Analog Sequencer/Sampler? (e)eprom

Jay Martin jmar at
Fri Aug 11 19:59:49 CEST 2000

There are problems with this approach to analog storage as well.

I am reminded of the old Luxor satellite receivers I used to service.
The Luxor worked with a down-converter that used a tuning voltage for
channel selection.  The Luxor receiver then had an analog memory system
not unlike the "flash" you are describing to remember the station voltage
for each transponder.  This gave it alot of flexability.  

As the parts would age, however, the voltages in the memory cells would 
begin to drift, requiring re-alignment.  Eventually, individual cells
would go bad completely, causing channels that could no longer be 

My $.02

-----Original Message-----
From: Martin Czech [mailto:czech at Micronas.Com]
Sent: Friday, August 11, 2000 12:35 PM
To: patchell at
Cc: synth-diy at
Subject: Re: Analog Sequencer/Sampler? (e)eprom

:::    As far as I know, you are not going to be able to store an analog
in a
:::bit of an (e)eprom.  If perfermance is what you want, I don't think you

Why not? Basically by writing ereasable devices you change
the threshold voltage of some NMOST. Be it Nordheim-Fowler
tunneling or hot electrons from drain.
I.e. you change the characteristic of an analogue device.
The read out value (drain current) is therefore an analog thing.

Of course, memory designers take a great pain in turning
this analog information into bits by elaborated read amplifiers,
but principally analog storage is possible.

What about Holtech chips? I think they store voice via these ideas.

Analog says nothing about discrete time or not.
Are BBDs digital??

There was a company in California ("San"), they proposed multi
level storage in FLASH, ie. each cell could have up to 256
different levels, this would mean 8 bits in the area of formerly one.

Of course, this was (too) demanding.


More information about the Synth-diy mailing list