Design

I've posted something similar to this before, but I drew it up again from scratch, so here it is.  The idea is to have the layout completely self-contained, and let it sit on a suitable base (probably file cabinets, at the moment).  A backdrop and lighting are critical.  Basically, it needs to be a show quality display.

Fortunately for me, the 160" length fits the 40" panels beautifully.  In fact, I designed the top to be four complete unmodified panels.  There are a total of ten vertical panels which support the backdrop and the top:

You can also see some of the bracing I added to keep everything rigid.  At the lower rear and upper front corners are completely enclosed box beams, which I hope will greatly stiffen the entire structure.  This cross section will give an idea:

The lighting trough at the top runs the full length of the layout for a little added stiffness, but there are dividers inside which cut it into three 53" long troughs, and which hopefully add even more rigidity to the structure.

Not included in my price evaluation are the backdrop material (FRP panel), lights (undetermined, probably fluorescent fixtures as drawn), or light diffuser panels.  The reason is that they would be needed for either material, so they are not really relevant to the comparison.

Price comparison

Once I had the basic design, I started laying the parts out on a grid of twenty foamcore panels, to see how they would fit.  Other than splice plates (which can be made from just about any scrap), I think I've got everything included.

So, total material cost here is around $110 US.  Other than the curved backdrop supports, everything is fairly straight and shouldn't take too long to cut out.  A sharp knife is probably the only tool I would need.

With that done, I tried laying the parts out on 4'x8' sheets of plywood.  It took more than I expected, actually.

All of the pieces on the left can be made 80" long instead of 40" as drawn - just ignore the seams there.  There's a fair bit of waste, though I can usually find a use for plywood in the shop, so that's not a horrible thing.  Other than the table saw, I would need a jig saw and a hole saw.  It still shouldn't take me too long to cut out, and might even go faster since I can set the saw up for repeated cuts fairly easily.  Because the wood is a bit stronger, I figured I could eliminate three of the backdrop supports, which helped get everything to fit.

The problem is, even cheap 1/4" plywood is about $20/sheet.  That's $120 in wood.  I could try to make the layout out of something like hardboard, which is half that price, but it doesn't seem like a really smart move.  Although, now that I think about it, it might actually work, given the design.
 

Pros & cons

I was actually surprised at how close the prices came out.  With less than $10 difference, I would call them a wash.  I assume that the wooden layout would be a bit stronger, but also a lot heavier.  Because of the extra weight, I wold presumably have to use reinforcements in the corners of the plywood version, whereas I might get away with construction adhesive or Gorilla Glue for the foam version.  The plywood version would require use of my woodworking tools, which involves more effort, planning, and time.  The foam version can be worked on at any time of night and in any convenient place without disturbing anyone else.

In the end, I am leaning towards foam.  The only real questions I have are, will it be sturdy enough, and what is the best adhesive for all those joints.  Something with a bit of gap filling seems capability seems like a good idea, which is why I thought of Gorilla Glue.

Just checking...

Dear Ken,

Are you genuinely asking for feedback on this proposal?

Happy Modelling,
Aim to Improve,
Prof Klyzlr

My experiences ...

... having worked with both (on the same project).  I used 5 mm foamcore (Kappa Board) and 6 mm cabinet grade plywood (which really only measured 4.5 mm thick ...).

• Foamcore is MUCH lighter than wood!  Really can't stress this aspect enough, as it relates to things such as "bending moments" while transporting the modules (i.e., only supporting them from their ends).
• I found foamcore to be "stiffer" than the plywood ... it tended to stay straight and "flat"; which made it much easier to assemble the parts.
• Foamcore is easier to cut (and quieter, and no dust); an important consideration for some folks.
• I used a furniture grade resin glue for both the foamcore and the wood.  After curing, it is resistant to moisture and temperature (i.e., will not soften under normal living conditions).  I think it is something similar to the "Carpenter's Glue" that you have in the States.  Others have used hot glue with success as well ... certainly faster.
• I found that by cutting the foamcore parts a little bit "tight"; I got a very good joint as the foamcore is compressible ... really, no "gaps" to fill.
• There are two areas where I think plywood has an advantage over the foamcore:

1. Plywood is MUCH more resilient to "bumps" and "knocks"; especially along its edges.
2. Plywood is easier to "mount" things to using fasteners such as screws.  I get around this on the foamcore by gluing plywood "pads" wherever I want to use a fastener of some sort to mount something.

edit: And OBTW; during the final phases of construction; I "overstressed" the profile boards by placing large amounts of weight on top of them.  It was the plywood profile board that failed (along an internal seam) ... the foamcore remained intact.

Lighting ...

... I do like the recessed lighting.  I considered something along those lines for my modules; however, I became concerned about the "shadowing" caused by the back of the lighting box and the uneven lighting on the backdrop as one moved upwards toward the sky.

I opted to mount my T8 LED "tube" lights below the sky-board and use a wider front valance to hide them from view.  The LED tube lights come with a reflector and diffuser built into the tube light itself, so none of that is required.

