• Our new ticketing site is now live! Using either this or the original site (both powered by TrainSplit) helps support the running of the forum with every ticket purchase! Find out more and ask any questions/give us feedback in this thread!

sharp curves + effect on wheels

Status
Not open for further replies.
Joined
12 Dec 2009
Messages
634
Location
Carmarthen
ok on road vehicles with tyres there is a mechanical device called a differential to allow the wheels to rotate at the speed they want to when turning. if you consider a very long, very heavy freight train slowly rounding a sharp curve, the solid axles on the loco and all the rolling stock prevent the wheels from turning the speed they want to. why doesn't this extreme slippage force cause the train to stall? there is an interesting description on wikipedia about "hunting oscillation" where the wheels ride up on to the larger circumference part of the wheel "cone" so the speed variation effect cancels out somewhat. is this exactly what happens to avoid this slipping?
 
Sponsor Post - registered members do not see these adverts; click here to register, or click here to log in
R

RailUK Forums

Old Timer

Established Member
Joined
24 Aug 2009
Messages
3,702
Location
On a plane somewhere at 35,000
Because there is no differential both wheels rotate at the same speed there will be sliding experienced by the wheel on the outside of the curve as there is more distance for that wheel to travel relative to the inside. However the amount of sliding varies with the radius of the curve and is not sufficiently pronounced as to be a major problem.

For a wheel of 3' 6" diameter, the minimum radius at which the wheel will round the track without sliding is 3,360 feet.
 
Joined
12 Dec 2009
Messages
634
Location
Carmarthen
from swansea station heading towards cockett is a very sharp left curve where the wheels can be heard screeching. there is a 3rd rail here on the inside curve leaving a 2-3inch gap for the flange to sit in. what is that for? there must be a heavy load on a bend like this for trains. the class 43 when valenta powered used to scream up this section.
 

Old Timer

Established Member
Joined
24 Aug 2009
Messages
3,702
Location
On a plane somewhere at 35,000
from swansea station heading towards cockett is a very sharp left curve where the wheels can be heard screeching. there is a 3rd rail here on the inside curve leaving a 2-3inch gap for the flange to sit in. what is that for? there must be a heavy load on a bend like this for trains. the class 43 when valenta powered used to scream up this section.
The squealing is generally the flange running against the outer or high rail. On very very sharp curves it can be the wheel being dragged round.

The "third" rail is a "check rail" and its purpose is to assist in preventing the outside wheel flange climbing the outside rail head.

The affect of curving on Pacers is exaggerated by the long wheelbase of the vehicle.
 

The Planner

Veteran Member
Joined
15 Apr 2008
Messages
19,652
Asks the question of which is the sharpest bend on the network ?? Syston North to East looks pretty sharp and the curve from Grand Jn to St Andrews is fairly tight.
 
Joined
12 Dec 2009
Messages
634
Location
Carmarthen
coaches and Bo Bo locos should pass sections like that ok as the wheel axles are close together. pacers as you say squeal coz they have just one axle at each end and the wheel does not hit the curve squarely. i've often wondered about a Co Co wheelset encountering problems on a curve and what about the old 9F's with 2-10-0 arrangement. there must be side play on the centre axles surely to cope with curves. (model trains use this side play)
 

Old Timer

Established Member
Joined
24 Aug 2009
Messages
3,702
Location
On a plane somewhere at 35,000
Asks the question of which is the sharpest bend on the network ?? Syston North to East looks pretty sharp and the curve from Grand Jn to St Andrews is fairly tight.
I cannot bring Syston to mind but having relaid the Bordesley Curve some years back, I can confirm that it has to have specially precurved rails fabricated on panels due to the radius of the curve.

I recall that St Andrews to the Grand has a larger radius, but without the Engineers drawings I could not say authoratively.



coaches and Bo Bo locos should pass sections like that ok as the wheel axles are close together. pacers as you say squeal coz they have just one axle at each end and the wheel does not hit the curve squarely. i've often wondered about a Co Co wheelset encountering problems on a curve and what about the old 9F's with 2-10-0 arrangement. there must be side play on the centre axles surely to cope with curves. (model trains use this side play)
We dont have sideplay on the wheels. What we do is to widen the track gauge on tight curves, thus reducing the degree of flange contact against the high / outside rail.
 
Last edited:

Wyvern

Established Member
Joined
27 Oct 2009
Messages
1,573
Right now. The OP mentioned a item in Wkipedia about hunting oscillation. Following posts have written mainly about flange contact causing squeal. This is at low speed and there willl be PSRs in place to prevent derailment.

This is not quite the same, however. Leaving out all the algebra, if you look at the conicity of the tread in the link I posted the moving wheelset will tend to stabilise with bearing contact with the rail in the centre due to the cone shape of the tread.

Now, as I understand it, sideways movement will cause one wheel to rise on its cone, the other to fall, Either the wheelset will correct itself or there will be a sliding contact due to the circumferential speed being different.

