ArticlePDF Available

Abstract

Bronze Age swords with a metal hilt can be considered the peak of Bronze Age casting technologies. To reconstruct the casting techniques used more than 3000 years ago, a metal hilted sword of the Schalenknauf type from Lower Austria was studied with the aid of macroscopic analyses and simulation of mold filling and casting solidification. A three-dimensional model of the hilt was created based on optical scanner measurements performed on a hilt recently discovered during archaeological excavations. Three different configurations of the gating system were considered, two on the pommel disk and one on the knob, and the effect of its location on the formation of casting defects was investigated. Three-dimensional computed tomography was used to detect internal defects, such as gas and shrinkage porosity, which were then compared with those calculated by simulation. The best match between actual and predicted hilt quality demonstrated the location of the gating system, which turned out to be on the pommel disk.
1 23
JOM
The Journal of The Minerals, Metals &
Materials Society (TMS)
ISSN 1047-4838
JOM
DOI 10.1007/s11837-015-1464-y
Casting Simulation of an Austrian Bronze
Age Sword Hilt
Annalisa Pola, Marianne Mödlinger,
Paolo Piccardo & Lorenzo Montesano
1 23
Your article is protected by copyright and all
rights are held exclusively by The Minerals,
Metals & Materials Society. This e-offprint is
for personal use only and shall not be self-
archived in electronic repositories. If you wish
to self-archive your article, please use the
accepted manuscript version for posting on
your own website. You may further deposit
the accepted manuscript version in any
repository, provided it is only made publicly
available 12 months after official publication
or later and provided acknowledgement is
given to the original source of publication
and a link is inserted to the published article
on Springer's website. The link must be
accompanied by the following text: "The final
publication is available at link.springer.com”.
Casting Simulation of an Austrian Bronze Age Sword Hilt
ANNALISA POLA,
1
MARIANNE MO
¨DLINGER,
2,3
PAOLO PICCARDO,
2
and LORENZO MONTESANO
1
1.—Dipartimento di Ingegneria Meccanica e Industriale, Universita
`degli Studi di Brescia, Via
Branze 38, 25123 Brescia, Italy. 2.—Laboratorio di Metallurgia e Materiali, Dipartimento di
Chimica e Chimica Industriale, Universita
`degli Studi di Genova, Via Dodecaneso 31, 16146
Genoa, Italy. 3.—e-mail: marianne.moedlinger@univie.ac.at
Bronze Age swords with a metal hilt can be considered the peak of Bronze Age
casting technologies. To reconstruct the casting techniques used more than
3000 years ago, a metal hilted sword of the Schalenknauf type from Lower
Austria was studied with the aid of macroscopic analyses and simulation of
mold filling and casting solidification. A three-dimensional model of the hilt
was created based on optical scanner measurements performed on a hilt
recently discovered during archaeological excavations. Three different con-
figurations of the gating system were considered, two on the pommel disk and
one on the knob, and the effect of its location on the formation of casting
defects was investigated. Three-dimensional computed tomography was used
to detect internal defects, such as gas and shrinkage porosity, which were then
compared with those calculated by simulation. The best match between actual
and predicted hilt quality demonstrated the location of the gating system,
which turned out to be on the pommel disk.
INTRODUCTION
Swords are one of the most precious metal objects
of the Bronze Age. Their importance derives from
both their material value and their function as a
weapon. Swords, in fact, were the first weapons
made to kill humans specifically. Furthermore,
carrying a sword indicated social rank and con-
tributed significantly to the importance of the
weapon. Swords were distributed throughout the
entire European Bronze Age world. We can distin-
guish two main types: swords with organic hilts and
swords with metal hilts. The latter are rarer and
their richly decorated appearance indicates a higher
value. Metal-hilted swords were usually made of a
separate cast blade, which was joined to a hollow
hilt by means of two or more rivets.
1
The cast hilts were characterized by a complex
geometry that indicates the high technological level
reached by Bronze Age craftsmen and have not yet
been reproduced with modern casting technology.
In the Bronze Age, several casting techniques and
mold materials were known. According to the com-
plexity of the object and the time period in question
(the European Bronze Age), these could be open
cast, casting in multipart molds, and lost-wax
casting. Sand casting cannot be excluded, although
it cannot be verified in the archaeological record.
The molds for casting blades and hilts were gen-
erally made of stone (i.e., sand stone and soap
stone), cooked clay, or bronze, as demonstrated by
archaeological evidence. In general, clay molds are
badly preserved; only a few mold fragments for
casting sword blades are known from Northern
Europe, whereas stone molds for blades have also
been found from the rest of Europe.
