Home Maritime Resources
Ship Models Machine Tools Other Stuff Site history

first published in autumn 2006
last updated 13/09/22

  Wespe-Class Armoured Gunboat: The Model 

Introductory Note

Work on this model began in autumn 2006 and then progress with various long interruption for personal reasons and due to diversions, such as the construction of tools and machnies described elsewhere on this Web-site. While you will find below a step-by-step description of building the model as it progresses, this is not a continuous 'blog', so watch out for the date on the beginning of paragraphs to identify new material. For ease of reference the following table allows to jump to the various sub-sections.

The hull and superstructures
The 30.5 cm Rk/l22 gun
Deck Furniture
Skylights, Companionways etc.
Ships' Boats
Completion and Assembly

updated 20/12/2021 updated 03/10/2020
updated 14/03/21
updated 03/10/2021 new 13/09/22
updated 26/05/2022


The scale chosen for the model is 1/160, which admittedly is somewhat unusual for a ship model. However, the reasoning behind this choice was that a large selection of N-scale railway figures is available that eventually will crew the ship. There are also space and portability consideration, which are important for someone, who has to move from time to time for professional reasons.
The model will  be a waterline model. This will allow a dioramic presentation of the finished model. Besides, the hull below the waterline is not quite so graceful. Above the waterline the hull is also more or less prismatic, with vertical bulwarks and virtually no sheer. These parameters together call for a bread-and-butter construction.
The building drawings are a combination of re-drawn Admiralty plans and scans thereof. These are printed to scale on the laser printer and the print-outs glued on top of e.g. the MDF board to serve as a guidance for cutting and sanding.


Choice of materials
I had been contemplating a variety of materials for the hull; for instance Plexiglas® layers with bulwarks made from brass foil. In the end, I choose MDF (medium-density fibre) board, which is available in thicknesses down to 1 mm from architectural model supply houses.  Other parts will be constructed from or covered with Bristol board, which is also available in various thicknesses (or rather weights per square metre). The bulwarks etc.. will be made from Pertinax® (phenolic resin impregnated paper, FR-2), which is available in thickness's down to 0.1 mm. Bristol board and then Pertinax® are easily cut with a scalpel, a razor blade or scissors and will not crease or dent as metal foil might. I currently have no facilities for photo-etching large parts, but if I had, perhaps I would have made the bulwarks from brass still. The other advantage is that Bristol board can be readily and permanently glued using white glue. Bonds between large areas of metal foil and Plexiglas® might become detached, though the plating on the steam-tug, made from copper foil, has lasted now for nearly twenty years. Pertinax® can be glued using cyano-acrylate or epoxy-resins. The dinghy of the steam-tug had received planking made from Pertinax® and glued with cyano-acrylate glue.
While I have been shying away from thermoplastics, such as polystyrene, on account of it being suspicious to be not 'permanent' (e.g. the articles by Dana Wegner), practical experience shows that plastic models built over 35 years are still intact. So I may reconsider my position in this respect. Polystyrene, of course, has several advantageous properties.

Some thoughts on Etched Parts
Some people refer to the process of making photo-etched parts as chemical milling and that is the way I view it; a process to cut out and shape parts that are too small or otherwise to delicate to handle conveniently with other manufacturing processes. Unfortunately, the employment of this process moves much of the modelling work onto the computer, as the patterns or masks now are produced with the help of a drafting program. These masks are largely developed by scaling the contemporary drawings and drawing the respective part over it in a different layer. These parts are then composed into the actual mask. Of course, 'left' and 'right' sides have to be drawn separately, if the part is to receive surface-etched detail. A strict procedure of copying and mirroring has to be adhered to in order to achieve a perfect line-up. Much thinking has to go into the best shape of parts and some experimentation. The etching process is not so well controllable, as a machine tool, at least in the simple set-up I am using. The best thickness of interlocking slots or the drawing size to achieve cut-outs of a specific dimension and similar features have to be found by trial and errors sometimes. Literary it is often 'back to the drawing board'.
This set-up is only suitable for dip-etching. Commercial companies use foam or spray etching, which work faster and produce less undercut. I decided to work with very small 'frets' only, the size of one or two large stamps. This reduces the cost of material, if something goes wrong and the smaller size seems to make it easier to get uniform results over the whole 'fret'. I bought second hand a UV-source for exposing printed circuit board. It has a timer and hence makes the process more repeatable. The developing and etching vessels are plastic film tins, coming from the standard film rolls (don't get any new ones since I have switched to digital, of course). The brass is bought in a ready-sensitised state, so no messing about with UV-sensitive lacquer is needed. Much experimentation went into a suitable way of making the masks. Eventually a newly bought ink-jet printer produced sufficiently uniform print-outs on overhead foil, but the resulting masks are not really perfect. However, I did not want to go a commercial photo lithography company for them.

The hull and superstructures

Autumn 2006 - The basic bread-and-butter construction of the hull is shown in the pictures above.
The Barbette mainly consists of a semi-circular breastwork armour, backed by hardwood and by an open space covered with thin plate. The latter presumably to retain splintering wood in case of an impact. Since no tube of suitable dimensions for the breastwork was to hand, I made a short laminated one from Bristol board glued together with white glue. The edges were soaked in thinned white glue before being trimmed down on the lathe. The tube then was varnished with filler for wood before the edges were sanded. Finally a half-circle was cut from the tube on the jig-saw. More wood-filler was applied before final sanding. After cutting in half it was glued into place. The inside of the barbette was lined with hard-paper to give a smooth finish.
The fore-deck has been covered in a sheet of thin Bristol board and the camber of the wooden decking built up with an additional piece of board and putty (I am using fast drying bodywork putty from car repair suppliers). The anchor pockets have also been lined with thin Bristol board, but Pertinax would have been better for this.
All surfaces that would have been iron plating, will be covered in thin sheets of Pertinax. The necessary holes for portholes and other opening will be drilled or cut before the sheets are fixed. In this way the barbette was lined with sheets of Pertinax, as was the deck-house.

Cutting the layers on a powered fretsaw

Sanding the sides of each layer vertical on an improvised disc sander
The first layers
The barbette and pockets for the anchors cut out
Milling slots for the rubbing strakes
Improvised drum sander to work the inside of the barbette
Shaping the body on the new shop-made disc sander
Tube made from laminated Bristol board
Trimming tube on the lathe
Drilling the hawse pipe on the horizontal milling machine
Deckhouse partially clad in Pertinax

January 2017
Most of the decks were plated and this plating was covered in oil-paint that was mixed with sand and cement in order to provide a certain corrosion resistance and above all a better grip in wet conditions. A modelling plan drawn by Wolfgang Bohlayer shows wood on some decks, but evidence that since has become available shows that this was not the case. Linoleum decking apparently was never used on these boats. However, as the model will show the boat in its original appearence, the plating was reproduced by engraving fine lines into then sheets of Pertinax. All decks, including that of the barbette will covered in this way. The exception is the deck above the foc'sle that a cover in planks, presumably to reduce wear, where the anchors were worked. This planking was laid-out in a radiant pattern, which seems to have been more resistant to the gun-blast than the more common parallel layout. The planks were also reproduced by lightly engraving the plank seams. In reality these seams would have been more or less flush with the deck, depending on the temperature and humidity, but a light engraving adds some life to the appearance.

Progress in constructing the hull
Engraving plating and planking
Excerpts from contemporary drawings Barbette
Toner-trasnfer printing of bulwark layout

The barbette - The floor of the barhette is partially covered in planking, presumably to protect the armour-steel deck underneath from the damage that might occur, when the heavy shells are handled. The steel deck underneath and in front of the barbette armour-belt is slightly sloping to deflect incoming enemy-shells from the ammunition storage-rooms. Within the barbette this is filled with timber to make a level floor. The interpretation of the various items that can be seen in the contemporary drawings is not straightforward. However, one can see a hatch that gives access to the crew's quarters (where also the hand-cranks for turning the gun-carriage is located). Then there is a round hatch for hoisting up the charges from the powder-locker below and a square hatch for hoisting up the shells. From the drawings it appears that these hatches were covered in steel-gratings. There is a further hatch with a double-lid that, according to a hand-written notice on one drawing is a man-hole leading to the ante-room of the shell-locker. However, as it is not drawn in the cross-sections we do not know its height. There are also a couple of racks for shells and some other rack-like features, the purpose of which I do not know - perhaps for tools needed in handling the shells. Unfortunately, there are no photographic images that show the rear of the barbette.  Stairs leads down from the bridge into the barbette. In addition two ladders allow quick access from the deck.
The floor of the barbette, which apparently did not have any camber, was built up from two layers of Pertinax one representing the steel-plating and engraved accordingly, the second cut out and engraved to represent the wooden flooring. The construction of the various hatches is described below.