When it enters a curve the outside wheel will have to travel further than the inside one, it will also be tending to move outwards, the outside wheel rising up the shape of the tread where the circumference is greater, the inside wheel moving down the shape of the tread where the circumference is less. Thus the difference in distance travelled tends to be evened out and reduce the tendency for sliding.

The amount of this movement will be controlled by controlling the speed of the train relative to the tightness of the curve and the amount of cant, or superelevation, in the track

However at near to the maximum speed you have a very delicately balanced system, where any perturbation such as a defect in the track can set it to swinging.

This of course is describing a wheelset on its own where it will tend to position itself parallel to a radius of the curve. However wheelsets are mounted in pairs in a rigid frame and so are kept aligned to the frame. Thus we have another source of instability which is what link http://www.railway-technical.com/whlbog.shtml describes.

All this was known about to railway engineers but until the middle of the twentieth century they simply bulit a design and ran it to find out the speed it would become unstable and set a speed limit somewhere below. Thus passenger trains generally did not exceed 80mph, and the typical four wheeled goods wagon less than 40mph. I once caught a train out of St. Pancras. It was about half an hour late and set out to make up time. I dont know what speed it was doing but the noise in the coach where I was was deafening. It isnt often I'm scared but I was then!

The first task facing British Railways in 1966 was to work out exactly what the wheelset on its own was doing and, more importantly, quantify it, and then design a computer to describe the most stable tread profile.

The next stage was to study the behaviour of various designs of suspension and derive computer models for their design. This is what the High Speed Freight Vehicle The work that was done formed the basis of every succeeding rail vehicle from the Sprinters to the Eurostar. It was the first chapter of any succeeding textbook of railway vehicle design, just as much a milestone as the firetube boiler and blastpipe had been acentury before.
 
Joined
12 Dec 2009
Messages
634
Location
Carmarthen
instability and noise is something i remember as a youth travelling on those DMU's that were in use in the early 80's (before the 150 sprinters). Along the sea wall between Burry Port and Llanelli the wheels used to roar for a mile or so at 40 - 50mph or so. this may be the effect we're discussing or something completely different. it's an odd phenomenon. like a sort of plagueing resonance
 

Hydro

Established Member
Joined
5 Mar 2007
Messages
2,204
"Roaring rails" is caused by railhead corrugation, I believe.
 

25322

Member
Joined
25 Feb 2009
Messages
201
Location
Staffordshire
The flange on the wheel is designed to hit the rail. Its the flange hitting the rail that causes the change in direction hence why they go round corners.
The squeal can be down the the back of the wheel hiiting the check rail but most of the noise comes from but in most cases. The flange squeal is caused by the fact that all the weight/force is transferred to one wheel ands so grinds away at the rail and tyre due to friction. Lubricators are fitted on some curves.

The longer the wheel base the more friction occurrs as the vehicle goes round the corner. Hence why pacers squal more than mark 1 bogies.
 
Joined
12 Dec 2009
Messages
634
Location
Carmarthen
yeah the flange will make the train turn the bend but what we're saying is about the cone shape of the wheel surface that keeps the wheelset straight without the flange coming into play unless it is necessary like when a straight section goes into a sharp curve at moderate speed then the flange forces the train in that direction
 

Hydro

Established Member
Joined
5 Mar 2007
Messages
2,204
Truckman has it, the flange ought not to contact the rail to any massive degree unless it's a very tight curve or S&C. Remember modern plain line rail is laid with the rails inclined towards the 4 foot at 1:20, to interact with the wheels conicity.
 

Old Timer

Established Member
Joined
24 Aug 2009
Messages
3,702
Location
On a plane somewhere at 35,000
The flange is NOT designed to run against the railhead edge. The wheel is slightly narrower between the flange front than the track gauge.

The inclination of the wheel tread and the railhead (1 in 20) causes the wheel to "centre" on the rail head, and we continue to achieve this be zero-ing the centrifugal forces acting on towards the outside of the curve by means of applying cant to the track.

On flat curves, and those of a small radius, we will install check rails to prevent the front of the flange making contact with the high/outside rail, The check rail is designed to be sacrificial, wear on the back of the flange has little impact on safe running.

The particular dynamic response of the vehicle to curving will relate to the precise wheel profile that is applied to the wheel, however at some point there will be a degree of slide which will depend upon a number of variable factors.
 
Joined
12 Dec 2009
Messages
634
Location
Carmarthen
yeah very interesting stuff. i can see exactly what old timer says. the 3rd rail is holding on to the inside of the flange to keep it in a position where the flange on the outer rail is held away from touching the curved rail. on my model rail i am always looking into why a wagon may derail on points and sometimes a 1/2mm error on the wheel gauge (or whatever the human eye can detect) is enough to cause one particular wagons' flange to catch the "frog". it is a living nightmare to try and alter this gauge to what is necessary to freely pass the points. (apologies for going off topic into model railways)
 

Hydro

Established Member
Joined
5 Mar 2007
Messages
2,204
on my model rail i am always looking into why a wagon may derail on points and sometimes a 1/2mm error on the wheel gauge (or whatever the human eye can detect) is enough to cause one particular wagons' flange to catch the "frog". it is a living nightmare to try and alter this gauge to what is necessary to freely pass the points. (apologies for going off topic into model railways)

Pretty much the same as the real thing to be honest. Striking the crossing nose normally deforms or breaks it off first, and over time it can wear nicely down to create a ramp that the wheel gets pushed up and over. The cover check rail which is laid against the opposite rail to the crossing ought to keep the wheels to the correct side to avoid this in the same way a check rail acts on a curve, but any gauge variation, particularly on the check rail, can prevent this. The same applies to the switch rails. If the switch rail that guides the flange off the straight path becomes worn enough, it forms a ramp the flange will just ride up and over. There are specific inspection regimes carried out regularly to check the condition of the rails within points and crossing layouts to avoid this.
 

