Recently one of our locomotive had a problem of Axle Breakage. The locomotive was in service for less than a year and the axle parted when train was pulling out from one of the stoppages.
What can be likely causes.
Rajeev
Almost certainly metal fatigue, and I'd expect the final break to occur precisely at a time of maximal torque.
On the other hand, if you had a progressive bending-stress fracture, it would also tend to fail at a time of maximal torque...
My father used to complain about how many so-called researchers did their statistical analysis of medical conditions with the hypothetical example "Studies show that most skiing accidents occur on the last run of the day". Regardless of the cause of the accident it's likely to be the last run of the day for the victim, isn't it? So that correlation alone doesn't really mean what you might think...* Same with the actual break here... we know it failed at maximal stress, but the mechanism predisposing it to fail might be very different (and, in fact, might well be a combination of multiple processes...)
First thing to tell us is where the axle failed. At one of the inside wheel seats? At the bull gear? At a journal?
Second thing to do is to provide some good high-resolution pictures of the face of the break, with the proper lighting and pre-treatment needed to analyze the type of metallurgical failure experienced. (There are some good online references that show different kinds of break and explain their significance; I won't recommend one over others.)
It is possible that this is a one-time failure due to an undetected flaw in the axle before or during fabrication. Something you might check is the QC and testing methods used by whoever fabricated the axle, and by your own people when doing inspections, to be sure they are correct in detail AND being carried out faithfully and correctly, with the results properly documented and reviewed.
* Yes, I know the researchers "probably" meant 'on the run latest in the day', where skiiers are more likely to be fatigued, developing incipient hypothermia, distracted by thinking about apres-ski, and so forth. But that is what they should have stressed... ;-}
An old auto mechanic friend of mine probably said it best. All other things aside, "It's metal. It could last for a thousand years, or it could break tomorrow. You never can tell."
Firelock76 An old auto mechanic friend of mine probably said it best. All other things aside, "It's metal. It could last for a thousand years, or it could break tomorrow. You never can tell."
That is why the various metal testing processes have been developed - to highlight signifigant parts that have potential catastrophic flaws - before they fail.
Never too old to have a happy childhood!
Rajeev Recently one of our locomotive had a problem of Axle Breakage. The locomotive was in service for less than a year and the axle parted when train was pulling out from one of the stoppages. What can be likely causes. Rajeev
From your name and due to the comparative rarity of 5000 HP locomotives, I assume you are talking about a Varanasi built EMD WDG-5 locomotive. This is a model GT50C MAC with EMD export type steering trucks.
The trucks, motors and axles would be the same as the much more common GT46C MAC. These have been successful and as an EMD design might be expected to be generally conservative regarding stresses.
My first thoughts would be to look for a manufacturing error in the axle resulting in a stress raiser in the area of the failure or perhaps some similar problem caused in the assembly of the axle.
In a locomotive that new you wouldn't expect the axle to have been touched since manufacture.
M636C
Another cause could be torsional oscillations between left and right wheel of the same axle. These oscillations can be heard standing track-side when the loco is running by. On your loco I would expect the frequency of these oscillations to be in the range of 80 Hz.
The Austrian locos type Rh1043 (ASEA design, same as Swedish Rc2 type) were reported to have one broken axle after some months of operation in the Alps - this was in 1971 or 1972. Operation in the Alps might have been heavier (steeper gradients, tighter curves) than in Sweden, where this type of loco had already been running in larger numbers.
The oscillations are induced by changing adhesion due to oscillating slip. This is possible, if the axle is operating at a steady slip velocity of more than 1.5 to 2 km/h (1 to 1.5 mph). The oscillations are not well damped - the torque in the wheelset axle can rise to 3 to 4 times the value you need for the full starting tractive effort.
Normally, the bending of the axle is responsible for about 90% of the stresses in the axle. But if the axle torque is more than 3 times the value for starting tractive effort, then it increases the resulting bending stress significantly. This can overload the axle - not, if it happens the first time, but after a larger number of such occurences, like starting a heavy train at medium adhesion levels (25 to 30 % adhesion typically).
If a loco is controlled in such a way that an axle (or a whole truck) can operate at sustained axle slip, it should have some countermeasures against these oscillations. The oscillations can be measured with the speed sensors of the traction motors. The slip control algorithm should turn down the allowed wheel slip if the oscillations start to rise. By such means, the oscillations were stopped within about half a second.
There is another sign you should look at. Heavy oscillations can even lead to a twist of the wheel against the axle. So, if you have markers on the wheel face and on the axle adjacent to it, you can detect a wheel twist when you find theses markers not in line.
If torsional oscillations lead to the break of an axle, I would expect the axle to break near the wheel seating. It should clearly show signs of metal fatigue.
oldfireboxIf torsional oscillations lead to the break of an axle, I would expect the axle to break near the wheel seating. It should clearly show signs of metal fatigue.
In fact, I would expect the pattern of metal fatigue to indicate torsional failure if this is the case..
If there are pictures (either original or as-etched) of both faces of the break, it would be good to see them.
Our community is FREE to join. To participate you must either login or register for an account.