1
In total, how-
ever, not more than 20 stone molds for casting
sword blades are known today. The only mold
known for casting metal hilts is the bronze mold
from Erlingshofen, Germany, of which the top part
was never found. It was used for producing hilts or
wax models of sword hilts of the Mo
¨ringen type
(c. ninth to eighth centuries BC) (Fig. 1).
2
Never-
theless, this mold dates from approximately
200 years later than the sword discussed next.
Because of the lack of further finds of sword hilt
molds, it is generally assumed
1
that cooked clay
molds were used for the production of these cast
parts. Unfortunately, these molds do not generally
leave many archaeological traces. Another argu-
ment supporting the theory of cooked clay molds is
the complexity of the hilt shape and the highly
JOM
DOI: 10.1007/s11837-015-1464-y
Ó2015 The Minerals, Metals & Materials Society
Author's personal copy
individual nature of every sword hilt, which is not
the case when metallic molds are used. This of
course might not exclude a clay mold for the con-
struction of similar wax models, which were then
used for making the final mold for several metal
castings.
Another significant argument for the usage of clay
molds for most of the Bronze Age metal-hilted
swords is the casting geometry. The hilt presents a
hollowed shape that normally narrows at the base.
To obtain this kind of undercut by foundry tech-
niques, cores are needed. Because of the shape of
the hilt cavity, metallic cores cannot be removed
once the poured alloy solidifies. Wood represents an
alternative core material. A wooden core gets par-
tially burnt during casting and might be completely
burned up during subsequent annealing. A cooked
clay core, meanwhile, can easily be broken into
pieces and, after watering, it can easily be removed
from the hilt, guaranteeing the proper final shape of
the casting.
As far as casting conditions are concerned, several
factors indicate that the inlet was placed on the top
of the hilt: (I) the hollow form of the hilts, (II) the
need to fix the clay core adequately to the base of the
hilt, (III) the only preserved casting mold for hilts,
and (IV) the study of radiographies of Austrian
Bronze Age sword hilts, which show the concentra-
tion of casting defects.
1
The gating system on the top of the hilt could
have been located either on the knob/pommel or on
the pommel disk. Macroscopic studies suggest the
location on the pommel disk because one of its sides
is always slightly thicker than the other; the pres-
ence of a circular, more porous zone on the surface
of pommel disks on some swords further supports
this assumption.
1
Additionally, as noted under point
(III), the bronze casting mold from Erlingshofen
shows the gating on the pommel disk and not on the
center of the disk.
To study the casting technique of Bronze Age
sword hilts and to better demonstrate the above
mentioned hypothesis about the gating system
design, a sword from Unterradlberg/St. Po
¨lten,
Lower Austria, was studied (Fig. 2). In particular,
the simulation technique was applied to investigate
the casting conditions used to produce the hilt. With
this aim, various gating systems and process
parameters were combined and analyzed. The
comparison between simulation results and x-ray
analyses enabled us to figure out the method
applied and to demonstrate the theoretical
assumptions.
EXPERIMENTAL PROCEDURE
The hilt chosen for the current study was recently
discovered during archaeological excavations in a
previously robbed cremation grave in Unterradl,
Austria
3
(Fig. 2). It is part of a sword of the Scha-
lenknauf type, which is associated with the Late
Bronze Age, and it dates from 1050 BC to 950 BC. It
Fig. 1. Late Bronze Age four-part bronze casting mold for sword hilts from Erlingshofen, Germany (after Ref. 2). The top part of the mold was
never found. The mold is the only one preserved for producing Bronze Age sword hilts: (a) model of the sword to be cast, (b) side part of the
casting mold (left: outside, right: inside), (c) core, (d) second side part of the casting mold (left: outside, right: inside), (e) view from the top, and (f)
bottom view.
Pola, Mo
¨dlinger, Piccardo, and Montesano
Author's personal copy
was found with the blade broken in the haft area.
This fracture was most likely the result of low
casting quality.
4
We were only allowed to sample the blade of the
sword in question. Its chemical composition was
analyzed by means of an electron-probe micro-
analyzer (EPMA) equipped (Cameca SX 100;
Cameca SAS, Gennevilliers Cedex, France) with
four wavelength-dispersive spectrometers and an
energy-dispersive x-ray spectroscopy system for
high-quality mineral and alloy quantitative chemi-
cal analyses (EMPA). It revealed a binary alloy of
copper with 10.4 wt.% Sn and traces of iron and
sulfur.
