Now that we are past 2000, we are getting to the point where some operating steam is nearing 100 years of age.

I am curious. Is there period of time that would pass that would cause steel/other metals to become structurally unstable? Would it matter if the steel was subject to operational stress/strain? If it would be subject to instability over time would there be some way to identify the degradation? Finally, would matter when the steal was manufactured?

Thanks in advance.

superheater@beer.com

Steel after a certain number of loading cycles succombs to metal fatigue. There is no set time limit or expiration date on a piece of steel, but after so many reversals of loadings the steel will fail. Just think of trying to break a piece of steel wire. Bend it back and forth so many times and then it snaps. Same with engine frames, boiler plate, rods etc. except it takes much longer to happen.

I don't think there is a way to predict or forecast when metal fatigue in a piece will occur. Computer models can predict to some extent, but for something that is old and has had an unknown number of load reversals it would be hard to predict.

I'm no expert, just my take on the matter.


jrowlands@neo.rr.com

WOW: that would be a rather large volume ... most of which can be picked out of the older editions of "Making,Shaping&Treating of Steel" by USS. Railroads used a full range of Iron & Steels over the years and different methods of manufacture affecting the properties of the product. We already saw a short blurb on Russia Iron a spectacular form of sheet iron. Before the advent of the Bessemer Converter production, steel was made by the crucible process which was limited to small amounts. Within the iron family you ranged from rather brittle grey iron to more flexible malleable iron to wrought iron. This was generally produced in cupola furnaces either by charcoal or coke depending on the location and date. I always assumed that parts that required a higher strength were made from forged crucible steel while the more mundane stuff was made from wrought iron, boilers etc. Forgings were a pretty standard high quality product while castings especially first tries were subject to a lot of guesswork for soundness. The advent of the Bessemer convertor ushered in a new era of cheaper steel used in many new applications to replace wrought iron. The major drawback to this product was high nitrogen caused by blowing air thru molten iron. This produced a brittle product that engineers had to be careful of but it was much cheaper than either crucible steel or wrought iron. The next advance was the Open Hearth furnace where steel was produced from iron and/or scrap. This produced a very uniform product with a more defined analysis. At the back end of the shop work was being done on ingot soundness. Crucible steel was what is known in the trade as "killed" steel that is gasses dissolved in the steel were removed by the time of the process while Bessemer steel was full of trapped gasses that left voids in the ingot that had to be rolled of forged out. Open Hearth steel could be made in a full range of oxygen levels depending on the end use. How well the ingot was made and the subsequent rolling practices make a difference in the end quality of the product. Having large quantities of uniform analysis steel opened the door for larger higher quality castings of great strength like locomotive and truck frames. Malleable iron could still be used for things like pumps and cylinders because of the abilty to take on weird casting shapes. Since quality control was limited the RRs depended on failure mode in operating to correct problems. Bessemer Rail was quickly found too brittle for heavy use and was replaced by open hearth product by edict. Accidents and failures were corrected by reengineering or beefing up already heavy safety factors. All of these products can fail by fatigue and fatigue life can be directly affected by internal defects (dirt, blowholes, porosity, corrosion) as well as external defects. In general the materials chosen and the sections used were based on a heck of a lot of experience by the RR people but each piece of equipment did have a given life. Using modern non-destructive testing methods any part can be explored internally for defects that will give you some idea of life expectancy ... like the new boiler UT requirements. One of the givens in this business is that all materials will seek the lowest energy state. Thermal cycling and operating conditions will cause all stresses to be relieved. Boilers will stretch, frames will sag, springs will break all on an unpredictable schedule. To answer the question of life expectancy I refer you to the nearest museum containing body armour from the middle ages. Here we see custom made iron and steel products, in a quiesent state, kept polished in a protected atmosphere (not out doing fan trips) that is quite ancient. This will definitely be the fate of any or all of the RR equipment we have today if we wish to preserve it "for the ages". In operation any of this equipment must be considered a comsumable and we beg the question at what place in the replacement cycle are we no longer looking at the original concept.

lamontdc@adelphia.net

> Steel after a certain number of loading
> cycles succombs to metal fatigue. There is
> no set time limit or expiration date on a
> piece of steel, but after so many reversals
> of loadings the steel will fail. Just think
> of trying to break a piece of steel wire.
> Bend it back and forth so many times and
> then it snaps. Same with engine frames,
> boiler plate, rods etc. except it takes much
> longer to happen.

