RPM DIE PROGRESSION THEORY
The
average Lincoln cent die produces approximately one million coins.
The coins struck at the end of the die's life are quite different in
their quality than those struck at the beginning. One million
strikes can produce a lot of wear and tear, even on a steel die. The
life of a die can be described by the wear and tear it undergoes
from its beginning to its end. There are two ways of describing die
progressions. The first focuses on die wear. The description of the
wear has been termed “die state” by Delma Romines, the developer of
the theory. In his theory, a die proceeds from a Very Early Die
State (VEDS) to a Very Late Die State (VLDS), with three additional
states in between (Early Die State, Middle Die State, Late Die
State). The die state is basically determined by the amount of die
wear as seen in the metal flow lines which develop as a result of
metal fatigue in the die itself. A VEDS coin has very sharp
lettering and design features with no metal flow lines. These are
the very first and the very best coins struck from a die. A VLDS
coin has very harsh metal flow lines and generally has a “washed
out” look to the lettering and design features. These are the very
last and the very worst coins struck from a die.
The
second way of describing die progressions focuses on die tear. The
description of die tear has been termed “die stage.” This term has
been in use for years and the name of the person who coined the term
has unfortunately long been forgotten. The theory behind “die stage”
progressions has seen several revisions through the years. The
general theory begins like this: When a die is first placed into
use, it is perfect, that is free from “tear.” Then as a die is
used, it develops imperfections that grow steadily worse and worse
until the die has to be pulled from the production line. These
imperfections consist mainly of die clashes, die cracks, die chips,
and die breaks. They may occur anywhere on the die but are most
frequently found where the die receives the greatest stress.
Both
descriptions of die progressions are necessary to fully describe a
die's life. It is entirely possible and indeed highly probable that
a die can proceed from VEDS to VLDS without any die tear. In such a
case die state is of utmost importance to determine where in the
die's life a particular coin should be placed. Likewise, under
normal circumstances severe die tear indicates a late die state and
thus a full production life for the die. However, severe die tear
can occur early in a die's life. Without knowing the die state, the
die stage gives the false impression that the die had a normal
production life when if fact it was cut short. While this general
die progression theory is accurate, there are several assumptions
that must be considered before the theory is actually put into
practice.
RPM Die Progression Assumptions
The
Perfect Die
The
first assumption is that of a perfect die. It is a well known fact
that master dies and hubs have their own set of progressions. Take,
for example, the 1936 Lincoln cents. One of the Working Hubs became
chipped and the left leg of the R of LIBERTY is missing on many of
the coins for that date. Another instance of this is the 1956
Lincoln cents where a small section of the lower 6 of the date is
missing on many coins. A less common example can be seen on the last
T of TRUST on some of the 1941 doubled die cents, where the left
cross bar is missing. If these rather large defects can occur, even
though only rarely, it seems only reasonable that more minor defects
find their way onto the working dies before they ever strike a coin.
This indeed is the case and thus a progression may begin not with a
perfect coin, but with a defective coin (that is a coin with a VEDS
but also showing die tear).
Die
Pair Mating
The
second assumption is that both the obverse and reverse dies are
placed together in the press at the beginning of their die life.
While this can and does happen, it simply is not standard operating
procedure. The reverse dies average just over a million strikes each
while the obverse dies average just under a million strikes each.
When the obverse die wears out, it is replaced with a new die. The
reverse die however, is left in place to strike as many as an
additional 400,000 coins. Thus a progression may begin with a VEDS
obverse and a VLDS reverse. Additionally, that VLDS reverse may show
die tear. When the reverse die wears out, it is replaced with a new
die. The obverse die is now in the middle of its die life. Carry
this sequence through several times and it becomes apparent that
more often than not the dies are paired together at different die
states as well as at different die stages.
More
Die Pair Mating
A
third assumption is that when the dies are placed in the press after
having been removed and abraded (new term for polished or sanded)
they are matched together with their original mate. While this
certainly can happen, it does not have to happen. In a quad press
there are eight dies in four pairs. Thus there is a lot of room for
movement between die pairs. While the dies need not be removed from
the press every time they are abraded, they certainly can be and
depending upon how often the dies are abraded, there could be
several die pair combinations. Abrasion also contributes to several
kinds of die markers. These are die scratches, die gouges, and other
kinds of die damage.
The
“Reverse” Progression
A
fourth assumption states that once a die develops an imperfection,
that imperfection remains on the die until it is retired from use.
Superficial die cracks, especially those on new dies (VEDS), can
“heal themselves” and disappear. As the die wears, the metal heats
and melts these superficial cracks together. Thus a coin without the
die crack is actually further along in the progression than a coin
with the die crack. Die clashes can be abraded away. Likewise, they
can also exhibit this “reverse” progression. Even heavy abrasion
marks or grinder digs can be removed, leaving little if any trace of
their previous existence (e.g. 1961-D 1c RPM-071).
