Copper Boiler Design

This forum is dedicated to the Live Steam Hobbyist Community.

Moderators: Harold_V, WJH, cbrew

User avatar
Carrdo
Posts: 944
Joined: Thu Mar 06, 2003 2:20 pm
Location: Toronto, Canada

Copper Boiler Design

Post by Carrdo » Sat Dec 08, 2012 3:58 pm

I guess this relates to my upcoming series related to the silver soldering of my 1" scale Yankee Shop Atlantic all copper boiler which will be fully superheated.

This will be quite a departure form the original Yankee Shop boiler design which is now over 60 years old. Mine will have a completely different tube/flue arrangement although the original design did show a couple of superheater flues.

To start with I want to define some terminology. I call all of the smaller firetube openings in the tubeplates "tubes" and all of the larger superheater openings in the tubeplates "flues". See photo.

Now to the issue at hand.

For this boiler design, I have used some of the information given in the most widely available model locomotive boiler design and construction books which have been published over the years namely, KN Harris, Model Boilers and Boilermaking, all of the various editions of Martin Evans Model Locomotive Boilers and Model Locomotive Boilermaking by Alec Farmer.

It should be noted that these are all UK publications where narrow firebox boilers predominate. While many of the empherical formulae and rules given in these publications are quite valid for wide firebox American boiler design, one still has to think critically about and evaluate the actual best design (and design and construction compromises) for the actual boiler being made.

I do also have the Chas Arnold publication on boilermaking for American type model boilers but it is not a design book and it is now rather outdated. Also, in my opinion, The Joseph Foster Nelson book does not give the necessary information to either design or build an all copper boiler.

I am not out to crticize anyone on this but the information just isn't there.

To proceed, the aspect of boiler design which is the subject of this thread is the rule govering the grate area of the boiler and what % of the grate area should the firetubes be?

I know about the million and one other rules concerning firetube spacing to have free circulation of water and gas bubbles in and around the firetubes, how far one should keep the firetubes away from the circumference of the boiler barrel and the edges of the tubeplates for scale buildup, the minimun "land" or amount of metal one has to have between adjacent firetubes tubes both horizontally and vertically depending on how one arranges the tubes and the greater "land" needed between the top row of superheater flues and first row of firetubes for boiler structural integrity purposes.

All of the above are in the books mentioned which take best UK practice and are totally valid here.

However, when you look at some of the other rules given things change. In the KN Harris book he states that the gas area through the tubes should be not less than 1/7 (14.3%) of the boilers grate area and without detrement be as high as 1/6 (16.7%).

This is OK for a narrow firebox boiler but quite impossible for a wide American firebox boiler. Martin Evans (wisely?) is silent on this.

The only other information I have relating to the above was from a private conversation with Peter Carr (no relation) of Kingswood Boilers in the UK. He has designed and built many copper boilers throughout his career.

Peter indicated a figure of 8% (and more if you could get it) but he did say this ratio is harder to obtain with a wide firebox boiler.

If you also want to avoid violating one or more of the rules mentioned above, it is next to impossible as, even at 8%, the boiler barrel ends up being chock full of firetubes.

Also , I don't know when calculating the ratio, does one use only the area of the firetubes or the combined area of the firetubes and the superheater flues? Can someone enlighten me on this?

Engineers love numbers so here they are for the Atlantic:

Net firebox grate area - 48 square inches ( 5.5" wide by 8.75" long),
Firetube net area - 2.03 square inches ( based on 16, 3/8" nominal type K copper tubes having 0.402" ID actual),
Superheater flue net area - 1.74 square inches ( based on 4, 3/4" nominal type K copper tubes having 0.745" ID actual),
Total net area of firetube and superheater flues - 3.77 square inches or 7.85% of net firebox grate area.
This drops to 4.23% if only the firetube net area is used.

Bill Shields, gwrdriver, any comments on this as you have built many copper boilers?

And anyone else who wants to comment. Am especially interested to hear what the UK/European/Canadian members have to say.
Attachments
48.jpg
Last edited by Carrdo on Mon Dec 10, 2012 7:36 pm, edited 1 time in total.

User avatar
Carrdo
Posts: 944
Joined: Thu Mar 06, 2003 2:20 pm
Location: Toronto, Canada

Re: Copper Boiler Design

Post by Carrdo » Sun Dec 09, 2012 1:34 am

To throw a few more numbers out on the table, the effective tube and flue lengths between the tubesheets is 16".

