I have built a few layouts using the box frame method and my current layout is my first L-girder experience. I don't know why you thought that the L-girders had to be perfect but they don't. Its all about using less material and wieght but keep the rigidity to better insure trackwork reliability. I use 12" centers for my supports and have no problems. I used 5/8" plywood for subroadbed and there is a support under all joints. L-girder is also very flexible, you can add supports just about anywhere.
John
ereimer wrote:there's no doubt the L girder method works well , and for a larger layout the wood saved by using L girders and cookie cutter subroadbed must be huge . i'm not convinced it makes so much difference on a 4x8 or smaller .
Having just built the benchwork for my sons 4X8 I would have to say it makes a difference when compared to using the box grid even on 4X8. I used less framing material, it's so much easier to put together and it's very rigid. On a large layout the difference would be huge as you mentioned.
I like domino construction as it allows my layout to grow and can easily be modified. I am however considering using L-girders to set my dominos on. I think this would be better in the long run versus constructing legs and braces for each domino.
JIM
Jim, Modeling the Kansas City Southern Lines in HO scale.
Bought Westcott's book years ago, but never built a L girder layout until I started the new one a couple of years back. Boy, am I impressed. All the stuff said here in support of the system is right on the mark. It's way easier than any type of box construction, uses less material, and is light weight. The flexability of the system has to be experienced to be believed. I've found that I've been able to make major changes in track routing, etc with ease. In the past I would have avoided doing this because of the difficulty in altering what I already had built.
As far as carpentry skills, it looks more complex than it really is. As others have noted the girders can be made by mating good edge to good edge, so if you goof on the ripping it doesn't matter. It's extremely forgiving. Warping of subroadbed has not been a problem. (Westcott gives distances between supports for various thicknesses of plywood. I used his recommendations) My only regret is that I didn't start using this system years ago
JBB
i suspect that L girder construction was used on the Troy Branch either to demonstrate how to do it for the article or because that's what the author always uses . looking at what's being built makes me think a flat piece of plywood attached to a box shape would have worked just fine . or a piece of 2" foam would be as good and give you the option of sculpting out some below grade scenery
there's no doubt the L girder method works well , and for a larger layout the wood saved by using L girders and cookie cutter subroadbed must be huge . i'm not convinced it makes so much difference on a 4x8 or smaller .
bogp40 wrote:Fred,Much of my reply was in reference to people using 1/2" CDX. You are using a decent grade of plywood. 5 ply AC cannot be compared to CDX by any means, especially today with all the inferior grades of crap that is forced on unsuspecting buyers. The 3 and 4 ply pine instead of fir sold by them is my main reason for the reply. There is still quality fir 5 ply CDX, you won't find this at Home center though. This better grade of plywood is still no comparison to the better quality found some years ago. An instance of this: today I actually salvaged the 5 ply fir CDX on a remodel and install of an Andersen slider. Instead of having to go to my supplier and PU a sheet of CDX, this 45 year old plywood was used to fill in the sheathing needed. Close to full 1/2" in dimension, stable and actually better than what I can buy today. The only thin wrong was 45 years of aging and discoloration.Also, any of these lesser quality plys will not retain any of it's integredy when cutting a radius and those surface grains become crosscut.
Fred,
Much of my reply was in reference to people using 1/2" CDX. You are using a decent grade of plywood. 5 ply AC cannot be compared to CDX by any means, especially today with all the inferior grades of crap that is forced on unsuspecting buyers. The 3 and 4 ply pine instead of fir sold by them is my main reason for the reply. There is still quality fir 5 ply CDX, you won't find this at Home center though. This better grade of plywood is still no comparison to the better quality found some years ago.
An instance of this: today I actually salvaged the 5 ply fir CDX on a remodel and install of an Andersen slider. Instead of having to go to my supplier and PU a sheet of CDX, this 45 year old plywood was used to fill in the sheathing needed. Close to full 1/2" in dimension, stable and actually better than what I can buy today. The only thin wrong was 45 years of aging and discoloration.
Also, any of these lesser quality plys will not retain any of it's integredy when cutting a radius and those surface grains become crosscut.
I agree with the above. I would not use CDX, nor anything less than 5 ply. If nothing else, there are too many unsupported voids when you cut it into ribbon subroadbed.
I may try 3 ply 1/4" luaun plywood on the layout in planning, but only if reinforced by laminating to 1" or thicker extruded foam. Or supported about every 8" or so.
