The document provides an overview of semiconductor memory, detailing the basics of memory arrays, addressing, and operations such as reading and writing data. It distinguishes between RAM (volatile) and ROM (non-volatile) types, explaining their characteristics, including SRAM and DRAM. Additionally, it covers various types of programmable ROMs and flash memory, including their operations and memory expansion techniques.
The document provides an overview of semiconductor memory, detailing the basics of memory arrays, addressing, and operations such as reading and writing data. It distinguishes between RAM (volatile) and ROM (non-volatile) types, explaining their characteristics, including SRAM and DRAM. Additionally, it covers various types of programmable ROMs and flash memory, including their operations and memory expansion techniques.
• Memory is the portion of a computer or other system that stores binary data • Units of Binary Data: Bits, Bytes, Nibbles, and Words • The Basic Memory Array: • Each storage element in a memory can retain either a 1 or a 0 and is called a cell • Memories are made up of arrays of cells Semiconductor Memory Basics • Location of a cell in a memory array can be identified by specifying a row and a column
The 64-cell array
viewed as either a 64-bit memory or an
8-byte memory Semiconductor Memory Basics
16 x 4 array A 64 x 1 array
A 16-nibble memory A 64-bit memory
Semiconductor Memory Basics • Memory Address and Capacity: • The location of a unit of data in a memory array is called its address the address of a bit in the 2- the address of a byte dimensional array is specified is specified only by by the row and column the row Semiconductor Memory Basics • The address depends on how the memory is organized into units of data. • The capacity of a memory is the total number of data units that can be stored Memory Banks and Ranks: • A bank is a section of memory within a single memory array (chip) • A rank is a group of chips that make up a memory module that stores data in units such as words or bytes Semiconductor Memory Basics Semiconductor Memory Basics Basic Memory Operations • Addressing is the process of accessing a specified location in memory • The addressing operation, selects the specified memory address. • The write operation puts data into a specified address in the memory • The read operation copies data out of a specified address in the memory • Data units go into the memory and come out of the memory on a set of lines called the data bus Semiconductor Memory Basics • Address is placed on a set of lines called the address bus The Write Operation Semiconductor Memory Basics Semiconductor Memory Basics
The Read Operation
RAMs and ROMs RAM • random-access memory • all addresses are accessible in an equal amount of time and can be selected in any order for a read or write operation • All RAMs have both read and write capability • RAMs are volatile memories. RAMs and ROMs ROM • read-only memory • data is stored permanently or semi-permanently • Data can be read from a ROM, but there is no write operation • The ROM, like the RAM, is a random-access memory • Traditionally RAM means a random-access read/write memory The Random-Access Memory (RAM) • The two major categories of RAM are: • the static RAM (SRAM) • the dynamic RAM (DRAM) • SRAMs generally use latches as storage elements - store data indefinitely as long as dc power is applied. • DRAMs use capacitors as storage elements – requires refreshing The RAM Family The Random-Access Memory (RAM) • Both SRAMs and DRAMs will lose stored data when dc power is removed • Data can be read much faster from SRAMs than from DRAMs • DRAMs can store much more data than SRAMs for a given physical size and cost Static RAMs (SRAMs) Memory Cell Static Memory Cell Array
The memory cells in a SRAM are organized in
rows and columns Static Memory Cell Array Basic Asynchronous SRAM Organization • An asynchronous SRAM is one in which the operation is not synchronized with a system clock. Basic Asynchronous SRAM Organization In the READ mode: the eight data bits that are stored in a selected address appear on the data output lines
In the WRITE mode:
the eight data bits that are applied to the data input lines are stored at a selected address
The data input and data output lines share the
same lines Tri-state Outputs and Buses • Tri-state buffers in a memory allow the data lines to act as either input or output lines • These buffers have three output states: • HIGH (1), • LOW (0), • and HIGH-Z (open). Tri-state Outputs and Buses • A bus is one or more conductive paths that serve to interconnect two or more functional components of a system or several diverse systems • A microprocessor is connected to memories and input/output devices by certain bus structures • An address bus: - to address the memories • A data bus: - transfer of data • A control bus: - control data transfers and timing for the various components Read and Write Cycles Read and Write Cycles Dynamic RAM (DRAM) Memory Cells • Dynamic memory cells store a data bit in a small capacitor rather than in a latch • Advantage • it is very simple • Allows very large memory arrays to be constructed on a chip at a lower cost per bit Dynamic RAM (DRAM) Memory Cells • Disadvantage • storage capacitor cannot hold its charge over an extended period of time • Charge is refreshed periodically • Refresh requires additional memory circuitry and complicates the operation of