> I don't think there is a way to predict or
> forecast when metal fatigue in a piece will
> occur. Computer models can predict to some
> extent, but for something that is old and
> has had an unknown number of load reversals
> it would be hard to predict.

> I'm no expert, just my take on the matter.

Cyclic loading of steal and fatique issues depend upon a lot of factors. Assuming a good chemistry and relatively low loadings a piece can have an infanant life. The cylcle you descibe is near the ultimate tensile strength of the material. Steel loaded to a quarter of this value can last for significant time periods. Aluminum, on the other hand, has a finite life no matter the loading cycle. The particular material at issue will determine a lot about its life. The life cycles of materials depends upon several factors both interal and external to the particular materials.


EPSW271@aol.com

> Ok, fair enough.

Does this mean that at some point frames and drivers become unusable from wear and tear- are we going to be reading about ow engine 99's frame broke with 300 passengers or how there was an "incident" because a driver's spokes gave way because of 1XX years of metal fatigue?


superheater@beer.com

Something to keep in mind is that building things with steel is still very new as compared to building thing with stone or bricks. Steel is at the very heart of thousands of buildings, bridges, and other structures.

Locomotives were build to have a lifecycle, after all the builders figured on selling more locomotives down the road. Just as the automakers have never intended to sell you a car that would last 100 years. There are industrial plants with equipment that is well over 100 years old and still operating. I know of one paper mill with boilers that are 80 years old. I know of another paper mill with operating paper machines that date to the 1870's.

Lifecycles can be extended and parts can be renewed. We only need to decide if the trade off for a living breathing locomotive is worth replacing every last part on the locomotive (or trolley, ship, or whatever you like). This of course is a matter of deep debate.

Tom Gears
Wilmington, DE

Forgotten Delaware
tom@forgottendelaware.com

The answer to your question could be yes if we don't look. Non-destructive tests will be used more and more to proof parts before they fail. UT on the boilers is only the first step. If you had the money you could X-Ray your frame and cylinder castings to see if there are internal defects and fix them before they break. Spoked wheels can be sand blasted (not shot blasted) and dye penetrant tested or magnafluxed for cracks. There are some interesting shots from England showing cracked drivers being tested and repaired. The plate frames of England are famous for fatiguing out but of course they are relatively easy to replace. Cast trucks of course stretch and must be realigned for Amtrak or other operations. And the beat goes on.

lamontdc@adelphia.net

When it comes to steam boilers I believe that corrosion plays a far more important role in diminished structual integrity than any metal fatigue can ever play. This is also so with many other components. Boiler failures as a result of fatigue failures were extremely rare though they did happen. The most common causes of boiler structual failures were improper operation and maintenance.
Roger Mitchell
Master Mechanic
Fort Collins Municipal Railway


Fort Collins Municipal Railway home page
n0mcr@netzero.net

You sound like you really know this stuff, thanks for the informative postings - I have always wondered to what extent does the normal heat cycle of an operating boiler actually heat treat the steel to reduce fatigue? It doesn't seem to me that we get hot enough to make a lot of difference in the outer shell, but too hot in the firebox, but just impressions. Might it be sensible to make baking the boiler after whatever repairs are made part of the 15 year program? 50 year program?

Dave

irondave@bellsouth.net

Dave:
I have a Met. degree and my experience had been in the manufacturing end, the time of which covered Bessemer, Open Hearth and BOP steelmaking, ingot production top pour and bottom pour a little of casting, heating, rolling and conditioning of billets and blooms and special shape and bar rolling.