Bulwarks, hull- and deck-plating installed
Making and installing the hawse-pipes
Scraper for half-round profiles
Rails and rubbing strakes installed
Toilet evacuation pipes
Milling the steps of jacobs-ladders

May 2019 - The main-deck plating, which had already been prepared a long time ago from a piece of bakelite (see above). The holes for the various fittings where marked out over a drawing and then drilled. The translucent property of the bakelite is very helpful for marking out. Once glued on, the deck was carefully sanded to the contour of the hull.
I spent a lot of time deliberating the best way to make the plating of the hull and the bulwark. The shape is quite simple, as the sides are vertical from just below the waterline (probably to facilitate the production of the armour plating that needed to be curved in only one direction). The original idea was to cut the plating in one piece from brass shim stock. This would have resulted in near scale thickness of the bulwark plating. I considered this too flimsy, even if the handrail was soldered on. Another option would have been to use 0.13 mm styrene sheet. Again I considered it too soft. Bakelite sheet of 0.1 mm thickness would have been closer to scale, but rather brittle. For practical reasons I decided to use 0.2 mm bakelite sheet.
The layout of the freeing ports, the location of stanchions, the ash chutes, toilet drain pipes, and port-holes were drawn onto an expansion of the bulwark that was developed from the original drawings. The drawing then was laser-printed onto an overhead projection foil (remeber these ?). This foil was taped to a piece of bakelite sheet and the drawing ironed onto it, using what is called the toner-transfer method.
The plating was cemented to the MDF hull using cyanoacrylate glue (CA). I am not very fond of CA glue, but it forms secure bonds with bakelite.
On the prototype, the bulwark plating was attached to the hull by an angle iron (8 cm x 8 cm) running along the top of the hull. I simulated the vertical part with a 0.5 mm wide strip of self-adhesive aluminium sheet into which a row of rivets had been embossed. The horizontal part would disappear under a thick layer of tar-based paint that was mixed with sand and onto which sand was dusted to provide a non-slip deck.
The hawse pipes were made from some 2 mm x 0.5 mm brass tube. First the angle with the hull was cut and then an oval ring from 0.4 mm copper wire was soldered onto this surface. The part was then taken into a collet on the watchmakers lathe and drilled out to 1.7 mm ID. Finally, the funnel shape was formed with diamond burrs and polished with silicone burrs. The hawse-pioe then was cemented in place and the end above the deck ground down in situ flush with the deck. The cover on deck is an etched part I made already several years ago. It was cemented on using CA and then another funnel was shaped with diamond and silicone burrs.

Next step was to install at the bows the fairleads for mooring cables etc. These were milled and filed from 0.8 mm thick sheet of Plexiglas®.
Then the rails on the bulwark in the rear part of the ship were installed. The rail also serves as a rubbing strake and continues to the anchor-pocket at the bows. At first the bulwark and rail (0.4 mm x 1.7 mm on the model) caused some head-scratching and concerns for the stability of the arrangement. I though about cutting a longitudinal slot into some rectangular styrne, but finally decided to make it in two, with the half glued inside and outside to the bulwark that have been designed higher for the purpose. In this way a 0.4 mm x 0.7 mm styrene strip could be glued all the way to the outside of the hull. A similar strip was glued to the inside. The half-round profile was shaped using a scraper made from a piece of razor-blade and held in pin-vice. The profile was shaped after attaching it to the hull, because it was easier to clamp the rectangular styrene strip while glueing. The glueing was effected by infiltrating CA into the joint between the styrene strip and the bakelite bulwark.
Arrangements varied somewhat between the different boats of the WESPE-class, but there was a WC for the officers in the deckshouse on the starbord side and a WC and pissoir for the men and petty officers on the port side. Each had a half-round evacuation pipe rivetted to the outside of the hull. The pipes were protected against damage by a wooden fender. After a few years of service, a strong wale/rubbing strake was added to the boats that also widened to a kind of sponson at the stern to protect the screws. However, this did not exist at the time in which the model is represented.
Steps ready to be installed
Jacob-ladders Fairlead for mooring hawser
Laser-cut lids for the freeing-ports Installation of frames and lids
Laser-cut doors 
Decks-house and back of the fore-deck with the doors installed

June 2019 - There are two jacob-ladders on each side of the hull, a wider one underneath the door in the bulwark and a narrower one a bit forward. The steps probably were made from wood and had slots towards the hull to prevent the water from collecting there and to prevent the wood from rotting. The steps are made from 0.8 mm thick Plexiglas® and the slots milled in. The sheet then was sanded down to the width of the steps and the ends rounded. Then individual steps of the right thickness were cut off on the lathe set-up with a mini saw-table. Unfortunately, the steps could only be cemented to the hull using cyanoacrylate glue, there being no positive locking. A bit of cellotape provided a guide for alignment. Nevertheless, the procedure was a bit nerve-racking.
Further, fairleads for the aft mooring hawser were installed. These were made from oval rings of copper-wire. The rings were formed over two 1 mm-drills taped together, cut off and closed by silver-soldering. The rings were sanded down to half their thickness and one each of these rings cemented to the inside and outside of the hull. The hole was drilled out and filed to shape.
February 2020 - Freeing Ports - Originally I had planned to surface-etch the lids and the frames on the inside of the bulwark. The drawings for the masks were ready, but I never got around to actually etch or have the parts etched. Since I now have the laser-cutter, these parts were cut from printer-paper (80 g/m2 = 0.1 mm thick). With a width of the frames of only 0.5 mm, the surface-etched rivets may not have come out anyway. The same for the rivets on the hinges of the lids. At least not with my somewhat primitve home-etching arrangement. If I had etched the parts from 0.1 mm nickel-brass, the overall thickness would have been reduced to a more correct 0.05 mm (= 8 mm for the prototype) The lids have no latches to lock them and the ports no bars across them to prevent items or people being washed over board. This makes their construction simpler.

Papers, even the smoothest ones, alway have a certain surface-roughness, at least compared to the bakelite of the bulwark. Therefore, the chosen paper was soaked in wood filler and spread to dry on a thick glass-plate that was covered in cling-film. The latter allowed to remove the paper without it rolling up. The surface was then smoothed with very fine steel-wool. The lids were cut from the thus prepared paper, but it needed several trials to find the right cutting parameters in order to arrive at parts of the correct dimensions. This is a disadvantage of such simple laser-cutters and their software. As the material is practically free, this is only a nuisance, but no other loss. Also the etching may not work out right in the first go, which may mean a considerable loss of money and time, if the process had been outsourced-
Unfortunately, it does not work for very small parts with the paper prepared as above. It turned out to better for the very small parts, including the frames, to cut them from unprepared paper. Perhaps I should switch to dark paper. Due to its lower albedo (reflectivity) it absorbs more energy from the laser. Unfortunately, all the coloured papers I have come by so far are quite rough on the surface.

I cheated somewhat for the freeing-ports. As I was afraid that I would not been able to cut them out cleanly and evenly, I abstained from it. Also, the bakelite-paper used for the bulwark for reasons of stability would have had a scale-thickness of 64 mm, when looked on from the side. Therefore, frames and lids were glued flat onto the inside and outside of the bulwark respectively. I hope one will not notice this too much, once the stanchions are in as well.

Frames and lids were glued on with zapon-lacquer. Little laser-cut rectangles of 0.3 mm x 0.5 mm were stuck onto lids to simulate the hinges.

Foredeck and decks-house were accessible through various doors. These were cut from 0.1 mm bakelite paper with the laser-cutter. The hinges were laser-cut from thin paper. In both cases various tries were needed with different cutting parameters and slightly altered drawings in order to arrive at the correct size. Die parts were assembled using zapon-lacquer. Zapon-lacquer was also used to glue the door into place.
On historical photographs I noticed that each door had a narrow step. These were represented by shaped and laser-cut tiny strips of paper.
Once the door were in place the hole for the bullseyes were drilled out. The laser-cut hole served as a guide. Once the boat is painted, the glazing will be installed in form of short lengths of 1 mm Plexiglas rods. The front of the rods will be faced and polished carefully on the lathe.
At a later moment also the door-knobs will be turned from brass and installed.

Best available image of the bow scrollwork and name-plate (S.M.S. SCORPION)
Only available image of the stern scrollwork (S.M.S. NATTER)
Artwork for the bow scrollwork Some examples of  (unused) laser-cut scrollwork and the name-plates Scrollwork and name-plate in place Stern scrollwork in place

