How Did RAMAC Work?
How Did RAMAC Work?
Storing data on magnetic tapes and drums was well established by 1957. IBM’s RAMAC team adapted the principle of magnetic storage to a spinning disk. Or rather, 50 spinning disks, each two feet across.
A moveable arm held two magnetic heads that read and wrote data—one head for the underside of disks above the arm, one for the tops of disks below the arm. Continuous airflow from a compressor prevented the heads from touching the disks.
It took, on average, only 0.6 seconds to retrieve a piece of data from anywhere on RAMAC.
RAMAC read/write head, 1956
RAMAC used compressed air to “fly” the read/write heads at a controlled distance from the disk surface. Modern drives use the airflow generated by the spinning disk.
View Artifact DetailRAMAC test platform
This engineering model from IBM’s San Jose, California lab proved that a rotating, magnetically-coated disk stack could be used to store data reliably.
View Artifact DetailRAMAC disk
This is one of 50 aluminum disks used in the RAMAC disk drive. A special coating allowed bits to be recorded as magnetized areas.
View Artifact DetailRAMAC access arms
An optional second arm reduced average access time. RAMAC could be purchased in two sizes: 5 million or 10 million characters.
View Artifact DetailCoating the Disk
The engineers were stymied. RAMAC disks required a smooth coating embedded with iron oxide particles. But spray-painting didn’t yield a uniform surface. The solution? Centrifugal force…and a secretary’s stocking.
Engineers poured magnetic paint onto spinning disks—filtered through a borrowed nylon stocking—and then let centrifugal force spread the coating evenly.
Pouring magnetic oxide paint
This shows the “spin-coating” technique IBM used at the early experimental stage to apply a magnetic surface to metal disks for RAMAC. More advanced and automated coating techniques were used once RAMAC was in production.
View Artifact Detail