The carbon steel used in most boilers is a forgiving material supplied from the mills in as rolled condition. It shows little effect of cold working (forming etc)and can stand a fairly wide range of temperature without changing its properties. The attached tensile strength chart (most boiler steels fall in around the dotted line) shows the tensile strength actually climbing up to about 500 deg. (blue brittle zone) and then falling dramatically at 1000 deg. When you get a crown sheet failure it is not only the flash in pressure but also the loss of tensile strength as the temp of the crown sheet rises. Things that cause early fatigue failures are stress risers like the corners of a square hole (which is why rivet are round) deep notches,areas weakened by corrosion (mentioned in the post above) and heavy temperature cycling. Boilers in steam locomotives were really designed to stay hot for longer periods in service then the normal museum operation so taking a longer time to to bring them to operating temps and a longer cool down period will reduce fatigue stresses. All this comes together in that you can rebuild normal boilers and do all the necessary things for repair and not worry about detrimental property changes in the structure. All bets are off however when it comes to "Alloy" boiler material. The reason so many problems were encountered when trying to repair boilers made from this product is that it is very sensitive to temperature and time. Heat this stuff up for welding and/or riveting and is becomes a different material, brittle and prone to cracking. This does not mean that it cannot be done but it will take working with a heat treating metallurgist and a facility to heat treat the whole boiler (big bucks).



lamontdc@adelphia.net

Thanks for more great information.

Dave

irondave@bellsouth.net

From the ICS book “Locomotive Frames and Cross-Ties”.

“The usefulness of a locomotive frame ends when cracks begin to occur in it at frequent intervals. A frame may crack several times during service but this does not mean that it should at once be renewed, because, with modern welding practices, the welds have been in many cases as good as the original frame, or better. However, a frame should be replaced as soon as it begins to crack continually after it has been in service for a number of years, because the stresses have then so affected the metal that the frames are no longer able to withstand the service for which they were designed.”

The frames of the M1 at the Railroad Museum have welds at about 1 foot intervals their entire length. Pretty well shot.


kelly@strasburgrailroad.com

> When it comes to steam boilers I believe
> that corrosion plays a far more important
> role in diminished structual integrity than
> any metal fatigue can ever play.

The exception to that would be the longitudinal seam of a lap seam boiler. Since by its nature it is an out of round shape, one that tries to become round every time pressure is applied, it is quite subject to fatigue failure over time. This is the reason the new FRA rules call for extra inspection of lap seams in boiler barrels, and many state codes have many restrictions on their use. Most locomotives in service today have butt seams in their barrels, which are not subject to this problem.

kelly@strasburgrailroad.com

> Might it be sensible
> to make baking the boiler after whatever
> repairs are made part of the 15 year
> program? 50 year program?

> Dave

Not such a good idea with a riveted boiler. All of the rivets would relax as much as the rest of the boiler, and you would end up with the worldÂ’s largest garden sprinkler. I think our best bet here would be to treat them as gently as possible in use to avoid thermal shocks. The usual fire up slow, cool down slow, and donÂ’t run the injector for too long at a shot.

kelly@strasburgrailroad.com

> The frames of the M1 at the Railroad Museum
> have welds at about 1 foot intervals their
> entire length. Pretty well shot.

You can only go so far with repairs to castings once they start to tear or separate internally. If you were to test the frames I bet you would find voids and be able to predict the next crack location. On a much smaller scale the steel castings that hold the lateral bearings in ge13031s axles were hammered out about 1/2" to 3/4" and cracked along planes of casting porosity in a poorly designed and cast sharp internal corner. To repair these it required heating and reforging the castings to original shape, grinding out the suspect corners (full of sand pits and voids) and reinforcing with weld. All visible cracks were V'd out to the base of the crack and welded. The backs of the aluminum-bronze bearings were shimmed out to original thickness and once we get some operating time to wear these in we will go to work on the front shims. These should last quite awhile in our limited museum operation and be stronger than the originals.


lamontdc@adelphia.net