Lots of pieces

   Seems like lots of work for a simple shelf layout?  Hollow core doors are about 80 " by 30" and if you have a building salvage/recycle center nearby they only cost a few bucks each. Just glue some foam on top if you want to carve ditches or other low spots then start building the layout. I wouldn't worry too much about weight on a shelf layout unless you have to carry it around all day, it's usually possible to find someone to carry the other end for occasional moves.  I don't think I'd bore those large holes in foamcore stringer since they are just foam and cardboard so not much strength would be left in the middle of each span.....DaveB

Not fit for purpose : See OP g&d's

Dear Dave B,

Factor in:
- Typical interior-grade single-person hollow core doors weight up-to 20 kilos apiece
(minimum 2 required for the 2D "footprint" area under consideration)

- The design from the outset has been stated as requiring full-proscenium, which means it's not "just some door-minos"

and in considering HC doors, we're already a long way behind the nominated starting position,
let alone on-our-way to the destination...

Happy Modelling,
Aim to Improve,
Prof Klyzlr

​PS a question, how does one run feeders/wiring on a HC-door layout?
- Over-the-scenery-surface requires pre-prep and creates "do not cut here" areas
(yes, wiring-run troughs can be routed/melted thru the XPS foam "top layer",
but once covered with plaster/glue/Chux/caulking/Woodland-scenics/etc, 
locating, avoiding, un-earthing, and debugging can be exceedingly painful...)

- Thru the door potentially leads to "loosing" the wire in the door cavity as it's fed thru
(Yes, I've been there, this is not just hypothetic)

- and even if one manages to get the wires thru the door first-time outta-the-gate,
the feeders and buss-wires, and any interconnects, are then fully-exposed and protruding below the "base/wear surface" of the module.

Sit the module on a flat surface (table) and you're crushing/stressing/pulling the wiring,
slide the module along a surface or edge (thinking table edge or lip-of-car-door during load-in/out),
and you're likely to sheer the wiring clean off where it exits thru the door skin
(again, I've been there, and been called-in to rewire HCD-based layouts after such disasters...)

Combination of the two

Food for thought,

Use both plywood and foamcore to take advantages of each material.  Use plywood for the parts that get bumped and require  screws and foamcore for light weight.  E.g., build the perimeter of the bench(base), bench top and ends of the valance with wood and use the foamcore for the interior bracing, backdrop.

Art    

Taking a punt....

Dear Ken,

OK, I'm going to preface the following by saying that you've obviously put a lot of thought thus-far into the mission, and far-be-it-from-me to deride that motivation or enthusiasm. That said, the following comes from over a decade of actively designing, building, and touring proscenium and 100%-foamcore designs. I hope it's helpful, as it's certainly not intended to be detrimental or derogatory in any way.

In order of appearance and occurance.

- I regularly source 25-sheet cartons of 60"x40"x3/16" foamcore for AUD$250.
At a per-sheet price of AUD$10/sheet,
and factoring that two of your 40"x30" sheets matches one of my 60x40s,
my source appears to be half your pricing.

FWIW, locally available 2400 x 1200 x 7mm (8x4 x 1/4") "structural" plywood goes for AUD$31
(and that's "CD" plywood, not furniture-grade "AA" or Luan!)

Keep this in mind, as it puts a serious dent on your comparable pricing specs,
(as per the thread title).

- The aforementioned 60 x 40 x 3/16" sheet of foamcore weighs in at 1.5pound/0.7 kilos.
The aforementioned 8x4 x 1/4" Structural plywood sheet weighs in at 20pound/9kilograms

​That translates into 
Foamcore = 0.11 pounds/sq-foot of 3/16" material
Plywood = 0.6 pounds/sq-foot of 1/4" material

Now, given that the proscenium design uses the same sq-foot area of material for each component, 
we can see that plywood is certainly heavier than Foamcore on a straight-material-vs-material basis.

It should be noted that the "monocoque" proscenium module design, for it's relatively large cubic volume,
(compared to a domino module of the same 2D footprint), is amazingly light. This can be simply understood when we realise the majority of the cubic size is air...

Oh, and for reference, suggest re-reading the original "$100 module" article by Keiran Ryan, from which the all-foamcore proscenium module a la "Brooklyn"/"Chicago Fork"/"Toorong" was developed.
http://www.krmodels.com.au/100dollarmod/100dolarmod.html 

Pay particular attention to the sentences directly below the "Conclusion" heading.
To wit: A 6' x 2'6" x 2'6" CD plywood version weighs in at between 35 and 40 kilograms, 
and that's just the naked (no track, scenery, lighting, wiring, or anything) module!!!

- I like the drawings, but I see a number of issues, or at the least "subtle design details which may or may-not perform as expected"

* Why 30" deep? Personally it seems excessive, and I have to admit I've not build a foamcore module deeper than 24" (never needed to, literally no scene I've met that can't be effectively rendered in a 20-22" deep scene
(scene depth + profile spine = total module depth).

Allied question, you nominate 30" deep, but I do not see a dimension for the profile spine?
(NEVER try to run a spine or any foamcore member < 2" continuous thickness).