4
The alloy composition of the sword hilt,
however, remains uncertain because sampling was
not permitted. In addition, the hilt is completely
corroded, so noninvasive surface measurements
would not be precise because influenced by the
presence of corrosion layers that cannot be removed.
However, CuSn alloys were regularly used for the
production of Bronze Age swords, as proven for
instance by Ref. 5. Therefore, the alloy of the hilt can
reasonably be assumed to be the same as the sword.
To investigate the casting quality, a dual-source
computed tomography system RayScan 250E was
used (RayScan Technologies GmbH,). In particular,
the hilt was analyzed making both frontal and axial
scans.
SIMULATION MODEL
A three-dimensional (3D) model of the hilt was
created starting from the measurements carried out
by using a 3D optical scanner (RayScan 250E—
Wa
¨lischmiller).
4
As shown in Fig. 3, the original hilt
is not complete: A part of the pommel disk is missing
and most of the surface is degraded by corrosion.
Thus, to obtain the full geometry of the hilt, the
detected shape was subsequently refined by using
SolidWorks 2013 (Dassault Syste
`mes, Waltham,
MA) and Rhinoceros 3D 2012 (Robert McNeel &
Associates, Seattle, WA). Its total height is 11.9 cm.
With regard to the internal cavity, the shape and
size were derived by x-ray measurements and
transferred to the computer-aided design (CAD)
software. The subsequent geometry was used for the
simulation, with the aim of clarifying the casting
conditions.
Simulation was performed by using the commer-
cial software ProCAST 2014 (VisualCast 9.0.2; ESI
Group, Paris, France) to predict mold filling and
casting solidification by solving numerically the
Navier–Stokes and Fourier equations.
A finite-element mesh of the hilt was created,
consisting of 204,705 tetrahedral elements. At least
three elements were defined at the thinnest sections
to ensure the accuracy of the fluid flow simulation.
Fig. 2. Late Bronze Age sword (a), hilt (b), hilt description (c), broken area (d), and detail of the broken surface (e). The sword was found broken
in its owner’s grave in Unterradl, Austria, and it dates from 1050 BC to 950 BC. The hilt of this sword was used for the casting simulations
discussed.
Casting Simulation of an Austrian Bronze Age Sword Hilt
Author's personal copy
As shown in Fig. 3, the gating system was designed
on the upper side, i.e., the side of the pommel and
pommel disk. Since three different sprues and run-
ning systems were considered (named case A, case
B, and case C; see Fig. 4), the number of total ele-
ments was slightly different for the three con-
figurations. In general, all meshes used in the three
configurations consisted of approximately 730,000
elements.
In case A, the pouring system is placed on the side
of the pommel disk. In case B, it is placed on the
pommel disc but rotated by 90°compared with case
A (i.e., right above the haft area). In case C, it is
placed directly on the knob. An internal clay core
was designed, based on the x-ray measurements
(85,376 tetrahedral elements). The clay mold was
created as a 10-mm-thick shell around the hilt, to
take into account the heat exchange with the alloy.
For the calculation, a CuSn10 alloy was used, in
agreement with the above-mentioned ranges of the
alloying elements. The temperature-dependent
properties of the bronze (conductivity, density,
specific heat or enthalpy, and viscosity) were cal-
culated by means of Computherm Database (Pan
Iron 5.0; Computherm LLC, Madison, WI) available
in ProCAST, as a function of the chemical compo-
sition and using the ‘‘Back Diffusion model.’’ Minor
elements were neglected in defining the properties
of the alloy.
The initial temperature of the bronze was fixed at
approximately 70°C above the liquid temperature
(equal to 998°C). Mold and core thermodynamic
properties are available in the software database.
The effect of the initial temperature of mold and
core on the casting quality was also investigated. To
this aim, three different initial conditions were
considered at 20°C, 500°C, and 800°C. A heat-
transfer coefficient of 500 W/m
2
K was imposed
between mold and bronze as well as between bronze
and core.
With regard to the boundary conditions, a heat-
exchange coefficient of 10 W/m
2
Kat20°C was
imposed on the external wall of the shell in contact
with air. Two different flow rates were also
evaluated: a constant one, equal to 0.08 kg/s, and a
time-dependent flow rate, ranging from 0.04 kg/s to
0.08 kg/s, which can fill the cavity in about 5 s.
