I've only seen two types of cores. Though I'm sure there are many others as I've never had a machine of my own which used it. The Rods you speak of may be RUBY? Hold it up to shine light through them. The ruby cores look mighty pretty. Regards, Jeff -----Original Message----- From: cctalk-admin(a)classiccmp.org [mailto:cctalk-admin@classiccmp.org] On Behalf Of Rick Bensene Sent: Thursday, February 06, 2003 11:38 AM To: cctalk(a)classiccmp.org Subject: RE: interesting find I have a strong feeling that this is a ROM rather than RAM. The rods form the core of a transformer. A bunch of 'word' wires each have a few turns wrapped around the (probably a ferrite material) rod in a clockwise or counter-clockwise to encode a 1 or a 0. Another winding around the rod is the sense coil. When a current pulse passes through one of the word wires, a current is induced into the sense winding. The induced current in the sense winding is different, depending on which way the word wire was wrapped around the rod. The current pulse in the sense winding is amplified and discriminated (to 1 or 0), and presented as the output for that bit. A number of similar architectures were were commonly used for read-only microcode storage on computers and even calculators until the mid-1970's, when IC-based ROM began to appear. The HP 9100A/B calculators use a similar architecture, using wire bobbins instead of rods, for a microsequence store. The Wang 500, 600, and 700-series calculators also use a similar architecture, utilizing "U"-shaped ferrite structures as the transformer core, for microcode storage. Definitely a neat old relic. Rick Bensene The Old Calculator Web Museum http://www.geocities.com/oldcalculators HP 9100B: http://www.geocities.com/oldcalculators/hp9100b.html Wang 720C: http://www.geocities.com/oldcalculators/wang720.html nventarnummer=I0105 It uses rather large magnetic rods to store data (4x25mm). The PCB is dated 1972 and it was probably used in a Nixdorf-Computer. It's called Staebchenspeicher in german (maybe rod-memory in english?). Does anyone have further information on this module? bye Thilo

On 06-Feb-2003 Rick Bensene wrote: Ahh, thanks that was the answer I was looking for... :-) This would explain why the coils are mounted on removable "cards" inside the module. And when nobody cared about the size, weight and power consumption of computers... ;-) Now I remember reading an article that mentioned this type of memory. The article was about radiation-hard memory used on early space missions. Yeah, another interesting item for my collection of forgotten memory technology :-) bye Thilo

The HP9100B was entirely discrete transistors [1], with normal R/W core memory, the core-on-a-rope microcde store and inductively coupled PCB tracks for the main program ROM. The HP9810 which replaced it was built from TTL chips, with 256 nybble PROMs for the microocde store (and the ALU, which was a couple of programmed PROMs), 512 byte ROMs for the main program store, and 256 bit DRAMs (1103s) for the R/W memory. Admittedly the 9810 had space for an internal thermal printer, and it had more user memory. But in the basic configuration it did less ('Math' functions, like SIN, COS, TAN were on a plug-in ROM module on the 9810, and bulit-in on the 9100). But the 9810 (the machine built with ICs) is larger than the 9100. [1] OK, there are 8 IC op-amps in the 9100B on the card reader PCB (read amplifier and comparator for the 3 data tracks and the clock track). But the machine will run without the card reader ;-) -tony