Wyvern

Established Member
Joined
27 Oct 2009
Messages
1,573
The flange on the wheel is designed to hit the rail. Its the flange hitting the rail that causes the change in direction hence why they go round corners.

Not under normal circumstances. Flange contact is something to be avoided.
--- old post above --- --- new post below ---
The particular dynamic response of the vehicle to curving will relate to the precise wheel profile that is applied to the wheel, however at some point there will be a degree of slide which will depend upon a number of variable factors.

And it is that profile which was first studied in the sixties - and still generates many scientific papers. Try putting "railway wheel profile" into Google.

The whole aim is to profile the wheel, design the suspension and lay the track so as to keep the rail/wheel contact in the centre of the tread. Apart from the relatively rare tightly curved sections, if track lubricators are needed it is a sign of problems.
 
Joined
12 Dec 2009
Messages
634
Location
Carmarthen
although you would think that railway grade steel, as hard as it is, would be fine hitting the flange against the railhead on parts of the journey. on a few sharp curves encountered, why is it a great issue?
 

Hydro

Established Member
Joined
5 Mar 2007
Messages
2,204
Wear on the inside edge of the rail (gauge face) called sidewear can reach an angle at which the flange climbs the inside of the rail and over the top of the railhead.
 
Joined
12 Dec 2009
Messages
634
Location
Carmarthen
it's weird. the flange is a massive angle increase from the normal wheel cone. you'd never expect it to climb over the railhead under anything other than rollercoaster type changes in sideways inertia
 

Hydro

Established Member
Joined
5 Mar 2007
Messages
2,204
Or in conjunction with the myriad of other faults that a track can be subject to, e.g. tight gauge, twist, poor top to name but a few.
 

Old Timer

Established Member
Joined
24 Aug 2009
Messages
3,702
Location
On a plane somewhere at 35,000
it's weird. the flange is a massive angle increase from the normal wheel cone. you'd never expect it to climb over the railhead under anything other than rollercoaster type changes in sideways inertia
If you take a closed railway system which has curves on it, you will find over time that both the inside of the rail head and the front of the flange wear and their profiles will alter.

The angle at which to flange is relative to the rail head edge is known as the angle of attack and is carefully calculated. As wear occurs this angle will change until under centrifugal force, as well as other forces, the angle is sufficient for the wheel to climb up the inside edge of the rail head and thus lead the vehicle into derailment.

This is one of the reasons why closed routes will turn their stock on a regular basis, alternatively a route may have the geometry specially developed such that we equalise curvature, indeed one of the jobs I am working on is to achieve precisely that.
 

Hydro

Established Member
Joined
5 Mar 2007
Messages
2,204
This is one of the reasons why closed routes will turn their stock on a regular basis, alternatively a route may have the geometry specially developed such that we equalise curvature, indeed one of the jobs I am working on is to achieve precisely that.


The Lymington Branch is an example of this. Being a single track line that is pretty much a constant curve, the stock is occasionally literally turned around by travelling Brockenhurst-Eastleigh, Eastleigh-Fareham via Botley then Fareham-Brockenhurst via Netley. Turning the stock ensures the wear is distributed on the wheels.
 
Joined
12 Dec 2009
Messages
634
Location
Carmarthen
but if the flange does not contact the railhead on a normal basis, why does enough wear and tear occur on this flange to the extent that there are such preventative measures on the sharp curves?
 
Joined
31 Oct 2009
Messages
19
Location
waiting for the road somewhere......
Wear on the inside edge of the rail (gauge face) called sidewear can reach an angle at which the flange climbs the inside of the rail and over the top of the railhead.


Always wondered what Sidewear actually was!! (just keep seeing the SW markings when out mashing the sleepers and never bothered to ask, gathered from its name it was wear on the side of rails, but did not know its implications!!)

Cheers

V
 

MrC

Member
Joined
20 Nov 2009
Messages
218
The Lymington Branch is an example of this. Being a single track line that is pretty much a constant curve.....

Um - hate to differ and all that but the Branch is quite s-shaped - from Brockenhurst the major curves are L-R-L-R-R-L-L-R. I'd have thought (but would love to be corrected :) ) that wheel wear would pretty well even out. I always thought the turning trips were to even out wear on the doors as all platforms are on the same side.
 
Status
Not open for further replies.

Top