One description of wheel squeal is included in the following article:
http://www.interfacejournal.com/features/01-07/wheel_noise/1.html
Note description in caption to Figure 2 which makes note of "...the bending resonances of the wheel.." Now my terminology may have not been completely accurate, but this indicates that the wheel is being bent (or twisted or torqued -- I'll have to check the dictionary on that last one), and unbent. And this is the vibration that causes the noise.
http://www.kelsan.com/reports/Publications/Published%20Papers/9.%20040301_DTE-MS-JK_Railway%20Noise%20and%20The%20Effect%20of%20TOR%20Liquid%20FM%20_CM%202003-Wear_Paper%20Published_Rev%200.pdf
The above paper says:
"Wheel squeal originates from frictional instability incurves between the wheel and rail. Stick-slip oscillations(more accurately referred to as roll-slip) excite a wheelresonance; the wheel vibration radiates noise efficiently.The accepted model involves TOR frictional instability underlateral creep conditions leading to excitation of out ofplane wheel bending oscillations. These are radiated andheard as squeal." (emphasis added)
Now to the size of the wheel contact area; in Railroad Engineering 2nd. Ed., William W. Hay, page 518, states that the wheel contact area for a 36 inch diameter wheel to be 0.210 sq. in. The diameter of a dime measured as 0.700". The area of a dime works out to be 0.38 sq. in. So, it appears that the wheel contact patch is smaller than a dime. The book doesn't say the initial conditions, but I assume the area given is for a new wheel on new rail. And wear in one or both of the two will alter this figure.
Take care.
marknewton wrote: 130MM wrote: When the flange of the wheel contacts the outer rail in a curve, the wheelset is slightly torqued or twisted. It then snaps back to normal. This happens continuously as the wheelset tracks around the curve, and at a high frequency; which causes the sounds that we all hear.Do you have any evidence to support this claim?
130MM wrote: When the flange of the wheel contacts the outer rail in a curve, the wheelset is slightly torqued or twisted. It then snaps back to normal. This happens continuously as the wheelset tracks around the curve, and at a high frequency; which causes the sounds that we all hear.
When the flange of the wheel contacts the outer rail in a curve, the wheelset is slightly torqued or twisted. It then snaps back to normal. This happens continuously as the wheelset tracks around the curve, and at a high frequency; which causes the sounds that we all hear.
Do you have any evidence to support this claim?
cacole wrote:I've heard it said that if our models had truly prototypically-sized wheel flanges we couldn't keep anything on the track.
From the far, far reaches of the wild, wild west I am: rtpoteet
130MM wrote:When the flange of the wheel contacts the out er rail in a curve, the wheelset is slightly torqued or twisted. It then snaps back to normal. This happens continuously as the wheelset tracks around the curve, and at a high frequency; which causes the sounds that we all hear.
When the flange of the wheel contacts the out er rail in a curve, the wheelset is slightly torqued or twisted. It then snaps back to normal. This happens continuously as the wheelset tracks around the curve, and at a high frequency; which causes the sounds that we all hear.
130MM wrote: Dave-the-Train wrote: Far more scarey than the smallness of the flange is the actual amount of wheel/rail contact... which at any one times for each wheel is something a bit less than the area of the palm of your hand.(stuff snipped)When curves are encountered the flange may be deflected with the movements of the cones as Tomikawa says with the resulting squeals of steal on steal. (stuff snipped)Centifugal force also comes into this and G forces. The thing that you do not want is for the G force on the outer wheels on a curve to become zero or less. If that happens you go straight on instead of round the curve.I have always been told that the wheel-rail contact area is the size of a dime.When the flange of the wheel contacts the outer rail in a curve, the wheelset is slightly torqued or twisted. It then snaps back to normal. This happens continuously as the wheelset tracks around the curve, and at a high frequency; which causes the sounds that we all hear.While zero G on the outer rail is bad, it can happen only when the train is moving slowly. What's really bad is the inner rail having zero g. This means the whole train rotates to the outside of the curve around the rail, i.e. ends up on its side to the outside of the curve. This is known as the overturning speed. And it requires the train to be travelling much faster than the normal design speed.DAW
Dave-the-Train wrote: Far more scarey than the smallness of the flange is the actual amount of wheel/rail contact... which at any one times for each wheel is something a bit less than the area of the palm of your hand.(stuff snipped)When curves are encountered the flange may be deflected with the movements of the cones as Tomikawa says with the resulting squeals of steal on steal. (stuff snipped)Centifugal force also comes into this and G forces. The thing that you do not want is for the G force on the outer wheels on a curve to become zero or less. If that happens you go straight on instead of round the curve.