Keep in mind that the 30" deep dimension materially plays into the weight and leverage effects of the proposed lighting rig (covered later in detail). Suffice to say, the deeper the module + thinning-the-spine + extreme leading-edge weight bending-moment = significant structural stresses to handle.

* The profile curved radii is too large, and there is not-enough sheer vertical backdrop dimension between the base and the curve. This will make intergrating the curved transition and the 3D scenery very difficult, far more than it needs to be. Equally, experience shows that the deeper the module, the tighter the radii can be without visually cramping the scene composition.
(Check the pics of the "$500 layout" HO version of Chicago Fork in the Sept 2013 MRH, VS the pictures of "Brooklyn" and "Toorong" available here on-forum. While the vertical dimension and viewing apeture did not change, the "$500 layout" version of Chicago Fork  at only 6" deep was left with a small tight-radii curve which always looked a little odd)

* The number of profile "C"s VS the sheet lengths and the overall dimension do not match. 
You appear to have 10 Profiles, which results in 9 "bays".

160" / 9 bays = 17.7" per bay

This dimension-per-bay concerns me for two reasons:
1: the optimum span length between profiles seems to wind up consistently around 16", at least when working with 4' modules and 60x40 foamcore sheets. Longer may well be do-able, but longer spans places more emphasis on correct sub-roadbed+roadbed lamination for dynamic load handling without deflection.

2: 17.7" does not appear to be a multiple of either 30 or 40", meaning it's very likely you will not have a profile acting as sheet<> sheet support. This could be materially affected and improved by using larger sheets from the outset (Check GregW66's "Foamcore Module" thread, I'm Very Big on Minimising the number of mid-member joints when using foamcore, or ANY sheet material for that matter!)

* I'm going to say that again, because I believe it is very structurally fundamental to any "sheet material based" design. Make Every Effort to Eliminate mid-member joints. Each and every joint in those 160" "continuous" members is a built-in Weak Point. If you absolutely must cobble what should-be a single-continuous member from a number of pieces/segments, using the largest spice-plates possible, with the maximum overlap surface, is your best approach. (I'm sure GregW's ears are burning right now.... this all sounds familar no? ).

Taking this "maximum overlap" concept to heart, arranging the roof sheets and front+rear stiffeners/fascias such that the joints do not align would materially help strengthen and stiffen the assembly.

* the "fully-enclosed" roof panels are un-necessary, and a waste of material. Sure, they will work great as a full-panel gussett, but an overkill waste of material is still a waste of material. One sheet should be enough to provide both Roof Fascia and Rear Upper stiffening lengths.

* I get and appreciate the box-girder arrays at top-front of the fascia and bottom rear of the base. However again, I fear this is engineering overkill for no real benefit. The weight of the proposed fluro mountings may well mandate some level of beefing-up. If the fluro ballasts are left integrated into the fluro mountings then things may get Very Interesting weight and bending-moment wise. However, from presonal experience with "Nine Mile", "Chicago Fork" (both scales), and "Toorong", either halogens or LEDs are more-than-capable of providing adequate illumination, with far less weight, and much less structural engineering required.

* For backdrop material, simply pickup 1.5 - 2mm matteboard. Enough "spring" to stiffen and support the profiles, thin enough to easily bend and be flexed into-position, available in the same sheet sizes as the foamcore, and takes paint/glue/scenery very well.

* The lower-rear box girder appears to have missed one of the critical details of the design, The Base Backdrop Support is deliberately taller than the base height, and it's Front Face is aligned such that it mates with the Rear face of the backdrop/curve/roof sheet material. failure to engineer this interaction will materially weaken the 90-degree joint between the base and the vertical spine of the profiles.

Now, onto some material cut-list/arrangement comments:

- The way the profile Cs are arranged is not optimal in material usage terms. Try something like this.
(I note you used such a technique on the Plywood cut-diagram, using it on the Foamcore cut-diagram gives back some very useful-dimensioned offcuts for misc supports, gussetting, and integrated storage compartments... ).

- The holes thru the foamcore members are not required for weight saving, see the above weight calcs. Wiring runs can literally be punched thru holes made by your fave screwdriver. Firthur, given the known design detail that "no foamcore member may be cut thru more than 1/2 it's dimension, or < 2", whicever comes first", a hole of the size shown would likely reduce the continuous foamcore cross-section too much.

In short, for the Foamcore version, there is no need for those holes, strongly reccomend removing them from the design.

- As above, the whole-sheet roof panels are overkill. That foamcore could be far-better-used as laminated sub/roadbed. (2x 3/16" foamcore thicknesses, laminated together _properly_, can easily support typical HO and On30 loadings).

Now, construction tips:

- Low-temp "craft grade" Hotglue is your reccomended go-to adhesive for all C, L, T, gussett, and short quick splice-plate lap-joint missions.

It's fast-acting, reliable, gap-filling, and creates bonds which can easily survive a "bounce" from 4" feet.
(ask me how I know this... ).

- Believe it's been documented here onlist previously, but a long ruler + trusty X-acto knife and LOTS of NEW blades are your go-to marking and cutting tools. 