June 2020 - Scrollwork and name-plates - As I had tried laser-engraving on cardboard for the gun-layer stand, I wanted to try out this technique also for the scrollwork and the name-plates. Originally, I had foreseen to develop the scrollwork by printing the design onto a decal-sheet and then build it up by sculpting it over the printed lines with acrylic gel. The name-plates could have been surface-etched in brass. One could have etched, of course also the scrollwork in brass and then complete it with acrylic gel.
It is not very clear what the scrollwork looked like when new and from what material it was made. The fact that it seems to have persisted intact over the whole life of these ships may indicate that it was actually cast in some metal, rather than carved in wood.
There are no close-up photographs of sufficient resolution in the black-white-yellow paint-scheme. Closer photographs are only available from a later period, when everything was painted over in grey and some of scrollwork may have been picked out in a darker grey. Originally it was probably painted in yellow-ochre with parts of gilded. In any case, available photographs are not clear enough to truly reconstract the scrollwork, so some interpretation was necessary.
In addition to the scrollwork per se, there was a shallow sculpture of the animal after which the ship was named, for SMS WESPE, of course, a wasp. Existing photographs only give a vague idea what these sculptures really looked like. In any case not for SMS WESPE.
There has also been some scrollwork at the stern, but pictorial evidence for this is rather scarce. There is only one known photograph that gives a full view of the stern of this class of ships and this was taken at the very end of their service life. Available copies of this photograph are not clear enough to really discern what the scrollwork actually looked like, so a fair amount of imagination is needed to recreate it.
Creating the basic artwork for the decoration was a multiple-step process. First a photograph of the respective section of the model as built was taken in order to give the necessary proportions. In the next step the best available photograph with the least perspective distortions was chosen and fitted over the model photograph. In another layer of the graphics software (Graphic for iPad) the scrolls were drawn free-hand (with the iPen) using the paintbrush-function and a good amount of smoothing. This artwork was saved as a JPEG. On the Internet I found a nice drawing of a wasp and turned this into a pure b/w image with a good bit of editing in Photoshop. Both, the scrollwork and the wasp were saved as transparent GIF. In my favourite CAD-program (EazyDraw), the parts were mounted together. This could have been done also in Photoshop, but I did have a scaled drawing of the bow-section in EazyDraw to which I exactly fitted the artwork. There were also some addtional parts to be cut.
The scrollwork was cut/engraved with the laser-cutter using the ‘half-tone’ function, which means that the laser is modulated to emit less power when a grey pixel is encountered and full power, when a black pixel is encountered. I had to play in several iterations with the settings of the laser-cutter in order to arrive at a satisfactory result.
In a first try the name-boards were made in the same way, but the half-depth engraving around the letters resulted in a somewhat fuzzy apearance of the letters. I, therefore, tried out a different idea. From previous trials it was know that the laser had no effect on transparent materials and very limited effect on translucent materials. Hence, I covered some cardboard with a thin layer of Pleximon 192 (essentially liquid, light-hardening Plexiglas). A thorough curing this sandwich was sanded flat and presented to the laser-cutter. The laser removes all the cardboard, but leaves the acrylic virtually untouched, with the exception of some light surface roughness. One ends up with a piece of thin acrylic sheet to which the letters and the scrollwork of the name-board are attached. Within the limits of the resolution (0.05 mm) of the laser-cutter the lettering turned out reasonably clear, perhaps not as crisp, as when photoetched though.
The scrollwork elements were attached to the hull using fast-drying varnish. The actual painting and guilding will be done, once the hull has been painted.

The aft part of a WESPE-Class-Boat (Lavverenz, 1900)
Etched and soldered together stanchions (they are about 5.5 mm high)
The bulwark-stanchions in place
Recessed slide and anchor release gear
Recessed slide with Inglefied-anchor put temporarily in place  View of the bow with the anchor stowage Plexiglas plugs ready for insertion
Glazed portholes
Glazed portholes

December 2020 - bulwark stanchions
- The bulwark in the aft part of the hull is supported by a number of stanchions that were cut from sheet metal and rivetted together. The looks for these stanchions is reasonably well documented on a number of photographs.
The stanchions I had drawn already years ago and depicted the rivetting by surface-etching. The material is 0.1 mm thick nickel silver. They were made in double as mirror images and soft-soldered together in pairs with soldering paste so that the rivetting appears on both sides. The location of the stanchions was marked on the bulwark before this was put into place by thermo-transfer of a drawing, i.e. a laserprinter printout was ironed on. The stanchions were cemented in place with fast-dryining varnish.
Already a short while ago I had fashioned the boiler-ash chutes by milling to shape little blocks of acrylic glass. They were cemented to the bulwark inside and outside at this stage too.
January 2021 - Anchor stowage and release gear - The Inglefield-anchors are stored on sort of recessed slides and released by a traditional form of gear. This gear consists of a rotatable iron bar with a couple of thumbs welded on over which the securing chains are hooked. The chains go around the anchor and the other end is shackled to the wall of the recess. The bar is prevented from rotating by lever that is also welded to it. The lever in turn is locked by a rotating claw at the end of a second lever. I suspected this mechanism from the available drawings, but wasn’t shure about it – a German colleague had better eyes than me an could confirm this indeed on the not very clear photographs.
The slide is protected by three T-rails on each from the weight of the heavy anchors.
The release gear was fabricated from 0.3 mm diameter tinned copper wire and assembled using varnish. The rails in turn are fabricated from laser-cut strips of Canson-paper that was soaked in varnish. They also function as bearing for the bar of the release gear. I suspect the bearings were a bit more elaborate on the prototype, but I don’t have more detailed information. The locking claw is also a microscopic laser-cut piece. As usual, I had to experiment with different variants of the drawings and settings of the laser-cutter until I managed to produce reasonably clean parts.
December 2021 - Porthole Glazing - Following the discussion on ways to make the porthole glazing further up, I looked over all available photographs and came to the conclusion that one does not actually seem to see the bronze frame from the outside. On the other hand, most photographs or their scans do not have sufficient resolution to really see such detail.
In order to make my life simpler, I decided to go for solid Plexiglas plugs. I did have 1 mm Plexiglas rod in stock and short sections were cut from this to make 2 mm long plugs. The plugs have to be a bit longer than their diameter, so that they can be inserted straight. The front face was turned flat on the lathe and the back-end was given a bit of a chamfer for easy entry into the pre-drilled holes after which it was painted black using a black permanent marker pen. The pieces were then transferred to the micro-mill for polishing the front face with a silicon rubber polishing bit.
In order ensure that the porthole plugs are set at equal depth, a little ‘tool’ was made, a punch with a recess of 0.3 mm depth around the rim.

The 30.5 cm Rk/l22 gun

Lower Carriage

February 2007 - There are some fixtures for the gun that need to go into their place in the barbette early during the construction, including the races for the gun carriage and the semi-circular toothed rack that is part of the gun-training machinery. I decided to make these from steel, even though ferrous metals in model construction are frowned upon by museums. My justifications were that it is difficult to represent cast iron or steel by paint and that there hundreds of models in museums around the world that contain iron. I have used steel it in models some twenty years ago and presumably due to the lacquering shows no signs of rust.
Cutting thin disks from round stock of sizeable diameter is a pain I wanted to avoid. Against my better knowledge I picked a suitably sized steel washer as starting material. Unfortunately, the steel used does not cut very well at all and lot effort was spent to avoid chatter marks while turning and to obtain a reasonably good finish. The various types of wheel collets available for the watchmaking lathe come into good use for working on inside and outside diameters of the disks.
I set up the hand-shaper for cutting the rack teeth, but had to throw away the first two attempts because of the poor material and because - again against better knowledge - I did not lock the traverse slide when cutting. The table was removed from the shaper and the home-made dividing head bolted on instead. For lack of a proper tool grinder (another project) I hand-ground a cutter for the rack tooth (0.1 mm at the bottom) from a rod of high-speed steel. For holding this tool-bit in the shaper, the old lantern-style tool holder from the watch lathe came very handy. The unwanted parts of the ring were cut away on the shaper using ordinary left and right hand lathe tools. Finally the necessary sections were trimmed off with a fine saw blade on the lathe's sawing table.

Roughing out the metal disk with the backing of a wooden disk

Grooving the races with a specially ground bit
Cutting out the inside of the large, backward ring
Trimming the outside of the small, forward ring
Shaper set-up for cutting the toothed rack
Cutting the toothed rack with a specially ground tool
Cutting away the unwanted part of the ring with an ordinary tool
The set-up showing the finished rack
The races and the toothed rack ready to be trimmed to correct length of arc
Base-plate and rails for upper carriage laser-cut from Canson-paper The basic frame of the lower carriage from the rear

2020 - The lower carriage of the gun was a rather complex construction from rolled L-profiles and thick steel sheet. Unfortunately only the drawings in GALSTER (1885) and the coloured synoptic drawing from the Admiralty have come to us. Many construction details are superimposed onto each other with dashed lines, so that the interpretation of the drawings is rather difficult in places. As aids to interpretation with have one close-up photograph, the large demonstration model in the navy museum in Copenhagen, and the preserved guns of Suomenlinna Fortress off Helsinki. The carriage for the Danish iron-clad HELGOLAND, however, differs from that of SMS WESPE in some details, being actually a turret-carriage. The carriages in Suomenlinna are Russian copies of Krupp fortress carriages, but they allow to verify certain construction details that are not clear from the drawings.
Originally I had planned to construct the lower carriage, like the upper carriage, from surface-etched brass parts. To this end I produced some time ago already the needed detail drawings. Surface etching is a very good process to simulate rivetting. In the meantime, however, I had purchased the laser-cutter, so that laser-cut parts would be an alternative. I had hoped to cut the parts from bakelite paper. Various trials with different cutting parameters unfortunately were not very successfull for the intricate parts. The 5 W laser ist too weak to burn the material fast enough. Burrs of molten and partially carbonised resin form. Therefore, I fell back onto Canson-paper, which is a bit over scale with its thickness of 0.15 mm.
The drawings for the etching masks had to be reworked for laser cutting. It turned out during assembly that I had made several mistakes or misinterpretations. If I had send them off for etching this would have been costly, as both masks and etching would have to be redone. When cutting paper with a laser such corrections can be made quickly and easily – and the material costs practically nothing.
The laser-cut parts were soaked in nitrocellulose wood-filler and once dry rubbed with very fine steel wool. To double up parts and for assembly zapon lacquer was used. This dries so fast that no special arrangements for fixing the parts are needed.
I did not take pictures of the different steps of assembly, as this would have rather impeded the process. First all parts to be doubled up were cemented together using zapon lacquer and weighed down to keep them flat during drying. The longitudinal parts of the carriage had slots cut into them, so that the transveral parts could be positioned exactly. The frame assembly then was cemented to the base plate (which in reality was not a plate, but rather the frame was put together from L-profiles and steel sheets). The racers, again in one piece, where glued on top of this assembly. Underneath the base plate the housing for the training gears (which will be very much simplified as they will be barely visible upon completion of the model).
One can see on the laser-cut parts marks for the rivets. These will be added as tiny spots of white glue. More details will be added in the next steps, but have not all been drawn yet.

