Seeking info

[sutty]

Cheers Ken we will only be using this boiler for a Stuart S50 engine but we're still looking for the specs for this engine pressure wise. I can change the firebox for 10 gauge, are the cross tubes ok at 10mm dia x 1mm wall ?

Thanks for your help and best regards sutty

 

[Mechanicboy]

Not a good idea to have a big pipe in the middle with other small pipes around the big pipe. Mostly all the heat is released through the large pipe instead of through the other small pipes. In other words: Great heat loss.

I has boiler with 31 smoke pipe 8 mm pipe in total length 4,8 meter long, it boil water fast due large hot surface.

 

[Steamchick]

Fair comment Jens. Also there are more efficient and more powerful designs in Horizontal boilers. Stuart use the humble "Smithies" horizontal with 3 or more water tubes beneath. A much better steam raiser and can take a bigger burner. I have been trying to work with the materials and ideas that Sutty has already bought, and keep him safe with the metal he has.
Here are pictures of the Smithies arrangement. (3-1C)

I would suggest that the original idea of 16mm cross tubes would have choked the flue considerably, so you could not get much exhaust gas up the flue, but with smaller cross tubes a compromise would have reduced heat losses in the flue. Of course, the latest design only has 1 central flue, with no cross tubes, so should be OK with the meths burner proposed, although not very powerful as a steam raiser. The cross-tubes in the firebox will do most of the heat transfer I guess?
K2

 

[Steamchick]

Hi Sutty, the 2" OD x 10 gauge firebox tube is just OK for 25psi NWP. (Safety full blow at 26.5psi). Compressive Hoop Stress = 901psi, limit 956psi. = just OK.
The 10mm x 1mm tube is in tension due to the water being inside it: Tensile Hoop Stress = 154psi, limit = 4552psi. (This tube is OK for NWP = 100psi. due to the small diameter).
You can see how the compressive stress limit of copper is so much lower than the tensile limit, compounded with the larger diameter of the firetube and "penetrations" in the firetube wall causing a stress concentration, thus increasing the stress in the copper, and deteriorating the NWP further.

The Stuart S50 is bore 5/8" and stroke 1 1/4", double acting: So you will need the following to run it continuously at 250rpm:
* About 200W of heat from the burners,
* 0.15cu.in of water per minute to be fed to the boiler (so it needs a feed pump, hand-pump or other.).
* 96cu.in of steam at 25psi (the NWP for this boiler)

As you make some hints to who is with you on this project, I guess you are teaching you Son some Engineering and Manufacturing skills?
I can do some more calcs, but have to go now (for today).
Cheers,
K2

 

[Mechanicboy]

With water tubes such as the Smithies/Stuart steam boiler is effective to boil fast. Same principle as the Babcock & Wilcox boiler.
Most effective is Yarrow boiler, but need attention to keep eye on the water level and the steam engine need water pump to feed the water in the boiler.


About Sutty's boiler.. are the boiler to demonstration ot planning to use in the model boat when we are thinking about the room to place the boiler in engine room in the model boat?

 

[sutty]

Jens, it is just a stationary engine set up, not driving anything .

Ken, thank you , we’re beginning to understand the technical side a bit more now, my apprentice is my grandson, his other grandad died when he was six days old. He was a model engineer and had built a 5” loco that was almost finished, it just needed the cab roof finishing and painting, we thought it would go to the grandson but it was sold off.
we’ve built a couple of half scale vehicles, a Willy’s jeep and a type 35 Bugatti but most of the work on those was to technical for him.
this project will be completed by him under my supervision. We have a good lathe with all the accessories a Tom Senior universal mill, again with all the kit and a comprehensive welding kit add to that a lifetime of my mistakes to learn from.
both are powered by mobility scooter motor and transaxles

 

[Charles Lamont]

I have not yet read this thread in detail, but at a quick glance through I supect Steamchick's has somehow got his figures way out.

Just as an example, taken from what I have immediately to hand, hundreds of Simplex 5" gauge locomotives have been built, tested, and run perfectly safely since the design was published over 50 years ago. This has a boiler shell of 4-3/4" (120mm) diameter and 10swg or 1/8" thick, with a hydraulic test pressure of 200psi and a working pressure of 110psi. The superheater flue, which is in compression, is 1-1/8" OD by 16 swg (0.064", 1.62mm). These boilers are still inspected and passed by qualified boiler inspectors under current regulations.