Far more scarey than the smallness of the flange is the actual amount of wheel/rail contact... which at any one times for each wheel is something a bit less than the area of the palm of your hand.
(stuff snipped)
When curves are encountered the flange may be deflected with the movements of the cones as Tomikawa says with the resulting squeals of steal on steal.
Centifugal force also comes into this and G forces. The thing that you do not want is for the G force on the outer wheels on a curve to become zero or less. If that happens you go straight on instead of round the curve.
I have always been told that the wheel-rail contact area is the size of a dime.
While zero G on the outer rail is bad, it can happen only when the train is moving slowly. What's really bad is the inner rail having zero g. This means the whole train rotates to the outside of the curve around the rail, i.e. ends up on its side to the outside of the curve. This is known as the overturning speed. And it requires the train to be travelling much faster than the normal design speed.
DAW
Um... I don't think so...
If I recall correctly a dime is about the size of our old sixpence. If the contact were that small a wheel would not spread a penny (old/real style - or modern 2p which is almost as big but not as thick) out to about twice its original size. My Brother still has one that he had squashed as a scout at Nine Elms Loco Depot back in the 60s. (In those days that was the example they used as to why you shouldn't trespass on the track... that and a clout round the ear if you got caught)
No. The wheelset is not significantly torqued in this manner. Constant action of this kind would fatigue the steel incredibly fast and break things. The wheelsets (which are always a minimum of two parallel pairs) are designed and set for the optimum operation in service using all the latest physics and digital technology.
I think that I pointed out that the actual settings used are a compromise between getting the wheelsets (in fixed pairs or larger groups) going both straight and round designed curves at designed speeds in the most efficient manner overall.
The squeel is produced by the wheelsets wanting to go straight on and resisting deflection. This causes the leading part of the wheel arriving on the railhead to make much more contact than normal and the noise is the result of two pieces of steel dragging across each other... just the same as if you drag a sheet of half inch 8x4 steel across another steel sheet. [i.e the noise is made by friction and can be reduced by adding a lubricant].
Torque (as I understand it) is a twisting action (force) which would normaly occur within a material... such as a shaft... such as the drive shafts in your car. Usually those shafts make no noise. Whenever I've known a shaft get out-of true it has resulted in a thrumming noise not a squeal. This isn't a direct comparison. When excess torques is applied a shaft sheers. Failing rail wheels do not sheer. Normally they either shatter or at least one of the componenets seperates from the others. (i.e. the tyre, web or hub break away... the whole wheel can break away from the axle as well). In a vertical wheel a sheering failure would be vertical. I've never heard of this occuring. But I have seen evidence and the results of all those I mention. Again... the noise you get when a drive shaft bearing squeals results from friction between the shaft and a bearing that has insufficient lubrication.
Also... A wheel rolling on a rail is exactly like a wheel guiliotine. The leading edge as it descends is like a knife blade. This is why you don't want to get any part of you in its path.
I suspect that I could go on further about the differences that would occur between friction causing the noise and torque when lubrication is applied by track flange greasing gear.
I fear that the G comments are way out.
