Crown Sheet Failure

The crown sheet was the ceiling of the firebox. Water in the boiler had to completely cover it. If not and it became exposed there would be a catastrophic failure and the engine would explode with tremendous force. Usually, with no survivors.

Sometimes the crown sheet would just collapse and water would douse the fire without a catastrophic explosion.  Some locos had fusible plugs that would melt to achieve this goal.  I think this is what happened on the Gettysburg RR in the 1990s, which led to increased oversight of steam locos by the FRA.  Hopefully Rich Melvin or someone who has in-cab experience will explain better.

Gettysburg RR Crown Sheet failure report:

https://www.ntsb.gov/safety/sa...ocuments/SIR9605.pdf

A deadly explosion of a steam tractor from a crown sheet failure years ago has spurred constant inspection of tractors at steam shows in Maryland.  We have to demonstrate proficiency at operating safely at every show.  A low water condition is very dangerous to your own health and everyone within a 100 ft radius.

When the crown sheet is not covered with water, the firebox heat will soften the steel and any weakness may rupture with an uncontrolled steam release.  The sudden loss of pressure within the boiler causes the remaining boiler water to instantly flash into steam.  Kaboom.

Tractor fire boxes are required to have a fusible plug in the crown sheet.  The metal in the plug should melt before the crown fails and allow steam to release at a rate slow enough to avoid flashing.

Re: low water ... other than inattention and failure of feed water injection, the biggest problem on steam tractors is the crud in the boiler that finds its way into small passages including the water level sight tube.  The crud can cause a false reading so we frequently blow down the tube.  Vigilance.

Ted S posted:

Sometimes the crown sheet would just collapse and water would douse the fire without a catastrophic explosion.  Some locos had fusible plugs that would melt to achieve this goal.  I think this is what happened on the Gettysburg RR in the 1990s, which led to increased oversight of steam locos by the FRA.  Hopefully Rich Melvin or someone who has in-cab experience will explain better.

Ted, I'm sorry, but your explanation is not quite correct. If the Crown Sheet fails, there will definitely be an explosion. If a Fusible Plug fails, there might be an explosion. They are two different things.

Water under pressure does not boil at 212 degrees F.  At 250 psi the boiling point of water is 401 degrees F. When a crown sheet fails, all that pressure is suddenly released. When the pressure is released, all that 401 degree water instantly turns to steam! That is what causes the violent explosion...the expansion of the water from a liquid to a gas.

A fusible plug is like a "soft plug" in an automobile engine block. If the crown sheet gets too hot, these plugs melt and theoretically allow water to spray into the firebox, thus putting out the fire. However, if there is no water above the crown to start with, there is no water to be sprayed into the fire! In this case, the steel in the crown sheet soon fails, with a resulting explosion.

Here's a video that will explain what a crown sheet is and where it's located in a steam locomotive.

This video, along with dozens of others, are available to all OGR Digital Subscribers in the OGR Video Digital Library.

Rich, thanks for the clarity. As I said, usually there are no survivors.

FORMER OGR CEO - RETIRED posted:

Here's a video that will explain what a crown sheet is and where it's located in a steam locomotive.

Hey!  Who was that handsome, clean cut young man narrating that most excellent video??  <grin>
He must have been in Kentucky, for he was standing in front of the C&O 2716!  <snicker>

LOL! That video was produced in 1996, back when I was young and skinny. It was shot in the Fort Wayne RR Historical Society shop in Fort Wayne. We had the 2716 on lease that year and ran it for a few trips.

The video may be old, but the information is timeless.

OK.  Suppose you have two locomotives of the same class.  One is at the ATSF Station in Galveston TX.  The other is at Union Station Denver.  Galveston is at 6 feet above sea level in the Strand.  Denver is a mile above sea level.  Would the locomotive in Denver produce less power because the atmospheric pressure is lower?  The pressure/boiling point relationship given above is based at sea level.   Still risk of explosion in case of boiler/firebox failure.  