RESULTS AND DISCUSSION
X-ray computed tomography images taken
frontally at different depths show three different
porosity defects (Fig. 5). In particular, a central
porosity can be observed in Fig. 5a, approximately
48 mm from the base of the pommel disk. This de-
fect seems to be the result of air entrapment during
filling. This is confirmed by the x-ray axial images
(Fig. 6), which show a rounded and an almost
regularly shaped hole, which is typical of air inclu-
sion. Moreover, as it starts from the internal surface
of the casting (in contact with the core), it cannot be
related to shrinkage phenomena.
Furthermore, the other two large porosities at
about 28 mm and 15 mm from the base seem to be
related to the metal flow (i.e., air entrapment), as
they are connected to the internal free surface and
the walls of the cavities seem almost regular (Fig. 6).
Some additional small cavities can be detected along
the thickness of the hilt, all with rounded shape; no
shrinkage porosities, characterized by irregular
contour, can be seen in the different slices.
These tomographies of the hilt can be compared
with the results of the simulation of mold filling and
metal solidification, which were performed in order
to understand the applied casting conditions.
In Figs. 7and 8, two subsequent steps of the
simulation are shown for the three inlet configura-
tions. The filling phase is reported in terms of the
Voids parameter; this allows us to see the parts of
the cavity that are still empty.
In all the reported simulation results, an x-ray
visualization mode is used for the mold to get a
better study of metal flow and solidification. Fur-
thermore, our results refer to the case of initial mold
temperature equal to 20°C. In the images shown,
this temperature has been chosen as representative
of all our investigations, as the filling was not sub-
stantially affected by the initial mold temperature.
At the beginning, the metal fills the cavity, due to
gravity, more or less in the same way, indepen-
dently of the position of the inlet. Only case A
reveals the formation of a small hole at less than
56 mm from the base of the pommel disk, quite close
to the defect shown in Fig. 5a (see the arrow). No
similar problems can be seen in the other two cases.
This agreement between simulation A and the
Fig. 3. Picture of the sword hilt and its 3D model: (a) original sword
hilt and (b) CAD model.
Pola, Mo
¨dlinger, Piccardo, and Montesano
Author's personal copy
experimental investigation means we can make a
first assumption about the position of the gating
system on the side of the pommel disk.
Further confirmation of this hypothesis comes
from the comparison between x-ray analyses and
simulations close to the end of the filling phase. As
shown in Fig. 8, in fact, once again case A induces
the presence of a small void (see the arrow in Fig. 8)
in the same position as revealed by the x-ray
tomography. On the contrary, in cases B and C no
pores are formed at about 15 mm from the base of
the pommel disk.
Fig. 4. Configurations of pouring systems: cases A, B, and C, and the core.
Fig. 5. X-ray frontal tomography images at different depths.
Casting Simulation of an Austrian Bronze Age Sword Hilt
Author's personal copy
The porosity at 28 mm from the base is most
likely a combination of shrinkage and gas entrap-
ment. In fact, as shown in Fig. 9, in all three cases
the middle-upper part is characterized by longer
solidification time, where shrinkage can easily
occur.
To better evaluate the location of shrinkage cav-
ities, a cut-off method was used to visualize the so-
lidification defect (Fig. 10). With the cut-off view,
also called x-ray view, the model is made partially
transparent in order to see features that are inside
the model itself, such as shrinkage porosity, and
Fig. 6. X-ray axial tomography images at different depths.
Fig. 7. Simulated mold filling in the three configurations, voids visualization mode.
Fig. 8. Simulated mold filling in the three configurations, voids visualization mode.
Pola, Mo
¨dlinger, Piccardo, and Montesano
Author's personal copy
that are within, above, or below defined value
limits.
As we can see, shrinkage seems to cause a small
cavity to form close to the position revealed by the x-
ray tomography (Figs. 5and 6, central image), but
only in cases A and B, and not when pouring the
metal from the center of the pommel (case C). This
additional result adds further support to the
assumption that the hilt was cast from the pommel
disk.
It should be noticed that the porosity at
approximately 28 mm from the base of the pommel
disk is placed on the internal surface of the hilt, as
shown by the axial tomography. This presents an
almost regular shape and is surrounded by small
bubbles. This combination of conditions lends sup-
port to the idea that this cavity is related to the gas
released by the core during metal filling and
solidification, as it can happen whether the core is
made in wood or in clay that is not baked long en-
ough. It is known that chemically bound water in
clay starts to be released at approximately
450–500°C, and most of the water is driven off at
about 650°C.
