28 Apr
1999
28 Apr
'99
12:31 a.m.
On Tue, 27 Apr 1999 Philip.Belben(a)pgen.com wrote:
The First "No" is supposed to be "Now"... Try typing on a -12 baud
telnet
link. Edit... took 8minutes to type this sentence and correct one error.
Hell will freeze before the capitalized "f" gets fixed.
<UNTIL you
get to about 7.5 amps or so, and then it looks like:
No you know what nuts dont work...
I don't usually have trouble with American English, but I must confess this
baffles me. Does it mean something like "Now you know why nuts don't
work"? :-) Hows this for an idea. If you find that the toroidal
ferrites that Siemens and
people make don't have enough hysteresis, why not go to the other extreme and
try the stuff magnets are made of. Ring magnets as used for loudspeakers and
things are probably a bit large :-) - I wonder if it's possible to drill a
hole down a bar magnet and then cut slices off. For mass production, I'm sure
the magnet manufacturers would sell you the stuff unmagnetised...
IT may work. It takes a lot to magnetize that stuff though.
Now can someone enlighten me: There will be a minor
glitch-type delay between
read and sense pulses with no transition, and a much bigger delay with a
transition. How do you tell the difference? Is this one of the applications
where a monostable really is useful?
The prefered digital delay is a delay line, monostables have far to much
timing jitter and drift with themperature to be reliable. They would work
for an experimental setup.
I think what he meant was: Rather than sending write
current through all core
planes and inhibit current back through those to which you don't want to write,
why not send write current only through those planes to which you want to write?
If you doing word width usually there is a single x and a single y driver
for the whole array with a seperate inhibit for the bit lines. It's
a matter of *n circuits.
The other property that affects this is size. The
smaller the core, the less
current it takes to magnetise it.
Yep.
Allison