Building the "Lilly Belle" Boiler

[gwrdriver]

01 - Introduction
Some time ago I was asked to build the copper boiler for a 7.5" gauge Central Pacific #173, also known as Walt Disney's "Lilly Belle." The bare boiler is approximately 24-3/4" overall length with a barrel OD of 6-1/8" and the completed boiler weighted in at around 70lbs. The original boiler design wasn't a true tapered wagon-top; the barrel was a straight tube and the taper was created by the wrapper. This made construction quite a bit less difficult than it would have been otherwise but there were still a number of difficult details to construct.

I used alloy C-110 copper sheet or bar throughout and all bushings are either C-510 phosphor bronze or SAE-660 bearing bronze, and all stays are C-510 phosphor bronze. The plate forms used are either hot-rolled steel, red oak, or yellow pine, depending upon their applications. The silver solder will be Harris Safety-Silv 45 non-Cadmium bearing with white flux.

The photography is not always the greatest, and no matter how careful I was to try to get a picture of every step, invariably some steps were missed, but there will be enough photos for you to get the picture. This thread is going to take some time and will be in installments as I can get to them.

I expect quite a few comments on why I did something the way I did, or why didn't you do it this way, and the simple answer is, given the size and construction requirements I did whatever I did to (a) simplify construction but retain or increase strength, and (b) to allow me to build the boiler, mostly unassisted, using the skill and equipment I have (which was not unlimited.) My main concern from the beginning was having sufficient heat, which I minimized (to some extent) by staging the soldering in sub-assemblies, but even though I have oxy/acetyl equipment a boiler such as this sucked up all the heat I could feed it. So, let's build a boiler.

[gwrdriver]

02 - Drawings and Templates
The first step was to review the original construction drawings (dated 1953), which were a rat's nest of lines (many dashed) and redraw the design in Cad to resolve, simplify, and where needed bring things up to current construction standards and flesh out the details, some of which are left for the builder to figure out, as well as nail down all dimensions.

With a Cad drawing in hand, I developed two sets of templates (or cutting sheets) for the steel and copper material. One set of templates was for the steel formers over which (or in which) the throat sheet and back head would be formed. The second set was for cutting the sheet copper blanks for the heads and wrappers. So far as I was able, I included the necessary machining, forming, bending, and trimming allowances in the templates. The copper templates were then arranged in Cad to produce the smallest sheet of copper which would make all the heads and wrappers. Material was ordered accordingly and the patterns were laid out on the plate and the head blanks were cut.

Photo 001 shows a paper template adhered to a blank of 5/8" steel plate which would become one of the throat sheet and back head former. This was made in steel for reasons you will eventually see.

Photo 002 is of a the 1/8" copper plate marked out for cutting.

Photo 003 shows a paper template adhered to a copper blank which will become the firebox front head. Sometimes the most expedient way of doing a job is the most primitive, in this case chain drilling out the hole in the blank.

[Builder01]

Thank you for posting your build thread. I hope to build a copper boiler sometime late next year. I look forward to seeing your build to completion.

[Bill C]

Agreed, thanks for taking the time to post this thread. Subscribed.

Bill C.

[redneckalbertan]

Looking forward to the continuation of this thread!

[gwrdriver]

03 - Beginning the Head Forming

Relatively long-term projects like this never run in a straight line for me. I always have two or three operations (or even projects) under way at any given time so if I get tired (or bored) with one, or if the work needs more thought or drawing before cutting metal, I can move to another task. I mention this because my sequencing may seem to bounce around a little bit but then that's the way it was built.

I'm going to skip ahead of the next logical step, making the head formers (or flangeing plates) and show you a few finished formed heads. These are the most difficult ones and should give you a better understanding of the nature of the job and why I made the forming plates as I did. The flangeing process isn't all that complicated, but for these considerable planning, specialized forming, multiple annealing, and lots of hammering was required to get them to the point where machining and other finishing could begin.

