jlassen

Functional Train Bridges

Recommended Posts

So I've been thinking about creating a train shelf and want to suspend my Lego tracks along the ceiling about 2 feet down.

 

I'm trying to come up with a functional tressel that will give me the most support... that is, the most lenght of tracks without needing support.   I'm more interested in showing off the trains, rather than making the bridge look "real" so I'm looking at pure functionality here.  Part of that, I suspect means light weight... what can I do with the fewest amount of bricks, that provides the most support at the track junctions.

 

I know I'm probably reinventing the wheel here... but here are a couple different techniques I've been pursuing

 

#1, my Orange and black bridge.

This makes use of tiles and facingn studs, as a way to hold bricks together horizontally. I have to admit, I was inspired by this technic because of the "Lego masters" tv show I watched with my kids... The Lego bridge that held over 1000K pounds used this method. 

ACtC-3cd0C8BmvKZwYl6S8PX-RstOE6jl9xWRT-X

 

ACtC-3eSv_hJ9y8eEc6EiTJ7JWb2x3lvh-83yucX

 

 

In the above pictures, I would expect the rail join to be directly above the one 1x4 brick with facing studs... I only made the span 1 brick and one plate, because the track provides most of the support, and any flex should be covered by the horizontal plates that are tying the bricks together. 

 

#2 Lighter weaight Axle/technic wight distribution.

ACtC-3eSGUmlBhNoFrHLeEHmlB4NZGWej_1K7bA5

ACtC-3dra0Y57Ougir_hnFbmYW0izS8lPG9u0wQn

I beleive the stenght in this one comes from the axles... if the two technics bricks are being pushed apart by downward stress at the rail join, that will transfer the force horizontally to the axles, which will push it to the other joins... thereby creating a feedback loop.. pressure in one spot has the strength of all of the joins together... 

 

 

This one I wonder about... As built now, I only have a axle hole bricks in the center.  I could probably prevent some axle flex if I put the 1x2 axel bricks halfway betwee the current bricks.

But 1) that adds more weight, and 2) the spress is going to theoretically be pushing from an end, and axle flex is most likely to be induced from downward pressure,  not pressure from the end, so the added weight might not be benificial. 

 

Any thoughts?

 

 

 

 

Share this post


Link to post
Share on other sites

Of these 2, number one is by far the better option.  The way number 2 would load would result in the top axles being in tension, which will effectively pull them out of the bricks with the axle holes.  Either way though, you're going to get flex just from the track itself.  Simply making the joints rigid won't prevent sagging in the middle.  You're going to need some additional structure to span the shelves.  How much is a matter of how far you want to actually span.  I have this going in my son's room:

26150684956_1d3d148030_c.jpg

Share this post


Link to post
Share on other sites

How would the axles be pulled out of the bricks?  There is a stopper on the axle end and the axels would be pusshing against other axles? I mean, I understand, under a faulure load, one of the joins would fail, and the axles would push out at that break point. But where are the axles going to move to?  That was the design intention, to put the weight and tension on the axels. instead of the track and bricks underneatht he track join.

 

Basically, I wanted to see how much of a span I could do with just this type of modular load spreading.  And then build up some kind of under deck cantalever support for any span that is two much for the "basic" structure. 

 

Ultimately, I know this is a build it, and find out kind of situation. But I do appreciate everybodies expierence and feedback. I know I'm reinventing a wheel here.

  

 

 

Also, I had contemplated an intial design that had the axles running THROUGH the track join section, but that would require more axels and more bricks at each 1/4 point, increasing weight, and would put the load downward on the axle... My goal was to have the Any speration force pusing lenghtwise against the axel, and every axel "anchoring" all the other axles, right up to the bridge towers which could be buttresses as much as needed, depending on the span and expected force... 

 

I don't know. Maybe I'm thinking about this wrong.  All critiques appreciated. 

Edited by jlassen

Share this post


Link to post
Share on other sites

Consider the picture of my bridge above.  In it, the top horizontal "girders" are under compression, whereas the bottom ones are under tension.  As such, yes, there are the stops on the axles, and that stop would be loaded against that technic brick.  As such though, the axle would be loaded in tension, leaving you only the little bit of clutch the technic brick w/ axle hole exerts on the axle to hold it in place.  And since it is so close to the bending moment of the system (think of it like trying to open a door by pushing on the hinge) it will have very little effective strength.  It doesn't matter that you have another axle there to back it up; the bottom portion will physically expand.  Add in there is no truss structure to aid it, and you won't be able to span very far at all.