Calculating the ratio L/D2 (where D2 is the external diameter in inches of the flue* squared - from Harris "Model Boilers & Boilermaking", page 132) the ratio works out to be 64 which is right in the suggested ideal range of between 60-80.

(* I think that there may be a misprint here in the book as from the examples given, he clearly appears to be talking about tubes rather than flues). Terminology is important.

On the other hand, Evans gives a similar formulae in his books (Model Locomotive and Marine Boilers, page 30) but says it is the length of the tube divided by the square of the internal diameter of the tube again all in inches. Using this formulae, the ratio for the Atlantic works out to be 98 which is somewhat above the suggested ideal range of 50-70. This suggests that one could possibly go up a size in tube diameter and have fewer firetubes.

Has anyone else noticed this discrepency between the authors?

User avatar
baggo
Posts: 130
Joined: Mon Jan 12, 2009 6:25 am
Location: Derby, UK

Re: Copper Boiler Design

Post by baggo » Sun Dec 09, 2012 4:29 am

D should be the INTERNAL diameter of the tube according to original Model Engineer articles by C M Keiller so your tubes are a bit long in relation to their diameter.

You should take into account the superheater flues when working out the total tube area but remember that the flues will be partially blocked by the superheater elements so their area will be reduced accordingly.

I always try to get the tube area to be at least 12% of the grate area but as you say, it is difficult or impossible with a wide firebox design. My thoughts are that you can get away with less but you need a stronger blast to get enough air flow through the grate due to the increased resistance of the tubes to the gas flow.

I occasionally run a loco where the ratio of tube area to grate area is only 6% (2.5" gauge LBSC Gwen Elms) and it's a pig to keep in steam. On the other hand another loco ( A 2.5" KIng) has a ratio of 13% and that steams very well.

John
Secretary of The National 2½" Gauge Association
Member of North West Liecestershire SME

http://www.modeng.johnbaguley.info

Asteamhead
Posts: 336
Joined: Tue Jul 31, 2012 1:59 pm
Location: Germany, Duesseldorf

Re: Copper Boiler Design

Post by Asteamhead » Sun Dec 09, 2012 6:49 am

Good morning Cardo,

You asked for answers from old Europe, so I dare to add my ideas.
First, in my opinion you are right regarding calculations and thaughts! And yes, engineers like numbers - so do I!
Your boiler design looks o.k.. Much more gas area can't be achieved at the given /small) inside boiler diameter not viloating rules for construction.
The difference between the typical elder UK and Prussian boilers with very small fireboxes and typical US designs is evident. This led to wider fireboxes in Europe, too.
As pointed out already, just the total f r e e gas area is the important number which is of interest! Thicker walls of copper tubes, flues and plates are reducing gas area in regard to the prototype.
But look, even superb locos like the NYC Niagara and N&W J prototypes are showing ratios of gas area/grate area of around 10 % due to their generous grate areas.
Be careful regarding your superheater! It may deliver more superheat you want in case diameter of superheater tubes are too small! I recommend using a pair of 5/16 outside (3/16 inside?) diameter or even slightly bigger. These will reduce free gas area by 0.61 square inch to a little more than 3 sqare inches in total. This will be about a mere 6.5 % of the grate area, indeed!
As pointed out the draft is to be increased a good deal due to the high velocity of the gas inside the flues and tubes. To the god side this will increase efficiency of both tubes/flues and superheater regarding convection!

Your boiler will be able to evaporate up to 40 pounds of steam per hour (about 1 kg per cm² of free gas area per hour). This might be the limit of the boiler without carrying over to much water, too.
1 pound of steam per pound of tractive force at the rim seems o.k. to me for any (model) locomotive.
So don't worry, just carry on in building yor locomotive!

Asteamhead (doing so with my N&W class A)

User avatar
Bill Shields
Posts: 4920
Joined: Fri Dec 21, 2007 4:57 am
Location: Somewhere in the World
Contact:

Re: Copper Boiler Design

Post by Bill Shields » Sun Dec 09, 2012 7:31 am

As said, consider the ID of the tubes and free area in the superheater tubes in the calculations.

Also, consider that you are going to have some build-up on the tubes, so 'clean tubes' is not exactly what you will have all the time.

