Super-Elevating Curves

BCR 570 Mar 2, 2009

  1. LOU D

    LOU D TrainBoard Member

    1,412
    2
    23
    Michael,there is a way to lower the inside of the roadbed,and that's the way I do it..I start by installing Midwest cork,then,after the glue is completely dry,I use a sanding block,and sand in the angle I want for the superellevation..By doing that,I actually lower the inside rail..And to do it smoothly,I always end up with at least a little transition banking on the straight section,if only an inch or two..And yes,I'm probably using 1/16 to 1/32 total..
     
  2. sp4009

    sp4009 TrainBoard Member

    803
    157
    22
    I was going to post last night, but figured I'd wait and pay a little more attention to the high banked curve I go around 5 nights a week...

    It's good for 50 MPH, though about 1,000 ft after, it reverses into a 30 MPH curve, so, unless you have a real light train, 42-43 is about as fast as I've been able to take it... Anyway... the super elevation starts on the tangent and increases through the easement... You don't notice it much at speed, but it's a little creepy at low speeds... not to mention it needs to be tamped... I came around it at 10-12 MPH last summer with a 4 axle unit... the condensation bin on the air conditioner was not draining properly... needless to say, I got a good cold shower...
     
  3. LTCTerry

    LTCTerry TrainBoard Supporter

    152
    26
    19
    But I am.

    Why do you "lean over" in a curve when driving a car? Your body wants to continue going in the same direction is was before the turn started. That's called inertia. Your upper body isn't really leaning, it's following Newton's first law of motion and continuing on in the same direction. The car's wheels (friction) supply a sideways force to change the car's velocity vector. There is friction between you and the seat, so most of that gets applied to you as well.

    The center of gravity is the point in an object where you can pretend all the mass resides. The CG in a car is above the the plane of the road where the sideways force acts on the wheels. So, the CG wants to go straight and the wheel want to turn. That puts a rotational vector on the car and it leans away from the turn. Ships do the same thing. So do trains...

    Cars are wider than they are tall and they have suspensions. Trains are taller than they are wide and the CG is quite high. The wheel flanges and wearing surfaces are very small. What makes a train tip over? The CG is normally centered between the rails. If you tip the car a little bit it will return; tip it a little more and still will return to upright. Why? The force vectors acting on the CG when the car is tipped have both a vertical component and a sideways component. The sideways component will cause the car to return to the upright position. However, if the inertia (force) trying to keep the train going straight as it makes the turn equals the horizontal component of the force through the CG then it won't return to normal. If it's stronger, then over she goes.

    So, if you slightly lower the inner rail, then you have "pre-tipped the train in the direction you want it go. It has to tip up before it can tip over. Isaac Newton is on your side.

    Force is equal to mass times acceleration. We can have "scale miles" and "fast clocks" but you cannot find "scale mass." Does it look good? Yes. Can trains tip over if they go too fast? Sure. But not at scale speeds and reasonable curves.

    A 6" drop is about 8.5 degrees. I think the angle in the pictures is probably more than that. If you tip in too much, gravity simply takes over and hurts you. You begin to lose contact on the outer rail rather than the inner rail (which is what super elevation is trying to stop).

    Super elevation is trying to counteract the effects of inertia in the curve. You should ease into super elevation as you are easing into the curve; neither should be abrupt, but they belong together.

    But, you ask, what about (race) cars and banked turns? Glad you asked, so this can get even longer. :)

    Tires don't have a tremendous amount of friction available and road conditions can only make things worse. Imagine a Hot Wheels track with a loop in it. the car goes through the loop because its front end is being lifted by the track and it has enough energy to continue. The road applied a force to the car to change its direction. In a banked turn the banked road applies a force "up" on the front of the car to make it change direction without having to turn the wheels (as much). The drivers at NASCAR races are not cranking the steering wheel way over when they make the high speed 180s on big banked curves. The road itself is applying the force not the wheels. The cars are making a "sideways loop."

    School's out. :)

    Terry
     
  4. seanm

    seanm TrainBoard Member

    282
    0
    15
    Nicely done Terry! Very nice description. I like the NASCAR reference and it makes it very clear... so if they could bank the racetrack 90% they would never have to turn! :)
     
  5. LTCTerry

    LTCTerry TrainBoard Supporter

    152
    26
    19
    Thank you. :)
    That is correct. Really. If the car is moving fast enough the down force created by the aerodynamics of the car would hold it there. Even upside down.

    Made up numbers - Car weighs 2,000 lb, downforce at speed is 5,000 lb for a total down force of 7,000 lb. Now, if you can roll the moving car and the road upside down (maybe in some sort of pipe?) you would have the car's 2,000 lb going down with 5,000 lb upwards for a net of 3,000 lb in the upward direction, more than enough to hold the car.

    I've made up the numbers 'cause I don't remember the actual ones, but the ratio is about right. This is from an article on "the science of NASCAR" a year or so ago.
     
  6. LOU D

    LOU D TrainBoard Member

    1,412
    2
    23
    Yep..Most racecars that have any kind of downforce,,even Nascar,actually use suspension stops to keep the car from being able to bottom out the chassis on the ground.Formula 1 cars weigh 1000 pounds or so,yet,they have several thousand pounds of downforce at speed.Actually keeping the chassis off the stops so the actual suspension still works,yet getting the car as low as possible at speed is one of the most difficult parts of car setup..
     
  7. SleeperN06

    SleeperN06 TrainBoard Member

    3,386
    50
    45
    Great photos and fantastic Loco! I like it and I think I’m going to try that myself.:thumbs_up:
     
  8. Boilerman

    Boilerman TrainBoard Supporter

    415
    48
    22
    Rather than sanding with a block (it was too slow), I tried my Vibrating sander with 80 grit paper, worked good, it is faster and I ended up with a smoother road bed over all and could control the amount of taper to super elevate my curves.
     
