In illustrations I’ve see of superheater installations, I see the steam tubes leaving the superheater header and running aft into the flues. But it’s unclear to me how the tubes/steam return to the header. Do the tubes make U-turns and exit the same flues they entered? Or do tubes run all the way back into the firebox, make U-turns there, and then return through different flues?
Still in training.
The elements make 'return bends' shy of the rear tubeplate ... sometimes well short of it, to avoid too much heat, which damages the metal ... and return back to the header. You may be interested to know that the return bends are specialized castings in most modern superheaters, not just 'tubing bends' of the sort made in more ordinary pipe.
Depending on the type, there can be one or two sets of passes and return bends (compare type A and type E superheaters); the more complex runs normally calling for larger flue diameter.
In general, putting active steam elements in an American firebox is a very poor idea. Steam doesn't take up heat very well, and stainless steel of the usual kind used in elements doesn't have very good heat transfer characteristics, so exposing elements to high radiant flux will sooner or later burn them -- you won't like anything that happens after that!
Thanks, Overmod. Much obliged.
One more reason for keeping the superheater -um- cool is preventing damage to the lubricant in the cylinders, too high a temperature will convert oil to abrasive carbon particles.
Thanks, Eric.
Are all flue-like passages thru the boiler called "flues?"
Or, are there smaller-diameter ones which are referred to as "tubes?" It seems that I saw reference to that somewhere. (But maybe that was a reference to superheater tubes in the flues. ??)
If there are two different sizes, why?
Flues are big pipes, tubes are small. (There is some controversy over exactly when a tube 'counts' as a flue, but in a boiler for standard gauge or greater 2" is almost always a tube, and 3" and greater is almost always (always, in my experience) a flue).
Every superheated boiler has a combination of the two; the tubes are optimized for heat transfer to water, the flues are optimized for best gas flow around superheater elements. (In part this is related to how closely you can 'pack' tubes vs. flues, vs. how flow occurs in relatively smooth tubes compared to likely-turbulent flow around the element structure).
As people began to recognize how much actual steam generation occurred in the waterspace and chamber rather than via the tubes, they began using relatively fewer tubes and more flues, to accommodate more and better elements. They also increased the peak size of a flue somewhat in some engines, to accommodate multipass superheater designs more readily and, to an extent, reduce required draft pressure (allowing a freer front end, or one more optimized for gas ejection rather than just steam exhaust efficiency).
Superheater "tubes" are called 'elements' precisely to distinguish them unambiguously from other use of the word 'tube' in a boiler. I use that same term for reheater elements when they are present, on principle.
See if you can find a cross-section of the Jabelmann Challenger boiler at its convection section -- that, to my knowledge, is representative of the 'last' modern practice in providing a relatively large number of flues at the 'optimal' length for standard flues (it's about 406:1, more commonly rounded up to 410, beyond which you are wasting weight for the marginal increase in convection-section steam generation).
I strongly suspect the Superheater Company's proprietary calculations of element size and configuration for a given locomotive design included equalizing gas flow through tubes and flues to give even stress in that part of the boiler structure. (Especially in a modern boiler without a defined superheater-damper arrangement.) But to my knowledge the secret Bene Gesserit lore has not been revealed, at least in the popular literature.
For fun, consider Besler tubes. These are passive re-radiators inserted in the bore of standard tubes (of whatever diameter), leaving an annular space relatively close to the wall for the combustion gas and making heat transfer a bit more positive in some significant respects (particularly if the combustion gas is relatively clear and soot-particle-free, as it ought to be in 'best combustion' conditions). It is difficult to find hard information on these outside the steam-automobile community, but the one table of comparative data indicated that actual heat transfer to water (for a given mass flow and quality of combustion gas) was about twice in a Besler tube what it would have been in the 'standard' equivalent. You can probably imagine the maintenance issues, and the likely solutions for them.
LithoniaOperator,
Overmod has laid out the basic construction of the superheater unit for you. Particularly the cast ends of the superheater tube bundles and the critical issue of burning the superheater units rendering them less than they were designed to be. This is a fairly critical issue today with the number of skilled locomotive engineers and firemen simply not available.
The Union Pacific steam program issued a video a few years ago concerning the UP844 boiler and this above mentioned problem - the burning of the superheater units causing them to sag and distort in the flues. Surprisingly UP was claiming to have developed a spring steel insert of their design that would hold the overheated superheater units in place within the flue allowing them to cool down into a shape resembling their indended form. A protection agains engineer abuse. Each superheater bundle would have several of these spring clips attached supporting the unit in the flue.