The basic frame of the lower carriage from the rear

The basic frame of the lower carriage from underneath with the housing for the training gears
The basis frame of the lower carriage with the upper carriage and the gun put temporarily in place
Working drawing for the parts of the hydraulic brake
The individual parts of the hydraulic recoil-brake
Dry-fitting of the recoil-brake into the lower carriage frame

Buffer beams on the lower carriage
One buffer dry-mounted

March 2020 - The 30.5 cm gun in pivot-carriage C/76 was one of the first guns in the Imperial German Navy that was fitted with a hydraulic recoil-brake, at a time, when compressors and brooks were still the standard.
The recoil-brake consists of a long cylinder with screwed-on cylinder-covers at both ends. The covers are pierced for piston rods and are sealed with packed glands. The piston rods are fixed at the front and rear end of the carriage respectivly. The piston is designed as self-opening one-way valve. The cylinder is filled with glycerine through a valve on top. The front-end cylinder covers acts also as cross-head and the upper carriage is linked up through two short forked connecting rods. The cross-head runs on a kind of slide to support the weight of the brake. The two piston-rods are only connected by the short piston, which also acts as valve, and that would not be able to support the weight.
When the gun is fired, the upper carriage slides back and the piston is pushed through the glycerine, converting the kinetic energy of the recoil into heat. The valve in the piston prevents the upper carriage from sliding back into firing position. In order to bring the gun forward, the rear end of the carriage is raised by turning the excentric bearings of the rear wheels and opening the valve in the piston. To facilitate this, the rear piston rod is hollow and a spring-loaded valve-rod extends beyond the piston-rod. The valve rod can be srcewed in and out by the aiming gunner using a long lever. In this way he can let the gun roll back into the firing position in a controlled way.
Unfortunately, not much of the hydraulic brake will be visible on the finished model, so that it was reproduced in a somewhat simplified way. It consists of five parts.
The piston rods were fashioned from clothes pins of 0.6 mm and 0.7 mm diameter respectively. Clothes pins are very suitable for piston rods, as they have a nicely polished surface. The eye of front piston rod was milled/filed from the head of the clothes pin.
The cylinder was turned in one piece together with the covers from a short length of 2.5 mm round steel. On the micro-mill a hole was cross-drilled for another short piece of steel that had the cross-head pins turned on. This piece was soft-soldered into the cylinder cover. The packed gland is compressed by a hexagonal nut, for which the hexagon was milled on in the dividing head in the same set-up.
The forked connecting links were laser-cut from paper and consist of three pieces each. The bronze housing for the valve spring was turned from 1 mm brass rod. The valve lever will be added at a later point.

Buffers and fastening nuts

Buffers and fastening nuts – the buffer have a diameter of 1 mm
Safety claw, pivot plate and drive shaft
Milling the loading crane
Fork for pulley
Milling the pinion and cog-wheel for the winding mechanisms
Part-assembled loading crane

Buffer beams - In order to limit the recoil and the running out of the gun, buffer beams are installed at both ends of the frame of the lower carriage. Each beam carries four buffers against which the front cross-beam of the upper carriage runs. The buffers are designed as pistons with piston rods screwed to the back of the beam. It is not completely clear what the elastic elements were. The drawings seem to indicate rubber discs with metal separating discs. On some of the guns at Suomenlinna fortress there are remains of rubber discs, while the demonstration model of the Danish navy seems to have spiral springs.
The bodies of the buffers were turned from 1 mm soft steel wire. The spring element was simulated by winding around it several turns of 0.15 mm tinned copper wire. Whether this is meant to meant to represent rubber discs or springs I will decide, when it comes to the painting stage.
The nuts that keep the buffers to the beam were also turned from 1 mm soft steel wire. First, the hexagon for a 0.6 mm spanner width was milled on in the dividing head of the micro-mill. On the lathe a 0.4 mm hole was drilled and 0.3 mm long nuts parted off. And no, I didn’t cut a 0.4 mm thread.
The parts of the buffer beams were laser-cut from 0.15 mm thick Canson paper and soaked in wood-sealer. They were folded and assembled using zapon varnish. In order to make folding more precise, a row of tiny holes were ‘punched’ along the folding lines with the laser-cutter, which weakens the material there. The rivetting was simulated by tiny drops of acrylic gel that was applied with a syringe and a fine injection needle. The needle was ground flat at the end for this purpose.
Safety claws - A heavy forged claw at each end of the frame hooks under the rail on which the carriage trucks run to prevent the carriage from lifting off the pivot. The profile of the hooks was taken off the original drawings and cut in multiple copies from Canson paper. These were glued together as a stack and sanded smooth – not a 100% satisfying solution, but filing such tiny but wide claws from the solid I found too fiddly. The lugs that attach the claws to the frame were also cut from Canson paper.
The gun is trained with the aid of a curved rack, a crown-wheel segment in fact. In to this rack made from bronze, a steel pinion engages that is driven by a shaft from a sort differential, which is powered by man-power from the deck below the barbette. After some consideration I decided not to make the pinion, though I would have liked the challenge, because it will not be visible once the gun has been installed on board. The driving shaft, which also is barely visible was fashioned in a simplified was from a clothes pin, the head of which was turned to shape.
May 2020 - Loading crane - Mechanically, the loading crane is a relatively simple affair, a rope winding drum driven through a pinion and cog-wheel, powered by a hand-crank, and for turning a worm-wheel drive equally powerd by a hand-crank. The console on which the crane rests is a quite complex part that was bolted together from several cast parts. My first thought was to mill the console from the solid or rather to solder it together from several milled parts. I finally decided to put the laser-cutter to work and fabricate it from several cardboard pieces. On the bottom line, this was the easiest solution and compatible with the rest of the under-carriage
The crane on the demonstration model in Copenhagen mainly consists of bright pieces of steel or cast-iron. Whether this was the case too originally on the prototype cannot be verified anymore, as no detail photographs exist. It is perhaps doubtful due to the continuous maintenance required to keep rust at bay. Although, the navy was not concerned about camouflage at that time, they were aware of the risk of early detection by the enemy due to bright metal part reflecting the sun. However, I allowed myself the artisanal-aesthetic license of bright metal, as I think it will be a nice contrast to the dark green of the gun carriage later.
The actual crane was milled from a 2.5 mm steel rod. To this end the thickness profiles in both dimensions were taken off the original drawings and ‚stretched’ out straight in the CAD software. After milling, the part was softened in the flame, so that it could be bent according to the drawing. The hole and slot for the pulley were machined afterwards, as the part could break there during bending. The final shaping was done with silicone-bound grinding bits.
Pulleys and forks form them are tiny parts machined on the lathe and the milling machine. The mechanism of the crane consists of a good dozen of lathe-turned parts, that were, apart from their minute size, were not particularly challenging.
The cog-wheel, the pinion, and the worm-wheel were turned together with their axes in one piece. On the photographs I counted 60 teeth on the large wheel, which gives, together with a diameter of 3 mm a module of 0.05. Making a single tooth mill seem to be too much work, so that I took the short-cut of just gashing the wheels with a 0.1 mm thick circular saw. It is only about the look and I did not intend to make these gears functional. Hobbing a worm-wheel of just 1 mm diameter was too big of a challenge, but at least I tilted the axis 20° when gashing it.
The final assembly can only be done, once the crane-console has been attached to the carriage and the whole thing is painted.