Dominic Mazoch posted:

....  The other is at Union Station, Denver.  ...

i smile every time i hear about the "mile high city"...  at 10,430', Leadville, Colorado is nearly TWO miles high... the AVERAGE elevation for the entire state is 6800 feet, 1¼ miles high.

Dominic Mazoch posted:

OK.  Suppose you have two locomotives of the same class.  One is at the ATSF Station in Galveston TX.  The other is at Union Station Denver.  Galveston is at 6 feet above sea level in the Strand.  Denver is a mile above sea level.  Would the locomotive in Denver produce less power because the atmospheric pressure is lower?  The pressure/boiling point relationship given above is based at sea level.

The inside of the boiler does not know or care about the altitude of the locomotive. It is the internal pressure in the boiler that does the work. The physical stress on the boiler shell in Denver might be a tiny bit higher because of the diminished pressure outside the shell, but I doubt that it could be measured or quantified.

The pressure/boiling point stats I cited above are absolute pressure numbers. Again, the external altitude of the boiler makes no difference in those internal pressure numbers.

Heres and old forum thread with interesting discussion regarding the photo below :

 Fireman Don't Let Your Crown Sheet Go Dry.

Rich,

Every year when we do our inspection, I'm usually on top of the boiler removing the steam dome lid. It's held in place by 20 studs and nuts. When we remove the lid, you can see how the studs have stretched. I've often wondered what would happen of that lid gave way while under pressure. Would the water turning to instant steam just shoot out through the open lid like a geyser, or would the all the water in the boiler, even at the bottom, turn to steam and explode outward in all directions?

That locomotive is definitely having a bad hair day.

Hopefully, since it appears the brunt of the explosion went forward, the crew survived.

Andre

Dominic Mazoch posted:

OK.  Suppose you have two locomotives of the same class.  One is at the ATSF Station in Galveston TX.  The other is at Union Station Denver.  Galveston is at 6 feet above sea level in the Strand.  Denver is a mile above sea level.  Would the locomotive in Denver produce less power because the atmospheric pressure is lower?  The pressure/boiling point relationship given above is based at sea level.   Still risk of explosion in case of boiler/firebox failure.  

A long time ago, I remember reading about some high altitude railroads had trouble with the power output of their diesels when they converted over from steam; ie, steam wasn't affected by altitude but internal combustion engines are. 

Thanks for the great info and great video. I'm sorry to say it, but in general steam locomotives do not seem to have been an especially safe mode of....uhhh...locomotion. Their sex appeal is, of course, timeless. But their designers really ended up pushing the safety envelope. I think. I hope everyone interested in a thread like this has read 

laming posted:

That locomotive is definitely having a bad hair day.

Hopefully, since it appears the brunt of the explosion went forward, the crew survived.

Andre

Unfortunately, no.

"The engine involved was a C&O T-1 class (2-10-4) No. 3020
The incident was a massive boiler rupture, apparently due to crown sheet failure. 
Three crew killed. 
Hauling coal to Columbus from Russel KY. 
The train was 156 cars and 12,884 tons
Running about 20 mph. 
Track was a .18% grade. 
Nearby residents thought the explosion was just thunder, as there was a good rain storm going at the time. 
The explosion blew out the superheater tubes out the smokebox. I have here a picture that shows a tangled mass of superheaters.
There was no time for the crew to do anything, so the throttle was still wide open, brake valves in the running positions, and reverser was full in the corner. 
This locomotive was working very hard. 
The engineer, fireman and front, or head end, brakeman were all killed, but not instantly. 
The engineer was lifted out of the cab later, the fireman was found trackside nearby, and the front brakeman was found walking back down the grade. 
About 100 yards away a fence was partially destroyed by the force of the explosion. 
The front brakeman survived long enough to tell something about the incident. 
In the ICC report, it's stated that the force of the explosion tore the rear of the boiler from the slides, and the right guide yoke waist sheet. The rear of the cab was twisted upward and the front of the cab rearward. 
The cab deck was bulged upward and the ash pan was blown out. 
Smoke box front was blown outthrottle box and superheater header were broken from the dry pipe connection in the smoke box and the superheater units were blown forward into a fan shaped exit from the now open smoke box. 
The feedwater heater was found about 345 feet ahead of the engine, and other parts were found in a 150 foot radius."