The
RPM Appearance
A
fifth assumption, this one specifically regarding RPMs, claims that
the strongest appearing RPM is the earliest stage. While this is
true a majority of the time, there are quite a few times when just
the opposite is the case. (The following material is an extract from
chapter 3 of The RPM Book (First Edition). The referenced
photos can be found there and below.) The specific coin that brought about the
discovery of this phenomenon is the 1959-D RPM-001. Ever since
Albert Kramer published his booklet on the doubled D cents, this
variety has puzzled the experts. There is a progression of die
stages in this variety causing the RPM to change greatly in
appearance throughout the life of the die. Al Kramer, and Jean Cohen
after him, listed the die stages so that the coin in photo #3 was
the first stage, the one in photo #1 was the second stage, and the
one in photo #2 was third.
|
|
|
Photo # 1 |
Photo # 2 |
Photo # 3 |
Their reasons for arranging the coins in that order were due to
conventional RPM theory and some unfortunate timing by the U.S.
Mint. First, by the theory they used, the coin in photo #3 had to be
the first die stage since it had the strongest appearing RPM. They
reasoned that as the die wore from striking thousands of coins, the
RPM would gradually become less and less distinct until it reached
the die stages represented by the coins in photos #1 and #2. In
fact, this is the way most RPMs develop. Second, they looked at the
reverses of each coin and found the reverse of the coin in photo #3
was perfect, while the coins in photos #1 and #2 usually had die
chips around the Memorial Building. Conventional theory says that
the coin in photo #3 is the earlier die stage.
However, there were questions not answered by arranging the die
stages in this order. For example, how could the very strong
vertical bar of the “D” to the East in
the earlier die stage be obliterated by die wear while the similar
vertical bar of the main “D” is not worn at all? This was a
key question, but everyone ignored it, hoping the question would
resolve itself because they had no answer. It was the new theory on
the 1880/1879 Morgan Dollar overdates that led directly to the
solution to the 1959-D RPM-001 progression.
Briefly, the theory by Leroy Van Allen was this: Some dollar dies
dated 1879 were discovered to be unusable late in the year. Instead
of scrapping some expensive dies, the engraver decided to try
re-impressing the dates on those dies with an 1880 date logo. When
the numbers 1 and 8 were re-impressed, there was little overlap.
However, when the numerals 8 and 0 were pressed into the die settle,
metal was caused to flow just as when a coin is struck. The easiest
direction for the flow to run in this case was into the voids in the
die that were created by the numerals 7 and 9. In most 1880/1879
varieties, the 79 was almost completely obscured by the metal flow.
This
metal flow theory of the overdates is directly related to the RPM
question. When this 1959 RPM was made, one “D” was punched into the
die and then, when the “D” was repunched, metal was displaced into
the cavity left by the first punching, leaving the surface
appearance of the first “D” quite weak (see photo #1). However, this
displacement from the second punch caused a phenomenon known as
“cold working,” with the greatest degree of working concentrated at
the boundaries of the two punches. Cold working hardens metal, but
at the same time it makes it more brittle.
So,
when the die was first put into use, the RPM appeared as in photo
#1. As the die wore during use, chipping formed around the Memorial
Building as seen in photo #4. Then at the brittle boundary between
the two punches, the cyclical stresses of striking coins caused the
metal displaced by the repunching to fail much earlier than
elsewhere in the die and to chip out. The result made the mintmark
appear as in photo #3.
|
Photo # 4 |
The
cause of a lot of confusion was that before the metal in the
mintmark area chipped out, the dies apparently were inspected and
the chipping on the reverse was discovered. The reverse die was
retired and a new one installed in the press. If the obverse die was
inspected at the same time, the RPM was obviously not discovered as
the obverse die remained in use. The timing of this inspection was
just bad luck for Kramer and Cohen, as the replacement of the
reverse die at that particular time was an important factor in the
way they listed the die stages. (The sequence in The RPM Book
(First Edition)
is also inaccurate. See the correct sequence
here.).
Incomplete Listings
One
final assumption needs to be addressed. No matter how many coins one
examines from a particular die progression, it is impossible to
examine every coin in that progression. Progressions are built on
the examination of a very small number of coins. Thus any
progression listing will have huge numbers of coins between the
various stages. Significant stages may even be missing. Exactly
where the dividing line should occur between stages is also a matter
of much debate and cannot finally be settled until large numbers of
coins from any given progression can be examined. A word of advice
about unlisted stages is helpful here. Just because a stage is
unlisted does not mean it has additional value. It simply means that
coins from that particular section of the die life have not yet been
examined. On the other hand, some stages are rare (at least for the
present) and should command higher premiums.