Drawing for laser-cutting - gun-layer stand

First Version with engraved surfaces of the platform for the gun-layer
Final Version of the platform for the gun-layer
Tea-bag fabric
The collection of gratings and steps
Caster wheels prepared for assembly Caster wheels in place

May 2020 - Gun operating platforms and gratings - The gun is mounted effectively on a turntable, so that platforms for crew are needed to give them access to the gun, while is being trained left or right. These platforms are made of wire gratings that are placed into angle-iron frames. The frames are suspended from the lower carriage by brackets. The pictorial evidence (photographs, drawings) is not detailed enough to fully understand what the brackets actually looked like and how and where exactly they were attached to the lower carriage frame. Some additional information is given by the Danish instruction model and the Russian clones in Suomenlinna fortress, but the carriages of these guns differ in detail from that on SMS WESPE. So the reconstruction of these platforms remains somewhat conjectural.
There are 13 gratings and steps in total, plus the platform for the gun-layer. The original plan was to photo-etch the frames from brass sheet, but with the arrival of the laser-cutter I changed this plan. The drawings were modified accordingly. The obvious solution to simulate the angle-iron frame was to design an open frame and then fold-up the vertical parts of the angle. However, it proved impossible to fold the narrow, 0.3 to 0.4 mm strips consistently and without distortions. Not sure this would have worked with the PE parts either. It was then decided to make the open frame and the vertical parts separately as narrow strips and glue them together with lacquer. After several iterations of drawings and laser-cutter settings to arrive a workable width of the strips etc. I arrived at an acceptable solution, albeit the ‘angle-irons’ are somewhat over-scale.
Assembly was a slow and nerve-wracking process. I did not manage to do more than one grating per evening and it involved a lot of (mental) foul language. Eventually, I got them all together. Zapon-varnish was used throughout the assembly. The finished parts are surprisingly strong
The original plan was to simulate the wire-mesh of the gratings by real wire-mesh and I obtained from wires.co.uk some really fine mesh in brass and steel. The idea was to pull every second wire in one direction, as the original mesh was rectangular. It proved, however, very difficult to cut such small pieces (sometimes only 1.5 mm wide) from the wire-mesh. Then a present to wife in form of a box with various (fruit) teas came to my rescue: some of the teas came in bags made from extremely fine but lightly woven fabric. I do not know what material it is, but as it dissolves in acetone, it is probably cellulose acetate silk or Rayon. Such fabrics are also used in silk-screen printing and I had not chanced upon the tea-bags, I would have looked there. This silk-screen or fabric can be precisely and easily cut with a new scalpel blade. The small pieces of fabric were dropped into the frames and fixed at the edges with a light touch of varnish.
The platform for the gun-layer is a more complex structure. A 5 mm sheet-metal armour shield is meant to protect him from shrapnel and small-arms fire. The armour shield is reinforced at the edges with rivetted-on metal strips. The original plan was to produce this as a surface-etched part. I realised that the laser-cutter interprets half-tone images as instructions to modulate the laser power so that it does not cut all the way through. Laser-engraving in other words. It did produce the desired effect, albeit with the engraved surface being rather rough due to the digitising effect. However, this part then was so thin and flimsy, that it would not stay in shape, when attempting to shape the round corner. I reluctantly accepted that it would be somewhat over-scale in thickness and cut the armour shield and the reinforcing strips separately. They were glued on top of each other with varnish and then the round of the shield formed over a rod. Folding and gluing completed the process.
I am not entirely happy with the result and tend to think, that etched parts may have looked finer. But then their assembly would have required a lot of very delicate soldering work – I don’t trust CA for metal/metal bonds too much. On the other hand, attaching the gratings to the lower carriage frame is likely to be easier for the cardboard parts than for brass parts. Before that can be done, I need to add the wheels, which requires a lot of handling ...
June 2020 - Caster-wheels - The (more or less) central pivot determines its rotational axis, but the weight of the gun is actually supported by four (kind of) caster wheels running on cast-iron rails bolted to the bottom of the barbette. The rails had been turned already a long time ago. The forks for the caster-wheels were fabricated from laser-cut cardboard. The wheels themselves are simple turned steel discs with a groove.
For the assembly, the rails were taped down onto an appropriately scaled print-out of the original plan of the vessel and carriage fixed with a clothes pin. The wheels and forks are temporarly united by axels made from short lengths of copper wire. The casters then were cemented under the carriage in the correct position with respect to both, the rails and the carriage frame, using again varnish.
The wheels will have to be removed again before painting the carriage, because they will be left in bright steel. I do not know, whether this is correct for the flanges of the wheels, but it gives the whole arrangement are rather ‘technical’ look. The axles with cylindrical end-caps have already been prepared from steel rod and will be installed during the final assembly.

Stiffening brackets added over the caster-rollers

Supporting brackets and rods for working the training gears Rollers in brackets to lead the running-in tackle The lower carriage with the gratings installed
Lower carriage temporarily placed into the barbette

June 2020 - More details on the lower carriage - While I was drawing some additional parts to be cut with the laser, I realised, that I had completely forgotten the stiffening brackets for caster wheels. They are essential elements in the construction, as the wheels each have to carry around 15 tons of the total weight of the gun. The brackets were fabricated from steel plates and forged(?) angles, fabricated on the model from tiny pieces of Canson-paper cut with the laser.
There were also two brackets needed for the operating lever including connecting rod of the gun training mechanism and for the clutch that connects the cranks below the barbette with the gun. The latter allows to connect gears for two different speed ratios, a high ratio for fine weather and a low ratio through as self-locking worm-gear for foul weather. A quite sophisticated arrangement actually, but as nothing of it will be visible on the model, it was ignored.
Connected to the gun training mechanisms is also a kind of capstan to help run-in the gun. A tackle is hooked into each side of the upper carriage and the runner lead by two guiding wheels into the lower carriage and onto the capstan. The wheels were turned from steel rod and their supporting brackets cut from Canson-paper. I meant to closely reproduce the original design, but in the end had to simplify it, because the parts were simply too small to laser-cut and handle. Because they are so flimsy that had to be put into place now and will have to painted over.
Finally the gratings were installed. Their brackets have flaps for glueing. The 'glue' used was again zapon-lacquer, which results in a surprisingly strong joint. The platform for the gun-layer was only put up for the photographs. It has not been attached yet, as it is too delicate and would impede the painting and the handling of the carriage.

The gun barrel and lock

March 2007 - Because there will various visible areas of bare metal, the material of the original, that is steel, was chosen.  A piece of round bar was faced, centred and rough drilled for the bore. This hole served as a protective counter bore for the tailstock centre during the following turning operations. In order to get a good roughening finish the automatic feed was set up. Unfortunately the minimum feed per revolution on the watchmaking lathes is still too high to get a 'mirror' finish. One day I have to construct some sort of reduction gear. The outer part of the barrel has slight taper (1 degree included angle) and the top-slide was off-set for this operation. For rounding off the ends of the rings the LS&Co. hand tool rest came to good use. The work was finished off with fine wet-and-dry paper (remember to cover ways!) and steel wool. The bore was bored to diameter using the slide-rest and micro-boring tool. I had originally envisaged to also show the rifling, but a quick calculation told me that for a 1 mm bore and 72 rifled fields I would need a tool edge just over 0.04 mm wide ...

Races and rack provisionally in their place inside the barbette

Facing and centring a piece of steel rod for the gun barrel
Rough drilling of the gun barrel
Turning the barrel using the automatic fine feed
Taper-turning with off-set slide rest
Rounding the 'rings' using a hand turning rest
Boring the barrel using a micro boring tool
Set-up showing for milling the seat for the lock

For drilling holes for the trunnions and milling the seat of the lock the diving head was set up on the slide-rest. I could have done this operation on the milling machine, but on the lathe the dividing head is centred automatically. The outer end of the barrel was supported by the arm with an appropriate centre fitted. The resulting shape from the milling operation looks like a keyhole, but something like a mushroom shape with sharp edges is required. This was achieved by hand filing. For the next operation the set-up had to be transferred to the mill anyway: milling the seats for the square trunnions. The trunnions merge in a concave curve with the barrel. The trunnions were turned up on the lathe as disk with two round stubs protruding from either end. In the dividing head on the mill the disk was milled square to the size of the seat (or rather the other way round). These parts then were soft-soldered to the barrel. Back on the mill the concave curves of the square part of the trunnion were milled using a miniature ball-head cutter, rotating the barrel in the dividing head.
Aiming a gun in these days was a rather primitive affair, using just simple sights. The sights (two of them on either side of the barrel) consisted essentially of a round bar with a sliding rod to give the elevation. The beads (mounted near the trunnions) were observed through a ring of inverted U-shape on top of the rod. The bar was screwed into a notch in the barrel. Now, drilling into a round at a tangent is nearly impossible without deflection and breaking the drill (0.3 mm!). Therefore, I ground flat a broken drill bit to make a make-shift micro-mill and sunk a start hole. This was finished with an ordinary drill.

Close-up of the milling operation in the dividing head with support

Working drawing and files used to finish the lock seat
Milling the square part of the trunnions
Milling the seat for the trunnions
Trying the trunnion
Milling the concave transition between trunnion and barrel
Milling the seat for the sights Drilling the seats for the sights
milling  the lock piece
Cutting off the finished lock piece

The next thing to be tackled was the lock piece. This 'wedge' has a rather complex shape with a flat front, but a round back and various recesses and cut-out. I decided it would be best to undertake most of the machining operations while it is still attached to some (round) material that can be easily hold in a collet. The round back was milled on the mill's rotary table after the various coaxial holes had been drilled and the flat sides milled, all in the same set-up. For machining the other recesses the piece had to transferred to the diving head on the mill. The large ring was also turned up and two holes drilled into it for seating the circular rack that forms part of the elevating gear.
The most time consuming part turned out to be the cover piece for the lock, which in the prototype was fastened by five hexagonal head bolts. It holds the moving and locking screws in their place. It took me four tries before I produced a half-way satisfactory piece. Soldering the microscopic bolts (0.4 mm head diameter) in place got me quite a few grey hairs. Finally a fake locking screw was turned up and the moving screw, which moves the lock in and out, was faked from a couple of drilled-together 0.1 mm copper wires, covered in a thin layer of solder to make them look like steel.
The various parts of the lock were assembled using lacquer and cyanoacrylate glue.