---------------------
Consider also the superheater problem relating to Norfolk and Western 1218. This massive articulated 2-6-6-4 locomotive ran fan trips back in the 1990's. Its boiler and firebox were worn out and an extensive boiler overhaul was needed. The NS Steam program began this project under the Claytor corporate leadership. New firebox sections were fabricated and the superheaters were pulled and brand new tubing ordered. This project came to a sudden end when NS decided under new corporate mangment to "pull the plug" on the restoration.
This left NW1218 in a serious condition of disassembly and repair. Parts scattered all over and question as to if it would ever be restored. The superheater tubing purchased was sold at auction and the new firebox sections were quickly tack welded into place leaving the entire project unfinished. If and when it would be restarted would all the parts be located? Including the all important superheater bundles ends and flanges. Better not to have started the repair than to leave is in such an unfinished state.
My final experience with superheater bundles was with the restoration of Pere Marquette 1225. Undertaking the restoration of the boiler was first attempted in the 1980's. This process involved removing the superheater bundles from the boiler thru the smokebox door. They were laid out on the grass because we were working out of doors on the Michigan State University campus! The superheater bundle was then attached to a custom made water supply with a hand pump. So we would pump up the superheater bundles and read the pressure gauge - then look for the water leaks - which were promply electric welded.
Later we came to realize that while this was a repair of sorts, the overall condition of the superheater tubes was of serious concern owing to the corrosion that occured from sitting in that cold boiler on display for years and years. There was really no recourse but to entirely replace the superheater tubing which involved rewelding the cast tubing ends and header flange onto new piping.
Without new tubing they just leaked and leaked and leaked and the performance and horsepower of the locomotive was constantly problematic. I believe even the "new" superheater units became a problem also - so that at one point they were removed entirely - and the engine run on only "saturated steam" not "superheated steam" - well we weren't pulling manifest freight trains like UP has been doing with the Challenger.
Consider the responsibilty Ed Dickens has with the entire UP steam fleet - UP844, UP3985, UP4014 and the successful storage of these engines for months at a time. Those unused boilers with flues and superheaters have to be dried out or the onset of corrosion is relentless. This was something the original railroad never had to deal with because they always kept the working steam boilers fired and hot all the time unless repairs were being made.
Hats off to UP and the best steam fleet in the world!
Dr. D
Dr D Consider the responsibilty Ed Dickens has with the entire UP steam fleet - UP844, UP3985, UP4014 and the successful storage of these engines for months at a time. Those unused boilers with flues and superheaters have to be dried out or the onset of corrosion is relentless. This was something the original railroad never had to deal with because they always kept the working steam boilers fired and hot all the time unless repairs were being made. Hats off to UP and the best steam fleet in the world! Dr. D
Don't want to stray too far off topic (how to properly lay up a steam locomotive deserves its own thread), the procedure I have seen for drying out the boiler is as follows, and is surprisingly simple:
1. Wash out and drain boiler, while it is still warm but not hot (we do this a day or so after shutting the fire off).
2. Blow EVERYTHING out with compressed air.
3. Leave washout plugs out, and store the locomotive where it will not be rained on or freeze.
Greetings from Alberta
-an Articulate Malcontent
To add to this a little:
Probably better in many cases, when laying up an engine, to keep the boiler FULL of water (properly treated to pH 11 with a healthy dose of deoxygenators) and then pressurize the steam spaces with dry nitrogen (ideally with the sort of line-pressure maintaining that NDG was describing in String Lining for telephone lines and that we used for waveguides). This saves beaucoodles of money for water treatment especially if you're using some variant of Porta-McMahon treatment.
The situation with drying out the headers is interesting, because it is slightly counterintuitive. When a modern locomotive is laid up, there is no moisture in the elements because all the poppets in the front-end throttle are definitively closed, and presumably lapped well enough that a small amount of oil will keep them sealed during storage. It can be a bit difficult to arrange for compressed air to blow the elements out from the header 'down' (which to me is more sensible than trying to loosen header connections in the field and blow upward) and it seems to me that a better approach -- with the flues and elements still as warm as possible -- will be to lay the engine up with the valve gear off mid, in a position where there is an accessible 'path' for moisture from the elements down through the valves to, say, one of the cylinder cocks, and then provide a good desiccator arrangement there. Over time any residual water in the elements, including any tendency for moisture to condense in the elements with the engine stored and naturally 'thermally cycling' in the day/night cycle that deposits dew or over seasonal changes, will be pulled out in the dessicant before major corrosion is particularly likely.
