THE FUTURE OF SCRATCH BUILDING?
From Gazette Issue 21 No. 5 November 2020
Robert Else
Photos by the author
unless indicated
The prototype on Platform 1 at Horsted Keynes station after a cosmetic makeover
I WANT TO GIVE MODELLERS an insight to my latest
creation in O gauge. Being an engineer at heart, my main
enjoyment out of modelling in O gauge is, like many,
actually building the railway, the rolling stock and
infrastructure that was found on a railway. The most
exciting item of rolling stock I had built was the 4MT by
Modern Outline Kits. However, this is not about my
experiences of building the 4MT, but of my other interest
in drawing in 3D CAD a series of individual parts, getting
these parts 3D printed, then building the model of a 45t
Ransomes & Rapier steam breakdown crane. I was initially
thinking about a Cowans Sheldon 70t crane but after some
research, and finding that the prototype for a Ransomes &
Rapier was easily accessibility at the local heritage railway,
there was no contest.
How it all started
I have been interested in breakdown cranes from when I
first saw the HO version that Marklin marketed many years
ago, the HO 1495 Krupp-Ardelt Crane. This project was
originally set aside for when I retired, to teach myself 3D
parametric solid modelling and at the same time design
something for my railway. However, with Bachmann
releasing their OO version, I thought I would give the O
gauge modellers one that they could build for their
railways. The challenge was on. I have used 2D CAD and
wireframe 3D CAD but using 3D Parametric modelling is
totally different, I had to throw away the old CAD rule book
and have a completely different approach to this design.
How do you start a project like this? Obviously, drawings
are paramount to this type of project, but where to get
these from? The initial reason for a Cowans and Sheldon
crane was that there are drawings available in the
Cumbrian Archives at a cost, but there are no drawings
from the Ransome & Rapier factory of this crane, unless
you know different!
The first port of call was the Breakdown Crane
Association ( http://www.bdca.org.uk
); yes there are others
passionate about breakdown cranes like me. I was told
about some excellent sources of information on
breakdown cranes compiled by Peter Tatlow. Peter, a very
passionate mechanical engineer produced three volumes
entitled Railway Breakdown Cranes.
Getting hold of
volumes 1 and 3 were no problem at all, they are readily
available from any good book shop or the internet, but
guess what, the information I was after, was in Volume 2.
This was a very rare book to get hold of. Luckily, the
Secretary of the BDCA had a copy of volume 2 at hand and
provided me with scans of the applicable drawings for the
45t crane I was after. This gave me the basic information to
make a start. I started with the carriage, the main frame the
crane sits on, it houses the self-propelling mechanism, the
ring of roller bearings that support the body with the gears
and winding drums. This gave me the overall dimensions
but the finer details were not included. Further
communications with the BDCA yielded no further
information.
Details of the lugs for mounting on to the carriage
Unbeknown to me, it so happened that the Bluebell
Railway had decided in 2012 to do a cosmetic restoration
of their 45t Ransomes & Rapier steam crane (RS1083/45 Ex
LNER) that had been sitting in the sidings at Kingscote
station since its retirement in 1995. In 2012 it was moved to
the sidings at Horsted Keynes but out of sight. It did have
a short working life on the Bluebell Railway, but like all
things that can blow up, it was subject to an insurance
inspection when its boiler certificate ended in 1995. Luckily
for me and my project, it was now in Platform 1 at Horsted
Keynes, nicely painted black and accessible. See the
Bluebell Railway website for details. https://www.bluebell-railway.com/
You can’t start a project with such complexity as this
without the prototype available. You need to spend hours
with a tape measure taking countless measurements,
numerous sketches of how it goes together and hundreds
of photos, in order to create accurate CAD drawings. As
with all CAD systems, you must draw it full size. So this
crane will end up 83ft 5½in long in the computer, but when
scaled down, only 592mm (23 ¼ in) long on the layout.
The final CAD model of the carriage for
the axleboxes but showing additional
brass components.
One discipline that must be ingrained form the start is
file management. The 3D drawings can take a long time to
create, and all the effort and hours of editing can be gone
in a flash, if you don’t make staged backups of the
drawings. I used a simple file naming structure: the name
of the main structure followed by the substructure and so
on, followed by the date e.g. CarriageFrontEndPlate
YYYYMMDD (hhmm). If the change was significant then I
would break it down further by following the file name with
the time. If the drawings were updated the next day, then
the filename would reflect the new date. This always
ensured you could recover a file with minimal disruption. I
learnt the hard way! Computer storage is cheap, your time
is not. You can always archive them at the end of the
project.
The first CAD drawings were of the carriage. With the
aim of myself producing this in kit form for modellers to
purchase and make themselves, the different parts had to
fit together with little knowledge of how the prototype
goes together. The advantage of using CAD is that you can
move the elements in one axis, X, Y or Z, add small lugs
and the corresponding holes and see how they fix together
by moving them back and forth.
The 3D printed jib head
& bottom third of the jib showing the cross braces at various angles
The most difficult and time consuming task was the
creation of the jib. The main issue here was the need to
position the cross members which joined each beam
together. These don’t line up with any axis, X, Y or Z. With
the main head of the jib there are a number of angles that
need to be catered for. It took me hours to understand how
the geometry of the jib lay. I’m not sure if it was the CAD
system or me, but to get the correct angles and distances
between each beam took hours of hard work until I was
happy.