Within the earths atmosphere there are only two ways (that I know of) of achieving zero G (weightlessness) or negative G (upward force without motive power). One is the parabolic curved flights used by NASA to train astronauts and the other is to whang an object round in a circle at sufficient speed as in a centrifuge (see James Bond... Moonraker IIRC). The result of this is in part to throw the object outward from the centre (see the water in your washing machine when spinning, a fairground ride, the wall of death etc).
As I understand it the forces within the object being spun become greater initially horizontally than the forces of gravity pulling it to earth. I believe that this means that the pull transfers from below the object up round the outside of the object on the outer side of the circle until it becomes a horizontal pull. this is not going to be a good thing applied to the side of a train at any time. I believe that the pull can move on up to the top of the circle round the object.
All this means that zero or negative G only happens to rail vehicles when they are travelling at speed and around curves. Unless some other force such as a side wind comes into play as in the Tay Bridge Disaster.
The example of negative G that occured near to my youth and where I later lived was a train smash at Eltham Well Hall in South London. Briefly a London bound train came onto a curve with a fixed soeed restriction of about 15mph at in excess of 70 mph. The HMRI report stated (among other things) that the G forces down through the wheels on the outside rail became negative (to -3G IIRC) with the result that the wheels lifted, ceased to have a guiding function on the direction of the train... and the train went straight on. In that case the train didn't fall on its side but went airborne into a disused coal yard some twenty plus feet below. There were no recogniseable parts left of the first passenger car behind the loco. One of the Inspectors I later worked with was one of the first to arrive at the scene. He told me that the devastation was like Dresden in 1945 in a small area.
I can't really figure out what is going on in 130MM's description.
It appears that he is saying that loss of downward pressure on the inside rail causes the curved profiled face of each inside -of-the-curve wheel to slide initially vertically and then sideways in a rotating motion around the curved profile of the rail head. This would certainly tip a car onto its side. It would land inside the rail though, toppling inwards. [But why wouldthe movemnet upward allowed by zeroG switch
Not in my experience. I have some 1/43 scale models with dead-scale tyre profiles and flanges. Providing that the track they run on has been properly laid, they run beautifully.
That chestnut probably originated as much in response to our generally sloppy (when compared to the prototype) trackwork as to unprototypical wheel profiles. The RP for Proto:nn modelers lists a scale flange depth of about 1.2 inches, which is very close to prototype, as well as a 1:20 tread taper. Of course, Proto:nn trackwork requires much more effort to install and refine than flex track. Consider also that our models are unprototypically light, which contributes to derailments regardless of flange depth. If mass scaled properly, we'd have fewer derailments, but it would be difficult to put anything on the track, and once there, our can-motor-powered locos would be unable to move it.
cnw400 wrote: How often is the measurement made? Is it routine maintenance, or the result of observation or an occurence? Where does the car or locomotive get sent for the measurement? How is the measurement done (manually, by machine, high-tech laser/ computer,...)? How much down-time? Tolerances? More info! Sorry for the Journalism 101 approach, but I think this is really interesting stuff, and I obviously don't know anything about it. Thanks.
How often is the measurement made? Is it routine maintenance, or the result of observation or an occurence? Where does the car or locomotive get sent for the measurement? How is the measurement done (manually, by machine, high-tech laser/ computer,...)? How much down-time? Tolerances? More info!
Sorry for the Journalism 101 approach, but I think this is really interesting stuff, and I obviously don't know anything about it.
Thanks.
In the UK there are stock inspections all over the place all of the time. Most times anyone not informed wouldn't know they were going on. Passenger stock gets examined at least once every 24 hours.
Apart from the modern technology, which often does things automatically, examiners learn what to look for... and where they don't already know a fault they know what looks different... the trick is to pick up on it and not just breeze by.
It's the same as when a cop walks down the street. He/she can spot regular trouble. A wrong car will stand out. When he wanders into his doughnut bar (she wouldn't she's watching her figure)(and very sensible she is too) () he has to catch on to a robbery going down fast not get blown away.
There used to be "wheel tappers". These men (it was a male job back then) examined more than just wheels but they tapped each wheel with a stick or hammer and knew what it was doing /how it was by the sound. they could do this in a busy/noisy yard or station. Experience was the big thing.