 

If you want to be able to see the trains, you can flip the truss over and put it beneath the tracks.  Doing so you could also then use the side of the shelves to push against the truss, giving you further support. 

Share this post


Link to post
Share on other sites

"leaving you only the little bit of clutch the technic brick w/ axle hole exerts on the axle to hold it in place."

No. the Adjacent axles hold it in place. The brick just prevents and horizontal or vertical movement, keeping the axles aligned.  My thought was the adjacent axles are what keep it in place. They are all lined up nose to nose, and stopper to stopper.

Are you saying there would be vertical force that would pull the axel and the brick is is through up, disconnecting it from the upper track bed? which woudl cause the axle "noses" to come out of alinement? Honestly, I hadn't considered that because I didn't think the force would be cause updward or downward, or sidewise force... but certainly "downward" would be the weak link, if that end of the axle applied downward force, due to load at the track connection end. 

 

"If you want to be able to see the trains, you can flip the truss over and put it beneath the tracks.  Doing so you could also then use the side of the shelves to push against the truss, giving you further support. "

yes. Something like this was definately my intention for when a span exceeded the support stenght of my modular deisgn... I appreciate you ephasizing this point

 

-Jeremy

Edited by jlassen

Share this post


Link to post
Share on other sites
3 minutes ago, jlassen said:

Are you saying there would be vertical force that would pull the axel and the brick is is through up, disconnecting it from the upper track bed? which woudl cause the axle "noses" to come out of alinement?

The sheer weight of the track & trains could (would?) very easily cause the bridge to sag, and those axles are only prevented from moving by "stoppers" at one end each: you can't in that setup protect against tension forces.

The solution @coaster has come up with is not only extremely strong, it's prototypical too. I'd vote for that. If you're worried the structure hides the train itself, you could build one of bridges that puts the truss under the deck:

Through%20and%20Deck%20Trusses.jpg

Share this post


Link to post
Share on other sites
2 hours ago, jimmynick said:

The sheer weight of the track & trains could (would?) very easily cause the bridge to sag, and those axles are only prevented from moving by "stoppers" at one end each: you can't in that setup protect against tension forces.

I really appreciate the feedback. certainly an undertress is the way to go on s lonhrt dpsn.

 

But I want to make sure I'm understanding you on the tension forces...

First... I'm assuming that from track end to end, the track itself is strong enough to support the tension forces for any given car that will be on that span.  It is when the tension forces are applied at the track join spots that the track needs "Support".

So that would be downward pressure at the join, that we are trying to accomdate... right?  

You and @coaster both indicate that there is nothing holding the axels in place.  My understanding of my design is the stoppers are not really holding anything.. The adjacent axle's stopper are what is holding the axles in place... 

 

I guess maybe I wasn't clear... this is a modular design..I plan to stack a bunch of these together...  LIke this...

ACtC-3dsdhaD6D9ZXqYAe2iM6RXUJoadpkFs0ug6

both you and coaster indicate that nothing is holdingn the axles in place... but my understanding is at each one of the rail joins, the axles are stopper to stopper. They hold each other in place.  and at the undetrack meeting of the two noses, they are facing each other, and holding each other in place.  at the end of the span, there will be a big block, as secure as need be, to hold the force being pushed against the it from the axles. 

 

From your responses, it seems like I am missing something .  The tension created by weight of train at the track joins pushes down on that joint, and the bricks below them, any tenshion that pulls those bricks apart will transferd from a downward force to a horizontal force in the axle... that force will then be sent the lenght of the span, across ever one of the axels, until it reaches the end of the span and the embudmant there... which we can assume for our purposes will be infinatley strong. ....

 

 

wait wait wait... hmmm.

so if the bricks are being pulled apart... the bricks are not actually PULLING on the stopper.. so the friction that coaster was mentiong was the friction created as the bricks pull apart... up against the side of the brick... But the axles still an't move t the left and to the right... (theoretically) and as those bricks pull apart... they are going to be putting up and down torqe on the axel... but there really isn't anything forciing that to go along the axels..

I think I see what you both were tellingn me now.

hmmmm.

Thank you. 

If I reversed it, and the stopper was on the outside instead of the inside... because the failure will be the bricks coming apart at the join... pulling away from each other... so if the stopper was on the outside...  that failure pull would THEN, via stopper, be transmitted to the axle.

Right?

 this is assuming the failure point is at the track join, and the failure is the two bricks at the top pulling away from each other.