6
This temperature range can easily be
reached by the mold when pouring molten bronze; in
Fig. 10, it can be clearly seen that just 1 s after the
complete filling of the cavity, with the bronze almost
fully liquid, the internal surface of the mold and the
core both reach temperatures higher than 650°C.
Another cavity can be obtained in the core of the
pommel when the gating system is placed on the
pommel disk, in both case A and B, whereas in case
C, this is concentrated only in the riser (Fig. 11).
Contrary to the previously observed defects, this is
only related to the alloy shrinkage that occurs
during solidification.
Fig. 9. Simulated solidification time (in seconds) in the three configurations.
Fig. 10. Simulated temperature profile of internal mold cavity and core about 1 s after the complete filling.
Casting Simulation of an Austrian Bronze Age Sword Hilt
Author's personal copy
Fig. 11. Shrinkage porosity prediction.
Fig. 13. Predicted shrinkage porosities as a function of the mold and core initial temperature: (a) 20°C, (b) 500°C, and (c) 800°C.
Fig. 12. X-ray tomography close to the pommel: (a) top of the pommel and (b) middle of the pommel.
Pola, Mo
¨dlinger, Piccardo, and Montesano
Author's personal copy
A similar defect can also be detected by x-ray
axial tomography, as reported in Fig. 12. In the
Bronze Age casting, the hole seems to be smaller
than the simulated one. This could be related to the
finishing operations performed on the hilt; in fact,
what probably happened was that the pommel was
deformed by cold hammering (i.e., hammering
below the recrystallization temperature) until the
desired shape was obtained, promoting the reduc-
tion of the porosity extension. This is backed up by
the undercut that exists in the actual pommel.
Finally, concerning the influence of mold tem-
perature on the formation of defects, practically the
same results were obtained whether the initial
temperature of the mold was set to room tem-
perature or to 800°C. In fact, the porosity at
approximately 48 mm from the base of the pommel
disk, which is related to the metal filling phase, does
not seem to be particularly influenced by the mold
temperature. The shrinkage porosities are slightly
larger or smaller according to whether the tem-
perature gradient between bronze and mold is lower
or higher, but the location remained the same, as
shown in Fig. 12 for the example of case A.
This correspondence in defect distribution, whatever
the initial temperature of the mold and core, further
sustains the hypothesis that the gating system was
placed on the side of the pommel disk (case A) (Fig. 13).
CONCLUSION
To study the casting technique of bronze sword
hilts dating to the Late European Bronze Age (c.
1300 BC to 950 BC), various analytical techniques
and casting simulations were used.
To clarify the actual location of the gating system,
casting simulations of a Schalenknauf-type sword
were carried out using the ProCAST software, while
also evaluating the effect of the initial clay mold
temperature. The alloy chosen for the simulation is
a 10 wt.% tin bronze, as it was the most commonly
used alloy during the Bronze Age and was also used
for the casting of Bronze Age swords. The results of
the simulations were compared with x-ray tomo-
graphies performed on the hilt.
The simulations clearly indicate that the best
match to the original hilt is represented by case A,
where the gating was placed on the bowl-shaped
pommel disk right above the shoulders of the hilt.
Both case B (gating placed on the pommel disk 90°
to the side of case A) and case C (gating located
directly on the pommel) showed different patterns
and distribution of porosities, which were not totally
in agreement from the actual casting.
The results we obtained will certainly facilitate
the reproduction of Bronze Age sword hilts and can
help significantly in the accurate reconstruction of
these weapons, proving, in this way, an optimum
insight into both production and usage of the earli-
est swords.
ACKNOWLEDGEMENTS
The authors wish to thank MSc. C. Viscardi, Prof.
V. Villa, and Dr. D. Salaberger for their helpful
support.
REFERENCES
1. M. Mo
¨dlinger, Herstellung und Verwendung bronze-
zeitlicher Schwerter Mitteleuropas. Eine vertiefende Studie
zur mittelbronze-und urnenfelderzeitlichen Bewaffnung
und Sozialstruktur, 1st ed. (Habelt, Bonn, Germany,
2011).
2. M. Wirth, Rekonstruktion bronzezeitlicher Gießereitechniken
mittels numerischer Simulation, gießtechnologischer
Experimente und werkstofftechnischer Untersuchungen an
Nachguss und Original (Aachen, Germany: Universita
¨t
Aachen, 2003).