- Use MRH search with terms "Foamcore cutting order of operation" for an existing how-to on getting confdent straight reliable foamcore cuts, first-time, every time.

- PVA/"white wood-working glue" is fine, perfect even, for laminating subroadbed and roadbed. 
(IE super-thin skreed layers between large laminated surface<> surface contact area). However, while it would work after an in-ordinately long working time, PVA has no real place for C/L/T and gussett joints.

- Get yourself a Great Planes Balsa Stripper, it is invaluable for cutting accurate tongue-in-groove/dovetail/lap joints in "inline", C, L, and T joints between members. It also allows strategic cutting thru specific-layers of multi-layer-laminate foamcore assemblies. 
EG I frequently laminate two thicknesses of foamcore together to form the module "landform base"/subroadbed surface, and then use the Balsa-Stripper to remove a 5mm strip around the outside of the lower layer only. This enables the resulting subroadbed laminated assembly to sit down inside the module base-members, making for an instantly-stronger assembly.

With all of the above having been said, I know Ken that you are a "man of action".
SO, with that in mind,

download This Document
(Yes, the words "This Document" are a clickable link to a PDF file)

print it out,
(ensure you set your printer to "NO RESIZE"!!!)

glue it to a sheet of cereal-box cardboard

cut out the components
(making sure to snip the noted "interleaved tab-in-slot" slots)

and build yourself a 1:10th scale 2x4x2 proscenium module to "get comfy with the concepts"...

Happy Modelling,
Aim to Improve,
Prof Klyzlr

PS With the exception of the extreme end-profiles, 
(which I've cut from MDF if required), foamcore is more than strong enough to handle bumps/knocks/jolts along the base and roof Fascia and end-panel positions, particularly if it is faced with automotive carpet...
(see pics elsewhere on MRH of my last show layout, "Toorong"). As the system is sitting on existing support structure, there is no obvious need to accomodate screws or other fixings.

Other Lightweight Materials

I saw your post about alternate light weight layout construction materials and this is something I have been testing for constructing Free-mo modules.  One of the biggest issues with wood is that it is affected by both temperature and humidity changes, so if we could use a material that is not affected by either that would be ideal.  

The aerospace industry has been using composite honeycomb sandwich panels for years, (probably 30+) for interior fabrication such as floors, cabinets, tables, and other places where added stiffness is required, such as bulkheads.  So why not use those methods for module construction.  Honeycomb panels have the strength like plywood and the weight of foam.

Where does one get composite honeycomb sandwich panels and how much do they cost?  This is one source: http://www.acpsales.com/Sandwich-Panels.html

They are not cheap, but you figure you could likely build 2 or 3 modules from one 4x8' sheet.

I was able to obtain some 1/2" thick phenolic material for testing purposes and while I have not constructed a module with it yet I plan to in the near future.  I did test it making a couple of 3" x12" x .5" coupons, which I bonded as 90 degree corners using two adhesives.  Gorilla Glue and 5 minute epoxy, 5 minute epoxy was best.  You could also use a fancy aerospace structural adhesive such as those made by 3M, Loctite, or Hysol, but you're going to pay $50-100 for a pint. 

You cannot cut or assemble this material using traditional cutting tools nor fasteners, so be careful and follow these tips.  You can however use some potted(bonded) threaded inserts if you must use fasteners, I haven't decided if this will be required or not.

***Wear a dustmask!  Breathing fiberglass dust is bad!  Wear gloves!!!  Fiberglass is sharp, and fiberglass slivers are not fun!

1. You will need a diamond coated saw blade, i got a cheap one from Lowe's which worked great in a circular saw.  If you try to use a regular steel blade, you may get 3 or 4 cuts before the blade will be useless.  Fiberglass is very abrasive.

2. Abrade all surfaces to be joined with 80 grit sand paper. Then clean with acetone or IPA.

3. Apply adhesive to both surfaces, and clamp them together to cure.  Be sure to wear latex or similar gloves while apply adhesive.

4. The open end of honeycomb can be filled with a filler such as Bondo to make it more cosmetic.  Or I have been thinking about using fiberglass .005 or .010" sheets as the fascia material.  

Let me know if you have any questions.

I've already received the "you're crazy" look from some fellow modelers while others feel it sounds ilke a great idea to test.  The way I see it is if we don't look a new methods how will we ever know if there is something better or not?  

Dan R.

@ Ken

I have to agree to with the Professor.  As I was looking at the design and keeping in mind your quest for lightweight I stumbled on the part about fluorescent tube lighting!  It has always been my experience that those lighting fixtures are not lightweight. The position you drew the lights in would make the whole thing top heavy and trying to fall toward the viewer.  So, there would always be stress there.

Get rid of the heavy fixture and design in for LED strip lighting.  There have been many discussions on the forum about  LED lighting.  

By the way, I like the effort you put into the design .  Once you get the kinks worked out I might have to try that.