I like to use 'slightly dirty' tubes as my baseline since after 20 minutes of running - that's where you are. You would be surprised how it changes the calcultions. I know this idea is contradictory to what the 'school' says, but unless you are running on oil or propane, it's real world.

Gut reaction from your numbers looks about right.

I am building a boiler right now for a 1" engine (5" shell) with 17 flues of 7/16 OD and one superheater of 1-1/8" diameter. Firebox is about the same size as yours.

In the long run, there is more heat made in the firebox than the tubes, so it isn't a issue of heat, but flow that you need to be concerned with.

We'll all see how they go.

User avatar
Carrdo
Posts: 944
Joined: Thu Mar 06, 2003 2:20 pm
Location: Toronto, Canada

Re: Copper Boiler Design

Post by Carrdo » Sun Dec 09, 2012 10:07 am

I would like to thank you all for the great advice and encouragement received so far on this. Please keep it coming.

If I were designing and constructing an entirely new boiler form scratch, I would consider going up a tube size for the firetubes to 1/2" nominal but since the boiler has alredy been constructed, am just trying to optimize what already has been done with this 60 year old design.

I am also considering the practical as well as the engineering in all of this from those who have had much more experience than I in actual building and operation.

They were saying that if one does not have the formulated minimum "land" between the tubes, the copper will be subject to cracking in this (or in any of these areas) from the boiler being heated and then cooled many times.

Also, they indicated when firing hard, there could be no water between the firetubes (only steam bubbles) and as a result, the tubes risk burning.

Also, if there is any buildup of scale within the boiler, there will be reduced or little circulation of water between the tubes. If the scale build up is such that it causes the tubes to touch, then the tubes will overheat and burn out immediately even with copper. They said they have seen it happen more than once.

I know we should all be perfect and prevent this through good boiler maintenance practices but, in reality, it doesn't always work this way. Wise engineering says that the design should be able to withstand a bit of neglect or even abuse from time to time.

All of the books I have read point out that water and steam circulation within our miniature boilers is not very good at the best of times so we should try and do everything we can to improve it.

Yes, I can run with a little blower on to force the draft a bit if found necessary.

Also might have to use a double petticoat as designed by Richard in his Boston & Albany to improve the draft when the locomotive is "loafing" which it will be doing most of the time. His locomotive is exactly the same old Yankee Shop design (and with superheat) only in 3/4" scale.

Richard's B&A boiler has the same plain slide valves as the Atlantic. He has not had any trouble with them so far but as the saying goes one swallow does not make a summer. There are ways of eliminating all of the problems associated with slide valves under superheat but I will let Richard explain how he does it if he wants to chime in on this. K.N. Harris says many of the same things in his book.

Yes, I hear the concerns about the very high superheat possible especialy if one puts in radiant type superheaters which go right to the back of this huge firebox. I was going to get into this in more detail with the boiler silver soldering talks.

To make a long discussion short at this point, we (the silver soldering team - there are 4 of us involved now) are going to hedge our bets. The design will be such that we can take the superheaters out entirely (and insert tubulators in the superheater flues to produce a more or less even gas flow through the tubes and flues - there are tricks here too) and/or we can make the superheaters any length we want as experience dictates.

But the more immediate concerns are the next set of problems I see coming up and that is the design and construction of the existing firebox crown girders and stays. I have lost enough sleep (and budget) on this already and we have only just started. Grrr...

User avatar
Carrdo
Posts: 944
Joined: Thu Mar 06, 2003 2:20 pm
Location: Toronto, Canada

Re: Copper Boiler Design

Post by Carrdo » Sun Dec 09, 2012 10:32 am

Just so you can see where things stand now. The boiler shell is 0.170" (a little less than 4.5 mm) thick.

I don't even begin to know where you would find such material today.

Bill, I can't get anthing in the hard copper tubing in any size between the 1/8" nominal increments that Great Lakes Copper stocks and even that was a huge struggle to locate. It has all become "unobtainium".

One of the reasons why the budget is gone - I bought an entire selection of their hard copper tubing as I know from experience that it is going to disappear never to be available again for small users like ourselves. And I don't have any industry connections.
Attachments
15.jpg
Last edited by Carrdo on Sun Dec 09, 2012 1:41 pm, edited 2 times in total.