Also, anyone can hear a wheel that has got damaged with "flats"... it thumps along. Bad ones can be heard miles away.
As Tomikawa says it is not the flange that holds the rail on wheel on the rail. The physics of four cones back to back (thick ends together) in two pairs in a fixed frame is what keeps a rail vehicle on the track. that and the vehicles weight.
Suspension also comes into it. Shocks imparted by rail joints and (if things are wrong) parts of switches are dampened and absorbed by the suspension. This acts against a vehicle being bounced off the track.
You think that that is small? it's changing constantly as the wheel rolls... at speeds up to 125mph and even 140mph in some places. Now that is "interesting".
If memory serves me as a wheel shop foreman (more decades ago then I care to count) wheel flange was close to 1 inch high when new or re-profilled, and not to be over and condemed at 1 1/2 inch in height. Dont confuse this with flange thickness as a thin flange was condemed at 1 inch in thickness, as this thinner flange could split a swich
Height is determined with a wheel gauge that is measured from the back (flat back) to the center of the 1 in 20 taper tread.
As .the rail wears the taperd tread, it will in time become concaved. Again measuring the wheel in its worn condition, the gauge drops down further as it rests in thes center part of the tread where the metal is worn away, as we now look at the gauge the flange is higher, compared to the tread, thus it will be condemed at 1 1/2 in height. ( the flange did not grow in physical size, as it is related in measurment to the center of tread.)
Clear as mud...LOL ...John
csmith9474 wrote: FJ and G wrote:It's a scary thought riding a typical VRE train to work that 1" inch on about 64 wheels is all that is keeping the train on the tracks. I'm sure someone can describe the physics in detail, but it's still scary; esp. since the train bounces up and down a lot. I wonder how much contact the flanges actually have with the rail on straight track. I was under the impression that contact is minimal, and engineered to be that way.
FJ and G wrote:It's a scary thought riding a typical VRE train to work that 1" inch on about 64 wheels is all that is keeping the train on the tracks. I'm sure someone can describe the physics in detail, but it's still scary; esp. since the train bounces up and down a lot.
I wonder how much contact the flanges actually have with the rail on straight track. I was under the impression that contact is minimal, and engineered to be that way.
What keeps the train centered between the rails is not so much the flanges as the 3 degree taper in the wheel tread. On reasonably well maintained track with reasonable curves the flanges won't contact the railheads at all.
To understand how that works, just lay a paper cup on its side and push it. It will roll in a circle, not a straight line. If you can connect two such shapes with an axle and roll them down two parallel rails, they will center themselves so that the contact lines have identical radii and circumferences. A wheelset has two identical conical forms, and will roll down the track on identical contact lines on the conical wheel tread, not on the radius that forms the tread side of the flange.
On a curve, the outer wheel moves away from the center of the rails and the inner wheel moves toward the center. That establishes two new contact lines, the outside one having a longer circumference than the inner one. If the curve is so sharp that the conical shapes cannot absorb the difference the outer wheel flange will finally contact the rail - with a screech that can be heard by anyone within several hundred meters. That's where the prototype will install flange greasers.
FWIW, note that tie plates impart a 3 degree slant to the rails, so the wheel will make contact across the entire railhead. That reduces wheel and rail wear, and reduces the kind of stress that can lead to transverse fractures.
Chuck
Andy Sperandeo MODEL RAILROADER Magazine
Mark,
Thank you very much. That is exactly the information I was looking for.
Paul
On the 1:1 prototypel how deep are the flanges typically? I realize there is much wear, and that the treads are re-profiled, new tires put on ect., but in inches how deep(tall) is the flange. I am just looking for a figure plus or minus an eight or so. I also realize the tread itself is tapered at about a 2 degree angle, and the radius between the tread and the flange makes it hard to measure. I was looking at the FRA and other websites but could not find the answer.
Thanks