  9. CaptainCarbon

    CaptainCarbon New Member

    4
    0
    12
    Seeing as how this is a popular subject, I shall throw down my two cents and watch as it ricochets off the table and rolls under the fridge, never to be seen again.

    On my last HO layout I attempted SE curves for the first time. I selected some string with a diameter of about 1.5mm, proper string, not the flat plastic stuff. I went around the outside edge of my curves, on the sub-roadbed, hot gluing the string down every 50mm or so. I then just glued the foam roadbed down on top, gently so as not to completely deform it over the string. The end result is not as great an inclination as BCR 570's example, but it is certainly noticeable.

    I have had no derailment issues relating to the super elevating and I don't believe it is having a significant effect on traction, but I haven't attempted to prove that. Unfortunately I don't seem to have any photo's of the string laying or the resulting inclination.

    I should also point out that my foam roadbed was not the commercial type, but a stiffer, denser material. Actually it is just strips cut off a camping bed roll/mat, a bit crude but very cheap.

    Douglas
     
  10. LOU D

    LOU D TrainBoard Member

    1,412
    2
    23
    Hmm..What the heck? I answered this post 2 days ago,it's gone..Anyway,I use a regular rubber sanding block with 40-60 grit paper,it cuts plenty fast enough.I start at one end of the curve,and make a complete pass with the block from one end to the other as many times as necessary to get the angle and easment I want,that keeps it nice and uniform.AS a matter of fact,I sand ALL my roadbed,that gives me extremely consistant roadbed,and nice,flat trackwork..
     

    Attached Files:

  11. CSXDixieLine

    CSXDixieLine Passed Away January 27, 2013 In Memoriam

    1,457
    0
    21
    Tim, great thread you have started here--I love superelevated track! In fact, I think it is about the best "bang for your buck" feature you can include on a layout to improve realism. Using the example in your initial post and the document Michael has posted on his RGW website, I superelevated a section of track on my hollow core door test layout. The results were GREAT and I will be using this method as I continue with the initial trackwork on my CSX Dixie Line layout. Here is a quick video I hacked together showing the results of the superelevated track on the test layout:

    [ame="http://www.youtube.com/watch?v=5LmWU8W3K1I"]YouTube - CSX Dixie Line in N-Scale: Superelevated Track Demo[/ame]

    Jamie
     
  12. BCR 570

    BCR 570 TrainBoard Member

    196
    4
    14
    I ddin't realize that this thread had kept going! Lots of very insightful and stimulating discussion here - thank you all for your contributions.

    Very nice video, Jamie - I will have to learn how to do that soon. You certainly don't seem to be having any problems with locomotive or train performance.

    I may have overdone it on the super-elevation a bit - sounds like 1/32" is what is needed. I wonder how many layers of masking tape that is . . .


    Tim
     
  13. CSXDixieLine

    CSXDixieLine Passed Away January 27, 2013 In Memoriam

    1,457
    0
    21
    Tim, I used eight strips of masking tape as Michael describes in his linked document on the first page of this thread. I stopped at six layers but that did not seem to be quite enough; eight seemed to be the sweet spot with regards to looks. I figure the masking tape I am using is about 5mil, so eight layers would be 0.04" thick. This comes out to be 6.4" of height on the prototype, which is way beyond extreme. However, it looks just right to the naked eye and through the camera at various closeup angles. As for performance, I have run plenty of 6-axle locos and some really finicky autoracks through (forward and reverse) without any hint of trouble. Next, I'll redo these same tests with some smaller 4-axle locos with 30+ coal cars next to see how they perform. Jamie
     
  14. RGW

    RGW TrainBoard Member

    2,018
    2,436
    64
    Tim/Jamie, you could use 7 layers if you feel that the combination of the tape you are using and 8 layers is too much. In your N scale example J, the contrast from the main to the siding is disctinct and looks really cool; especially when the two trains are side by side. Did you try 7 layers and see if that worked, or was it just not enough visual lean?

    One thing I believe to be true is that the visual lean in any scale can be slightly overdone and still be very prototypical, even if the dimensions are a bit on the overdone side (6" vs 6.4" like you mention). If you look at this picture, the visual lean is impressive when at speed on a 70 mph curve. You would never guess it's only 6". Great work! Michael
     

    Attached Files:

    Last edited by a moderator: Mar 25, 2009
  15. sandro schaer

    sandro schaer TrainBoard Member

    2,020
    87
    43
    i use evergreen styrene strips between the track and woodlands track bed.
     
  16. CSXDixieLine

    CSXDixieLine Passed Away January 27, 2013 In Memoriam

    1,457
    0
    21
    I did not try 7 layers but did start with 6. The 6 looked ok but not great so I added two more layers to get the 8 like you stated in your document. That really seemed to get the "wow" effect I was looking for. Jamie
     
  17. subwayaz

    subwayaz TrainBoard Member

    3,222
    106
    44
    Yes I found your article & this thread quite useful and easy to follow. Thanks for it.
     
  18. FlamesFan

    FlamesFan TrainBoard Member

    176
    0
    17
    Degree of Angle

    I would like to have super elevated curves on my new layout but the questions I have is what degree should the angle of the track in the center of the curve be?? Or do you just go by trial and error??

    I really like the look of super elevated curves so if someone could help that would be great.

    Cheers
     
  19. Chaya

    Chaya TrainBoard Supporter

    1,095
    2
    23
    I have read that the maximum super-elevation of prototype track is 6".
     
  20. Grey One

    Grey One TrainBoard Supporter

    8,903
    3,622
    137
    I'll be using gift cards. As a cashier in a large chain I have access to large quantities of these which are generally thrown away after they are empty.
     

Share This Page