Wing rails and Track Geometry

[Andrew_au]

Looking at NMRA standard S-1:

I notice that as the wheel transitions from the closure rail to the frog via the wing rail, that as it crosses the gap the effective wheel radius decreases because the edge of the wing rail is further from the flange and thus on a smaller radius of the wheel surface. This could lead a slight drooping of the wheel and thus a distinct "bump" as the thicker part of the wheel makes contact with the frog.

I see two ways of compensating for this:

• roll off the point of the frog slightly so the incoming wheel can roll up a slope rather than contact the point. You will still get a "dip" but not a "strike".
• gradually and slightly increase the height of the wing rail as it diverges from the opposite closure rail / frog rail so as to compensate for the shape of the wheel surface.

Have I even understood the issue correctly? How do real railroads handle this?

 

[Greg Elmassian]

the difference in diameter is technically true, but not the issue.

Properly designed, the wheel tread is supported at all times, but with our models, the wing rail flangeways are too wide, and the frog cannot be made to keep support of the tread.

LGB compensated by making a flange-bearing frog.

better quality turnouts come close to continuous support.

Your issue is not the difference in diameter, it's the overly sloppy tolerances we have to endure, for several reasons.

real railroads don't have to handle this, they don't have our sloppy tolerances, and in some cases they do use flange-bearing frogs. If you look closely, higher numbered frogs should have more overlap between the wing rail and the frog.

Greg

 

[Andrew_au]

Primarily interested in real railroads.

For example, consider the turnout pictured here, and a train heading toward the camera along the mainline (our left branch). Just before a wheel reaches the point of the frog, it is in contact with the wing rail close to a track width away from where the rail edge would nominally be. (See blue line on image below for deviation). If the surface of all relevant track was perfectly flat and level, the wheel profile would allow the wheel to lower at least a mm or so before it reached the point of frog.



Things I observe:

• Unlike the surrounding track, the frog area has a solid metal base. Is this purely to maintain spacing, or do the wheel flanges actually run on this base?
• The wing rail has two distinct surfaces (join is highlighted in yellow on diverging route's wing rail). The "outer" surface appears to be a continuation of the closure rail. What is the role of the inner surface?

 

[Software Tools]

For example, consider the turnout pictured here,

That appears to be a self-guarding crossing (aka frog). That is a type of crossing which does not require guard rails on the opposing track.

 

[Andrew_au]

That appears to be a self-guarding crossing (aka frog). That is a type of crossing which does not require guard rails on the opposing track.

I did not realise such a thing existed. Thank you for broadening my knowledge!

That said, the turnout as a whole is not primarily self-guarded. If you follow the link to the full picture, you will observe some quite massive guard-rails. That said, nothing stops the engineers from combining a self-guarding with guard rails - possibly to reduce wheel wear by relegating the guard-rails to a secondary role?

 

[Greg Elmassian]

If you read about it:

Self guarded frog


A Solid Manganese Steel Self–Guarded Frog is constructed with guides or flanges above its running surface. These flanges contact the tread rims of the wheels and safely guide the wheel flanges past the point of the frog. Thus, the self-guarded frog does not require the use of guard rails.

Better picture of a real self-guarded frog, you see the raised flanges that control the tread rims

 

[DafyddElvy]

If you read about it:

Self guarded frog


A Solid Manganese Steel Self–Guarded Frog is constructed with guides or flanges above its running surface. These flanges contact the tread rims of the wheels and safely guide the wheel flanges past the point of the frog. Thus, the self-guarded frog does not require the use of guard rails.

Better picture of a real self-guarded frog, you see the raised flanges that control the tread rims

View attachment 302907

As a track engineer I can tell you I don't like the idea of turnouts without check rails, self guarded crossings I believe have a severe speed restriction applied to them.

A much easier way to deal with the gap between the crossing nose and crossing knuckle is to use a swing nose crossing, I have been thinking about this type of crossing for my own garden tramway to get around the issue of coarse scale wheels on visiting models.