 

 

 

 

Edited by jlassen

Share this post


Link to post
Share on other sites

Hey @jimmynick  and @coaster

I really apprecaite your patience in walking me through the failure of the the red and black design. I do understand what you are talking about, regaurding the downward force not being actually pushed out along the axles.

I am going to work on a couple of alternate ideas, to figure out how to best transmit that force down the lenght of the axles. 

 

 

 

version 2 of the red and black bridge.

I need to transmit the force of the two bricks pulling apart down the axels. That is the assumed break point when downward force is applied at the rail join.  

 

so in this design, as the bricks pull apart in failure... two of the four axels will have that force applies to their stoppers... that will Push it downward, against the other axles.

 

This design requires alternating top to bottom, or left to right.  I feel like top to bottom will provide more reliable strength at any given join.

ACtC-3dpC37n12KxDIIkWTUVgrcocZCtxmHssK7i

In this picture, a failure at the right join will cause the bottom cables to take that pressure... that will then transmit that force, along both bottom axles, to the left... which will then push it up against the head to toe lenghts of the axles.

and a failure at the left join will cause the oppoite.. the top two axels will push to the right, along the lenght of the spans axles.

In addition to properly trasmiting the tension horizontally, this also puts an axel across the possible seperatoin/break point of the two breaks. Not sure if that helps at all, given they are not inside an axle hole, but rather inside a pin hole.

 

The other thing that should probably happen is the 'Guid bricks"  1/4 of the span on each side should be moved over next to the join bricks... so there is a block of 4 bricks at each join, instead of the two... 

 

2 4x4 plates should acomplish this task... hmmm... more to come.

 

Edited by jlassen

Share this post


Link to post
Share on other sites

version 3.

Okay, this went from "Looks kind of lightweight and elegant" to Built like brick shithouse.

ACtC-3fPRpXpddj47hWE8E396BW84UuK84lRvTS3

But this now connects the break point across 4 brick widths and at has the axle stoppers pushing the force down the length of the cables. 

I can't wait to see how many jars of pennies this will hold up. :)

Time to have a competition with my daughters to see which design holds the most. Light weight black and orange or red black and yellow brick s* house.

 

Also time to stop playing with virtual legos, and go make dinner. Thanks again for your feedback, coaster and Jimmy.

 

DInner is being made...

Overkill

ACtC-3cdRftog16oZIaISVg7Q5HtEccGxODvwRkc

 

Edited by jlassen

Share this post


Link to post
Share on other sites

So to answer some of your questions here, The technic axles solution isnt ideal because the clutch (grip strength) of the axles in the 1x2 brick with cross hole is pretty low. You have a stop on the one end but the middle has no stop. 

Here is a quick example of the basic physics. 

Say this is two track sections and the stops are also the connecting ends represented as vertical lines:


  |-----------------||-----------------|


You will have these forces as you put a train on:

      THISISATRAIN      THISISATRAIN
  |-----------------||-----------------|
          ||                 ||
          \/                 \/
Weight (mass and gravity) pulls down at the arrows

This will caause the force to be transmitted like this

      THISISATRAIN      THISISATRAIN
  |-----------------||-----------------|
      <=       =>        <=       =>

All this to say that the technic axles dont have the clutch power you would want for your train bridge. 

 

I would recommend looking at some bridge design articles and also picking up a free bridge builder game on your phone to learn about how this works in an interactive way. 

Share this post


Link to post
Share on other sites

I think you are having some difficulty understanding the direction of the forces applied.

Look at the span as a whole.
As weight is added, the span sags.
Everything at the top of the bridge is under compression forces, being pushed together.
Everything at the bottom is under tension forces, being pulled apart. The bridge is being stretched, pulled to make it longer.
The issue with the axles is there is only a stopper on one end of it. So while the end with the stopper is nice and secure, the other end is still able to pull out. The axle stopper of the next section will not help, as the axle is being pulled away from it, not pushed towards it.
Your design has good compression strength, but low tension strength, and tension strength is what we need at the bottom.
Also the axles themselves are a bad choice. They are thin, and bend very easily. The track sections are also able to bend fairly easily. So even if we fixed the lack of tensile strength, the bridge itself would still sag due to its flexibility.

You need a beefier, rigid structure. One that is not able to bend or flex. One with multiple layers of overlapping parts at all levels. Coaster's bedroom bridge is a good example of this.
There's a reason bridges are designed the way they are in real life, take note of those designs, you need to copy those elements in lego form.

Share this post


Link to post
Share on other sites
9 hours ago, davidzq said:

I think you are having some difficulty understanding the direction of the forces applied.