Dr D LithoniaOperator, Overmod has laid out the basic construction of the superheater unit for you. Particularly the cast ends of the superheater tube bundles and the critical issue of burning the superheater units rendering them less than they were designed to be. This is a fairly critical issue today with the number of skilled locomotive engineers and firemen simply not available. The Union Pacific steam program issued a video a few years ago concerning the UP844 boiler and this above mentioned problem - the burning of the superheater units causing them to sag and distort in the flues. Surprisingly UP was claiming to have developed a spring steel insert of their design that would hold the overheated superheater units in place within the flue allowing them to cool down into a shape resembling their indended form. A protection agains engineer abuse. Each superheater bundle would have several of these spring clips attached supporting the unit in the flue. --------------------- Consider also the superheater problem relating to Norfolk and Western 1218. This massive articulated 2-6-6-4 locomotive ran fan trips back in the 1990's. Its boiler and firebox were worn out and an extensive boiler overhaul was needed. The NS Steam program began this project under the Claytor corporate leadership. New firebox sections were fabricated and the superheaters were pulled and brand new tubing ordered. This project came to a sudden end when NS decided under new corporate mangment to "pull the plug" on the restoration. This left NW1218 in a serious condition of disassembly and repair. Parts scattered all over and question as to if it would ever be restored. The superheater tubing purchased was sold at auction and the new firebox sections were quickly tack welded into place leaving the entire project unfinished. If and when it would be restarted would all the parts be located? Including the all important superheater bundles ends and flanges. Better not to have started the repair than to leave is in such an unfinished state. --------------------- My final experience with superheater bundles was with the restoration of Pere Marquette 1225. Undertaking the restoration of the boiler was first attempted in the 1980's. This process involved removing the superheater bundles from the boiler thru the smokebox door. They were laid out on the grass because we were working out of doors on the Michigan State University campus! The superheater bundle was then attached to a custom made water supply with a hand pump. So we would pump up the superheater bundles and read the pressure gauge - then look for the water leaks - which were promply electric welded. Later we came to realize that while this was a repair of sorts, the overall condition of the superheater tubes was of serious concern owing to the corrosion that occured from sitting in that cold boiler on display for years and years. There was really no recourse but to entirely replace the superheater tubing which involved rewelding the cast tubing ends and header flange onto new piping. Without new tubing they just leaked and leaked and leaked and the performance and horsepower of the locomotive was constantly problematic. I believe even the "new" superheater units became a problem also - so that at one point they were removed entirely - and the engine run on only "saturated steam" not "superheated steam" - well we weren't pulling manifest freight trains like UP has been doing with the Challenger. Consider the responsibilty Ed Dickens has with the entire UP steam fleet - UP844, UP3985, UP4014 and the successful storage of these engines for months at a time. Those unused boilers with flues and superheaters have to be dried out or the onset of corrosion is relentless. This was something the original railroad never had to deal with because they always kept the working steam boilers fired and hot all the time unless repairs were being made. Hats off to UP and the best steam fleet in the world! Dr. D
Overmod See if you can find a cross-section of the Jabelmann Challenger boiler at its convection section -- that, to my knowledge, is representative of the 'last' modern practice in providing a relatively large number of flues at the 'optimal' length for standard flues (it's about 406:1, more commonly rounded up to 410, beyond which you are wasting weight for the marginal increase in convection-section steam generation).
Overmod,
What does this ratio (406:1) represent?
Scott Griggs
Louisville, KY
sgriggsWhat does this ratio (406:1) represent?
It's nominally in the same terms as the Wagner ratio, which is defined as the ratio of 'area of free space' (a percentage of the tube area corresponding to the effective orifice through which gas passes in induced flow) to inside tube surface area: in Wagner's JILE paper (253) this was defined as A/S. (Of course in typical fashion the proportion is backward, probably to accord with a semantic picture of tubes in situ, diameter to length). A basic principle here is that the tube walls induce drag, which applies all down the length of the tube. Lawford Fry (in The Study of the Locomotive Boiler) discussed 'mean hydraulic depth' (which is, if I recall how he defined it correctly, nominal sectional area / gas-swept perimeter) but didn't include a drag parameter relative to tube length: his basic recommendation was that bore area:length shouldn't exceed 1:100.
I use the 1:406 because that evolved out of some calculations indicating an optimum there instead of 1:400 or 1:410; of course any analysis of a boiler using older methodologies so heavily relies on empirical factors that the "best" tube length could only be determined by experiment, and probably only chosen subject to fairly strict initial and operating conditions not characteristic of 'normal' service.