 

[Rhinochugger]

How do real railroads handle this?

There are two issues:

1. The wheel being guided through the frog
2. The wheel dropping into the frog

There's been plenty of response on the first issue, and Greg alluded to the second.

Basically in the 1:1 world, the ratio of the wheel diameter to the gap in the frog is much greater (due to the finer tolerances that Greg mentioned) and the wheel doesn't drop to any noticeable extent.

Not sure how the really small narrow gauges manage this though, where they use much smaller diameter wheels.

 

[Greg Elmassian]

They go slower and their track is noticably "rougher", at least here.

 

[PhilP]

They go slower and their track is noticably "rougher", at least here.

I'd second that..

Riding a slate wagon, tests the integrity of your dental work, as well as control of your muscles!


PhilP

 

[Andrew_au]

Basically in the 1:1 world, the ratio of the wheel diameter to the gap in the frog is much greater (due to the finer tolerances that Greg mentioned) and the wheel doesn't drop to any noticeable extent.

Some example math:

To figure how much a wheel could drop:

• k = size of gap
• r = radius of wheel
• d = maximum deviation (or drop)

The gap forms a chord across the circle formed by the wheel.

• perpendicular distance from chord to centre = sqrt(r^2 - (k/2)^2)
• perpendicular distance from chord to edge of circle = d = r - sqrt(r^2 - (k/2)^2)

Wheels on NSW railways are approximately 900mm diameter (450mm radius). If we assume a 10cm (100mm) gap, then the maximum 'd' is 2.78mm. That's not insignificant, but it is only 0.6% of the wheel radius.

If we decrease the gap, the deviation drops rapidly due to the very flat angles involved. For example, reducing the gap to 80mm reduces 'd' to 1.78mm (0.4%).

All this assumes the wheel can "fall" all the way into the gap without being arrested by the wing rails.

* I don't actually know what the gap is. Though NSW Standard ESC-250 section F (design requirements for the "nose" of the frog) is fascinating in its level of detail. For example, the standard requires that the nose be 8mm below the wing rail height at the point it is 16mm wide and gradually rise from there.

The wing rail has two distinct surfaces (join is highlighted in yellow on diverging route's wing rail).

Apparently this is a manganese steel insert. Basically upgraded rail material at the most important point.

 

[dunnyrail]

Some example math:

To figure how much a wheel could drop:

• k = size of gap
• r = radius of wheel
• d = maximum deviation (or drop)

The gap forms a chord across the circle formed by the wheel.

• perpendicular distance from chord to centre = sqrt(r^2 - (k/2)^2)
• perpendicular distance from chord to edge of circle = d = r - sqrt(r^2 - (k/2)^2)

Wheels on NSW railways are approximately 900mm diameter (450mm radius). If we assume a 10cm (100mm) gap, then the maximum 'd' is 2.78mm. That's not insignificant, but it is only 0.6% of the wheel radius.

If we decrease the gap, the deviation drops rapidly due to the very flat angles involved. For example, reducing the gap to 80mm reduces 'd' to 1.78mm (0.4%).

All this assumes the wheel can "fall" all the way into the gap without being arrested by the wing rails.

* I don't actually know what the gap is. Though NSW Standard ESC-250 section F (design requirements for the "nose" of the frog) is fascinating in its level of detail. For example, the standard requires that the nose be 8mm below the wing rail height at the point it is 16mm wide and gradually rise from there.


Apparently this is a manganese steel insert. Basically upgraded rail material at the most important point.

Wow at least with a spreadsheet we can get to grips with the original query, I think!

 

[Greg Elmassian]

I come from a Physics background. If you ask me "does stirring this glass of water increase it's temperature?" I will say "of course, absolutely"
But if you ask me "how much warmer", I would tell you an amount we cannot measure from a practical perspective,


So, does the wheel drop? Of course.
How much? Not enough to make any difference to the passengers.

Greg