Look at the span as a whole.
As weight is added, the span sags.
Everything at the top of the bridge is under compression forces, being pushed together.
Everything at the bottom is under tension forces, being pulled apart. The bridge is being stretched, pulled to make it longer.

I appreciate this concise description of the problem and the forces at work.  Than you. 

Share this post


Link to post
Share on other sites

I've found that a sandwich of plates and panels makes for a strong and lightweight bridge, e.g., layers going from top to bottom:

 

track

6x8 plates

1x4x3 panels

6x8 plates

 

If cost and aesthetics are not an issue, there is one person on this forum who makes some really long spans just by making stacks of 4-6 sets of track segments.

 

Then again, if it is functionality you are looking for, perhaps use non-lego for structure and then have a cosmetic bridge facade hang over the edge.

Share this post


Link to post
Share on other sites
2 hours ago, zephyr1934 said:

Then again, if it is functionality you are looking for, perhaps use non-lego for structure and then have a cosmetic bridge facade hang over the edge.

This would be the go to for me if it was "just" a shelf for display

 

Share this post


Link to post
Share on other sites

I really appreciate all the feedback I've received on this thread. 
And of course, "The right tool for the right job"  principal shows axles ain't gonna get it done. clearly.

 

Lift arms and pins is where I need to go.

ACtC-3f7r22N_hMFQEJNxBbvr1Adi-Bi29kh6dbc

 

ACtC-3c4zPvj3uInznqOz26z14LdV5ZUtkwM8p-R

 

I feel like maybe *I* might not be able to stand on this, but my daughter probably could. Thoughts?

I'm using both the arched lift arms and the interconnects between the top arm and bottom arm to transfer the contrasting sag forces (compression at the top, extension at the bottom) against each other. The interconnects between tracks (and modular cantilever sections) are being held together by the 4 lift arms, and the 4x4 plate. 

and with this narrow cantilever, I could encase it if I wanted to change up the aesthetics... 

ACtC-3e6qanLTTFFAt2ZTJAjUf1OnQ_9U8RfimVf

Edited by jlassen

Share this post


Link to post
Share on other sites
On 9/17/2020 at 5:19 PM, zephyr1934 said:

Then again, if it is functionality you are looking for, perhaps use non-lego for structure and then have a cosmetic bridge facade hang over the edge

That is what I would do/have done as well - as my first choice.

It really is a tough task to do what you want to accomplish with a pure LEGO solution. When the trains are supposed to "show", then - as you are attempting - the support structure needs to go below the track.

Did you study a bit how "real" bridges (relatively wide-spanning) are designed this way? There are not many - for a reason, laid-out above. I believe though this study is a good starting point. And then transfer that to the LEGO world - with all the options and restrictions you have "brick-wise".

Good luck and all the best,
Thorsten

 

Edited by Toastie

Share this post


Link to post
Share on other sites
1 hour ago, Toastie said:

 

It really is a tough task to do what you want to accomplish with a pure LEGO solution. When the trains are supposed to "show", then - as you are attempting - the support structure needs to go below the track.

With my narrow cantilever design right now the primary lift arms that are attached at either end of each track junction-pieces *are* the primary track support. The Two vertical arms connecting the lower arms and the upper arms also serve as track supports at the center. These leaves at most 3 studs worth of track on either side not directly supported.

I am wondering if it is a net positive or negative in this design to not have the tack attached to most of the support structure... so any rotational force applied by train movement on the tracks won't be transmitted to the cantilever structure... except at those 4 attachment points. that force would let the track move and flex and expand/contract without most of that force being transmitted to the cantilever. Sort of an "earth quake proofing" technique.

Conversely, I could put a Tile across either end, which would fit in the open space between the rails at each point. That might prevent more movement and shaking of the track if it was more tightly fit.

 

Honestly the fun of this is trying to do it in Legos. And I'm learning a lot. Honestly The geometry of Lift arms and technics pieces is not my strong suit --Thats why my first attempts were so brick focused.

So this is forcing me to learn in.

 

Edited by jlassen

Share this post


Link to post
Share on other sites

So an external shell, completely isolated from the cantilever structure, except via the track.

Something like this:

ACtC-3cHV8ePDmvv-LDSQ8_J-P0O6iA04JEr8UNY

 

Kind of prototypical of an open deck plate girder bridge.  Hell. Despite the added weight, the exterior shell might make the overall structure MORE stable.

And for those suggesting I take a look at real bridges.... here is some serious Bridge porn i found.   Enjoy!