Fred W
fwright wrote: bogp40 wrote: ....In cookie cutter construction a 3 to 4" strip of 1/2" plywood is not that strong on 16" centers. The use of 3/4" would be a better choice. This is especially true when cutting out the radiuses from the ply. I would recommend that you don't allow the surface ply graining pass a 45 degree through the cut. Reorient and make a splice. Many will say that plywood doesn't have grain, the material is strongest when the outer plys are 90 degrees to the support. Try cutting a thin strip against grain and see how weak it is.My experience doesn't square with what is cited. I have used 1/2" AB 5 ply plywood, generally cut about 3" wide, supported on 16"-18" spaced joists, as recommended in Westcott's book. I didn't bother re-orienting the joists so that the support on my 18" radius curves was even further apart. Sagging was never an issue. What was an issue was not having the risers and cleats spaced closely enough to take out the twist where the curve was on a grade. On a 180 degree 18" radius curve on a 4% grade, I had some significant negative superelevation on the 1st half of the curve and positive on the 2nd half because of the rigidity of the plywood between the supports. Extra risers and significant force had to be put in to twist the plywood back to level across the track all the way around the curve.just my experiencesFred W
bogp40 wrote: ....In cookie cutter construction a 3 to 4" strip of 1/2" plywood is not that strong on 16" centers. The use of 3/4" would be a better choice. This is especially true when cutting out the radiuses from the ply. I would recommend that you don't allow the surface ply graining pass a 45 degree through the cut. Reorient and make a splice. Many will say that plywood doesn't have grain, the material is strongest when the outer plys are 90 degrees to the support. Try cutting a thin strip against grain and see how weak it is.
....In cookie cutter construction a 3 to 4" strip of 1/2" plywood is not that strong on 16" centers. The use of 3/4" would be a better choice. This is especially true when cutting out the radiuses from the ply. I would recommend that you don't allow the surface ply graining pass a 45 degree through the cut. Reorient and make a splice. Many will say that plywood doesn't have grain, the material is strongest when the outer plys are 90 degrees to the support. Try cutting a thin strip against grain and see how weak it is.
My experience doesn't square with what is cited. I have used 1/2" AB 5 ply plywood, generally cut about 3" wide, supported on 16"-18" spaced joists, as recommended in Westcott's book. I didn't bother re-orienting the joists so that the support on my 18" radius curves was even further apart. Sagging was never an issue. What was an issue was not having the risers and cleats spaced closely enough to take out the twist where the curve was on a grade. On a 180 degree 18" radius curve on a 4% grade, I had some significant negative superelevation on the 1st half of the curve and positive on the 2nd half because of the rigidity of the plywood between the supports. Extra risers and significant force had to be put in to twist the plywood back to level across the track all the way around the curve.
just my experiences
Modeling B&O- Chessie Bob K. www.ssmrc.org
bogp40 wrote:....In cookie cutter construction a 3 to 4" strip of 1/2" plywood is not that strong on 16" centers. The use of 3/4" would be a better choice. This is especially true when cutting out the radiuses from the ply. I would recommend that you don't allow the surface ply graining pass a 45 degree through the cut. Reorient and make a splice. Many will say that plywood doesn't have grain, the material is strongest when the outer plys are 90 degrees to the support. Try cutting a thin strip against grain and see how weak it is.
Doug T wrote: The floor joists in your house have 16 inch spacing and floor covering of at least 3/4 to 7/8 of an inch. If you go up into your attic you will see that your roof on your house has either 1/2 plywood or 1/2 oriented strand board (OSB). The exact measurement may be 7/16 and not 1/2. The roof is supported by rafters every 16 or 24 inches. People do walk on the roof to put down the shingles. The roof does not collapse. My question to you is, How much weight do you plan to put in a 16 inch square? Are you planning to crawl up onto the layout?
The floor joists in your house have 16 inch spacing and floor covering of at least 3/4 to 7/8 of an inch.
If you go up into your attic you will see that your roof on your house has either 1/2 plywood or 1/2 oriented strand board (OSB). The exact measurement may be 7/16 and not 1/2. The roof is supported by rafters every 16 or 24 inches. People do walk on the roof to put down the shingles. The roof does not collapse.
My question to you is, How much weight do you plan to put in a 16 inch square? Are you planning to crawl up onto the layout?