User avatar
LVRR2095
Posts: 1292
Joined: Sun Mar 04, 2007 6:50 pm
Location: Maine, USA

Re: Copper Boiler Design

Post by LVRR2095 » Sun Dec 09, 2012 10:46 am

Carrdo wrote:Just so you can see where things stand now. The boiler shell is 0.0170" thick.

.
Gee....it looks thicker than 0.0170" in the pictures! :D

Keith

User avatar
Carrdo
Posts: 944
Joined: Thu Mar 06, 2003 2:20 pm
Location: Toronto, Canada

Re: Copper Boiler Design

Post by Carrdo » Sun Dec 09, 2012 11:01 am

Hi Keith,

My camera angles distort things.

And if you look at the photo of the firebox front tubeplate, the land between a couple of supeheater flues and tubes looks about the same as between the firetubes. It isn't, it is more, but the holes are slightly chamferred for the silver solder to flow into and this appears in the camera photo to be less metal.

User avatar
LVRR2095
Posts: 1292
Joined: Sun Mar 04, 2007 6:50 pm
Location: Maine, USA

Re: Copper Boiler Design

Post by LVRR2095 » Sun Dec 09, 2012 12:06 pm

Carrdo wrote:Hi Keith,

My camera angles distort things.

And if you look at the photo of the firebox front tubeplate, the land between a couple of supeheater flues and tubes looks about the same as between the firetubes. It isn't, it is more, but the holes are slightly chamferred for the silver solder to flow into and this appears in the camera photo to be less metal.
Hi Carrdo....it still looks like it is thicker than seventeen thousandths of an inch to me! :D
Are you sure the copper isn't 0.170" thick?

Keith

User avatar
Carrdo
Posts: 944
Joined: Thu Mar 06, 2003 2:20 pm
Location: Toronto, Canada

Re: Copper Boiler Design

Post by Carrdo » Sun Dec 09, 2012 12:14 pm

John (Baggo),

The numbers you quote for the 2.5" gauge Gwen Elms and the King are for narrow firebox boilers?

I admire anyone who can design and operate any 2.5" gauge locomotive well. Everything there is going against you due to the scale factor. I note in the national association's newsletter, how many locomotives have to retire due to this or that issue.

I hope this is isn't going to turn into a major thesis (just like most things I get into).

I understand the need to have larger diameter fire tubes for combustion gas passage area but not so large as to start to lose an excessive amount heat into the smokebox (a compromise as usual).

But with larger fire tubes doesn't the free space between the fire tubes and the fire tubes and the boiler shell/tubeplates need to increase as well for the practical and design reasons cited? The land between the larger fire tubes also will need to increase slightly with a larger fire tube size for boiler structural integrity reasons.

So now, with slightly larger fire tubes (which increases your % gas area through the tubes/grate area), we may loose a tube or two due to the above (which will lower again the increase) and we will now also have less heat transfer through the tubes as fewer larger tubes transfer less heat than a greater number of smaller tubes.

I see that Bill thinks this is all hooey but,

These requirements are all in direct conflict with one another so a (best) compromise has to be determined.

I guess what I am asking is; one will have to give on some thing(s) but what is better? To optimize the free flow of water/steam bubble path/clear space/ between and around the firetubes or to optimize the % of firetube to grate area and the gas flow in the firetubes? Or to compromise on both as this design seems to do? You can't do it all. The grate area on these American fireboxes is huge compared to the space one has available to play with in the boiler barrel.

This is a conundrum which is the faced by all boiler design over here and not just with copper boilers.

Thinking about it, it seems to me that, within reason, it is six of one and half a dozen of the other but I may be mistaken as you have done far more boiler design/building than I have.

On the model Big Boy, even having a combustion chamber, the ratio of the firetube length to the dia. squared is going to be in the hundreds and yet they seem to work quite well. But I guess we are now talking about outliers.
Last edited by Carrdo on Sun Dec 09, 2012 2:45 pm, edited 1 time in total.

User avatar
Carrdo
Posts: 944
Joined: Thu Mar 06, 2003 2:20 pm
Location: Toronto, Canada

Re: Copper Boiler Design

Post by Carrdo » Sun Dec 09, 2012 1:44 pm

Hi Keith,

Whoops, how did I miss that. Got an extra 0 in there. Yes, it is a 0.170" thick boiler shell. That's why I need people like you to proof read and peer review my threads. Starting to be too many sleepless nights.
Last edited by Carrdo on Sun Dec 09, 2012 1:46 pm, edited 1 time in total.

Post Reply