Also in the ICC report is a statement that the injured brakeman said several times that he "knew it was going to happen" and "the water was too low." 
Also, very telling, is "I told him (the engineer) that he had water and to put some in the boiler." And, "The water was gone!"
The brakeman then stated to his parents at his side in the hospital that "He (the engineer) ran for ten miles on low water."

Railroading has a way of visiting grave consequences to those that lack of diligence to one's responsibilities.

Andre

Just went through the old thread posted above, and found it very interesting despite some attempts at derailing the thread...

There was one photograph of a letter regarding testing the PRR did in 1868 to try to get a crown sheet to fail on purpose. Does anyone have any suggested reading on how the current best practices for steam boiler operation came to be through the 19th and 20th centuries? There seems to have been a significant amount of trial and error to get to today's standards...

RickO posted:

Heres and old forum thread with interesting discussion regarding the photo below :

 Fireman Don't Let Your Crown Sheet Go Dry.

 

That's actually a rare failure of the front flue sheet.  If the crown sheet failed, the boiler would have likely left the frame.

Crown sheet failure due to inattention:

Rusty

Rusty Traque posted:

That's actually a rare failure of the front flue sheet.  If the crown sheet failed, the boiler would have likely left the frame.

Crown sheet failure due to inattention:

Rusty

 FWIW, I posted the ICC report above stating it was a crown sheet failure. Regardless, its horrific.

smd4 posted:

Rich,

Every year when we do our inspection, I'm usually on top of the boiler removing the steam dome lid. It's held in place by 20 studs and nuts. When we remove the lid, you can see how the studs have stretched. I've often wondered what would happen of that lid gave way while under pressure. Would the water turning to instant steam just shoot out through the open lid like a geyser, or would the all the water in the boiler, even at the bottom, turn to steam and explode outward in all directions?

Check the studs with a thread gauge.  If they have actually stretched - replace them.  If you have concerns insist that the studs are checked for cracks by an ultrasonic testing technician.  

Most likely you would have a leak at the gasket prior to anything else.  But if all the studs failed the lid would get launched like a mortar,

The pressure coming out would be a terrifying fountain of water and steam.  The sudden pressure change would likely destroy or damage most of the tube and in the process damage the flue sheets and staybolts.  In short, a catastrophic loss of the lid would likely condemn the boiler.

Remember when steam expands into the atmosphere it generally takes up 1700 times the volume it had when under pressure.  So a boiler that as an example held 400 gallons of water or roughly 53 cubic feet, when released that same amount of water now instantly wants to occupy 90,100 cubic feet.  

Safety is no accident, always maintain a questioning attitude!

RickO posted:

Rusty Traque posted:

That's actually a rare failure of the front flue sheet.  If the crown sheet failed, the boiler would have likely left the frame.

Crown sheet failure due to inattention:

Rusty

 FWIW, I posted the ICC report above stating it was a crown sheet failure. Regardless, its horrific.

After reading that report, I said to myself what a third rate operation.  The sight glass valves were so caked up with deposits there's no way they were reading correctly.  Gauge missing, no way to know if the injector was putting water in the boiler.  Bad record keeping, lack of training, lack of maintenance.  So lucky no one died or more hurt.  The rules that came about after the accident I have a found new respect for

"Just went through the old thread posted above, and found it very interesting despite some attempts at derailing the thread..."

I must be dense. I'm not seeing any derailment of this topic. Water boiling point vs pressure of same, elevation considerations, inspections, discussion about other crown sheet failures and such is what I'm seeing.

Steve aka SMD4:

I'm a'feared the boiler would instantly turn to steam and thus: Ka-boom.

Take care of yourself out there. Railroading is railroading, regardless of what you're hauling or why.