Once the beams were mastered, it was how am I going to get this printed? Although it seemed to fit physically in
the 3D printer, once scaled down it was still 300mm long,
the file size was quite large and I couldn’t upload it. I
decided to break it down into three managable sections
for printing. Another good thing about CAD is it’s easy to
split models. You just give it a plane or flat surface and say
split! I added little bits of plastic on the inside but out of view,
which aligns the parts and keeps them together. One
deviation from the normal approach to CAD drawing was
that I had to create the whole assembly, i.e. the whole
carriage , from muliple parts drawn while in place. It would
be nearly impossible to draw all the individual parts and
hope they fitted together. The only assemblies that were
done in isolation were the boiler and components, the
control levers and the small boiler water feed engine.
Manufacturing
I had always planned to have this 3D printed. I had used 3D
printing with my previous employer and was fascinated by
it; now was the chance to use it with an end goal in the
frame. I explored various 3D printing services which would
manufacture the parts for my crane. I also looked at
different materials and processes. This kind of fine detail
modelling can’t be done using the Fused Deposition
Method (FDM). The resolution provided is too coarse. You
also have to consider the layer thickness. For this model it
had to be down in the tens of microns.
The eight axleboxes
for the carriage linked
together on a sprue
& 3D CAD model of the main roller bearing upon which the superstructure revolves. All the rollers rotate about their axis
The target printer also has an effect on the design. Each
printing process has limits it works to. If you don’t take
these into account in your design work, then you’ll end up
missing crucial detail. I needed to make the bolts and rivets
slightly over size so the printer could pick them up. The
steel plates had to be thicker in the 3D model than they
would have been in the prototype etc. Back to the process:
for this to work the medium had to be either nylon or a
resin. I opted for the ‘Fine Detail Nylon’ medium offered by
Shapeways which, like all 3D printing processes, is built up
in layers; in this case layers of powder fused together with
a laser, a process called Selective Laser Sintering (SLS).
This did provide the required level of detail and at a
reasonable cost.
The final assembled Carriage complete with movable props, you can
see the detail in the rivets and the plates. The gear
on the top was done in resin using the SLA process
I decided on a company that gave online quotes and
reported on errors in the designs, of which there were a
few, mainly bodies not connected to each other. 3D
printers don’t like floating rivets and provide a set of
online tools to check your designs. You also got
immediate feedback on the costs of different
materials so you could make comparisons. I did
some test runs using the resin, or SLA
(Stereolithography) process. These turn out very
well. Some of the desktop printers, like the
Anycubic Photon, do some very fine detail.
The axleboxes were rather small and if done
individually would have cost a fair amount. So, I took a leaf
out of the Airfix book and made up a spru with the eight
axleboxes for a single print. Much more cost effective. I
eagerly waited for my first prints and within a one week,
the parts arrived hot foot from Eindhoven, Holland. What
service!
The first parts back and assembled into the
carriage of the crane, everything went together
perfectly and the level of detail was astounding. You
could see all the rivets and bolts I had included.
The prototype O gauge
model produced so far
from 3D printed parts.
Now to test the 3D printer’s capability with the main
carriage bearing: 38 rollers configured in a ring and linked
together between an outer and inner ring. With careful
consideration to the printers capabilites in clearances, I
created this bearing. When the print came back, the rollers
actually rotated allowing the superstructure to rotate
exactly like the prototype.
Assembly
Now to assemble them together to make the carriage.
Everything went together extremely well, everything was
square and true, just what you would expect from CADdrawn
and digitally-manufactured parts. I chose standard
‘off the shelf’ steel wheels of the correct diameter for all
vehicles as there is plenty of choice.
Every assembly would have to be built in the CAD
software from the measurements and photos of the
prototype and then disassembled in to sub-assembles
that can be 3D printed and then assembled together in a
sensible order to recreate the final assembly.
After lots of hours learning new CAD skills, taking
numerous visits to the prototype for photos and
measurements to create the various specific parts, of
which there are lots, I have ended up with a full 3D CAD
model of the whole breakdown crane.
I have had many parts manufactured during the project to end up with all
the vehicles to make the crane set. These consist of two
relieving trucks, used to take some of the weight off the
main carriage to spread the axle load, a match truck to
hold the jib in place and the carriage itself.
Small parts are destined to be manufactured as a brass
casting, e.g. safety valves, water sight glasses, etc. This is
because the 3D printing process could not cope with the
resolutions required. They can however be designed in
exactly the same way.
Details of the boiler and associated components,
safety valve, steam supply valve, etc.
An exploded view of the parts that were 3D
printed and how they would fit together. Note the pegs and
holes for alignment.
Into production?
As I mentioned at the start, my intention was to complete
the 3D CAD model, get the parts 3D printed, use what
commercially available parts I could obtain, write some
instructions and sell it as a kit. I even had the box sorted.
What this has taught me is that nothing is as
straightforward as how you first envisaged it. All was well
to start with. I could reduce my costs at the 3D printing
stage but extras started creeping in: brass etchings, white
metal or brass castings for the small parts, brake gear and
tread plates and many more. It was becoming increasingly
evident that this project was becoming very large and the
costs were increasing, which could potentially make the
final cost to the modeller too much. Even if I did go down
this route, with still working full time, this project would go
on for another year. A change in direction was called for so
I have decided to offer the CAD drawings to a manufacture
that would be interested in taking this on and seeing it into
a production mode.
I would like to hear from any manufacture that would be
interested in taking my design work and producing either
a kit or a Ready To Run model so that other O gauge
modellers can have the delight out of owning this excellent model.
representation of a 1943 45t Ransomes & Rapier Steam
Breakdown Crane. No railway should be without one.
If any company would be interested in seeing it to a full
production model, I would love to hear from you.
robert@else.org.uk
