Results

Cold Striking and Pure Silver Coinage

The time it took to cold-strike 30 pure silver coins was 8 minutes and 55 seconds. I judged an “acceptable” coin to have minimal or no multiple striking. Also, the obverse (heads) legend should be readable. Most of the facial features should be clearly transferred. Only about 3 out of the 30 struck were acceptable in terms of quality (Figure 19). Thus, it took 18 seconds to make one coin. For every 178 seconds, one acceptable coin was made. Many coins were unacceptable due to the bouncing and subsequent shifting of the coin blank and upper die. Each time the upper die was struck, the coin blank and upper die moved slightly before the next hammer strike. This shifting caused multiple images to be impressed onto the coin blank (Figure 18). This error was not uncommon in the Roman times. It took anywhere from 4 to 10 blows to deeply transfer the image from the dies to the coin. I noticed that after a certain number of blows, usually around 5 or 6, the coin blank hardens so much that it becomes difficult to transfer the image onto the coin blank with subsequent blows. After this point, bouncing is all that occurs.

I found that my small hand was somewhat of a disability in striking because of the length of the upper die. I held the upper die firmly, but this did not stop the bouncing and shifting of the upper die and coin blank. However, when my advisor Jim Mathieu struck a coin, I noticed that the upper die shifted less because his hand was bigger and could afford a firmer grip braced flat against the anvil.

Hot Striking and Pure Silver Coinage

The time it took to hot-strike 30 pure silver coins was 14 minutes and 30 seconds. Thus, it took 29 seconds to make each coin. Out of the 30 coins struck, 12 of them were acceptable (Figure 20). For every 73 seconds, an acceptable coin was made. It took 16 minutes for the furnace to heat up to 600 degrees Celsius. These coins were of much better quality than the cold-struck silver coins. The heat makes the silver more malleable, so that most or all of the force is transferred onto the coin blank and not back to the upper die. Thus, there was less bouncing of the upper die.

Cold Striking and Debased Silver Coinage

The time it took to cold-strike debased coins was 13 minutes and 20 seconds. Only 1 out of 30 was acceptable (Figure 21). Thus, it took 27 seconds to make each coin. For every 800 seconds, one acceptable coin was made. Cold-striking debased coins led to the most bouncing and, thus, the poorest quality. The cold debased coin blanks had the highest hardness out of the four types, so most of the forces transferred to the coin blank bounced back into the upper die. I also sustained an injury because of the bouncing problem. When the upper die bounced, it would sometimes wobble from side to side significantly. If the upper die is displaced too much, the hammer could miss the upper die altogether and hit my hand. This was exactly what happened. The upper die wobbled to the side, and because of the rapid nature of striking, I did not bring the die back into alignment before I swung the hammer again. The hammer scraped my left thumb knuckle.

The appearance of the struck coins still had a slight coppery tone.

Hot Striking and Debased Silver Coinage

The time it took to hot-strike 30 debased silver coins was 13 minutes and 15 seconds. Out of the 30 coins struck, 11 were acceptable (Figure 22). Thus, it took 27 seconds to make each coin. For every 72 seconds, an acceptable coin was made. The hot-struck debased coins were of much higher quality than the cold-struck debased coins. The higher quality was due to the increased malleability afforded by heat. Thus, there was less bouncing. As with the silver coins, the softer the metal is, the more force it can receive. The harder the metal is, the more force it transfers.

I put the coin blanks in the furnace and heated them up until they reached 600 degrees Celsius. In the furnace, they acquired a gray ashy surface of oxidation. However, when I struck them, the oxidation flaked off, leaving a silvery surface. This was an interesting effect because I had wondered how the Romans made their debased coins look so silvery on the surface. My replica blanks were 40 percent copper, but they looked orange. However, when I put them in the furnace and then struck them, they gained a silvery appearance.

A summary of the numerical results can be found in Table 3 in the appendix.

Dies

The dies deteriorated faster with the striking of alloyed coins. Furthermore, the end of the upper die that received the blows mushroomed (Figure 23). This effect is also seen in ancient dies (Figure 24). After the minting experiment, which produced roughly 200 coins, considerable wear to the design of the dies was evident (Figure 25).

Minting errors

There were two common errors in my replica coins that are also present in real ancient Roman coins. The first is the phenomenon of multiple images on one coin. With each slight shift of the upper die, the image of the coin shifts to the direction of the bounce, creating a trail (Figure 18). This is seen in ancient Roman coins also. The second phenomenon is related to the first. Sometimes, the bounces or shifts of the upper die are dramatic. When this happens, two widely spaced impressions are made (Figure 26).

<<PREVIOUS PAGE NEXT PAGE >>

Fig. 18: Coin error type 1: Multiple impressions.

Fig 19: Cold striking produced these pure silver coins.

Fig. 20: Hot striking produced these pure silver coins

Fig. 21: Cold striking produced these debased silver coins.

Fig. 22: Hot striking produced these debased silver coins.

Table 3: Summary of results

s=seconds	Cold striking silver	Hot striking silver	Cold striking debased	Hot striking debased
Efficiency (s/coin made)	18s	29s	27s	27s
Efficiency (s/acceptable coin made)	178s	73s	800s	72s

Fig. 23: Dies after experiments. Note mushrooming on upper die.

Fig. 24: Ancient die with mushrooming effect.

Source: Zograph p. 38

Fig. 25: Impression of lower die before and after experimental striking.

Before	After

Fig. 26: Coin error type 2: Severely displaced double images.