Unless the thinking has changed lately, the Australian Miniature Boiler Safety Committee codes are the generally accepted standards.

 

[sutty]

Charles, I think you’ve come in half way. We’re not building a loco, we’re building a small stationary mill engine, what you read was a brief history.
we know nothing about steam but a lot about engineering and I needed a project to fire up the grandson and bring him from his virtual world of screens and keyboards to the real world of making stuff.
Steamchick has offered guidance and some understanding of the way to go.

best regards sutty

 

[Charles Lamont]

Sutty, the application does not affect the calculation of tube thickness.

 

[Steamchick]

Hi Charles.
In my defence... I shall explain where I get my conclusions.
I follow the "latest" ASME information that I have - an article written by Kozo Hiraoka and published in Live Steam and Outdoor Railroading magazine in December 2006... Possibly this is out of date, or "not applicable" to where you are...?
But I have been a professional engineer for over 40 years, including some time with higher pressure air systems (NWP 28 Barg - 10in dia piston motor). During my time on that design work I experienced British standards for pressure vessels and Company standards, which didn't always match. The Company standard needed updating from its 1950s source, as the British standards were tighter limits for some odd bits. (I can't remember the details - just information in the making of pressure vessels...). So it doesn't surprise me that there are thousands of old boilers working satisfactorily to the "standards" of their time, yet do not comply with latest Standards and Regulations.
So with this background, and having collapsed a flue tube on a boiler that "shouldn't have collapsed", I set about checking the compressive strength of Copper, and confirming the hoop stress calculations (from the interweb...). I found that typically, at the temperature of steam at 100psi, the compressive strength of copper is only 21% of the tensile strength at that temperature (~200deg.C.). (A university paper - graph of strengths with temperature). There are also many empirical methods of calculating the deterioration of compressive strengths of tubes due to the distortion of the tube (Elliptical, distortion of round, etc.). And there is a Stress Concentration Factor (I was advised by someone who's job it was to know... but don't know which ASME clause, etc.) which must be considered when there are penetrations in the shell (both in tension and compression). The standard SCF for ASME is 3.3, to be used to determine the stress in materials in the design of pressure vessels. (I had been using SCF of 2.5, up until then, based on my "job standards" from 1980s).
I have been able to compile a spreadsheet which takes into consideration:
The Hoop stress based on:
* the calculation of hoop stress given by:
Shoop = σc = [(pi ri2 – po ro2) / (ro2 - ri2)] – [ri2 ro2 (po - pi) / (r2 (ro2 - ri2))]:
* A SCF = 1 - if without any penetrations, or = 3.3 if with penetrations (Irrespective of how they are reinforced).
* NO consideration for non-uniformity of the roundness of tubes in compression, as "straight and true and undamaged" tubes will be good, and any damaged or distorted tubes will fail by collapsing - probably during the hydraulic test (as did mine!). (Studies and calculations of deformation effects are complex! Straight and true is good, anything else isn't.).
The calculated max stress shall be less than the limits as defined:

* The permissible Tensile stress (based on ASME limits - Kozo's paper) based on temperature of the steam in the boiler.
* A reduction from the Tensile stress limit to 21% of that limit to account for compressive stresses instead of tensile stresses, as appropriate.

Now I appreciate that British standards are different to ASME, Australian etc. but I understood that ASME was effectively the same standard as that applied in Australia? But you would know, and I can only assume such.