RPM Die Progression Practice
Up
until now, the discussion has centered on die progression theory. It
is now time to place to that theory into practice. But, where does
one begin? The procedure is two-stepped. First, begin with the RPM
itself. Check the RPM direction (North, South, East, West, Rotated,
Tilted etc.). Compare this information to the RPM pictures to narrow
down the scope of the search. Next begin a clockwise search around
the obverse of the coin being careful to notice both the die state
and then any die markers that could determine die stage. Repeat the
search procedure for the reverse of the coin. Match the die markers
against those listed for the RPMs in the narrowed scope. The vast
majority of the time this will bring into focus only one RPM. For
those few instances where markers are similar or lacking, check the
mintmark position relative to the date both vertically and
horizontally. By this time the RPM number should be firmly
established.
The
second step, determining the stage identification, is all that is
left for an accurate RPM attribution. Match the die markers against
the stage listings until a stage is found that matches the markers
exactly. If the markers do not match exactly determine where in the
progression the coin should be placed. Remember to use all the clues
given in the die markers and the general outline of die progression
theory as noted above.
RPM Die Progression Terminology
Two sets of terms need to be clarified. First
are the terms describing die markers.
Die Clash - When the dies come together
without a planchet between them, part of the design from each die is
transferred to the opposing die. Coins struck by those dies will
show the transfer of the design. The transferred design will show
upside down and incuse on a struck coin. For the Wheat Cent look
under Lincoln's chin for an upside down N of CENT on the obverse and
between the E of ONE and the N of CENT for Lincoln's tie on the
reverse. For the Memorial Cent look for the vertical columns of the
Building both in front of and behind Lincoln's head as well as for
the horizontal building lines through the 1 of the date. On the
reverse look in the first 3 bays of the Building for an upside down
RTY of LIBERTY. Also Lincoln's head can be seen through the letters
of ONE CENT.
Die Scratches - These are raised thin
lines usually on the fields of a struck coin (but may occasionally
be seen on the design) which are created by the sanding or abrading
of a die. It is the pattern of the die scratches which make them
important for stage determination. The pattern acts like a
fingerprint, pinpointing an exact position in the die progression.
Die Gouges - These are raised thick
bulges usually on the fields of a struck coin (but may occasionally
be seen on the design) which are created by the striking of a hard
object against the die. This object could be another die, a hammer,
or even a grinder.
Die Cracks - These are raised irregular
lines found usually at the design stress points. Thus they may be on
the design itself or on the fields. Usually they extend from the
design onto the field. Since die cracks usually occur in the same
general areas on every coin (as do also die chips, and die breaks),
they cannot be used to absolutely fix a particular RPM. However,
added to other die markers they serve to tilt the weight of the
evidence in one direction or another.
Die Chips - These are small raised
irregular blobs found usually at the design stress points. They are
most often found in the recessed areas of certain letters or numbers
(e.g. B & R of LIBERTY and the 9 & 5 of the Date). This is because
the recessed areas of the letters and numbers on the coin are raised
areas on the die, which look like little islands. It does not take
much stress before these raised pieces of the die chip and break
off, leaving a raised area on the coin where a recessed area is
expected.
Die Breaks - These are large raised
irregular blobs found usually at the design stress points. They are
most often found in the fields between the design and the rim. After
repeated strikings under tons of pressure the dies begin to crack.
When these cracks meet each other (as in a circle) or when they
extend to the rim, the metal contained within the borders of the
crack begins to break away from the die itself. The broken piece may
not at first be loose enough to fall away. In such a case the coin
will show a depression where the die chip or if it is large enough
the die break has occurred. If it is the anvil die, that is it is
the die which is on the bottom (for Lincoln cents that is usually
the reverse die) then the collar may hold it in place. The coins
produced from such a broken die with show the portion of the die
which has broken away as a raised area like a plateau. Such coins
which have raised areas extending to the rim are called retained
CUDs. Complete die breaks which extend to the rim are called CUDs.
Second are the terms describing the placement
of those die markers. Refer to the following charts for a pictorial
representation of the terms used to describe the placement of die
markers on the Lincoln Cent.
Die Progression
Listings
The
die progression listings should normally be read from bottom to top.
While this at first seems backwards, it will readily be seen that
this is the fastest way of determining the die stage of a given RPM.
This is because the listings provide only the new markers for a
given stage. The markers showing up in previous stages are not
repeated. By beginning at the bottom, each stage can be eliminated
by matching that entry's description against the coin's die markers.
When the markers match the description, then that is the RPM stage.
Practice makes perfect. Thus, after a few matches, the procedure
will become second nature.
|