Synopsis of Wagner's paper and some early discussion is here:
http://www.steamindex.com/jile/jile20.htm
Thanks to a confederacy of dunces (and what can best be described as dog-in-the-manger greed) it remains impossible to access Fry's book via the Web, even via HathiTrust; it was published in 1924, just after the 'copyright hole', and any lapsed rights have not been updated.
Thank you for your reply. I have started to read through the Wagner paper. For the heck of it, I took a crack at calculating S/A ratio in an excel spreadsheet I have that contains boiler design parameters and calculated evaporation for several U.S. types of interest. In so doing, it became obvious that there are several ways to calculate S/A. You could calculate it separately for flues (containing superheaters) and tubes. In that case, my calculations for flues ranged from 549 to 866 and for tubes ranged from 136 to 183. As I said, these were for medium to large American locomotives with tube lengths between 19 and 24 feet. The other calculation method I used was to determine the summed tube/flue/superheater surface area, and divide by the free gas flow area which I had previously calculated. In those cases, the resulting 'composite' ratios ranged from 503 to 833. I think I need to adjust my calculation method, because I did not take into account the fact that superheater tubes do not pass through the entire length of the flues. Maybe with that adjustment, the S/A ratios would drop a bit more with some approaching 400.
What is your preferred method for calculating S/A?
I am on this but running around between computers all day.
Tube and flue calculations are wildly different, and in the case of flues populated with elements, very difficult to 'calculate' with any degree of actual relevance. I have maintained for a long time that nondeterministic mechanics is likely the only valid methodology for that.
It would be nice if we could just partition the rear tubesheet into areas and calculate A/S relative to the gas flow "at" the partitioned area for each tube. (Neglecting flow through the flues at this point in the analysis). This gets around the "A" problem you see referenced in the JILE discussion of Wagner's paper, what the 'effective' mouth area of the tube from a hydrodynamic standpoint should be. One reason gas mass flow is a meaningful metric is that it is conserved over time, as volume or pressure or heat content would thoroughly not be, but that doesn't help much with modeling the actual gas speeds (and the various "frictional" and other drag forces on the gas flow) in the tube given the hopefully dramatic heat transfer during the pass.
If I had to give a version for nonpartitioned measurement, I'd tend more toward adjusting Fry's measure to give an empirical areal equivalent for the tube mouth (for each analyzed speed and load condition; you'd want several to approximate the behavior under 'automatic action') and then see what expected and then under test actual gas flow turns out to be.
S could be determined, again quasi-empirically, by inducing flow in a test gas of appropriate 'combustion-plume' characteristics through a test tube of appropriate length jacketed by overcritical water at appropriate pressure (and steam-removal arrangement to maintain that pressure net of heat transfer). I don't agree with some of Wagner's assumptions on the actual cause of what he thinks is surface drag, but I also think that a reasonable good modern interpretation can be developed using either nondeterministic or traditional hydrodynamics.
Something else that might be considered -- and you are one of the people qualified to do it -- is to revisit the USGS boiler testing circa 1910, which used much higher nominal gas speeds to "scrub" the tube walls and observed much greater nominal heat transfer as a stated result, and comment in terms of more modern understanding.
This would likely be in line with much of the observed efficiency of forced as opposed to induced draft in a convection boiler, as has occasionally been tried (with almost uniformly dirty results) in locomotive practice. It also should work dramatically better with Besler tubes, since the higher gas speed also sweeps the re-radiator surface for better mass contact.
I’m not familiar with the USGS boiler testing of 1910—where does one access a report?
sgriggsI’m not familiar with the USGS boiler testing of 1910—where does one access a report?
Now that the Government funding follies are, however temporarily, addressed, most if not all the USGS reports are here:
https://pubs.er.usgs.gov/browse/Report/USGS%20Numbered%20Series/Bulletin/
It is a little difficult to navigate the listing as it is periodically out of numerical order by 'bulletin'. Important items here include 367, 373, 402 and 403, and 412.
As a geologist, I am familiar with USGS Bulletins, but I never knew of these boiler tests. I guess the tie-in was the coal fuel.
I came across the testing in a magazine description, possibly National Geographic, many years ago, and couldn't understand how USGS could possibly be associated with technical testing of combustion efficiency, let alone innovative research. In those days it was impossible to find the primary material without physical library searches. The 'germ' of the effort is almost certainly at the Louisiana Purchase exhibition, where USGS had a coal-efficiency plant and PRR had the locomotive testing plant; all the practical papers seem to date from the Roosevelt administration, when activist government could support practical research with long-term implications for resource utilization. (Note the research into briquetting, a means of reducing smoke and utilizing fuel not suited for direct combustion in some plants.)
Our community is FREE to join. To participate you must either login or register for an account.