Den

"Factor in: - Typical

  Hi Prof,  I read it as he wants a light layout of show quality(whatever that means?) but that he plans to set it on some existing furniture and only move if necessary so I don't see that making it a touring layout would be the most efficient use of time and materials?  My thought is that extreme light weight is not the first thing that should be worried about as normal layout sections are relatively light compared to everyday household goods that we move around. If you need to move a section you go ask the neighbor kid to give you a hand, or ask a friend from work to stop by on the way home. Likely the layout will just sit there in the same place for years and the extra weight will be a positive adding stability and ruggedness. Since I have other options I've never built a hollow core door layout but the guys who do tell me they thread feeders thru with a stiff wire like they do thru foam sheets. I guess the small diameter holes don't hurt the skin enough to cause trouble. I guess one could add a couple of bottom stringers to create a wiring basement, then add a fascia to cover the foam ,door,and stringer face. Lighting and backdrop would most efficiently be attached to the wall behind the doorminos , something like Ian Rice shows in his small layout book. Basically stick up a backdrop on the wall,set the layout in place, add a lighting valence supported by cheap shelf brackets and forget about it. If you ever move you reverse the process and take the pieces to the new house and set them up again. The great bulk of time and money goes into everything on a layout other than the benchwork so I wouldn't worry so much about the cost or spend any extra time on the construction as so much more lies ahead to be done. I do worry about excess weight when building my modules but only in the sense of don't use a 2X4 when a 1X3 will do the job, and the result is sections up to about seven feet long that are easy for me to move around when needed.Some of the sections I'm arranging for my new CCT layout are on their third or fourth layout now.....Dave

One long reply

Absolutely!  And I see that I have quite a bit already.  I was especially hoping to get some good feedback from you, Prof, and you have not disappointed me.  I will address a few questions, and take what you've said to heart.

I can get them for $10-$15 apiece, and my original design included them.  However, Prof has pointed out most of the issues with hollow core doors, and those issues have made me re-think my original plan of using them.  It's not out of the question, but I suspect there may be a better way.

That sounds like an excellent idea, and one I will consider.  I know I can make a wooden beam 160" long which will be plenty strong enough, so replacing the two box sections with wood might be the simplest way to make the layout plenty strong.

No worries, Prof.  I put a lot of time into my designs to help ensure that I am satisfied with them long term.  Any advice or experience you wish to share is greatly appreciated.

The layout I wish to build will be 24" deep.  While it does not absolutely have to be that deep, it turns out that I can just fit the prototypical track arrangement and a complete depot into 24" without any compression.  That appeals to me a lot, and since I do not intend to transport the layout on a regular basis, I feel comfortable trading a little convenience to get room for my prototype.  Also, using 30" wide material, 30" seems like a convenient size.

That's another reason for the 30" depth.  With 24" of layout and another 1.5" of space behind for backdrop lighting, that leaves the thinnest portion of the spine 4.5" deep.

Honestly, the curve was just a guess, based on the goal of getting the best appearance and reducing bending stress in the backdrop material.  Also, since I propose to stop the scenery 1.5" in front of the backdrop (and presumably have a short profile backdrop attached to the scenery), I thought that it would leave a little more room for the curve.  That said, I will re-evaluate the curve.  One of the goals was to keep the backdrop material 24" wide (easily cut out of a 4' wide panel), so tightening the radius will bring the top of the backdrop toward the back, and allow me to move the lights back as well - a desirable thing.

That's the reason I included a longitudinal brace between each profile, beneath the roadbed.  The result is a 17.5" x 13" open space.

I was actually thinking that a joint over a support would me a weaker point, since the foam is so thin that it would only support a very narrow strip of the sheet above.  Much like the logic which suggests putting rail joints between ties instead of over them.  Obviously, I have only theory, so I will defer to one with actual experience.

My thought is that a suitable splice plate properly glued with wood glue would be as strong as the rest of the material, and possibly stronger.  Again, I respect real-world experience, but I wonder if it is possible to eliminate that weakness?

I do have a few pieces which span those joints, and was thinking that the splice plates would do the rest of the job.  I will consider re-arranging the joints, though.

From a structural standpoint, yet.  However, I know that sooner or later, something will be set on top of the layout, so a fully enclosed top seems like a smart move for a relatively low cost.  Also, I have cats in the house, so keeping dust and hair out of the layout and as easy to clean up as possible seems like a good goal.

I thought about that as well.  Given that there's room elsewhere in the design, I would probably relocate the ballasts to reduce weight and possibly save space as well.

I am considering LEDs as an option, but I wanted to keep my options open.  I would hate to learn after it's too late that the only way to get adequate light from an LED is to spend a lot more money, when I could have designed the layout to accept an inexpensive fluorescent fixture the first time around.

I wasn't sure, and it's easy to not cut them out.

That's both interesting and useful!  I would not have thought of hot glue, but I can definitely see how it would have a number of advantages.  I like the speed and price, too.

I'm afraid I don't really understand what you're saying here.  My idea was to cut the profile Cs out of a solid piece, and then glue the backdrop in place (with the backdrop constrained top & bottom, so that it must fit into the desired curve).  Thus the backdrop becomes a structural member as well, and any residual tendency to flatten itself out would try to raise the center of the front beam.  Then the track base can be glued in, and glued to the backdrop so that it stiffens the lower straight edge.  Have I missed something?