Notwithstanding: as an ex-professional engineer, I cannot in all good faith, suggest anything less safe than the safest to the best of my knowledge and experience, and in accordance with suitable Regulations... What the standards were when the Simplex boiler was designed is immaterial to my calculations for helping someone make a safe design of boiler for someone to make. I suggest that with a FOS of 2, the Simplex boiler would be safe in almost all conditions of normal safe working. But that would not meet current regulations for the design of copper boilers for the USA. (ASME). I guess the ASME is based on a FOS of 8 or thereabouts?
I'll be glad if anyone corrects my assumptions and applications of the calculations, or simply corrects any errors I have made. (I am prone to dyslexic typing! - and thinking!). We are all here to learn from each other, so please ask questions, from which we may get better answers.
I hope this and the attached paper I wrote will help any understanding of where I am coming from... so you can correct me where I am wrong and teach me the right thing to do.
Steam safely.
K2

Copper:
Tensile strength = 210MPa. (Deteriorates with temperature)
Compressive strength MIN 45 MPa, (Deteriorates with temperature)
Extract from Kozo's article on Copper Boiler design:

 

[Steamchick]

Hi Charles, for the benefit of anyone interested, and with the greatest respect for Martin Evans, who has designed many locos and boilers. All that I know have functioned well and have happy owners.

My Simplex boiler drawing is dated 1968: which pre-dates the AMERICAN regulations by 38 years, so that is probably where the difference lies.
ASME Regulations - determined by the most able Engineers in that country, and suitable for the society of legislation and litigation that existed at that time, determined that "Silver-soldered Copper boilers shall not exceed 100psi NWP". To achieve this decision they probably had a lot of technical expertise of materials and their properties in use in copper boilers. And the consequences and liabilities that Insurances would want to avoid. (The sort of things that lawyers love when they think there is a large pot of money to be exploited from which they may get their 10%?).
ASME does not apply in the UK - as far as I understand. But I don't know the Design Standards for the UK.... just the testing standards. (As published by the Southern Federation of Model Engineers).
Hence, to interpret Martin Evans design with my calculations as I know to meet ASME regulations, I have calculated the shell to have a hoop stress (Tensile) and FOS of 3.8 - to the ASME limits. The superheater tube is better with a FOS of over 5. Not a surprise, as in 1968 these ASME limits did not exist, or apply to a boiler for the UK? Of course, to the UK owner of an existing loco, "an approved design" when tested to 2 x the NWP is classed as "OK", providing all inspections are good. (Am I crazy to think that is a bit "old fashioned"?).
But I understand the UK limits for NEW designs to require a FOS of >8. - More than twice as safe as the Simplex Design? And if this was in the USA it would be derated from 110psi to 100psi NWP to meet the Regulation max pressure...
But then at 100psi and this shell size it would have a FOS of >13 without considering the stress concentration factor of 3.3... But considering that factor it becomes No Good (FOS = 4.1 !). And should be re-rated at NWP = 60psi. (The engine might not run very well!).
So the discussion point is whether or not the Stress Concentration Factor of 3.3 is required? - As an Engineer, I say "of course", because it then makes the boiler safe enough to withstand repeated cycling and any fatigue stress development that could occur at the penetrations and reinforcement points. And "of course" if defined in the Regulations somewhere!
I recall that in the 1960s there were aircraft grounded because of stress concentration issues. - The plane was a pressure vessel that cycled the external pressure by flying at high altitude. The stress cycling caused the shell to fail from fatigue stress where there was a local high stress concentration. So it probably was "new Engineering" when or after Martin Evans published his designs?
Charles, I have not heard that the Australian regs apply in the UK. Because I don't know what they, or the UK Standards, include, can you advise please? I had only heard someone state the Australian Regs are basically the same as the ASME Regulations. I have yet to determine the UK Design regulations for copper boilers. Just the rules for testing...
I'll be most appreciative if any professional can assist in clarifying these issues?

If this point is not appropriate to this thread, then I'll start a new one? - Suggestions please?
Thanks,
K2

 

[sutty]

Ken, all good by us, it’s interesting to hear the differences plus it’s giving Ed an insight to material choice and design.

regards. sutty

 

[Charles Lamont]

I should emphasise that, although I also have an mechanical engineering degree, I am no expert in this field. I just wanted to drop in a 'hang on a mo' to the discussion because some of the suggested thicknesses and safe pressures seemed unusual in comparison with the many designs I have seen in Model Engineer over the last half century.

As I understand it, because silver soldered copper boilers end up with the copper in an annealed state, when hydraulic tested they have plenty of give to accomodate stress concentrations and will adopt a small, and perfectly acceptable, amount of permanent set. Also, as I understand it, the safety factor of 8 is intended to allow a safe boiler to be designed using simplistic calculations. A copy of the AMBSC codes would help, but I don't have one.