A Context For Common Historic Bridge Types NCHRP Project 25-25, Task 15

Edited by jlassen

Share this post


Link to post
Share on other sites

This looks much, much stronger. It also looks attractive with the panels on the outside, as @zephyr1934 suggested. I'm slightly concerned about the integrity of the joins between these modules, but hopefully the combination of pins and 4x4 plate will do the trick. It may be time to build it in the brick and to tinker with it to ensure the structure is sound.

Share this post


Link to post
Share on other sites
5 hours ago, jlassen said:

So an external shell, completely isolated from the cantilever structure, except via the track.

Something like this:

ACtC-3cHV8ePDmvv-LDSQ8_J-P0O6iA04JEr8UNY

At that point you are sort of building nested bridges. A potential problem with the technic design is that the pins have a bit of play, so a long bridge will likely sag.

Brickshelf is back up, first, here's the bridge I was talking about earlier

3img_1818.jpg

Though I have yet to test it under extended use it is 64 studs long. The track is just connected to 6x8 plates below. On the left is this design,

viaduct1.jpg

over a similar span. The key with this build is that every tie in the track is connected to the side walls with a bracket. The side walls provide significant rigidity. You could easily flip the walls so that they were below the track but then you are getting towards the boxbeam above.

Meanwhile, PennLUG had an even longer boxbeam bridge that I believe they still use at many of their shows, only differences are that it was better decorated and used 1x6x5 panels, here's a not so great shot of it,

img_2081.jpg

Share this post


Link to post
Share on other sites
14 hours ago, jlassen said:

Something like this:

I believe this is simply not necessary. I believe the entire Technic assembly will bend much more than the outer shell. It looks very good, but it makes not that much physical sense. The very narrow tolerances on the stiff larger pieces of the shell will lead to much less bending. Well, I believe. I have actually made a bridge spanning about 1 m (1 yard, 3 feet) from only 4 + 4 (on each side) monorail straight track:

The fewer pieces used to gain span, the less bending will occur. The looks however is equally important as well, for sure!!!

Best
Thorsten

P.S.: Phew - that document you referenced is NICE!!! Thank you very much for the link. I have securely saved that one! It does point on many occasions into the "add stiffness" by panels direction, doesn't it? Figure 3-108 is a nice example, isn't it?

 

Edited by Toastie

Share this post


Link to post
Share on other sites

I agree with @Toastie, that whole Technic construction inside looks seriously over-engineered.

@zephyr1934's design is essentially a box. The panels put the bottom layer of plates at a distance from the tracks, so that if the track wants to sag it would have to stretch the bottom plate layer, which it can't. This is a box girder bridge: https://en.wikipedia.org/wiki/Box_girder_bridge

To make it even stronger the best weight to strength you'll get from any of the truss bridges that @jimmynick posted. They're all based on the engineering principle that a triangle doesn't deform.

The box girder should be strong enough to hold any Lego train but for longer spans you probably want to turn to a truss bridge.

I found this one:

99232537de72c8ffc6ffb12bce39a2c3.png

It's very simple but should be strong. I would make 2 changes:
- Offset the connections of the track pieces and the top level Technic bricks (shift the track 8 studs either way).
- Add a layer of 1x plates at the bottom. I think there's a studless beam inside there to hold the bottom row of Technic bricks together but I'd prefer plates.

Share this post


Link to post
Share on other sites
24 minutes ago, Duq said:

They're all based on the engineering principle that a triangle doesn't deform.

Very true - provided there is no "play" at the "corners", i.e. they are welded or close to that. All Technic pieces (beams, bricks with wholes) connected with pins introduce some play in addition to their "wobbliness" - they are somewhat soft. This is certainly not an issue here; and the more of those are used, the better. 

However, I believe in the LEGO world - no welding, no riveted girders - reducing the number of pieces used per length and height segment and exploiting the very narrow tolerances/clutch of bricks/plates leads to a construction with the highest stiffness, i.e., the box girder approach, you mentioned, @Duq.

I made a two track drawbridge, which gets its stiffness from an assembly of six 8x16 "tiles" connected with plates: It does not sag at all - it simply breaks when the load is too high. Which it never gets with trains on it.

Best
Thorsten

Share this post


Link to post
Share on other sites

Create an account or sign in to comment

You need to be a member in order to leave a comment

Create an account

Sign up for a new account in our community. It's easy!

Register a new account

Sign in

Already have an account? Sign in here.

Sign In Now

  • Recently Browsing   0 members

    No registered users viewing this page.