The replica bells and cymbals – an update!

We have an update on the status of our metal instrument replicas! To recap, local Canterbury jeweller Justin Richardson is making replicas of a number of the metal  Roman bells and cymbals in the Petrie Museum, using alloys and manufacturing methods as close to the originals as possible. The aim is to play these replica instruments to reveal the sounds of Roman Egypt, and to see how different production methods and materials might have affected their sound. We recently met up with Justin to get a progress update.
Justin Richardson with some of the bell and cymbal replicas. Photo: Ellen Swift.
The process began with the 3D laser scanning of our selected objects – this data was used to make wax models which were then used to create casts in metal. This, along with sand-casting (discussed in our other blog post here), would have been the most common methods of casting metal in Roman Egypt.
There were however a few problems encountered along the way. Some of the objects we scanned created models where the walls of the bells and cymbals were very thin. This is because during their lives the surfaces of the instruments had worn down through use. However, this thinness resulted in castings that had large holes and with sections missing. We therefore had to edit our scanned models to increase the thickness of the instrument walls in places, thus ensuring a successful cast during the next attempt. This does however mean that the finished replicas will represent the instruments as they had been when new, rather than the older worn versions we now have in the museum collection.
The replica bells, cast using moulds made from the 3D models of the laser scanned originals. Photo: Ellen Swift.
Another challenge was presented by the clappers inside the bells – these are the articulated pieces of metal that cause the bell to ring when they touch the walls. The clappers could not be laser scanned as they moved around, and in a couple of examples were no longer extant at all. Instead we had to rely on what we could see by eye both in terms of style of clapper and material. It appears that both bronze and iron were used for clappers, with several different styles in use – therefore we have tried out a range of examples in the replica bells to reflect this. The cymbals also presented us with a range of different production techniques and materials. Brass, bronze, and copper are all represented, as are the techniques of both casting and hammering. For our replicas, there will be a pair made in spun brass, and a pair cast in bronze – this will allow us to see what kind of difference there is in sound quality.
One of the original Roman cymbals with a thinned wall (UC33268B). Photo: Ellen Swift.
 Justin and his team are now adding the final touches to the replicas. One of the bells had a ring that was originally gilded (as identified by the XRF analysis) so that element will also be replicated. The cymbals also need metal attachments added in the form of U-shaped metal handles. Finally, several of the bells were attached to metal bracelets, which are also being made. These bracelets are made of iron, but we cannot source iron of suitable dimensions to use for this. Instead Justin will be using steel, which is more difficult to work with than the original iron.
The bell that originally featured gilding – this decoration will also be included on the replica. Photo: Ellen Swift.

Once the replicas are completed, our next step is to create a recording off them being played. We will be using software that can mimic the acoustic qualities of any space, so we hope to replicate the sound of music as played in typical houses known archaeologically from Roman Egypt!

An Expert Visit: identifying production techniques

Our project relies upon collaboration with a number of different professionals whose expertise we are lucky to benefit from. Canterbury jeweller Justin Richardson recently joined us up at the Petrie to view the instruments first-hand; Justin will be hand-making replicas of some of our selected metal bells and cymbals using authentic materials and techniques, so he needed to identify the methods of manufacture used to create the originals. Many of the processes used in Roman Egypt are still employed by metalworkers today, and looking closely at the artefacts can reveal tell-tale marks on the surface of the metal left from specific manufacturing techniques.

Cymbal with pitted surface. Photo: Ellen Swift.

The first thing Justin noticed was that some of the cymbals have a distinctive pitted surface in places. This texture was created by the ‘sand casting’ method of manufacture. This process involves creating a mould by pressing each side of the object to be copied into damp sand, thereby leaving a perfect imprint. Molten metal is then poured into the cavity, leaving a distinctive mottled surface in the process. Sand casting requires an original version from which the mould must be made; this was likely forged by hand, with any hammer marks on the original obliterated by the sand casting process rather than being transferred onto the copies. Sand casting is a cheap and easy method of production, and produces a lower quality object than ‘lost-wax casting’ – making it ideal for an object as simple as a small metal cymbal.

Hammered and lathe-turned cymbal, with caulked edge. Photo: Petrie Museum.

In contrast, the surface of another one of the cymbals revealed hammer marks. It was also significantly lighter than the others, leading Justin to conclude the object had been made by hammering a flat disc of metal over a hemispherical mould to create a dome shape. He also identified a caulked edge – caulking is a metal working technique where the edge of a piece of metal is hammered, end-on, to thicken it up and make it appear more robust than the item actually is. Further identifiable traces included concentric circles on the surface along with a perfectly centred hole, revealing further work on a lathe. Justin’s technical knowledge also revealed some unexpected surprises. The two main copper alloys used for the instruments are brass and bronze; our modern understanding of these materials and the kinds of objects they make resulted in us expecting any brass objects to be handmade,  and bronze objects to have been cast – however the exact opposite of this appears to be the case! The manufacture of a bronze bracelet with a bell attached also caught us out.

Cast bronze bracelet with dot decoration (UC58537). Photo: Ellen Swift.

It would seem logical to make such a flat bracelet by cutting out a strip of metal from a sheet, and then bending it to shape, however closer inspection revealed the object to be cast; the decorative motifs appear soft edged rather than cut into the surface of the metal, and the indented dots revealed no ‘ghosts’ on the reverse. Ghosts are marks left on the reverse of a metal surface when tools are hit against it to produce decorative marks. However, as no such ghosts exist on the bracelet, it again supports the case that it must have been cast. Whilst unexpected, casting such bracelets makes sense as it allows mass production of identical objects with little effort or additional expense, additionally suggesting this bracelet was not a one-off, but in fact one of a batch produced.

We learned a lot from Justin’s visit that we can use when looking at other metal objects in the Petrie collection. We look forward to seeing the replicas and will post an update on them soon. If you’d like to know more about Justin’s work, you can find his website here.