I have asked a mate who has recently become a boiler inspector at his local model engineering society, and he is going to check the current design standard with a well known professional model boiler maker who is also a member of the club. That may take some time.

As an aside, in the UK we still have full size locomotives working in preservation with (riveted) copper inner fireboxes, with working pressures up to 250psi.
The copper used is C107 arsenical copper which has higher strength at higher temperatures.

 

[Steamchick]

Thanks Charles, Seems like your prompt was useful anyway, because I checked my spreadsheet, and had used the wrong part for the "tubes in tension"... so the results I had quoted were a bit worse than they really are, but not a problem anyway. It was the "tubes in compression" that were worst in all the designs I have studied. (Backed-up by some "burst" boilers I have been given to fix!).
The real problem is the stuff I have not found in ASME, or anywhere, and that is
# how to manage the stress concentrations,
# Copper tubes (Regulation stress limits) in compression, and
# how to cope with "out-of-round" tubes (Limits, factors, etc.) which can be found in Engineering texts, but I have not yet found the Regulation stipulations.

But in the mean time, any boiler with tubes in compression needs proper hoop stress calculations and consideration of the MIN stress limits (IMHO) and SCFs need to be included in the Tension and Compression hoop stress calculations when there are ANY penetrations in tubes... The "Simple" boiler making books do NOT cover these calculations. (Good though they are for what they do cover). While their boilers probably won't fail immediately, they will fail a proper analysis if judged against ASME Regulations. Just "out-of-date" thinking. (Not even Engineering!). I am not big enough to challenge the Regulations of any country.
As you point out, the copper is annealed when first manufactured (from the silver soldering process). So the MIn compressive strength safe limit must be applied to any tube in compression as otherwise it is likely to deform (and possibly start a crack) during the Design hydraulic test or steam test! Extra thickness of copper (to reduce the stress to safe limits) while in the calculation stage, - before anyone buys thinner copper tube - is a simple solution so all the work of making a boiler makes a good long lasting model, instead of suffering early life failures.
Cheers!
K2

 

[sutty]

Ken, ED asked me what the abbreviations stand for, I don’t want to guess or get them wrong, if we’re going to do something we want to do it right, I’m having a hard enough time answering his questions as it is .

best regards sutty

 

[Steamchick]

Hi again Sutty, Ed.
Here's a few I throw into the Engineering conversations...
FOS - Factor of Safety
SCF - Stress concentration factor.
IMHO - new-speak for "In my humble opinion".
ASME - The American Society of Mechanical Engineers.
I MechE - the British Institute of Mechanical Engineers.
I Civ E - the British Institute of Civil Engineers.
I Elec E - the British Institute of Electrical Engineers.
I Chem E - the British Institute of Chemical Engineers.
IDE -the British Institute of Design Engineers. (I nearly joined that one!).
psi = Pounds per Square Inch
barg - Bar Gauge (above atmospheric) - as opposed to Bar , or barA, = Bar Absolute = pressure above Vacuum.
Stupid = Most Engineers like me.
Clever = Most everyone else who is not like me.
Mathematician. A guy who works it out with a pencil, or develops formulae so we can work it out.
Engineer = Someone who applies Ingenuity to resolving problems. (A Greek taught me that - NOTHING to do with Engines really!)
Engine: a converter of the forms of power/energy. As in something that converts "push" into "lift", something falling into something turning, or chemical energy to Kinetic energy, or potential energy for use "later", something that converts "electrical or magnetic" energy into kinetic energy.. There are lots of Engines around us.
Steam - gaseous water - invisible - NOT fine droplets of water suspended in air like Clouds, mist, or the wisps you can see above a boiling pan of water. This gaseous water can carry a lot of energy as "pressure and temperature" combined.
NWP = Normal Working pressure - The pressure at which a boiler is certified. It MUST NOT be exceeded in normal use. The point at which the Safety Valve must FULLY open is NWP + 6%. So on a 50psi boiler, "thou shalt not exceed 53psi" by the application of the maximum heat from the heat source, under any circumstances. - But of course, to be able to perform leak tests at 1.5 times NWP (NOT with steam!) and Hydraulic tests at 2 to 2 1/2 times the NWP (NOT with heat) the SAFETY valve must be replaced with something else.
Stress: What makes an Engineer become Stupid! - In "material" terms, it is "pounds per square inch", - or Metric use "Bar", Pascals, Mega Pascals, etc. = "Force per Area" in every case. In a coffee cup, the stress on the bottom of the cup is the weight of the coffee divide by the area of the bottom of the cup (a pressure - uniform in the liquid and acting on the bottom of the cup), applied to the thickness of the ceramic/plastic of the bottom of the cup by some clever equations, so you can find out how thick to make the bottom of the cup - from one side to the middle and on to the other side. But all that "stress" is also applied to pushing the bottom of the cup away from the sides - where the handle is holding it - and in the corner where the bottom meets the sides, the stress is concentrated to 2 or 3 times the stress in the "flatter" parts that come towards the corner. The "2 or 3" is the SCF. Then the stress is evenly divided around the sides of the cup - in tension - until it gets to the handle, where it is "horribly" concentrated at the 2 points between the cup side and the handle. Then it tries to bend the handle, so a bending stress is happening all through the handle - and the sides of the cup close to where the handle joins the body. The Engineer uses Mathematics to calculate all these stresses, then decide how much or how little material he need to make the cup - so it doesn't break in normal use of holding hot coffee while you drink it. - then he tests his design to find out what he forgot to consider, re-designs it and makes a lot of money making cups that work reliably! The "non-engineer" makes lots of cups, all with slight changes, and scraps the designs that break in use, or cost too much. Having selected the best, "IMHO" he sells them and doesn't care if lots of people drop hot coffee down their body to parts that don't like hot coffee!

