another derailment
 

Spyke wrote the following in:




"According to sources at the crash site the incident was caused by
a broken rail, and a member of the rail staff who saw the rail
said it was "rusted" before it fractured."

Out of curiosity (and I accept there may be a perfectly good
answer to this), how can you tell a rail is rusted just by looking
at it? Being steel there's always going to be a fair bit of
corrosion on the outside, with only the tops being polished clean
by passing trains.

That's what I was wondering. It sounds a bit like someone saying "the
road was rough and covered in some sort of tarmac like material priar
to the accident".

--
message by Robin May, consumer of liquids
Hello. I'm one of those "roaring fascists of the left wing".

Hacker is to computer as boy racer is to Ford Escort.

another derailment
 

In article ,
(Spyke) wrote:



"According to sources at the crash site the incident was caused by a
broken rail, and a member of the rail staff who saw the rail said it
was
"rusted" before it fractured."

Out of curiosity (and I accept there may be a perfectly good answer to
this), how can you tell a rail is rusted just by looking at it? Being
steel there's always going to be a fair bit of corrosion on the
outside, with only the tops being polished clean by passing trains.
--
Spyke (Whose own tube tracks are thankfully made out of aluminium!)
Address is valid, but messages are treated as junk. The opinions I
express do
not necessarily reflect those of the educational institution from which
I post.

I wondered that, but from this mornings BBC updated site, it appears that
it was indeed a broken rail and that the rust was showing in the broken
end of the rail. Presumably cracked for ages and hence the rust. The rust
wouldn't have been seen from above.

"A rail worker at the scene told BBC News the steel rail was rusted three
quarters of the way through.

If correct, this would suggest it had been cracked for a considerable time
before it finally snapped under the weight of a train."


Roger

another derailment
 

In article ,
(Richard J.) wrote:

I see the Evening Standard's accuracy is at its usual level:

"between Hammersmith and Barons Court stations .... Passengers were
being
rescued from the tunnel ..."


--
Richard J.
(to e-mail me, swap uk and yon in address)


Could be possible if the traction current was discharged between Barons
Court and Earls Court substations. However there's no mention of
passengers being walked from tunnels on the BBC site and I assume they
would have mentioned it if it was true.

Roger

another derailment
 

Spyke wrote in message ...


"According to sources at the crash site the incident was caused by a
broken rail, and a member of the rail staff who saw the rail said it was
"rusted" before it fractured."

Out of curiosity (and I accept there may be a perfectly good answer to
this), how can you tell a rail is rusted just by looking at it? Being
steel there's always going to be a fair bit of corrosion on the outside,
with only the tops being polished clean by passing trains.

Quite right.
There's almost always some corrosion on rails (identified by the
typical orange colour, normally in the web of the rail). In very
unusual circumstances (less than 1 in 1000) pitting from corrosion can
lead to a rail defect/break. It takes close inspection by an expert to
spot the spot.
Rails are tested ultrasonically at regular intervals, and should
identify the defect - a small crack in the rail which will eventually
grow until it becomes a rail break. It takes quite a while for the
crack to grow (months, even years) - testing is every 2 months, to
pick up cracks before they become breaks. It is concerning that the
break occurred, since the testing regime should have prevented this.

another derailment
 

On 18 Oct 2003 01:40:09 -0700, (JDikseun) wrote:

Spyke wrote in message ...


"According to sources at the crash site the incident was caused by a
broken rail, and a member of the rail staff who saw the rail said it was
"rusted" before it fractured."

Out of curiosity (and I accept there may be a perfectly good answer to
this), how can you tell a rail is rusted just by looking at it? Being
steel there's always going to be a fair bit of corrosion on the outside,
with only the tops being polished clean by passing trains.

Quite right.
There's almost always some corrosion on rails (identified by the
typical orange colour, normally in the web of the rail). In very
unusual circumstances (less than 1 in 1000) pitting from corrosion can
lead to a rail defect/break. It takes close inspection by an expert to
spot the spot.
Rails are tested ultrasonically at regular intervals, and should
identify the defect - a small crack in the rail which will eventually
grow until it becomes a rail break. It takes quite a while for the
crack to grow (months, even years) - testing is every 2 months, to
pick up cracks before they become breaks. It is concerning that the
break occurred, since the testing regime should have prevented this.

May I suggest that it is not quite as simple as all that.

Generally there are 3 methods of fracture in steel. Ductile (shows
typically as a cup and cone fracture surface) Brittle (very grainny
surface) and stress corrosion (cracking).

Not the 'posh' name for it. Cracking is considered as a corrosion
defect.
Generally to initiate a crack , some form of starter is required. This
as stated above could be a pit, or a sharp change is shape (corner) or
any other surface defect. Then for best effects cyclic loading is
needed but not absolutely necessary. The high stress concentration
causes the material at that point to be in a different state from the
rest of the material which it has effectively stress relived. That
causes galvanic corrosion. The product of corrosion being bigger than
the parent material causing the stress to increase and the next bit of
the crack is initiated.

The starter concentrates the stress in the material to that point
which causes intergranular failure (steel being a granular structure a
bit like a sugar cube). This bit of failure then weakening the item a
bit further causing another little failure. The result being that
there is a wave form pattern on the failure surface, which when traced
back through the arcs pinpoints the starter.

So some corrosion of the surfaces can be expected. Where they have
been rubbing, which is usually the case, then that corrosion is
obviously cleaned off.

Now so far as examination and testing is concerned. It is always a
problem when one method is used and relied on. Each method has its
limitations. In my own field (Boiler and Pressure Vessel) we quite
often specify maybe three different exams.

Ultra-sonic is good, but:
1. It can not detect surface defects within up to 2mm deep. This is
because of interference by slight imperfections in the surface causing
spurious echos.

2. It can not detect a crack under compression very well, if at all.
This is because the system relies on passing ultrasound through the
material and listening for back echos from the surface of a defect or
crack. If the crack is closed under compression the sound will travel
through with little or no echo.

3. Angle of probe. Probes can be made with several angles so that the
sound is passed through the material at an angle. If a 90 deg probe
was used it would not pick up a crack. (They are usually used for
thickness testing). Likewise a 30 deg may not depending on the angle
of the crack, or a 45 or whatever. so several passes may be needed
unless the nature of the defect to be found is known. ie., orientation
etc.

We would usually back up ultrasound with Magnetic particle testing
(cover the surface with a background contrast paint and spray with a
magnetic fluid whist holding a strong magnet on the surface) or in
non-magnetic situations use Dye Penetrant. (similar sort of idea).

Simple hammer testing often highlights problems as with simple
standing back and looking.

In every case cracking is the second worst thing to try and find. Nice
and easy when it is gaping open for any of the methods, but when
closed or hidden away from direct sight or sound probe a shear sod.
Skip wave probing is a special art which most testing engineers will
not touch.

(The worst thing is called smooth corrosion whereby every part of a
surface equally wastes. Common in Coal fired furnaces. the surfaces
appear as if nothing is happening until you see light come through or
catch the steam when it comes flying past you or through)

Regards
Keith J Chesworth


www.unseenlondon.co.uk
www.blackpooltram.co.uk
www.happysnapper.com
www.boilerbill.com - main site
www.amerseyferry.co.uk