Andre

Dominic Mazoch posted:

OK.  Suppose you have two locomotives of the same class.  One is at the ATSF Station in Galveston TX.  The other is at Union Station Denver.  Galveston is at 6 feet above sea level in the Strand.  Denver is a mile above sea level.  Would the locomotive in Denver produce less power because the atmospheric pressure is lower?  The pressure/boiling point relationship given above is based at sea level.   Still risk of explosion in case of boiler/firebox failure.  

Pressure gauges in a locomotive boiler show "gauge" pressure, which is the pressure in excess of the local atmospheric value. The "absolute" pressure includes the atmospheric pressure and therefore is higher than the gauge pressure. Stresses in the boiler are dependent on gauge pressure - which is the difference between the absolute pressure and the atmospheric pressure. Safety valves operate based on gauge pressure. Boiler structural failure would occur at a particular value of gauge pressure and would not be affected by altitude. Having said that, the steam temperature at a particular value of gauge pressure does vary with altitude - and is higher at sea level than at altitude. Therefore, at least in theory, more energy would be released in a sea level failure than at altitude.

MELGAR

I've often wondered how much and how long the crown sheet would need to be heated without water covering its top surface for it to finally soften and break way of the crown bolts and implode on itself. Not saying that this is something I would love to try of course, but I believe the Pennsy or some other major railroad during the steam era actually experimented with this by purposefully overheating the crown sheet with a low water scenario and studying to see what would happen over time to finally make the crown sheet fail. In fact if I recall correctly they even injected "cold water" on the overheated crown sheet to study if, in theory, it would instantly cause a boiler explosion from the "cold" water hitting the red hot crown sheet, vaporizing, and imploding the crown sheet. And, again if my memory serves me right, nothing happened. Hopefully Kelly Anderson or someone familiar with this experiment that I am thinking of can chime in. 

Here is a link to a thread on Frisco.org concerning a boiler explosion of a locomotive in 1933:

http://www.frisco.org/shipit/i...ark.5035/#post-36324

WHY it happened is not clear (and debated among folks who suggest inexperience, fatigue, and violation of Rule G), but WHAT happened is pretty clear.  One of the 1100-class Ten-wheelers in St. Louis suburban service was tied up at Valley Park, MO, overnight under that care of an engine tender, who, for whatever reason, was not sufficiently vigilant.  He entered the cab late at night, discovered no water in the sight glass, and turned on the injector, which apparently sprayed cold water into the boiler and onto the bare crown sheet.  The resulting explosion ripped the superstructure off the frame of the locomotive, and threw the boiler some 50 - 75 feet down the line, the cab some distance beyond that, and the body of the engine tender some distance further yet.

palallin posted:

  He entered the cab late at night, discovered no water in the sight glass, and turned on the injector, which apparently sprayed cold water into the boiler and onto the bare crown sheet. 

There is simply no way to know this.

First of all, the check valves are not placed directly over the crown sheet. They are placed far forward on the boiler, precisely so that "cold" water from the injectors (which isn't as cold as you think, having been heated by the steam in the injector body) has a chance to mix with the hot water and not thermally shock the boiler. You can see the check valve on the boiler side below the bell--about as far forward as you can put it.

And what direction does "cold" fluid go? Down, of course.

Even if the injector was found with its valves on, there is simply no way of knowing what the hostler (not "tender") was thinking or what he saw upon entering the cab. The steel could have become ductile even before an injector was started, leading to an explosion even before any "cold" water hit the sheet.

See the letter near the bottom of this page:

Red-hot crown sheet

What is known is that the explosion occurred shortly after he entered the cab.  No one alive is certain of the sequence of actions he took, but he was in there, and the valves were open.  That the crown sheet didn't fail until he was in the cab seems like a stretch, but it is, of course possible.  That there was a connection between his actions and the explosion is the conclusion of the investigators at the time.  I certainly can't insist upon a specific sequence, but the officials then believed there was a fairly clear one.

