Recycling at Longmorn
Until about 1949, Ruston 48DS shunters had no doors on the side of the cab. So driving Queen Anne in winter with a cold east wind blowing across the fields of Morayshire must have been a bit chilly.
When dismantling the cab, we noticed that a curtain rod with curtain rings had been fitted to each side of the cab. There was however no trace of the curtain. It looked as though this might have been a modification by the staff at Longmorn. When the curtain rings were cleaned up on the grit blaster, they were found to be made of copper.
Recently Jim, who works for Forsyth’s (they make most of the copper stills and fittings for Scottish distilleries) caught sight of the rings and recognised what they originally were. Apparently they are made by winding copper wire onto a former and then slitting lengthwise to produce the rings. The rings are then linked together to form a piece of “chain mail” about 1 metre square. This is then fitted inside the copper still and rotated to knock back any foam generated by the boiling wort.
Some of the foam-busting curtain rings |
Because the rings form a link with the locomotive’s whisky heritage, the curtains will be reinstated when the cab is rebuilt.
Recycling at Aviemore
A lot of riveting is currently taking place on Aviemore’s Black 5, 5025. The running boards are mainly held on by 5/8 inch rivets. These rivets as supplied are about 4 inches long, and have to be shortened to the thickness of the plates plus an allowance for forming the rivet head. As a result, there are a lot of 2 inch lengths of 5/8 inch diameter rivet-grade steel left over.
These leftovers have been machined into 3/8 inch dome head rivets suitable for Queen Anne’s cab. Tests have been carried out, and they can satisfactorily be cold riveted. The original rivets were simply flattened to a “pancake” shaped head on the reverse side, and suitable rivet dollies to do this have now been machined. Riveting will be done using the hydraulic riveter, which is basically a big G-Clamp with a power-operated hydraulic ram.
Bolts on the left, rivets on the right |
The other cab fasteners were 3/8 inch Whitworth bolts with a domed and slotted head. Whitworth bolts are now very expensive, and almost impossible to get with this type of head. So new bolts have been machined out of EN316 stainless steel for this purpose.
Other News
Cab Steel
The steel to join together the water-jet-cut steel cab sheets has now arrived. This consists of angle, flat bar and “half-round feather edging”. As it will be a while before we have time to start rebuilding the cab, to prevent the steel from corroding, the angles and flat bars have been cut to length, rubbed down and painted with red oxide primer. The feather edging has been rubbed down and treated with an oil-based preservative.
Cab strip and angle sections painted |
Feather-edge strips with protective oil coating |
Battery Box
Here are some pictures of the battery box, which was restored some time ago. It is currently used to store nuts bolts and other small parts for the locomotive.
Battery Box after cleaning up a bit |
The toolbox had been repainted with beige paint, which was removed using a hot air gun, revealing the original “Ruston Green” colour. The toolbox lid was warped, and was flattened using clamps and a hot air gun. The rotten lower halves of each side were cut off and new sections attached using a half-lap joint and Gorilla Glue. A new bottom board was made and reinforced with steel angle brackets so that the box could be fixed slightly above the floor. All original screws and nails were removed and replaced by brass screws. A new battery compartment door was made from wood reclaimed from an old T&G door and fitted with new hinges and a catch. The many holes and imperfections were filled with two-part wood filler and the whole box painted in grey undercoat.
New bottom sections and battery door fitted |
The restored Battery Box |
Another view of the box |
Engine News
Although the engine runs, it produces a lot of white smoke (unburnt fuel) and this has been traced to cylinder no. 1, which is suffering from low compression. More details will be given in a future blog.