On Wed, 12 Jul 2006 14:45:22 -0000, Xor wrote:


"One turn effect" in magnetic core computer memory?
Miniature toroids single-stitched and woven with wi
http://www.fortunecity.com/marina/reach/435/coremem.htm
What's the Amidon number of these cores?

Different animal. Those cores have a high remnent magnetism
and a very hard BH curve. The initial ui is far higher than common
ferrites used in RF today. Some were nickel based steels in very
thin foils wound as 50-200mil toroids. The idea is you write them
with a lot of current (for a 50mil core around 2-300ma) and read them
by "writing" them with an apposing polarity pulse, the size and
timing of the pulse returned is a 0 or 1 after amplification and
slicing. The key is you always get a pulse during read but it's about
3-5x bigger and delayed in time (25 to 300ns depending on core size,
temperature and material) _if_ the core was written in with the
opposing magnetic polarity. Typical cores such as used in DEC
pdp-8E had a memory cycle time of 1.6uS (read, modify, write) as a
read is destructive and requires a write cycle to restore the data.
That was typical speed for core memory of the day (1970).

If you want to try this for yourself (1bit memory) a nail or better a
peice of hypersil (transformer core material) with a few turns of
wire, compass and a scope will demonstrate it. The test will be to
find how much DC current will magnetize it then find out how much it
takes to reverse the magnetism (compass helps here). Then you add a
few turn sense winding and watch the pulse that results when you
magnetize it with a given polarity, repeatedly. Then reverse the
power and hit it and the resulting pulse will be later in time and
bigger.

Some electric guitar tube amps used perpendicular
point-to-point wiring in 3-dimensions between
terminal strips to minimize inductive and capactive
crosstalk. They are amazing to see.

Different animal, thats called cordwood construction and its
also for space savings and affords mechanical ruggedness.
Early aerospace systems were built that way for size. Problem
is they are impossible to maintain and heat is problematic for
larger cordwood.

Neither directly relate to the concept of an tuned toroid inductor
in proxmetry to other conductors/metal cases/ inductors.


Allison