In Photo 004, heads A and B are the back head and the rear inner firebox head, both of which were required to have a forward offset of 11/16" in the water leg and maintain a minimum water space of 3/8". The reason for this offset was never fully explained ("Mine was not to reason why, mine is but to do and die" - Tennyson) other than the owner wished to replicate the Disney boiler. Other than this offset the forming of the back heads was fairly conventional, as was the forming of the front firebox head and the front flue sheet.

The Throat Sheet (head C) was another kettle of fish. As you can see it was made in one piece and this required a fairly complex pair of forms and positioning which I'll describe in upcoming installments. The raw forming plate blank and the sheet copper blank for the Throat Sheet are shown in photos 001 and 003 above.

[Marty_Knox]

Thanks for posting this, Harry. I know how hard it is to build something, let alone document this. I've tried recording boiler construction several times, but I always reach the point where I need to 'Get it out the Door!'

[Benjamin Maggi]

It is interesting to watch a 7.25" gauge boiler being built out of copper. Most people build steel (for various reasons, usually economics) but large copper boilers seem rare these days.

[gamh44]

Thanks for posting! Great to see the attention to detail in your posts.

Watching with interest.

Geof

[Dick_Morris]

Thanks Harry.

[gwrdriver]

04 - Head Forming Basics

Since I already had a finished steel former from previous 6" copper boiler project I formed the front flue sheet (A - Photo 005) first.
Disregard the other blanks and formers as those are for a boiler for my own loco project, but that's a story for another day (or year.)

Most of you will be familiar with the flangeing process, but for those who aren't the first step is to thoroughly anneal the copper blank by heating it to a cherry red. A barely visible red color isn't enough heat IMHO, and a carrot orange is too much! Annealing is a necessary part of the process and will need to be repeated several times during the forming of each head. From time to time there is discussion about the best way to cool copper after annealing, whether to quench in water or let it cool in the air. The answer is it doesn't matter, the results will be the same, but I prefer a water quench for two reasons, it's faster, and the thermal shock created by the flash of water to steam will blast off much of the dark oxides created by heating. After bringing it to red heat the blank is quenched by quickly plunging it into cold water, and this means fully submerged in cold water, not just a quick splash under the tap.

In Photo 006 the front head blank is clamped in place on the former and is ready to begin flangeing. What this photo doesn't show is the steel backing plate on the back side of the blank. A backing plate is required for almost every flangeing operation to prevent the sheet copper puckering or rolling up when it needs to remain flat. The backing plate is clearly visible in the next photo.

Photo 007 shows the blank (and the backing plate) and the state of the flange after the first hammering. At this point the sandwich is dismantled and the blank is annealed. Correctly realigning the blank on the form is usually not a problem as after the first beating enough copper has been pushed over to form a register.

The initial annealing will normally allow the flange to be laid over about 1/3rd of the way. The second annealing and hammering will take the flange to 2/3rds of the way, the 3rd annealing and hammering will usually close it to the form. Occasionally on a head such as this one where excess metal will "gather up" at the free edge of the flange, I'll do a fourth annealing to insure the excess metal has been compressed and everything lays tight against the form.

Copper is a forgiving material but it will quickly work-harden from bending and mustn't be forced beyond the point where it won't bend further under normal hammer blows, and 1/8" copper will definitely let you know when it doesn't want to move any more. At this point it needs to be annealed again. The annealing and hammering is repeated until the flange lays completely over and is tight to the former. Photo 008 shows the front head completely formed and ready for machining.

The same process will be used to form all the heads although the more complex heads usually need to be annealed more than three times. The hammer I use for copper (Photo 008a) has replaceable rubber and plastic faces (Vaughn Mfg. #SF6) which can deliver quite a blow and I discovered it has a kind of Fail-safe attribute. It isn't heavy enough to force 1/8" copper beyond the point where an annealing is required, so when it stops moving the metal it's time to anneal. I rarely use a steel-faced hammer on copper, and then only a broad-faced "fender" hammer to snug and finish an uncooperative area around the former. This combination works well for me, but your mileage may vary.

CP008a.jpg (15.13 KiB) Viewed 5779 times

[SilverSanJuan]

This is great information, Harry. Thanks for sharing.

Todd