Final shape

Following a series of trials boiler proper took its shape...

This is a cross section of the boiler from CAD. The burner's flame is surrounded by a copper (fire)box, that will be soldered to the boiler barrel. Hot gasses are supposed to run between another copper sheet and boiler tube. The steam runs through the dry pipe with a diaphragm inside, manifold, superheater pipes soldered to the firebox sides and below the boiler tube. All this is supposed to be insulated and wrapped in a brass sheeting.

The trottle is hidden inside a dry pipe. It consists of a diaphragm that can be blocked using thick seewing needle.
into which the steam flows from the dome.
The needle goes through a slightly undersize hole made in a 0.5mm thick wall in a steam manifold piece.
These are the valve assembly parts. In the final version the diaphragm divides the dry pipe, so that its location is better conrolled.
Boiler after the steam pipes have been soldered.
With the sheeting in the place.

Trials

The problem still to be solved is an efficient burner. It turned out, that the firebox is too confined and the flow below the boiler barrel too much restriced. To remedy the first problem, I removed the front-bottom wall of the firbox. But the second problem is more difficult, since there is simply not enough place available within a resonably-to-scale diameter.

For the time beeing I decided to use gas soldering iron as a burner, at least until my model runs on steam which is yet to come.

This is how the (unsuccesful) burner "assembly" looked like. I use a cigarette lighter as a gas container, already equipped with all the valves. Shortened a bit it will fit in the tender.

Burner, Ø3x10. It gives an inch long, only a bit shining flame. The nozzle inside was drilled using 0.3mm drill and then compressed with three-jaw chuck. This works, but you cannot achieve such a momentum of gas as is commercial burners (which are oversize). Commercial burners use a aluminium foil in which a tiny hole is made, I should adopt the same approach at some point.

Thinking about an R/C controlled steam flow, I fought about making an electromagnetic valve, such as shown on the right. The ball B either sits in the nozzle closing the flow when attracted by the magnetic field of the coil C or it is repelled by the gas pressure and opens the valve.
Unfortunately it didn't work. The magnetic field of an air coil is far too weak.
I decided to give the idea another try. Making the valve body of mild steel (grey parts) should help by proving a magnetic circuit that the ball would close when closing the flow. The ball is to be located in the cavity inside, while the gas outlet is towards the reader, in the removed part.
These are the parts of the valve ready (O.D. 3mm). I have tested the valve on compressed air, without a coil and coil core, actuating it with a magnet. It works great, you need just to bring the magnet within an inch distance to the valve and the ball is attracted to the nozzle and closes the flow. It is even likely stuck. Unfortunately I was quite at rush when making it and the socket of the ball is a little bit uneven, so that the valve is never truly closed. Also, as it turned our after some more investigation, the ball is always stuck to the inner sidewall of the enclosure, so that it won't ever close the nozzle. So I decided to leave the electromagnetic idea.
Somewhat obsolete CAD section