The question as to the strength and load plywood takes for the 1/2" roof sheathing is, the entire 4x8 panel is attached 16" on center. Yes if you deck the entire layout surface, then you can easily get away with a decent quality of 1/2" CDX. In cookie cutter construction a 3 to 4" strip of 1/2" plywood is not that strong on 16" centers. The use of 3/4" would be a better choice. This is especially true when cutting out the radiuses from the ply. I would recommend that you don't allow the surface ply graining pass a 45 degree through the cut. Reorient and make a splice. Many will say that plywood doesn't have grain, the material is strongest when the outer plys are 90 degrees to the support. Try cutting a thin strip against grain and see how weak it is.
One thing about L girders that nobody has mentioned yet - the short, not particularly heavily loaded joists can be made of ANYTHING! Planks split from larger planks, ragged edge up. Left-over lengths of whatever from remodeling projects. Odd lengths of 2x4. Even steel studs (my preferred material.)
Likewise, the risers can be made from anything thin enough to allow it to be screwed to the side of a joist (or the flat side of an L girder.) They don't have to be any exact length; long enough to reach (close to) the bottom of the joist is fine. If too long, cut off the excess with a saber saw if it annoys your sense of aesthetics (or is long enough to make another riser somewhere else.)
There is always some kind of screwing flange on the top end of the riser, since the roadbed is supposed to be anchored from the bottom up. Westcott's method drove all screws either upward or sideways, never downward - and especially never downward through roadbed and/or subgrade, where the screw could end up buried in scenic cover (or, worse yet, under the track itself.) I like to make risers from pieces of steel stud material. All it takes is a little creative work with tin snips and bending pliers.
The roadbed is expected to be cookie-cut plywood or equivalent. Large flat surfaces can be supported by multiple risers, but leveling them could become an issue. IIRC, the yard on Linn Westcott's home layout had risers connected at the tops with extended cleats.
Where L-girder benchwork really shines is when you want to reproduce a line clinging to the side of a cliff, with really deep gorges dropping well below the bottoms of the joists at the fascia line. Some of my scenery (like its prototype) will resemble that.
Chuck (modeling Central Japan in September, 1964)
Amen on the L-girders. I have used them now on two layouts and I just love how easy it is for a klutz like me to produce fast and easy benchwork. As the above posts have said, changes in elevation or post-building corrections are very easy to make as well. I say unless you're building a portable layout, L-girder construction is the way to go. In addition, the shape of your subroadbed or table can be any shape or size and the joists (and risers, if necessary) can go any place to accommodate any scenic features and/or track plan.
A word of advice, get yourself some good clamps if you don't already have them. Construction will go much smoother, especially when building the girders and assembling the benchwork.
16" would be about right for 3/4" plywood from what I have read. Its what I bought for the next layout. Maybe some of the others with larger layouts can comment with more direct experience.. thats the distance I plan to use on average on my next layout for the rather pragmatic reason that it will more easily let a power drill between the joists then my current 12" spacing. Linn Wescott's book recommended a maximum of no more then 24" I think.
When you are supporting a larger area like a yard or town you can use more risers or use wider risers as needed. Whatever you need.
You will run into the same consideration if you are using butt-joint construction. You still have some distance between joists that is supporting a sub-roadbed of some kind.
Chris
Thank you, all.
Your points are all well-taken. I still have concern, though, that the plywood sub-roadbed would tend to want to sag between the small "footprints" created by the 1x4 risers, wouldn't it? Isn't that why floor joists in homes are placed approximately every 16" apart?
First off L girder construction is nothing new. Linn Westcott (former editor of M.R.) was the man behind it's use back in the 1960s or 1950s. He put together a soft covered book on how to build it and how to use it. That book or a similar book I believe is still available from Kalmbach.
I will attempt to tell you how I use L girder construction. I use the lumberyard dimensional lumber. I try to buy only straight boards and get lumber with 4 sides having a 90 degree edge. I use 1/2 inch plywood or CDX as my subsurface. I then use 1/2 homosote for my top surface. I place my support stringers for the plywood and homosote on top of the L girders and space them every 12 inches. There is no sag!!!!!!!! I am using this type of construction on my G scale train layout in my basement right now. Prior to this I used it for HO layout.
Have you built a layout and then wanted to make a change to a portion of it? Once the old layout was built and sceniced all the screws you used were covered over. Trying to locate these screws to undo what you had done can be a bit of a headache.