Milling square and hexagonal bolts

Facing the locking screw in special protective brass collet
The (almost) finished gun barrel with its lock
Part view of the drawings for the photo-etched upper carriage frames Surface etched frames for the upper carriage Filler and covering pieces laid out for soldering Assembled side pieces and ties laid out

The upper carriage

Throughout 2008 - Much time has been spent on re-drawing the carriage as templates for etched parts. After the etching process has been more or less 'mastered', surface etched parts of sufficient quality were produced.
February 2009 - The side pieces have been assembled. A filler was sawn from 0.8 mm brass sheet and the etched covers soldered on. Then 'rivetted angle-irons', from etched parts were soldered on. These will connected by tie-plates. The frame is also strengthend by horizontal ties. These are composites from several etched parts in order to show the rivetting. The horizontal ties were soldered to the side pieces, while the bulkhead-like ties were glued in because it would have been to difficult and risky to bring the heat for soldering at the right places. The covers for the trunnion-bearings were bent from an etched part and soldered together.
The upper carriage was further kitted-out with wheels, the gears etc.  The front and rear rollers were turned from steel to give them a real 'steel' appearance. On the prototype the rear rollers sit in excentric bearings that allows them to be brought into to contact with the rails on the lower carriage: when being fired the upper carriage slides back on these rails, the rollers allow it to roll back into the firing position.
Assembled carriage from the rear
Assmbled carriage from the front
Carriage with the barrel in place. Note the trunnion bearings cover (not yet trimmed to lenght)
Added the rollers plus the sockets aft for the lever that is used to turn the excentric bearings of the rear rollers

March 2009 - The gears were cut from brass stock in the milling machine with the help of direct dividing head and different division plates. The shape of the teeth is not exactly correct, because I used a disc-shaped burr as cutting tool. However, at this module (0.06), where the teeths are merely pitched 0.1 mm apart, this is hardly noticeable. The gear wheels are parted off from the stock on the lathe. The gear segment that will be attached to the barrel was produced in the same manner.

Cutting the gears for the gun elevating mechanism using different division plates
Cut-off wheels before further machining
The elevating gear train in GALSTER (1885)
The elevating gears on the instruction model in Copenhagen
Krupp factory photograph (TU Berlin) The step-wise forming of the dished handwheel

July 2020 - Completing the upper carriage - With the lower carriage basically ready for painting, I turned my attention back to the upper carriage. The structural elements made from photo-etched parts had already been constructed many years ago. Dito some of the details had been fabricated more than ten years ago, or at least partially. The elevating mechanism consist of a double reduction gears and is driven by a deeply dished handwheel with six spokes. These reduction gears are duplicated on each side of the carriage. The last wheel in the drive has a pinion on the inside of the carriage, which acts on a gear segment that is attached to the gun barrel. How the gear segment is guided is not clear from the available drawings and the model in Copenhagen. On the Russian Krupp-clones the arrangement is slightly different.

There is a friction-brake on the axle of the last large wheel of the gear train, which is worked with a cross handle. How this functions is not clear, but it presumably just pull the gear onto the frame via a short thread that is cut onto the end of the axle. On the starboard side of the gun there is a brass disc and an indicator lever that somehow shows the degree of elevation and presumably the range of the gun with different kinds of projectiles and charges. Again, how this indicator disc is coupled to the elevating gears is not clear, as I do not have any suitable photographs. In any case, the respective gear train will not be really visible on the model.

The dished handwheel started life as parts photoetched from 0.2 mm brass. In order be able to bend each spoke into the dished shape, a former was turned from some round steel and set up on the watchmakers ‘staking tool’. The spokes were pre-bend by hand and then finally pulled to shape using a hollow punch. The parts then were chemically tinned and soldered together with the aid of some flux.

The remaining parts, such as the axles, are simple parts turned from steel rod for strength, as they are quite long compared to the diameter.

August 2020 - The gear segment for the elevating mechanism of the barrel was produced by turning a short piece of copper pipe that I happened to have in stock to the correct inside and outside diameters. The teeth then were cut on the micro milling-machine using the dividing head in a horizontal position. Then slots were sawn at the angular distance required and then a slice of the required thickness parted off. The ends of the segments were finally filed to shape. The copper then was tinned in self-tinning solution to resemble steel. For the brackets with which the gear segment was attached to the reenforcement ring of the gun barrel a piece of brass rod was turned out to the correct inside diameter. On the mikro-mill with the dividing attachment in upright position the other faces were milled to shape. Finally, the individual bracket were sawn off with a circular saw at the correct thickness. The parts, which are just over 1 mm long, were chemically tinned to adapt them somewhat to the steel colour of the barrel. As they will not have to withstand any mechanical forces, they were glued to the reenforcement ring with zapon lacquer.
There were still a few details missing on the upper carriage, for instance the indicator disc for the elevating mechanism. How this indicator is coupled to the elevating mechanism I was not able to find out. It is not shown on the drawings, it is not visible on the model in Copenhagen, and the respective parts are missing from the guns in the Suomenlinna fortress. There was probably a gear train on the inside of the carriage. For this indicator disc a piece of 2 mm brass rod was faced off and a mock gradation engraved with a toolbit turned onto its side in 6° steps. There is a steel indictor lever (the function of which is not clear to me, either the disc turned or this lever, probably the former). For this a steel disc was turned with a short arbor and transferred to the micro-mill, where the shape of the lever was milled out. This indicator disc seems to have been fitted only to the starbord side of the carriage.
Furthermore the brake-handels for the elevating mechanism were missing. A short piece of 0.25 mm diametre copper wire was flattend in the middle with a 0.8 mm diametre punch in the watchmaker’s staking tool. The resulting round flat part was soldered to a short distancing bushing and turned cap glued on from the other side.

Progress in homeopathic doses: I realised that I forgot the the two steps at the end of the upper carriage. So, the parts for the frame were laser-cut, pieces of tea-bag mesh inserted and the assembly attached to the carriage with lacquer.

(Almost) all the parts of the elevating gear laid out
The elevanting gear provisionally assembled
Engraving the indicator disc for the elevating mechanism on the lathe
Steps for the gun-layer

September 2020 - Assembly of the gun
I realised now that I had assembled so many tiny parts for the gun, that it became difficult to not loose them and to remember what they were for. Some of the parts indeed had been made years ago. Therefore, I will proceed now to paint the parts and to assemble the gun, which then will be placed as a whole into the barbette, once the model is getting close to be finished.

The gun carriage will be painted green, as evidenced by some contemporary builders’ models and a somewhat later instruction manual. The hue of the green is another issue. It was probably based on chrome oxide green.
The barrel of these breech-loading guns was scraped clean, then wiped with vinegar until a brownish oxide layer developed. The process was repeated several times and any loose ‘rust’ wiped off. Finally, the barrel was rub down with lineseed oil, effectively producing in situ a paint with ferric oxihyroxide and ferric acetate as pigment. The resulting colour would be something like caput mortuum. This is the way the barrel of the demonstration model in Copenhagen seems to have been treated. Moving parts and mechanically relevant surfaces were keept clean carefully, of course. I will, therefore, lightly spray the barrel in Schmincke caput mortuum.
All parts temporarily assembled had to be taken apart for painting first. After selecting a green for the carriage, all the parts were given several light coats with the airbrush until a uniform colour and sheen was achieved. Not so easy on some of the complex parts. After letting it thoroughly dry, the paint was scraped off from those parts that are meant to be bare metal, but could not be masked off, due to being difficult to access.

The assembly then proceeded from the inside out on the lower carriage. First the parts for the hydraulic recoil brake were installed. I decided to deviate from the prototype and not to install the protective tunnel over the piston of the brake in order to show the metal-work. I think this small bit of artistic license is permissible. All parts were put together with small blobs of zapon-lacquer, which dries up quite invisible.

Next the spring buffers were installed. Putting in the tiny hexagonal nuts required a very deep breath each time.

Flipping the carriage over the caster-wheels were put back, but this really taxed my patience. The wheels are held in place by little flat-head pins inserted from both sides. A simple through-pin would have been easier to install, but wouldn’t be quite prototype fashion.

The lower-carriage was very difficult to handle due to the flimsy and delicate grilles and steps. One was broken off in the process, but luckily attached nicely again. 

The rail on which the upper carriage runs would be bare metal. Here the limitations of using cardboard as structural element shows its limitations. If I had used etched brass parts, I would have chemically tinned them before assembly and now could have just scraped off the paint or masked the area before painting to reveal the metal. Now I had to simulate it with paint and a soft lead pencil. I am not entirely satisfied with the result, but can’t do anything about it now anymore.

Overall, I am somewhat ambivalent as to the merit of using cardboard. The surface and cut edges simply are not as smooth as those of metal or plastics, such as bakelite paper or styrene. Unfortunately, styrene could not be cut with my small laser-cutter.

When proceeding to the upper carriage, I noticed a couple of mistakes I made years ago, when putting it together. Two of the transversal members were installed at a wrong place. The wheels of the carriage would have not touched the rails otherwise. When trying to rectify this, the whole assembly gave, but luckily I managed to put it back together without permanent damage.

Another issue also arose: one should not work from drawings alone, particularly in a project that streches so long as this one. It turned out that the carriage was a couple of tenths of milimeters to narrow and would not fit over the lower carriage with its guiding plates. I should have properly verified this, when developing the parts for the lower carriage. With a bit of bending and tweaking it could be made to fit, but cobble-jobs like this leave parts behind that are not as crisp as they should be.

