline of most of my layout.  I don't recall for sure, but I believe the curves where bridges are now located were mostly 34" radii or larger.

To build the bridges in the photos, I cut out the roadbed where the bridges were to be located, then flipped-over the cut-out sections of roadbed and constructed the bridges, upside-down, atop the plywood curves.
Determining the height of the support towers and piers was a trial-and-error process.
The tighter the curve, the shorter the spans will be, and for tight curves, a deck girder bridge is the best choice for accommodating longer locomotives and rolling stock.

Bridges on such curves, with above-the-track superstructures, will need to be overly wide if you're running longer equipment.  Of course, the choice is up to the modeller.

Wayne

Subscriber & Member Login

Login, or register today to interact in our online community, comment on articles, receive our newsletter, manage your account online and more!

Users Online

There are no community member online

Search the Community

ADVERTISEMENT
ADVERTISEMENT
ADVERTISEMENT
Model Railroader Newsletter See all
Sign up for our FREE e-newsletter and get model railroad news in your inbox!