Using L girder construction all of the screws are driven in from the "underside" of the layout. If you decide to make a change and add a something new to your layout. It is just a matter of removing any buildings and track and unscrewing the section from the underside of the layout.
L girder construction is cheaper to build and lighter in weight. I have been using this type of construction for about 30 years. I use to build my benchwork similar to what you are doing.
Having the above mentioned book explains and answers a lot of questions regarding L girder construction. The article you are referring to does not go into a lot of detail as to the reasons for using this type of construction.
If you were to build one 8 ft long section of L girder construction, you would see how easy it really is. You mention the stringers and keeping their length the same. You simply cut as many as you need. Then you place a stringer at each end, making sure they are set the same. You then could run a string along the front edge the length of the L girder section from one stringer to the other stringer. I just eye ball mine when I set them in place.
I do not or have not ever worked for Kalmbach. I did see their old layout back in 1991 during a regional open house at The Silver Rails to Milwaukee convention.
Jeff But it's a dry heat!
I used the "box" or "grid" method in the past. Then I read a Linn Westcott book that included a section on L-girder benchwork. I'll likely never use the grid method again.
A grid may be better for modules or "dominoes" where the layout is intended to be taken apart and moved, but otherwise I would prefer L-girder hands down. Many of the reasons why I prefer L-girder were mentioned in a previous post above.
I have used both open grid and L-girders successfully. I would say that both methods work well, depending on your needs. My current layout is modular, or, perhaps more correctly, "dominoes" a la David Barrow. Two layouts ago, I used the L-girder method. To me, the choice is the amount of overhang or unsupported area you wish to span.
Each of the dominoes is no longer than 6' and no wider than 3'. But if you are going to have a longer span between legs, or something that projects off the side of the layout, the extra construction effort and weight may well be worth it.
John Timm
I recommend Linn Wescott's How TO Build Model Railroad Benchwork for a much better discussion of the whys and hows of L-girder vs Box.
L-girders main disadvantage is it is "thick" and not as appropriate for modules, in my opinion. I used it on my small 3x5 "test" layout and am quite pleased. In the image below you cann see the large section that will be Whitefield Junction and ribbon of roadbed that will be a steeply hilly slope on the other side. Picture a backdrop running down the middle in the final product seperating the two scenes.
http://www.homefry.com/mrr/image/WhitefieldProgress-3.jpg
bnycrail wrote:I confess I have never used L-Girders, opting exclusively to build my benchwork using traditional box frames. I acknowlege the structural advantages of L-Girders, but frankly I've never acquired the necessary confidence in my modest carpentry skills to be assured of producing a quantity of L-Girder beams, each possessing consistently perfect 90 degree angled corners.
I confess I have never used L-Girders, opting exclusively to build my benchwork using traditional box frames. I acknowlege the structural advantages of L-Girders, but frankly I've never acquired the necessary confidence in my modest carpentry skills to be assured of producing a quantity of L-Girder beams, each possessing consistently perfect 90 degree angled corners.
I find L-girder to be much more forgiving then box. The 90 degree takes care of itself when you screw the flange in. With box, all your dimensions need to be perfect or you end up with odd gaps.
bnycrail wrote:Having confessed my ignorance of L-Girder construction techniques, I remain confused by Mr. Popp's decision to employ this method in the building of the "Troy Branch." According to his consstrution notes and accompanying photograph, the joists are placed laterally across the L-Girders at the base of the risers, with the plywood sheet serving as sub-roadbed resting on the relatively small 1x4 (actually 3/4" x 3 1/2 ") "footprints" of the scattered risers.
Having confessed my ignorance of L-Girder construction techniques, I remain confused by Mr. Popp's decision to employ this method in the building of the "Troy Branch." According to his consstrution notes and accompanying photograph, the joists are placed laterally across the L-Girders at the base of the risers, with the plywood sheet serving as sub-roadbed resting on the relatively small 1x4 (actually 3/4" x 3 1/2 ") "footprints" of the scattered risers.
The strength comes from how often the supports are present. You really shouldnt need such fancy plywood. 3/4" 5-ply is fine, I used 1/2" on my 3x5 layout since the distance between risers was only a foot. I get the pine stuff sanded on one side. By having risers with the joists on the bottom you gain maximum scenic flexibility. Of course, with their industrial layout, this is less of an issue and less readily apparent then in more mountainous scenery.
The L-girder construction also gives you more control on where the risers are present by flexible joist location.
