From: "Dave Emery" <die@pig.die.com>
Date: Thu, 19 Oct 95 14:46:41 PDT
To: cypherpunks@toad.com
Subject: Re: DalSemi: Add-Only Memory for Storage of Digital Cash (fwd)
Message-ID: <9510192146.AA00991@pig.die.com>
MIME-Version: 1.0
Content-Type: text/plain


Forwarded message:
From die Thu Oct 19 17:44:12 1995
Subject: Re: DalSemi: Add-Only Memory for Storage of Digital Cash (fwd)
To: loofbour@cis.ohio-state.edu (Nathan Loofbourrow)
Date: Thu, 19 Oct 1995 17:44:12 -0400 (EDT)
From: "Dave Emery" <die@pig.die.com>
Reply-To: die@die.com
In-Reply-To: <199510191346.JAA10841@colon.cis.ohio-state.edu> from "Nathan Loofbourrow" at Oct 19, 95 09:46:23 am
X-Mailer: ELM [version 2.4 PL24alpha3]
MIME-Version: 1.0
Content-Type: text/plain; charset=US-ASCII
Content-Transfer-Encoding: 7bit
Content-Length: 2547      


> 
> I seem to remember PROMs actually undergoing physical, rather than
> electrical, state changes (that were presumably nonreversible). Am I
> recalling old technology, or am I just plain mistaken?

	There are three technologies here (as I am sure you know),
masked proms programmed at birth by metalization pattersns and not
subject to non  destructive change after they leave the foundary, fuse
programable proms (which is basically a technology of the  70s to 80s)
and  CMOS proms that depend on patterns of  trapped charge injected into
insulating layers (crudely EPROMs and lots of cousins such as flash
EPROMs, eraseable PALS and FPGAs etc).  

	The old fashioned titanium fuse proms indeed did operate by
having the fuse elements literally blown open by the programming process
using higher voltages and special prgramming logic activated by the
higher voltages that allowed a given fuse to be addressed individually
and blown open (literally vaporized).    This process is obviously one
way (except in defective devices) and once a fuse has been blown (making
a 1 in a blank device a zero for example or enabling a circuit path in a
programmable logic device) there is nothing that can be done to reverse
the process.

	Fuse programmable parts are usually bipolar devices.  And for
the most part fuse programmable proms are small and very fast and
somewhat expensive and use lots of power.

	CMOS eprom technology (which has spread to many many kinds of
programmable logic devices as CMOS has become faster and faster due to
finer line geometries) depends on injecting pockets of charge into a
silicon glass insulating layer.  The electric field associated with
these pockets of charge acts as the gate of FETS built on top of the
insulating layer - turning on those transistors where there is charge
and leaving those off where there is not.   The classical UV eraseable
version of this technology is erased by exposing the device to UV light
which renders the silicon conductive which shorts out the trapped charge
sites.  A modern varient induces a controlled avalanche breakdown
through the silicon insulator region which also shorts out the stored
charge.  The later version is eraseable without UV light by electrical
action and is the technology used in many EEPROMs and Flash EPROMs.

	Thus the modern flash EPROM technology and its cousins in
programmable logic allows many (hundreds to tens of thousands)
of erasure and reprogramming (reinjectiong charge) cycles in modern
prom devices.

						Dave Emery


  > 
> nathan
>