3. C. Blesl, M. Mo
¨dlinger, T. Ntaflos, and D. Salaberger,
Fundber. O
¨sterr. 48, 47 (2010).
4. M. Mo
¨dlinger, Insight 50, 323 (2008).
5. J. Riederer, Die Schwerter in Ostdeutschland, ed.
H. Wu¨ stemann (Stuttgart, Germany: Franz-Steiner-Verlag,
2004), pp. 259–329.
6. H. Fraser, Ceramic Faults and Their Remedies, 2nd ed.
(London, U.K.: A & C Black Publishers Ltd., 2005).
Casting Simulation of an Austrian Bronze Age Sword Hilt
Author's personal copy
... Il est en outre possible de visualiser la surface des parties inaccessibles de l'objet, telles que la languette ou la cavité interne de la poignée (figure 30). Ce type d'examen et la production de modèles 3D dans lesquels les défauts à l'intérieur du métal sont précisément localisés ouvrent en outre la voie à d'autres types d'analyses, telles que des restitutions numériques de la coulée de la poignée (Pola et al. 2015). ...
Thesis
Er zijn verschillende regionale syntheses gepubliceerd over vollegreep zwaarden uit de bronstijd, die deze zwaarden beschrijven voor een groot deel van het Europese continent. Voor Frankrijk en de Benelux ontbrak een dergelijke studie echter. Dit proefschrift beoogt in de eerste plaats deze leemte op te vullen door een inventarisatie van de zwaarden uit deze regio. Door deze thesis dan te combineren met informatie uit bestaande studies, is het mogelijk de verspreiding van zwaarden met een massieve greep op een Europese schaal te bestuderen. Het inventarisatiewerk is een essentiële voorwaarde om de verspreiding van deze wapens over grote afstanden te bestuderen. Dit dient gepaard te gaan met modellen om de organisatie van hun productie en de verspreidingswijze ervan te karakteriseren. Het vereist eveneens dat er vragen worden gesteld over de status van de ambachtslieden, hun organisatie en hun plaats in de maatschappij. Tot nu toe hebben veel auteurs gekozen voor een relatief eenvoudige benadering, die erin bestaat types zwaarden te definiëren op basis van hun morfologie en versieringen en vervolgens hun verspreiding te bestuderen. De concentraties op de verspreidingskaarten worden beschouwd als het productiegebied. De productietechnieken van vollegreep zwaarden zijn tot nu toe weinig aan bod gekomen bij de vraagstelling over de productiegebieden en hun verspreiding. Een van de eerste belangrijke punten van deze thesis is te reflecteren over de oorsprong van de morfologische, versierings- en technische kenmerken van deze wapens. De productietechnieken zijn eerder het resultaat van de keuzes van de ambachtslieden. Het is onduidelijk of de vorm en de versiering van de zwaarden bepaald zijn door de gebruikers, de producenten of beiden. Daarom lijkt het zinloos om aan de hand van stilistische kenmerken types te definiëren en toe te wijzen aan productiegebieden, aangezien ruimtelijke concentraties op zijn best overeenkomen met regio’s vollegreep zwaarden werden gebruikt. Dankzij beeldvormingstechnieken konden we bepaalde aspecten van de zwaardproductie bestuderen, zoals het gieten van het gevest en de manier waarop het aan de kling is bevestigd. Op basis van deze waarnemingen werd een typo-technologie geconstrueerd, waarmee verschillende technische tradities konden worden geïdentificeerd die overeenkomen met verschillende groepen van producenten. Door de studie van de weinige bekende gietmallen van vollegreep zwaarden, evenals van de fenomenen van imitatie en hybridisatie, die op lokale producties kunnen wijzen, hebben wij potentiële productiegebieden geïdentificeerd alsook de organisatie van de ambachtslieden en de circulatiepatronen kunnen bepalen. Dit werd eerst gedaan per type alvorens de synthese per periode voor te stellen. Tijdens een groot deel van de midden en late bronstijd werden de meeste vollegreep zwaarden eerder in opdracht geproduceerd in gespecialiseerde werkplaatsen, die eerder weinig in aantal waren en waarschijnlijk gecentraliseerd. Parallel aan deze productiecentra moeten minder gespecialiseerde en meer veelzijdige ambachtslieden enkele wapens op een minder gestandaardiseerde manier hebben gemaakt. De situatie verandert abrupt tijdens Hallstatt B2/3. De laatste fase van de bronstijd wordt inderdaad gekenmerkt door grote stilistische en technische veranderingen en door een herconfiguratie van de uitwisselingsnetwerken. Dit weerspiegelt waarschijnlijk een vernieuwing onder de gebruikers van deze wapens en een diversificatie van de productiecentra.