Finally, but having everything up on joists, it makes it very easy to wire and install switch machines. I don't have to dodge joists underneath.
bnycrail wrote:Furthermore, while one may, with a table saw, set the fence to ensure that the risers are all cut to a uniform length, great care must still be taken to make sure each and every riser is installed precisely.
Furthermore, while one may, with a table saw, set the fence to ensure that the risers are all cut to a uniform length, great care must still be taken to make sure each and every riser is installed precisely.
Complete non-issue. Just use a level to set the height you want from the riser relative to the next riser. I use scrap pieces of lumber for risers of all different heights. Changes are easy to make.
bnycrail wrote:Mr. Popp explains that risers were employed in order to allow the plywood to be cut away and permit "sub-track level" scenes. With a bit of planning, couldn't this still be done using a box frame? One could off-set the joists a bit, turn the "risers" upside down and build a pocket for the "sub-track level" scene. It may not be able to carry as much weight as having that scene supported directly on the L-Girder, but do you really need all of that extra load-bearing capibility for a little plaster and ground foam?
Mr. Popp explains that risers were employed in order to allow the plywood to be cut away and permit "sub-track level" scenes. With a bit of planning, couldn't this still be done using a box frame? One could off-set the joists a bit, turn the "risers" upside down and build a pocket for the "sub-track level" scene. It may not be able to carry as much weight as having that scene supported directly on the L-Girder, but do you really need all of that extra load-bearing capibility for a little plaster and ground foam?
You could, but its a lot easier with L-girder.. and a lot easier to change your mind. As for "Extra load bearing" its all about maintaining rigid construction using the least amount of lumber. You will use less lumber with L-girder since you can get away with smaller dimensions.
One thing to keep in mind.. its NOT about load bearing.. its about rigidity. THe L-girder will remain rigid over a longer distance with smaller lumber. The flexible joists are easy to attach because of the flange. Exact cutting is not as important because you dont have to butt anything like you do with Box.
Again, if you are really interested in the pros and cons, I suggest the Wescott book, he does a much better job showing the pros and cons of both then I ever could with my limited experience and I think he provides a little better depth then the recent MRR articles have.
I have skimmed the recent posts; please accept my apologies if my questions have been previously addressed.
I have been re-reading the January, 2008 MR article about the construction of the "Troy Branch," and in particular, the construction notes appearing on page 50.
I confess I have never used L-Girders, opting exclusively to build my benchwork using traditional box frames. I acknowlege the structural advantages of L-Girders, but frankly I've never acquired the necessary confidence in my modest carpentry skills to be assured of producing a quantity of L-Girder beams, each possessing consistently perfect 90 degree angled corners. All of my box frame sections have been assembled, face down, on the smooth and level concrete floor of my workshop, where I can at least know that the relevant side of the frames will remain smooth and even regardless of any irregularities in the dimensional 1x4 boards, and then hauled individually down stairs for installation. I do concede a certain waste of material, as each section ends up with a redundant joist when connected to the next section; however, these sectional frames are very easy to make and are very easy to level and adjust before securing final installation.
It would seem to me that, with such a small area from which the plywood could receive support, the sub-roadbed would remain vulnerable to sagging even if -- as Mr. Popp hastily adds -- one uses a rather expensive, 7-ply "cabinet grade" sheet of birch plywood. Furthermore, while one may, with a table saw, set the fence to ensure that the risers are all cut to a uniform length, great care must still be taken to make sure each and every riser is installed precisely. Any deviations in either the length of the riser or the point at which the riser is attached to the joist would result in the plywood having to either bend up or down in order to seat properly on top of these small columns.
In looking at the photo on page 50, it seems what is is needed is yet another set of joists, installed laterally at the top of the risers (and perhaps even logitudinally, depending on the width of the layout board) in order to guarantee the plywood sub-roadbed does not sag. But once that is done, haven't you wound up using at least as much material as you would have had you just built a box frame in the first place?
Granted, the L-Girder construction design suggested in Mr. Popp's article may indeed provide superior structural support, but it seems to me to be a bit of "overkill," as a simple box frame design provides, at least in my experience, more than adequate support for an HO-scaled model railroad. Furthermore, at least in the construction of this relatively narrow "Troy Branch," use of a traditonal box frame would further minimize the chances that the plywood top would warp or sag.
L-Girder Enthusiasts: What am I missing?