Painting the gun barrel turned out to be a major nightmare. I did not want to prime the steel in order to not loose its metallic appearance. Usually, acrylic paints dry so fast that there are not serious issues with rust formation. When I first applied the first coat it looked ok, but the next morning it had developed a mottled appearance. The same phenomenon reappeared after each coat, but somewhat less. I attributed it to the fact that the bottle of paint was actually almost 25 years old and it had not been sufficiently mixed. In the end I cleaned off the paint and began again, but with the same result. Once more I took the paint off and then sprayed it, but without agitating the bottle, thinking that some of the pigment might have coagulated – same result. Finally, I decided to lightly prime the barrel with zapon-lacquer to isolate the steel. This forms a very thin and virtually invisible layer. This did the trick, but the priming was not done carefully enough and some spots were left bare – with the result that those areas appeared mottled again. I tried dipping, but this leaves a too thick layers in corners etc. Eventually, I managed to obtain a reaonably even layer – one has to work very fast and going over areas already treated is virtually impossible due to the rapid drying. It is also very difficult see, whether one has covered the whole surface. In conclusion, I think the pigment of caput mortuum, which probably is the mineral haematite (Fe3O4) has reacted with the steel (Fe0) leading to the mottled appearance. However, I managed to reproduce the appearance of the barrel of the demonstration model in Copenhagen reasonably well, considering the small scale.

A few of the flimsy and easy to break off details have not yet been installed and some levers to work the mechanisms still have to be fabricated.

The close-up photographs also show a lot of dust and fluff that need to be cleaned and that the paintwork has to be touched up here and there.
The painted and (part) assembled gun

October 2020 - Ammunition and ammunition handling
Thanks to the book published in 1886 by Carl Galster, we are relatively well informed about the ammunition of the German naval artillery of that time. The WESPE-Class was the only class of ships fitted with the Rk 30,5 cm/l22. According to Galster, three types of projectiles were available for these guns in the late 1870s/early 1880s: a) armour-piercing shells, b) shells with a time-fuse, and b) dummy shells for gun-drill.
All shells had two copper guiding rings that would be squeezed into the rifling. One ring sat shortly above the bottom and the second ring where the cylindrical part would transit into the ogival part of the shell.
The armour-piercing shells were cast in a particular way to harden the steel from which they were cast. They were hollow, but with only a relatively small chamber for powder in the rear part. The nose was cast solid. However, at that time functional impact fuses were not yet available, so the shells were filled with a mixture of sand and sawdust to give the approximate weight distribution as a powder charge would give. The threaded hole for the fuse in the bottom was simply plugged. Armour-piercing shells were painted blue.
The ordinary shell had thinner walls and consequently a larger power-charge. The nose was threaded for time-fuses. It is beyond the scope of this building-log to discuss the fuses in detail, it suffices to say that these were made from brass. Shells were painted red and when actually charged with powder marked with a black ring around the nose.
Dummy shells were ‘seconds’ of ordinary shells filled with a sand-sawdust mixture to give the same weight as a real shell. The hole in the nose was closed with a wooden plug. They were painted black all over.
Powder charges were supplied in cylindrical bags. Each bag weighed 46 kg. Up to two bags could be loaded, allowing to adapt the firing range. The bags were stored and handled in cyclindrical boxes lined with zinc sheet or where made from German silver.
A total of five shells were kept ready in the open barbette. I would assume that these would be only the armour-piercing and drill ones, as the fuse of ordinary shells would be rather exposed to the elements. I set out to make six shells in total, three armour-piercing and two drill-shells, that were stored in their respective racks in the barbette. The sixth is an ordinary shell to be placed in the shell-cradle under the crane.
My preferred steel in the workshop are copper-coated welding rods. The copper-coating is very convenient here, as their diameter of 2 mm is exactly the scale diameter over the copper guiding rings. The nose was turned free-hand with my special Lorch, Schmidt & Co. graver holder. The shells are 4.8 mm long. For the live shell, a little brass button was turned and inserted into a pre-drilled hole in the nose.
Shells in handling cradles Powder bag
Free-hand turning of the shell
Gun drill, showing the cradle
The finished ready-shells

It not clear, how the heavy shells (weighing around 330 kg) were handled inside the ship and hoisted to the level of the barbette floor. The crane on the gun-carriage does not actually reach over the access-hatch to the shell-store through which the shells presumably were hoisted. The drawings are not clear on the various hatches in the barbette and over the shell-storage, because of various elments being hidden behind others and therefore not drawn. I will have to live with this ignorance.

On the decks, the shells were wheeled around in trolleys. In the Rigsarkivet in Copenhagen a blue-print (in the true sense of the word) for such a trolley has survived. The trolly forms a cradle that can be hoisted by crane to the breech of the gun. At the rear of the gun two hooks are provided (not realised on the model) into which the cradle hooks. The shell then can be pushed into the gun with a rammer.
The parts for the trolley where laser-cut and assembled using zapon lacquer. Effectively the trolley was built around the shell for rigidity. A hole was drilled into the shell to secure the hoisting ring.
The racks for the ready shells were laminated together from laser-cut pieces and painted white. The retaining bar was made from flattened pieces of 0.3 mm diameter copper wire that was chemically tinned. In theory, each individual shell should have had its own retaining ring (keeping in mind how important it is to restrain these 300 kg beasts in anything but the slightest sea), but after several attempts to put these into place without damaging the paint-work on the shells too much, I gave up. Flattening the wire reminded me of another pending workshop project, namely a micro-rolling mill to produce metal strips of consistent width and thickness from soft wire.

Deck Furniture


May 2007 - The ships was fitted with four pairs of bollards of square cross section; two at the rear and two on the raised quarterdeck. Luckily a good rather close-up photograph of the real specimen is available (see main page). The bollards are milled from round brass stock. Round stock was chosen as a starting point rather than e.g. flat stock, because it can be held easily in the lathe for turning a spigot on which, by which the part can be held for further machining. Otherwise it would be difficult to mount such small a part on the miller for machining five sides. The spigot is also a convenient reference for machining and for fastening on the model eventually. From the lathe the raw part is transferred to the dividing head mounted on the milling machine. After each pass with the tool, the part is turned by 90º or 180º depending on requirements. Thus a square and symmetric part is produced. For a final machining step the part is transferred back to the lathe and the dome shaped head formed using a very fine file on a roller-filing rest. The job is completed by rounding off the corners using a not-too-hard rubber-bonded abrasive wheel (CRATEX) in the mini-drill. Remaining machining burrs are removed by offering the part to wire brush wheel.

Turning the raw bollard

Mounting the raw bollard in the dividing head on the milling machine
Milling operations: first squaring, then producing the waist
Rounding off the cap
The roller filing rest
Finished bollards and part of working drawing
Drilling the holes for the bases

September 2008 - The base for the double bollards were intended to be a surface etched parts, but I was not happy with the results. So I decided to make them from solid brass. Solid brass was easier to handle for machining than brass sheet. Nevertheless the envisaged machining operations prompted me to make a couple of gadgets, fixtures, for the mill and the lathe.
Milling around the edges or on top of flat material always presents work-holding problems. Worse, if several identical parts have to be produced. Hence I divined a work-holding block with several clamps and stops running in a T-slot. Similarly holding small parts for cutting off on the circular saw is tricky and best done on the lathe with a special saw table clamped to the top-slide. This saw table allows parts to be safely clamped down for cutting.
The three parts of each bollards were soft-soldered together.

Drilling set-up
Milling the beading
Sawing off surplus material
Parting off the individual bases
Milling a bevel
Parts of double bollards
Work holding for soldering
Bollards, chain stoppers and spill


May 2007 - One pair of chain stoppers is located immediately behind the hawse pipes as usual. A second pair is placed above the chain locker, which is located immediately in from of the armoured barbette. The bodies of the stoppers are rather complex castings, calling for some complex machining operations in model reproduction. The same basic technique as for the bollards was used. Given the complex shape, however, machining is not possible in one set-up. for certain operations the axis of the spigot has to be perpendicular to the milling machine, while for others, such as drilling it has to be parallel. For the latter and for milling the various slots, I choose to transfer the dividing head to the lathe. This has the advantage that its centre line is at the centre of the lathe spindle.
The slots were milled using a micro-tool made from a broken carbide drill, the end of which was ground flat. This results in a non-ideal clearance of 0º, while the cutting angle and side rake are that of the original drill bit. However, not much metal is removed so that this doesn't really matter here.

Milling the profile of the chain stopper

Milling the slots on the lathe
Milling bits and product
Squaring the part on the upright collet holder
Round-milling on the rotary table

One set of stoppers was milled from brass, while for the other one I used PMMA (PLEXIGLAS®, PERSPEX), the main reason being that I ran out of brass stock. However, genuine PLEXIGLAS®, is pleasant material to machine and easy on the tools. It holds sharp edges and it easier to see what you are doing than on the shiny brass. Acrylic paints seem to key-in well - basically its the same molecule, of course. On the downside one may note that small and thin parts are rather brittle. Using diamond-cut carbide tools gives a nice smooth finish, but normal CV- or HSS-tools can also be used, of course.
While for the bollards and the front pair of stoppers the spigot could be on the geometric centre of the part, making it easy to measure while machining, for the after stoppers I had to place the spigot to the centre of the pipe down to the locker, so that the concentric rounded edges could be milled. The pictures show this operation.