Article
Dans les années 1960, une épée à poignée métallique datant de la seconde moitié du Bronze moyen (1450-1300 av. J.-C.) fut draguée dans une gravière du Rhône à Champagneux (Savoie). Il s’agit d’une épée à fusée octogonale (en allemand : Achtkantschwert). Ce type, caractéristique de la région nord-alpine et du sud de la Scandinavie, était jusqu’alors inconnu en France. Des analyses par spectroscopie de fluorescence X (XRF) et tomographie haute résolution par rayons X ont été réalisées afin de comprendre comment la poignée a été coulée puis fixée à la lame. Ces méthodes nous ont ainsi permis de reconstituer la chaîne opératoire de cette épée en réalisant une inspection minutieuse de la structure interne de sa poignée. Nous avons par ailleurs été en mesure de visualiser la surface de la languette de la lame, insérée à l’intérieur du manche, qui présente une série de cinq marques disposées verticalement. Des éléments similaires n’étaient jusqu’à présent connus uniquement sur des outils et parures de du Bronze final, essentiellement dans la partie orientale de l’arc alpin. Bien que leur fonction exacte demeure inconnue, ces marques pourraient être des marqueurs laissés sur les objets au moment de leur fabrication. Dans le cas de l’épée de Champagneux, il pourrait s’agir d’un moyen pour l’artisan d’identifier rapidement dans son atelier la lame et la poignée devant être assemblées. Ces marques nous permettent ainsi d’obtenir de nouvelles informations l’organisation de la production de ce type d’arme.
Full-text available
Article
A 3000-year-old sword with a broken blade of the Schalenknauf-type, that has been found in a Late Bronze Age grave near St. Plten, Lower Austria, was analysed using different methods to get information about the manufacture and usage of Bronze Age swords. Among the methods used were X-ray, micro-X-ray computer tomography, EPMA, metallographic and surface analysis. The micro-X-ray computer tomography was done at the University of Applied Sciences in Wels, Upper Austria; it was the first time this method was ever applied on Bronze Age weaponry in Austria. It was possible to detect the cause of the breakage, assess the quality of the casting and locate the pouring gate. Furthermore, a detailed description of the casting method, the construction of the haft and the casting mould could be made. The break of the blade and the absence of other visible destruction marks (which are often found on intentionally destroyed swords) are arguments that support the theory of the use of the sword as a weapon. As the metallographic examination of the blade sample shows, the sword blade was annealed and cold-worked. Afterwards the blade was sharpened. A high percentage of European research on Bronze Age swords still considers them as unused or unusable weapons; this denies thousands of years of continuous development towards ultimate effectiveness.
Article
Archäologische Funde interessieren uns nicht nur als Ausstellungsobjekte. Erkenntnisse über ihre technischen Eigenschaften und den Herstellungsprozeß gewähren uns vor allem Einblicke in die Handwerkskunst, das Leben und den Entwicklungsstand unserer prähistorischen Vorfahren. Fundsituation, Erhaltungszustand und konservatorische Restriktionen ermöglichen jedoch oft nur selektierte Untersuchungen und archäologische Experimente, sodass systematische und technisch fundierte Arbeiten bisher selten blieben. Hier bietet die Bronzezeit, aus der im Gegensatz zu anderen Zeitperioden zahlreiche, meist gegossene, Metallgegenstände überliefert sind, eine sowohl in archäologischer als auch in technischer Hinsicht außergewöhnlich gute Forschungssituation. Vor allem die einheimischen Bronzeobjekte zeugen von einem weit entwickelten Metallhandwerk. Die Untersuchungen dieser Arbeit richten sich insbesondere auf die Gießerei- und Schmiedetechnik der in der Bronzezeit gebräuchlichen Randleisten- und Lappenbeile sowie den Einsatz und die Herstellung ihrer Gußformen aus Lehm, Stein und Bronze. Ein abschließendes Kapitel behandelt die Herstellung bronzezeitlicher Schwertgriffe, mit denen ein Objekt berücksichtigt wurde, dem neben der praktischen auch eine repräsentative Aufgabe zukam. Äußere Merkmale wie Bearbeitungsspuren oder Gußfehler geben meist erste wichtige Hinweise auf die Herstellungsweise der Originale. Weitere Details verraten Untersuchungen der inneren Strukturen wie Porositätsausbildungen oder Kornverteilungen. Hierbei werden zum Erhalt des Kulturgutes möglichst zerstörungsfreie Methoden, z. B. die Röntgenographie, eingesetzt. Simulationen des Formfüll- und Erstarrungsvorgangs ermöglichen auf unkomplizierte Weise, gießtechnisch