the DRAM Dynamic RAM (DRAM) Memory Cells • Typical DRAM cell consisting of a single MOS transistor (MOSFET) and a capacitor Dynamic RAM (DRAM) Memory Cells (Write Operation) The transistor acts as a switch A LOW on the R/W line (WRITE mode) enables the tri-state input buffer and disables the output buffer A 1 to be written into the cell, the DIN line must be HIGH The transistor is turned on by a HIGH on the row line The transistor then connects the capacitor to the bit line. The capacitor to charge to a positive voltage Dynamic RAM (DRAM) Memory Cells (Write Operation) To store a 0 A LOW is applied to the DIN line If the capacitor is storing a 0, it remains uncharged, or if it is storing a 1, it discharges When the row line is taken back LOW, the transistor turns off This disconnects the capacitor from the bit line, thus “trapping” the charge (1 or 0) on the capacitor Dynamic RAM (DRAM) Memory Cells (Read Operation) R/W (Read/Write) line is made HIGH, enabling the output buffer and disabling the input buffer The row line is made HIGH The transistor turns on and connects the capacitor to the bit line and to the output buffer The data bit appears on the data-output line (DOUT). Dynamic RAM (DRAM) Memory Cells (Refreshing) For refreshing the memory cell, the R/W line is HIGH The row line is made HIGH The refresh line is made HIGH The transistor turns on, connecting the capacitor to the bit line The output buffer is enabled The stored data bit is applied to the input of the refresh buffer This produces a voltage on the bit line corresponding to the stored bit, thus replenishing the capacitor The Read-Only Memory (ROM) The Mask ROM • It is permanently programmed during the manufacturing process Programmable ROMs • The difference is that PROMs come from the manufacturer unprogrammed and are custom programmed in the field to meet the user’s needs. • Once they have been programmed they are the same as mask ROMs. • A PROM uses a fusing process to store bits, in which a memory link is burned open or left intact to represent a 0 or a 1 • The fusing process is irreversible Programmable ROMs
The fusible links are manufactured into the
PROM between the source of each cell’s transistor and its column line Programming process: a sufficient current is injected through the fusible link to burn it open to create a stored 0 The link is left intact for a stored 1. Programmable ROMs • Three basic fuse technologies used in PROMs are: • metal links, • silicon links, • and pn junctions EPROMs • An EPROM is an erasable PROM • An EPROM can be reprogrammed if an existing program in the memory array is erased first • An EPROM uses an NMOSFET array with an isolated-gate structure • The isolated transistor gate has no electrical connections and can store an electrical charge for indefinitely • The data bits in this type of array are represented by the presence or absence of a stored gate charge EPROMs • Erasure of a data bit is a process that removes the gate charge. • Types of erasable PROMs are: • the electrically erasable PROM (EEPROM) • ultraviolet erasable PROM (UV EPROM). EEPROMs • An electrically erasable PROM • Erased and programmed with electrical pulses • The EEPROM can be rapidly programmed and erased in-circuit for reprogramming. • Two types of EEPROMs are the floating-gate MOS and the metal nitride- oxide silicon (MNOS). The Flash Memory • Flash memories are high-density read/write memories • Flash Memory Cell: A single-transistor cell in a flash memory Flash Memory Cell • The stacked gate MOS transistor consists of: • a control gate • a floating gate • the drain and source • The floating gate stores electrons • A 0 is stored when there is more charge and a 1 is stored when there is less or no charge Basic Flash Memory Operation • There are three major operations in a flash memory: • the programming operation, • the read operation, • and the erase operation Programming • Initially, all cells are at the 1 state because charge was removed from each cell during erase operation • The programming operation adds electrons (charge) to the floating gate of those cells that are to store a 0. • No charge is added to those cells that are to store a 1. Programming This attracts electrons to the floating gate
Once programmed, a cell can
retain the charge indefinitely without any external power.
A sufficient positive voltage to the control gate
with respect to the source Read Operation • A positive voltage is applied to the control gate • The amount of charge present on the floating gate of a cell determines whether or not the voltage applied to the control gate will turn on the transistor. • If a 1 is stored, the control gate voltage is sufficient to turn the transistor on. • If a 0 is stored, the transistor will not turn on because the control gate voltage is not sufficient to overcome the negative charge stored in the floating gate. Read Operation Erase
During an erase operation, charge is removed from
all the memory cells
A sufficient positive voltage is applied to the
transistor source with respect to the control gate.
This voltage attracts electrons from the floating gate
and depletes it of charge Memory Expansion • Memory can be expanded to: • increase the word length • increase the word capacity • Or Both • Memory expansion is accomplished by adding an appropriate number of memory chips Word-Length Expansion Word-Length Expansion Word-Length Expansion Word-Capacity Expansion Word-Capacity Expansion End