October 2008 - The stoppers have now completed with etched brass releasing levers, etc. The fore stoppers were also soldered to surface etched base plates.

Undercutting using a micro saw bit Stoppers compared against a 5 Euro-Cent coin Drilling the hole for the release lever
Finished after stopper
Etched fret with stopper base plates (bottom left) and levers (bottom right)
Finished fore and after stoppers (right column)

Anchor capstan

August 2007 - One component that always has puzzled me somewhat as to their manufacture in a model has been the sprocket on capstans. While the geometry on horizontal windlasses is quite simple, with suitable depressions for the chain links around the circumference, the sprocket on a capstan is a complex affair. In any case the capstan head cannot be manufactured in one piece. So I broke it down into three pieces: the spill head, the sprocket and the base drum with the pawls. The whole capstan has more pieces including four guiding rollers and a finger to pull the chain off the sprocket. The cast base on the prototype will be reproduced as a surface-etched part.
The sprocket started out as a 2.5 mm brass rod taken into the dividing and into five notches were milled to produce something like a five-pointed star (these sprockets typically have five or six arms). The notches for the horizontal links were cut on the lathe with a forming tool. The sprocket then was faced and drilled to fit onto the capstan stem. The next step is cutting it off. This produces some burrs that need to be taken off. Luckily I have collected over the years almost every type of work-holding device that was ever made for the watchmakers lathe. Here the insert jewel chucks came handy to hold the 2.2 mm by 0.6 mm sprocket for facing-off.

Milling the sprocket, 1st step

Milling the sprocket, 2nd step Cutting with a forming tool
Drilling the sprocket
Facing-off the sprocket in a jewel chuck
Capstan head ready for cutting off

The capstan head is a simple turning job. The curved surfaces are pre-cut with appropriate lathe tools and then finished with very fine files. Incidentally, the implement shown on the appropriate picture is a rare miniature micrometer, also coming from the watchmakers toolbox and very handy for measuring narrow recesses and the likes. They came in sets of three, the other two are a depth-micrometer and one for measuring the width of notches respectively.
Finally, the three parts are soft-soldered together.

September 2008 - Again the guiding rollers are a simple turning job. The shapes were produced with a free-turning graver and by rotary milling in the dividing head. In the meantime various etched parts had been produced, including the base plate made up of two different superimposed parts and minuscule pawls. Also a chain separator from 0.3 mm copper wire rolled flat was produced. The various parts were soldered together.

Assembled capstan head
Shaping the head of the rollers by rotary milling
Set-up for shaping the rollers using the geared dividing head
Etched fret with capstan base plate (top left) and pawl (bottom centre) Finished Capstan (bottom left)
Engine-room telegraph drawings, original in the Norsk Maritimt Museum, Oslo, and the two telegraphs on the model

October 2019 - Engine-room telegraphs
On the ‘official’ lithograph of SMS WESPE from the early 1880s an unsual form of engine-room telegraph was drawn. It has a horizontal dial. In the earliest known photography of the ship during fitting-out, the telegraphs had not yet been installed.

A short while ago I discovered during a visit to Oslo in the Norsk Maritimt Museum a very similar telegraph on display. Unfortunately, the legend is not readable on my image. I seem to remember that the inventor or patentee was named. A search on the Internet and in my library did not produce anything, so I would be grateful, if anyone has an idea, who the inventor or patentee might have been.

The telegraph was redrawn from the lithography in order to serve as a working drawing with measures to guide the lathe operation.

The whole telegraph seems to have been made from brass and accordingly the model was turned from brass. The indicator arm and follower were made from flattened brass wire and the ‘wooden’ handle built up from PVA glue.

SMS WESPE had two telegraphs, one for the starbord and port engine each, of this early twin-screw naval vessel.

Binnacles from the 1880s lithograph
Working drawings for the binnacles
Milling the octogonal columns
Milling the glass hood in the shape of an octogonal pyramid
Cleaning up after painting
The parts of the binnacles
Binnacles temporarily assembled

November 2019 - Binnacles. SMS WESPE was originally equipped with three binnacles, one on the bridge, the mother-compass on a sort of pole in front of the engine-room skylight, and the third one in front of the emergency steering-wheel at the stern. In the 1890s a fourth binnacle was installed on a platform atop the engine-room skylight, but is left off here. As SMS WESPE was built in 1876 the original binnacles lack the conspicuous compensation spheres, that were only invented in the 1880s by Lord Kelvin. Also other type of compensation gear is not visible on the lithographs and the earliest photograph. A photography of the early 1890s shows a much more substantial binnacle in front of the emergency steering-wheel, which preumably now houses the compensation gear and also sports the compensation spheres. Originally, the compasses must have been illumanted by petroleum lamps, but from the lithographs it is not clear, where these lamps would have been attached. At least there are exhaust funnels on top of the binnacles, which have disappeared in later photographs. This seems to indicated that electrical illumination might have been introduced, when a dynamo was installed on board in the early 1890s for a search-light.

For the model the individual binnacles were redrawn from the lithograph in order to serve as a basis for working sketch to guide the lathe- and mill-work. One needs to keep in mind that the total height is somewhere between 10 and 15 mm.
The columns presumably were made from mahagony and were turned from brass rod before being transferred to dividing head on mill to cut the octogonal shape.
The actual compass was made, as usual, from brass and so on the model. Body and funnel did not provide a particular challenge, not considering the small size. To the contrary, the glass hood with its narrow frames of perhaps 15 mm width on the original. The body was roughly turned from Plexiglas and then transferred to the mill. Here the octogonal pyramid was milled. Using a 0.3 mm ball-head burr narrow grooves were cut into the edges and these grooves filled in with brass paint.
Once the paint had thoroughly dried, the faces were very lightly milled over, which resulted in sharp narrow brass strips at the edges. This is a technique that I copied from making engraved scales.
Originally I had the crazy idea of placing a miniature compass-card underneath the Plexiglas hoods, but even without it, assembling the binnacles was fiddly enough.

January 2020 - Steering-wheels. All the boats had two sets of steering wheels, one on the bridge and the emergency steering-wheels at the stern. Both stands had double wheels that worked in the traditional way on drums and ropes. There is a rather good photograph of the emergency steering position, which allows to deduct the details of the wheels. On the model these wheels are rather delicate affairs of only just under 10 mm diameter overall. I had been considering many different ideas for different kind of materials for fabricating them. Machining the slender spokes seemed a daunting task. Photo-etching and assembling them from different layers seemed a more realistic proposition. It then appeared to me that laser-cutting might be also an option, as I had recently acquired a cheap, small machine.

After some tests with the laser-cutter, I finally chose 120 g/m2 Canson-paper, which is 0.15 mm thick and has a smooth surface. It cuts well with the laser-cutter, as it is not ballasted with inorganic material, such as barytes. Some trials were needed to determine the right cutting parameter combination of contrast, laser-power and cutting depth. One should assume that for a simple B/W-picture the contrast should be 100%, but somehow changing the contrast setting changes the width of the cuts. For this reason the final dimensions of the parts depend on the contrast setting. Laser-cutting is contactless and the cut-out parts are not moved during the cutting process. Therefore, it is possible to cut them out completely and in contrast to the photoetch-process they do not need to be attached to some frame. When designing the image with which the laser-cutter works, one needs to consider all these factors that sometimes can only be determined by trial and error.

The wheels are built up from five layers in order to simulate the joinery work and to arrive at the necessary 3D-rendering. The core part was thickened by two more layers, the outline of which was drawn a bit smaller to simulate the profiling of wheels and handles. A further layer on each side simulate the rim and hub. The individual layers were glued together with zapon-lacquer, which impregnates and stiffens the paper. Unlike many other glues, this lacquer only forms a very thin layer, not adding to the thickness of the wheel, and the parts can be adjusted, as long as the lacquer has not dried. 

The prototype steering-wheels were re-enforced by brass-rings screwed onto each face. My intention was to make these rings from real brass shim (remember: only real metal looks like real metal ...). However, I did not manage to cut so narrow rings from 0.05 mm brass-shim. In the end, I bored out a piece of round brass stock to 6.8 mm and turned down the outside to 7.2 mm. From this tube with 0.3 mm wall thickness, slices of 0.1 mm thickness were parted off. After a few trials to get the settings right, this worked fast and repetable. The rings were deburred on 600 grit wet-and-dry paper, ground finely on an Arkansas-stone and polished on a piece of paper with some polishing compound. The brass rings were glued on with lacquer.

The axle including drum for the steering rope were turned from brass. The wheels will be spray-painted painted all over and then the paint rubbed off from the brass rings. This will nicely simulate the rings let into the wood as per prototype.

Laser-cutting machine
Laser-cut steering wheels Components of wheels
Steering-wheels and brass reenforcement rings
Assembled wheels and components
Gratings: JPG-mage as input for the laser-cutter Steering-wheel pillars: JPG-mage as input for the laser-cutter Machining the bearing caps in a ‘jewelling’ collet Shaping the covering cap of the wheel-axle using a cup burr