relevante Parameter wie Gießzeit, Anschnittsystem, Formmaterial oder Gieß- und Formtemperatur im für die Bronzezeit wahrscheinlichen Rahmen so lange zu variieren, bis das Erscheinungsbild des Originals optimal angenähert ist. Ergänzende Gießversuche verifizieren die aus den Objektuntersuchungen und der Simulation erhaltenen Erkenntnisse und erschließen die über die Simulation nicht zugänglichen Bereiche wie Formaufbau, Schmelztechnik und Nachbearbeitung. Über die systematische Hinterfragung der Entstehung einzelner Objekte konnten mit Hilfe moderner Methoden der Gießereiforschung weitreichende Erkenntnisse zu den bronzezeitlichen Herstellungsprozessen gewonnen werden. In Verbindung mit den Ergebnissen aus den archäologischen Recherchen eröffnen sich darüber hinaus auch Einblicke in die zeitlichen und räumlichen Entwicklungen, sowohl des bronzezeitlichen Metallhandwerks als auch der bronzezeitlichen Gesellschaft. Letzterer können wir neben einem bereits großen technischen Wissen und der Beherrschung komplizierter Fertigungsverfahren ein Streben nach effizienten Lösungen bescheinigen. Archaeological findings normally arouse our interest in well arranged exhibitions. But thorough knowledge about their technical properties and their production process can reveal hidden secrets about the handcraft, the living and the social evolution of our prehistoric ancestors. Unfortunately the finding situation as well as the state of conservation and restrictions due to aspects of preservation often only allow very strictly selected investigations and archaeological experiments. Thus, until now there are only few systematic and technically reliable studies within this field. At this point, in contrast to other prehistoric periods, our knowledge about the Bronze Age takes advantage of a big amount of preserved, mainly casted metal artefacts. This circumstance provides an unusually good bases for archaeological and technological research. Especially the local bronze objects testify a well developed metal handcraft. This thesis basically investigates the casting and forging technology of flanged and winged axes. Both were common in the Bronze Age. Further researches are treating the use and production of the casting molds, made from clay, stone and bronze. A final chapter covers the production of Bronze Aged sword handles, objects that in addition to their practical use were also worn for representative reasons. Surface characteristics like working traces or casting defects often give first useful hints about the production process of an original artefact. Further details are revealed by investigations on inner structures like the development of porosity and grain distribution. Here, with respect to the preservation of the cultural treasure, mostly non-destructive research methods are applied, for example the X-ray photography. Virtual casting or, strictly speaking, the computer aided calculation of the mold filling and the subsequent solidification process offers an uncomplicated tool to alter most technologically relevant casting parameters until the expected results are approximating the original's appearance. These parameters can be the casting time, the gating system, the mold material or the casting and mold temperature. Complementary casting experiments verify the results gained from the investigations on the objects and the simulations. Further more they give assessment to the fields that cannot be covered by simulations, such as molding and melting techniques and surface finishing. Using up to date scientific investigation methods of casting technology, these systematic investigations of the production of single artefacts could reveal hidden secrets and new aspects of Bronze Aged metal handcraft. Furthermore, in correlation with the results obtained by archaeological researches this knowledge also gives insight in the chronological and local developments of both, technology and society of the Bronze Age. Obviously our prehistoric ancestors already had good technological knowledge and readiness in sophisticated production processes, always aiming towards efficient solutions.
  • M Mödlinger
M. Mödlinger, Insight 50, 323 (2008).
Ceramic Faults and Their Remedies
  • H Fraser
H. Fraser, Ceramic Faults and Their Remedies, 2nd ed. (London, U.K.: A & C Black Publishers Ltd., 2005).
  • C Blesl
  • M Mödlinger
  • T Ntaflos
  • D Salaberger
C. Blesl, M. Mödlinger, T. Ntaflos, and D. Salaberger, Fundber. Österr. 48, 47 (2010).
Die Schwerter in Ostdeutschland
  • J Riederer
J. Riederer, Die Schwerter in Ostdeutschland, ed. H. Wü stemann (Stuttgart, Germany: Franz-Steiner-Verlag, 2004), pp. 259-329.