IIRC, the man was the fireman, left to "tend" the engine--it was not in a yard, and he was not, at least by title, a hostler.  I can recheck my sources to verify (not that it's important).  In any case, that is why I used the term "tender," not as a technical term but rather as a description of his duty at the time.

palallin posted:

What is known is that the explosion occurred shortly after he entered the cab.  No one alive is certain of the sequence of actions he took, but he was in there, and the valves were open.  That the crown sheet didn't fail until he was in the cab seems like a stretch, but it is, of course possible.  That there was a connection between his actions and the explosion is the conclusion of the investigators at the time.  I certainly can't insist upon a specific sequence, but the officials then believed there was a fairly clear one.

IIRC, the man was the fireman, left to "tend" the engine--it was not in a yard, and he was not, at least by title, a hostler.  I can recheck my sources to verify (not that it's important).  In any case, that is why I used the term "tender," not as a technical term but rather as a description of his duty at the time.

Why would it be a stretch? If anything it was simply coincidental. A boiler explosion could have happened whether or not he was in the cab at the time. We simply do not have enough information to assume causation (at least regarding the injector causing the explosion). It's pretty clear the fireman let the water get down (and truthfully, even that sounds fishy to me. A standing engine doesn't use a whole lot of steam if no appliances are being used. It would have taken a very long time to get to a low-water situation, unless it was low to begin with, with generator and air pumps running, and a very leaky system).

In the photos in the link you posted, "cold" water would never have reached the crown sheet. The check valves are too far forward--almost to the smokebox. By the time that water reached the crown sheet, it would have been nearly the temperature of the rest of the water.

Also, as explained in other posts, it probably would have been smarter to drop the fire ASAP, before adding water.

As you clarified, if it was the fireman, you can simply say "the fireman was tending the engine."

If memory serves, a SP 4-10-2 had a crown sheet failure in Arizona, killing everyone but a conductor in the cab of a diesel helper ahead of the locomotive. Fire brick could be found as late as the 90s over 100 feet away from the site and parts went way further than that. The boiler sailed and the cab was found a decent distance from the site as well. The S1 diesel helper's rear cab wall was pushed forward substantially and the crew was blasted out through the windows. The steam engine crew, I assume they were in pieces.

I knew someone who grew up along the N&W and he swore there was a crown sheet failure and he'd found body parts of the crew scattered in the trees as a kid.

There was also a 2-6-6-6 that blew up like that, and it's still debated among the C&O fans as to how it did.

That Gettysburg incident changed the game for everyone. It's amazing they didn't blow up that 4-6-2. It had something to do with how the firebox was designed, and even though I have long ago read the NTSB report, I can't say I really 'get' why it didn't simply blow up. But as much of a 'bubba' operation that was, it was just a matter of time before something bad happened.

I've always wondered if there's been any crown sheet failures with large live-steam engines, like on the 7.5-inch gauge stuff. I know that the crown sheets are proportionally way thicker than a full-sized engine is, but has anyone ever had one of those blow?

I'd also assume that back in the day, steam tractors blew up every now and then but people didn't hear much about it. That one in Ohio killed a few people and from what I recall, launched at least 10 feet into the air at the time?

samparfitt posted:

Dominic Mazoch posted:

OK.  Suppose you have two locomotives of the same class.  One is at the ATSF Station in Galveston TX.  The other is at Union Station Denver.  Galveston is at 6 feet above sea level in the Strand.  Denver is a mile above sea level.  Would the locomotive in Denver produce less power because the atmospheric pressure is lower?  The pressure/boiling point relationship given above is based at sea level.   Still risk of explosion in case of boiler/firebox failure.  

A long time ago, I remember reading about some high altitude railroads had trouble with the power output of their diesels when they converted over from steam; ie, steam wasn't affected by altitude but internal combustion engines are. 

