|1987|| Conner CP340 Family|
|The definitive 3½ inch HDD|
Why it’s important Conner Peripherals through innovative designs and unique customer-investor relationships established the definitive model for the 3½-inch generation of disk drives. Technical innovations included microprocessor controlled functionality, embedded servo positioning technology, self testing capability, and adopting an emerging industry standard disk size which simplified infrastructure and reduced cost. In addition to design innovations, Conner utilized an investor-customer relationship with Compaq to achieve instant market presence at high volumes. Conner thereby created a mass market product which ultimately established 3½-inch form factor as the dominant HDD form factor for decades to come.
See also: Conner CP341 Drive for Conner Peripheral's involvement in the commercialization of the ATA interface.
Discussion The 3½ form factor HDD was introduced as early as 1983, see Rodime, but the form factor had achieved limited commercial success in the HDD market. Drive manufacturers had focused primarily on cost per megabyte, which favored larger form factors with bigger disks. The 3½-inch disk size has about 1/2 the storage capacity per disk than the larger 5¼-inch so adoption of smaller media wad deemed not economically desireable by drive makers. Higher performance mechanisms, such as a closed loop dedicated disk servo system, could provide greater capacity favored by the market, but were too expensive. Thus prior to the Conner CP340 family the 3½-inch HDD form factor had failed to achieve market success even though it offered advantages such as reduced size, lower power consumption (less heat to remove), greater mechanical robustness, and a form factor which matched the 3½-inch FDD allowing manufacturers to "mix and match" FDDs and HDDs.
Finis Conner, Vice Chairman and a co-founder of Seagate announced a 3½-inch HDD at a trade show in early 1984 and subsequently left Seagate in September. Seagate made a public commitment to the new size after Conner's departure but also failed to succeed in the market, again because their low cost mechanism could not deliver the capacity then required in the broad market, i.e., 20 megabytes. In his first year after leaving Seagate, Conner "spent most of his time on his yacht and the pro-am golf circuit" [1998WSJ] and then with a small group of engineers developing a smaller drive using his own resources, which he abandoned after less than a year [1985EN].
In June 1985, MiniScribe veterans John Squires and Terry Johnson founded CoData, which began in Johnson's Colorado guest house. Johnson described Squires' product as, " ... a disk drive like one I had never seen before and I’m not sure the industry had ever seen anything like this before". It was totally microprocessor-centered, compared to mechanically controlled drives. The servo system utilized pre-recorded "embedded servo" tracks on the disk itself which allowed the track density to be increased beyond the capability of conventional low cost drives (e.g. stepper motor) and thereby first achieved in the 3½-inch form factor the capacity point required for broad market success, then 40 megabytes. While not the first to use this technology, see DEC RL01, Conner was first in this form factor.
In January 1986, wanting to strengthen the management team Johnson and Squires approached Finis Conner about joining their team. Conner, between positions, having just left Computer Memories Inc after about one month (Sep - Oct 1985), liked what he saw but it was apparent to both Squires and Johnson that Conner "had to have his name on the door." In a rather astute gesture, Johnson agreed to step aside and in Feb 1986 CoData merged into Conner's shell company, Conner Peripherals. Conner did get his name on the door, and when Conner Peripherals went public, Johnson owned 7.1% of the company [1998 Prospectus].
Conner's marketing mantra was "Sell, Design, Build" to insure that a new product would be a guaranteed success. He found a willing customer at Compaq computer, who took a 49% stake in the company with a series of investments, and guaranteed to buy all the drives he could build at that time. It should be noted that the first drive Compaq purchased had already been designed and built to a prototype level. With money, product, and a significant customer in hand, Finis obtained additional investments, built foreign factories, and Conner Peripherals became the fastest growing company in America's history [1990 Fortune].
Compaq Computer at that time was shipping a new intelligent interface (today Parallel ATA) using a bolt on controller from Western Digital with drives made for them by MiniScribe and CDC. In perhaps one of the more interesting HDD management blunders, CDC declined to do a 3½-inch version thereby creating an opportunity that Conner Peripherals was quick to exploit (see The History of CAM ATA and Conner CP341 Drive). Conner Peripherals offered versions both in Parallel ATA (desktop and mobile markets), and SCSI (enterprise market; including narrow, fast, and wide SCSI).
The Conner Peripherals drive could also adjust itself, measuring the acceleration and deceleration of a positioner, and optimizing the access speed (plus controlled acceleration and deceleration) to reduce overall positioning time. Another revolutionary change was using the processing power within the drive to test itself during manufacture. Conventional practice before this innovation was to attach each drive to an expensive and relatively slow test stand, which was a bottleneck in production and very expensive, with factories requiring hundreds of testers, millions of dollars invested, waiting time for tester deliveries, and significant floor space plus manpower. The Squires method utilized the drive to write multiple data patterns and read-back verify to assure correct operation. The only requirement was a power supply, usually one large supply on a rolling rack was used to test over 100 drives at one time. The only drives requiring human intervention were those which failed self-test, lighting an indicator signaling need for technical help.
In 1996 Conner Peripherals was acquired by Seagate.
[1985EN] Riser, J., "Computer Memories Names New CEO," Electronic News, September 23, 1985, p. 21. The venture was Akers-Carroll Manufacturing.
[1998Prospectus] Prospectus offering 5,000,000 shares of Conner Peripherals Common Stock, April 19, 1988
[1998WSJ] Gupta, U., "Conner's Peripheral Vision Is Still Sharp," Wall Street Journal, April 23, 1998, p. 37
[1990Fortune] "AMERICA'S FASTEST-GROWING COMPANY," Fortune Magazine, August 13, 1990, p. 48+
Bill Carlson was the original "author" of this article with Rev 24 dated July 22, 2011 being the last version under his control.
An extended version of this article was circulated for review by the Storage SIG at its March 28, 2012; discussion was tabled pending additional input