"I know I can make a wooden

Composite construction would be my first choice but I wouldn't make the beam any longer than the modules, say 6' to 8 foot max length. The house floor serves as your beam once the sections are installed on their bases so making the beam longer just makes it harder to carry.  If I was building this layout first I'd consider a separate base with backdrop wall mounted and lighting valence wall hung to take advantage of the house wall strength. or if it had to be one self contained unit  I'd do something like this sketch with wooden front and rear beams and cross supports for backdrop and lighting spaced about 24 inches or less .The layout roadbed and scenic base could be any light material you prefer and wouldn't need to be any stronger than necessary to hold up the trains and scenery. I forgot to show a top covering but it could just be hardboard stapled on top to keep out the dust....DaveB ​

lost feeders

A long time ago I built...

something like Ken has shown, but with cabinet grade plywood that made a more perfect C on the inside.  I even made some Es for a dual deck.  It was nice. It wasn't too heavy.  It was designed to use shelving brackets. You should have seen the corner modules!  Those where great!  Sorry, it was so long ago I wouldn't have any photos of it.

Den

Waffle Bottom and Questions

@Ken - have you considered a waffle bottom al la the Sipping and Switching Society  http://s-ss3.home.mindspring.com/id1.html ?  Seems like that would be a natural for a foam board module and although it will use a bit more material it would provide a torsionally strong base without building the box you show in your drawing.

The foam board proscenium design would make it easy to lay the module on its back or top to do underneath wiring.  And you will note that in the Toorong/ Coramba module the Prof notched out the base for a river and bridge.  This is a very flexible design.  

The sign supply distributor, Regional Supply in Salt Lake, that sells 4' x 8' styrene also has 3/16" x 4' x 8' Fome-Cor brand foam board for $14.01 per sheet.  Lends itself nicely to a 4 ft module or a seamless 8 ft module if you have a way to transport it.

@Prof, a couple questions:

Qubelok seems to be an Australian product.  Are you (or is anyone) aware of a similar product in North America?  How do you attach the foam board to the Qubelok frame?  Also it appears from the photos in your earlier reply that you use Qubelok for your 'A' frame legs.

What material do you use for the finish fascia on the Toorong display?

Do you use the DAEX25 exciters in your Toorong display?  If so where are they mounted?  Do you typically mount exciters to the bottom of the roadbed surface, and if so what effect does track and scenery have on the sound?  Seems that mounting exciters on the backdrop would be a natural but I wonder how matte board would hold up over time?  Perhaps with 'lean-in volume' it is a non-issue.  Some .040" styrene might be a viable alternative.  Would love to hear more discussion on exciter placement and suggested materials for the sound board.  I really like the idea of hidden speakers.

Regards,

Dave

Evaluating the Curve

Dear Ken,

Some pics for your consideration:

First, a historically-critical image of the cut-out from between the profile "C"s from Toorong. The scribbled dimensions are "flexible", but are quite close to my "go to" values for most missions
(all dimensions in mm)

Second, an end-one shot of some 4x1 modules I built back in 2009 for demo purposes.

NOTE: 
- The "sky" is deliberately raked UP,
or said another way, the backdrop<> sky curve is < 90-degrees
- these dimensions could still work for a deeper (20+" scene) as-is,
the leading-edge of the roof structure would simply end up "thinner" the furthur it projects forward

...and some more misc pics of these early demo "Heavy-duty-touring" grade modules FYI...

Something that also "tweaked me" about your design, which didn't fully-become-clear in mind until yesterday, you appear to be using lots of small/short components, rather than tabbing-and-slotting the "strips" to the "profiles".

It is this tab-n-slot/interleaved-slot technique,
along with minimising joints (using longer sheet as the starting material),
that gives the assembly a significant amount of it's strength.

Indeed, at clinics I've "slotted together" a module from it's component parts,
and with a small piece of gaffa-tap at each corner,
(where the joints are "L" dovetails, not inherrently-self-holding tab-in-slot "T" or "X" joints),
proceeded to lift and swing it above my head, with no adhesive in play.

The point being, with your existing "lots of little pieces" config, you'll be relying almost wholely-and-solely on the adhesive. Adjust the design to more closely reflect the original KRModels-style with tab-in-slot/interleaved-slot construction, and the entire assembly will be much more inherrently structurally solid and rigid.

It may not make a lot of sense in text, so I again recommend you relook at the original KRModels site and associated diagrams, and download and assemble the 1/10th scale version linked to earlier in this thread.
(It's a one-night project, needing only an A4 sized piece of cereal box cardboard,
and the "actually building it" process will help it make sense...)

Happy Modelling,
Aim to Improve,
Prof Klyzlr

Backdrops, supporting pieces, and how-it-fits together

Dear Ken,

Um, yes, there does appear to be an air-gap in proceedings.

Please download the 1/10th size example linked above, and look closely at the rear of the profile "C".

Now, per the instructions on the PDF download, the RED lines are "slots" for some other component's "tab".