Hope this helps?
K2

 

[Steamchick]

P.S. I hope the "tongue-in-cheek" bits don't offend anyone, and post #36 is in the spirit of this site of helping people understand (my perspective of) the "Engineering" bit...?
Good Engineering makes Good Designs of Models. As a model is usually a representation of a real object - or system - but at a scale not equal to 1, often the factors of a design that are "non-linear" cause things to fail, or need to be changed in order to function, a suitably represent the "real" thing. (e.g. the size of a strut - like a pushrod - or bearing, or size of a hole that steam is passing through!).
K2

 

[sutty]

Thanks for that Ken, going to Print them off and hang ‘em in the workshop, I’ll start Eddie of on some simple equations.

best regards sutty

 

[cds4byu]

So with this background, and having collapsed a flue tube on a boiler that "shouldn't have collapsed", I set about checking the compressive strength of Copper, and confirming the hoop stress calculations (from the interweb...). I found that typically, at the temperature of steam at 100psi, the compressive strength of copper is only 21% of the tensile strength at that temperature (~200deg.C.). (A university paper - graph of strengths with temperature).

K2, do you have a reference for this data? I've never heard of a metal having compressive strength that is 20% of the tensile strength. Metals (especially ductile metals like copper) have nearly equal tensile and compressive strengths, in my experience. I'd like to learn more about thjis phenomenon.

Cylinders collapsing when they "shouldn't", on the other hand, I have lots of experience with. That happens because they don't fail in yield; they fail under elastic buckling. And elastic buckling is very sensitive to geometric imperfections.

Anyway, if you can point me in the direction of that strength data for copper, I'd be grateful.

Carl

 

[Steamchick]

The standard data tables quote compressive strength min from that level to max about twice the tensile strength. I have looked for the university study that went into this in great depth, but I've lost it. It was a graph showing temperature versus compressive strength of annealed copper, as a graph with various curves for states of hardness. But we steam test boilers as new when annealed from the silver soldering, so must design in that state. Curiously, older boilers get stronger - by a lot - but unless your calculations consider the strength deterioration of copper with temperature, and considervthe annealled copper condition, you will probably not get certification to ASME Regulations. It may depend how knowledgeable the Engineer is who certifies the calculations.
I will admit I was surprised when I read the Paper, but it was real... by an American university I think? But it was an obscure thing hard to find on the web - which I find pretty useless and getting worse for good technical stuff. Even ASME Regs etc. are near impossible to find from the UK. And it gets worse week on week!
K2