How did they mass produce those old cylinder records?
How did they mass-produce those old-fashioned cylinder records? A conventional molding press, like they use for discs, would leave some sort of line where the two halves met, which would show up as a click or thump when the cylinder was played. How did they make 3-D moldings of such accuracy in the 1890s? Or did the artistes just make the same recording over and over again?
They sure did, at least at the beginning. This is why you didn't see a lot of albums selling 18 million copies in 1887. The need for a cheap and easy method of reproduction was one of the first problems the early recording industry faced, and the problem you describe was one of the reasons why cylinders lost out to discs as the principal recording medium.
In the very beginning, of course, a little thing like a seam on the recording surface didn't matter too much. On Edison's original phonograph, the ends of the tinfoil sheet that recorded the sound were just tucked into a slot that ran the length of the metal cylinder that the foil was wrapped around. You did get a click this way, but since you also got an indescribable barrage of burps, wheezes, and rasps, the first recording devices being a little on the rustic side, it seems probable that you did not object to the clicks so much. Later, the recording blanks were made of wax, which could be cast in one piece, eliminating the click, if nothing else.
When records first began to be sold commercially, the only way to make additional copies was to have the artistes make the same recording over and over. You would hire, say, a brass band, which you would surround with a phalanx of recording machines loaded with blank discs, and you'd get some guy with a suitably stentorian voice to go around to each machine, flip it on for a second, and holler the title of the piece into the speaking horn. Then you'd turn on all the machines at once, and the band would play as much of any given tune as would conveniently fit onto the cylinders, which was generally about two minutes' worth. Then you changed cylinders and started over. Apart from being stupefyingly monotonous for the performers, this method was very slow.
Eventually somebody hit on the idea of recording additional cylinders off a master cylinder by means of a pantograph, which was an arrangement of levers and wires that transmitted the sound vibrations from the stylus on the master disc to that on the receiving disc. This was faster and less boring, but the masters tended to wear out quickly, and then the band had to go at it again.
Finally, around the turn of the century, Edison's phonograph company developed a reliable method for mass production. They coated the wax master with a thin layer of gold by an electrical process, coated the gold layer with a copper layer for strength, then melted out the original wax. This left a negative metal mold. Then they put a wax blank inside and applied heat and pressure. When the wax cooled, it shrank a little. In addition, the master and blank were tapered slightly--one end was slightly wider than the other. The combination of shrinkage and taper was enough to let them slip the master off the copy without (a) damaging it or (b) leaving a seam.
Actually, this method had occurred to Edison and his buddies fairly early on, but the first recording styli gouged out such deep grooves that the shrinkage wasn't enough to enable them to clear. The development of the sapphire-tip stylus, which made shallower indentations, cleared up this problem.
Unfortunately, by the time they got all this worked out, cylinders were beginning to decline in favor of discs, which were longer playing, among other things. So it was all for nothing, as is often the case in the record business.