NB the Slot coming UP from the base,
NB the short slot coming DOWN from the vertical part of the profile, into the base.

NB that they are NOT aligned with each other! This is NOT a drawing mistake, but a key detail!

The key is that:

- The Backdrop Lower Support member will be => the height of the base
- Thus it will end up protruding above the subroadbed surface, with the front-face flush with the Profile vertical edge (Purple hand-drawn line below)

- The backdrop material (Green hand-drawn line) will lie flat against the profiles, with it's lower-edge anchored in the slot shown (Hint: this slot also allows you to cut the backdrop sheet material slightly over-size, it will absorb the excess once the sheet is curved and snapped-into-place).

- The resulting overlap between the Lower Rear backdrop Support member (foamcore) and the backdrop material (matteboard) will ensure the backdrop lower edge stays flat, and doesn't "bow" or "gape" backwards between each pair of profiles

(The gap shown between Purple and Green lines is due to dodgy Photoshop drawing on my part,
in actual assembly these 2 sheet-pieces will be lying flat against each other...)

Does that make any more sense?

Happy Modelling,
Aim to Improve,
Prof Klyzlr

Qubelok, legs, and exciters

Dear Dave,

In order of appearance:

- "Qubelok" is the product name of the joiner system.

http://www.capral.com.au/Qubelok-Accessories

It is manufactured by Capral, and integrates with standard 1" aluminium tubing.

http://www.capral.com.au/Qubelok-Extrusion

A competing product called "Connect-it" is manufactured by RCR

http://www.rcr.com.au/metalmate/metal-mate-diy-connect-it-system

The closest thing I've found thus-far in the US is by Brunner Enterprises

http://www.brunnerent.com/Tools/Portfolio/frontend/itemlist.asp?type=2&size=0&lngDisplay=2&strMetaTag

although I'd be Very Very interested to hear of any other similar US-based product lines...

- "Toorong" was the first layout I've used A-frame leg systems on, and Qubelok seemed to be a natural fit. The resulting legs pack flatter, lighter, and smaller in the transportation vehicle than equivalent pine or similar wood A-frames. 

Admitedly, I didn't like A-frames until the "Toorong" build, I'd seen too many show layouts with poorly engineered and constructed (wood and steel) A-frames. Weight, packed-for-transport size, and ability to handle typically uneven show floors were all considerations.

My previous layouts ("Brooklyn : 3AM", "Nine Mile", "Yallah2"), used collapsible trestle-table legs, mounted to a narrow plank. with the module sitting and indexing onto the "flat-pack" support. Leg extensions C/O MR-esque electrical conduit pipe with inner wood dowel core for rigidity. Carried like a briefcase, and very stable.

That said, I'm all about challenging my own assumptions, and the Qubelok A-frames have answered most-if-not-all of my concerns about A-frame legs, and proven very "road ready"

NB:
- chain not shown for clarity
- continuous threaded-rod pivot + aluminium tube + nylocs replaced with 2x bolts + nylocs
(cheaper, easier, lighter)

As designed, with a Foamcore module having 150mm tall base-section,
results in 1200mm (4') tall A-frame and 1350mm (4' 6") floor<> track display height.

RE Layout Fascia treatments

I have finished all my layouts since 1995 with Charcoal grey car carpet. This gives:
- a nice "warm and fuzzy" feel to the layout,
- presents it well with an overall darker "picture frame" effect as one would expect from a Proscenium system,

and protects the layout from
- significant impacts, bumps, continuous rubbing-up-against by punters,
- rubbing between modules in the vehicle whilst in transit
- and even "piercing projectile" attacks.

I have to admit I really really really really really want to build a layout someday with the opportunity for "rusted riveted iron" fascia a la Tim Warris's "Bronx Terminal". I have seen various textured and "rust effect" paints at my local hardware store, and am sure there has to be an appropriate application for them in a show-layout build! I'm also not afraid of injecting a bit of "period weathering" into layout signage and similar if the layout theme warrants it...

- "Broughton Vale Tramway", being an Aussie NG Logging layout, use a beaten-up piece of rusted corriugated iron as it's signage. All new steelwork was "weathered" with muriatic acid to provoke real-rust patina.

- "Nine Mile" used a series of planks from one of the show-crew's rotten backyard fence, assembled on a aluminium L-channel frame, and hung with a genuine 6' cross-cut saw above the layout. Again, all "fresh and shiny" steelwork using to build the sign got the muriatic acid treatment...

- and of course, modern-era "Brooklyn : 3AM" just wouldn;t be NYC without an appropriately spray-painted (on foamcore! ) sign

However, (says Prof, recognizing that he's wandered a looooooong way OT),
my fear is that any such "solid patina" layout fascia will actually not handle the kind of rubbing/buffing/scraping/scratching that a show layout frequeenctly is subjected to as well. If you have to keep "patching up" the layout fascia, then that's a hint that all is not well...
(although in the Tim Warris "rusty tetanus" case, such "new patches" may well visually work OK...

It should also be noted that used car carpet allows layout skirting, "Can you find?", and info boards to be attached directly to the layout with velcro. Fast and easy setup wins the points yet again...