Correct.  Internal combustion engines are certainly affected by altitude and will produce less horsepower at higher elevations.  Turbocharging a diesel goes a long way to alleviate these issues, and was one of the reasons railroads like Union Pacific made a BIG push to get EMD to turbocharge the 567 diesel, doing a vast amount of R&D work, some in conjunction with EMD by applying turbochargers to a group of UP GP9s. EMD had issues early on with the turbo compressor wheels exploding, but eventually overcame the problem and introduced turbocharged models like the GP20 and SD24 to their model lineup.  If you follow drag racing, you may know that the cars that compete in the NHRA Mile High Nationals at Denver typically produce less horsepower in the high altitude, even the supercharged ones. 

Yeah, but the air is thinner so they go faster.

A thought.  In the world of automation/safety first.  Crown sheet protection could be minimal with an automatic low water feed.  IMO.  A lot of human error has been eliminated from our world.  

NYC engines, at least engines in road service, were fitted with a low water alarm, described as a "swimming valve" that, if not covered by boiler water, would vent to atmosphere with a loud noise (shriek) that could be heard in the cab of even a hard working engine.  In spite of this, the most famous NYC boiler explosion is probably that of Hudson 5450 on the 20th Century Limited, which failed in the fall of 1943.  The engine was doing over 70 mph at the time, and the boiler separated from the frame.  None of the crew survived.  I believe, without referencing the report, that there was an injector malfunction, possibly compounded by the fact that an incorrect boiler water level gage was used, OR that it was not installed at the correct height on the boiler backhead.

A famous D&RGW boiler explosion occurred in Oct 1952. 4-6-6-4 #3703 blew up south of Denver, killing 4 men in the cab (1 was a railfan, it has been reported).  Low water was the cause. The boiler and the rear 6-4 left the track due to the violent explosion. The front boiler-less 4-6 continued rolling forward and went about a mile down the track, eerily.

I was in Durango in Aug 1961, and a K28 was the assigned switcher in town. I asked the engineer if I could catch a cab ride. He said no, as " a locomotive blew up about 9 years ago and the company forbid any non-employee cab rides" !

A Rio Grande 4-8-4 blew up just south of Denver Union Station in the '40's, wherein a hostler had 2 4-8-4's coupled, returning to Burnham Yard. The water on the 2nd locomotive got low, due to the engine being un-manned, and suffered crown sheet failure.

Mike CT posted:

A thought.  In the world of automation/safety first.  Crown sheet protection could be minimal with an automatic low water feed.  IMO.  A lot of human error has been eliminated from our world.  

I think people that have never been involved, hands-on, with real steam locomotives don't have a good understanding of what it takes to operate them safely. This lack of understanding (coupled with threads like this), leads to unnecessary fear.

Probably at least 10 Steam locomotives are operated every day of the year in this country, and during some parts of the year, many, many more. They are sometimes operated by very young, very inexperienced crew; they are also sometimes operated by well seasoned veterans.

Tell me--exactly how many catastrophic boiler failures have there been in the last 24 years?? 

Mike CT posted:

A thought.  In the world of automation/safety first.  Crown sheet protection could be minimal with an automatic low water feed.  IMO.  A lot of human error has been eliminated from our world.  

To follow up on what Steve said in the post above this one, there is nothing “automatic” on a steam locomotive. The only devices on a steam locomotive that deliver feedwater to a steam locomotive boiler are manually operated, heavy mechanical devices (injectors, feedwater pumps, etc.)  that would be extraordinarily difficult (and expensive!) to “automate.”

The steam locomotive fraternity is dealing with this blending of old and new tech with the advent of PTC (Positive Train Control.) The PTC rules specifically call for closing the throttle in case of a signal violation. On a modern diesel, that’s a couple lines of code in the computer program. But how do you do that on a throttle that consists of a long, ratcheted lever operating 6 big throttle valves via a bell crank 30 feet from the cab? Needless to say, the main line steam guys are working with the FRA to come up with a solution to this!

When operated by a competent crew, a steam locomotive is no more dangerous than any other big piece of machinery. 

Can anyone describe the operation of the Nicolson Thermic Syphon?