[ Ref. Paul Bellamy in 161002 MMDigest ]

Much has been said about the differing tuning scales for the
24.5" Polyphon and I feel the scale printed in Graham Webb's book,
The Disc Musical Box Handbook, is probably pretty close to, for
sake of argument at this point, the standard 24.5" Polyphon scale
that we usually encounter.

I say 'pretty close' because the scale would have been recorded on
paper from perhaps this 'standard' Polyphon tuned in this key.  For
most people with any experience in comb tuning, they will know that
the combs, whilst in the process of being tuned, do not always conform
exactly to any standard pattern.  There are definite variables found
within the steel, so this non-conformity is probably due to the carbon
content and presence of other alloys in the combs structure.

We must remember that in historical terms the Bessemer process for
steel production was drawing to a close at the time Polyphon got into
the full swing of mass production.  The cost of a low phosphorous,
good quality iron ore was extremely expensive.  A lower grade and much
more affordable quality, still giving excellent results, would have
been economically more viable for the competitive industrialist, but
variables were encountered with ore rich in phosphorous and carbon,
hence the lack of consistency.

Which, for example, could be echoed on the discs.  When trimming up an
old disc back into playing condition, some projections are found to be
soft and malleable and will straighten up readily, while at the other
end of the spectrum, you get extremely brittle ones that break with the
merest touch.

The temper?  Highly doubtful; the manufacturing process in pressing
them would have been a constant, so in all probability I suspect it's
the carbon content within the steel.

Does this have a relevance on tuning?  The answer is yes, I think so.
I suspect that it's universally accepted, that unless you are tone
deaf, no two Polyphons really sound the same.  It is found, when tuning
a Polyphon comb, from my experience, you first rough it out, then get
it somewhere near and then the fine tuning begins.

At this point just as a note is close (or indeed bang on what it should
be) either this tooth, or another tooth, be it higher or lower on the
scale die, they won't resonate and instead of delivering a pleasing
resonant ring (despite being tuned correctly) the frequencies
interfere and cancel out.  The result is a brief 'plop', 'plonk' or
'plink'.

To address this, a mere scraping of the lead, whilst altering the notes
frequency by as little as a cent or two, brings the affected teeth back
into full song.  Hence some Polyphons being able to sing out, while
others are as dull as ditchwater.

This is the aforementioned standard 24.5" Polyphon, and contrary to
Mr. Bellamy's assumptions, this machine was not expensive, but extremely
competitively priced, as can be seen, and comparisons noted, within all
the catalogues of the day.  It was priced to sell, and sell it did.  It
was obviously a highly successful commercial machine because it sits
second only to the trusty old 19-5/8" in respect of surviving numbers
(of commercial Polyphon models).  One cannot refute the math.

To date, to my knowledge, there is nothing concrete in terms of serial
numbers that could contribute to our understanding of manufacturing
output, but if we look at the bigger picture, and consider the sheer
size and might of the Polyphon factory, served with its own rail line,
and consider further the sheer amount of surviving discs -- where we
find music catering for the English, German, French, Russian, Polish,
Hungarian, Serb/Croat and other markets -- then indeed it's safe to say
that a conservative estimate would be in the tens of thousands, if not
hundreds of thousands.

Thirty years ago I came across a 24.5" Polyphon which was tuned
in a different key to what is the standard we usually encounter.
It outshines every other Polyphon of this size; immeasurably so.
It sports a fine and original stained glass door panel, choice veneers,
and one or two more elaborate touches, namely to the keys, escutcheons
and base handle.  Having heard many examples prior to this, and many
a score since, I can confirm only one other playing in this key has
surfaced in the UK that I am aware of.  On discussing this anomaly
albeit somewhat briefly with Mr. Ord Hume some years ago, he told me
that he too had encountered a 24.5" Polyphon tuned in a different key.

Was this anything to do with changing tastes in music as Mr. Bellamy
suggests?  Most probably not; the music played was the same irrespective
of the key.  The musical taste is reflected in the discs title, and
just about all tastes were catered for.

Was it due to the economical use of lead in manufacture?  Highly
doubtful, for this material was indeed plentiful and cheap and the
difference in size of the resonators would be minimal.

The answer to this mystery is probably in the grade of steel used for
the comb.  It stands to good reason that a more expensive quality of
steel would have been used for this second tuning scale, due to its
purity, the steel, while resonating, would conform, and beat in a set
pattern, due to a consistent carbon content.  This could not be achieved
in a lower grade material.

So yes, Polyphon could and did produce a better sounding machine, but
needed a constant with which to work; unfortunately, it was probably
prohibitively expensive to manufacture.  These better sounding examples
were produced in exceptionally small numbers, based upon the scarcity
of surviving examples, and as such, I suspect these were exhibition
pieces, produced with the sole intent to bring in the orders.

To add further support, the quite exceptional Polyphon I mentioned
earlier was exhibited at the Worlds Fair Expo in London 1899.  It can
be seen and heard here.  However the way in which the bass notes
'float' in the air cannot be demonstrated by digital media.

  https://youtu.be/iOedqz-GpCM 

Therefore, I feel the scale encountered on what I refer to as an
'exhibition piece' would have been the ideal for Polyphon.  But the
standard was the economically viable compromise, where we find good,
bad, and indifferent sounding examples.

To add a little credence to this argument, as stated, the Bessemer
process was in decline by the mid 1890's and the improved Thomas
Gilchrist method was now online.  With new technology comes higher
initial costs, and this method was much better at removing impurities,
as it became more financially viable, and as thus more widely used,
it remained a state-of-the-art method of steel production, almost
exclusively, for the next half century.

In relation to comb tuning, it was the Gilchrist method that produced
commercially viable steel with a high standard of consistency.  This
production method would have been in full flow by the time the likes
Regina and Mira entered the fray, and that, I suspect is why we know
exactly what to expect from one of those latecomers.

Mark Singleton
Lancashire, England