In the case of all previous layout, the Foamcore "core" was faced with thin MDF, over which the carpet was applied. (Do you still "lay" carpet if it's applied to a vertical layout fascia?!?!?!?)

However, in the case of "Toorong",
building on top of the previous "Chicago Fork" experiences and "Light Touring" foamcore ethos
(as opposed to "Heavy Duty Touring")

the layout fascia was formed by a 2nd-skin layer of foamcore, mounted over the core "Foamcore module",
and the carpet was applied directly over the top
(IE no heavy-grade ply or MDF layer between the carpet and the module-proper).

RE DAEX Exciters

No, unfortunately the build schedule for "Toorong" was too tight to include proper LayoutSound deployment,
(Yes, I am very disappointed with myself, and have given myself a good talking-to about this major oversight...)

However, I have used DAEX25s, mounted to either/or/both the backdrop (1.5 - 2mm matteboard) and the sub-roadbed (multiple thicknesses of 5mm Foamcore), and powered with a small 12V car power-amp, for both "ambient" and "spot FX" deployments. 

I have also used exactly the same exciters and mounting-technique under standalone structure-demo foamcore modules.
(For the astute, any guesses how the funky elipse-shaped module shown below came about?
HINT! You've seen that shape cut out of foamcore earlier in this thread!!!).

Respectfully, you're approaching the problem from the wrong end. Even the exciters webpages and instructions make the point "Exciters, unlike traditional speakers, are largely dependent on the mounting and material to achieve a given audio specification performance".

IE do not think "mount the exciter somewhere arbitrarily,
and then deal with whatever unexpected co-lateral effects the additional of track/ballast/scenery/structures may cause",

rather, get the resonant structure reasonably complete, 
start playing some familar music thru the exciters,
and then move them around on the surface until you get a suitable aural reproduction result...
(Funnily enough, this tends to hold true for traditional cone-type speakers being mounted up thru the benchwork too.... )

That exciters turn the entire structure into a speaker "cone" makes the game of "locate the layout-sound speaker" almost impossible to win, the sound is coming from everywhere and there is no really-tight "point-source" on which the layout viewer/listener can focus-on.

For me, I'll use either "foamcore base" or "matteboard backdrop" mounting as the layout project leads. IE either approach can work and "make a noise", but it's your ears which will tell you plainly if one approach "works better for this scene/layout/module" than the other. That said, I haven;t yet seen any sign of matteboard backdrops somehow "structurally failing" under constant Exciter vibration, and that includes when I've been testing using Metallica, SoundGarden, and various flavours of "Doof Doof"...

Hope this helps...

Happy Modelling,
Aim to Improve,
Prof Klyzlr

Thanks Prof Klyzlr

 Yes, it helps immensely.   I appreciate the experience, knowledge and wisdom that you share with others.  I have learned a great deal about a how to implement a light weight module and some of the extra things that can really enhance a project.

With deepest gratitude,

Dave

Re-thinking

Thanks for all the comments, guys.  With the advice given here, I am re-thinking the logic behind using foamcore.  I know, without any doubt, that I can build a more conventional wooden structure which would be more than strong enough.  I can probably do it cheaper as well, by carefully and intelligently using dimensional lumber.  At the very least, that would be within my existing skill set.

With all that in mind, I have just about made up my mind to build a pair of 80" long wooden modules, each with the base, back, and top in a solid piece.  I am also looking at LED lights, as I think I can get the job done more easily there.  Of course, everything is still up for debate at this point, but it's looking like that may be the most practical option.  After all, theory is great, but the ultimate goal is to build a layout which I know will work and which I will not have to worry about in the future.

Re Rethinking

Hi Ken,

At the outset you said weight was going to be a key issue...

So how are you going to save weight in a material that weighs in a 5x heavier than foamcore by unit volume? Where are the design changes that you will have to make to get that type of a weight saving? If the structural plans are going to be changed to get that type of weight saving, can the revised structure be built in foamcore? If not why not?

Remember here that a layout for home use ony has to support the weight of the rolling stock, track, scenery and lighting rigs, It does not need to be "ruggedised" for show use, where repeated set-ups and tear-down cycles have to be designed for. 

The LED lighting strips offer significant weight saving and you can get their heaviest component, the powerpacks, down below track level, thus minimising any loading on the valence. This load above track level should only come in at only a few pounds per square foot unless you are modelling the Rockies with real rock. This light type of a loading does not need a massive superstructure to support it, This is really where the foamcore slot-and-tab structure comes into its own.  

Yes I have a foamcore layout, It was built in 2005 by Professor Klyzlr and myself. It is still dimensionally stable, has had almost zero work done on it since commissioning. It has a fully enclosed top shelf on the valence made from 2 layers of foamcore laminated together ,  This has supported a lot of junk on top of the layout over the years without distortion.  For a 30"' span on the upper shelf you might have to fit some additional cross bracing above the lights and maybe go to a 3-layer foamcore laminate shelf. I have seen a 3 layer laminate shelf support a 10lb O scale locomotive with zero deflection.

I know it can work. Perhaps another re-think?