AMD AM28F020A-90FI 2 megabit (256 k x 8-bit) cmos 12.0 volt, bulk erase flash memory with embedded algorithm Datasheet

FINAL
Am28F020A
2 Megabit (256 K x 8-Bit)
CMOS 12.0 Volt, Bulk Erase Flash Memory with Embedded Algorithms
DISTINCTIVE CHARACTERISTICS
■ High performance
— Access times as fast as 70 ns
■ CMOS low power consumption
— 30 mA maximum active current
— 100 µA maximum standby current
— No data retention power consumption
■ Compatible with JEDEC-standard byte-wide
32-pin EPROM pinouts
— 32-pin PDIP
— 32-pin PLCC
— 32-pin TSOP
■ Latch-up protected to 100 mA from
–1 V to VCC +1 V
■ Embedded Erase Electrical Bulk Chip Erase
— Five seconds typical chip erase, including
pre-programming
■ Embedded Program
— 14 µs typical byte program, including time-out
— 4 seconds typical chip program
■ Command register architecture for
microprocessor/microcontroller compatible
write interface
■ On-chip address and data latches
■ Advanced CMOS flash memory technology
■ 100,000 write/erase cycles minimum
■ Write and erase voltage 12.0 V ±5%
— Low cost single transistor memory cell
■ Embedded algorithms for completely self-timed
write/erase operations
GENERAL DESCRIPTION
devices within this family that offer Embedded Algorithms use the same command set. This offers
designers the flexibility to retain the same device footprint and command set, at any density between
256 Kbits and 2 Mbits.
The Am28F020A is a 2 Megabit Flash memory organized as 256 Kbytes of 8 bits each. AMD’s Flash memories offer the most cost-effective and reliable read/
write non-volatile random access memor y. The
Am28F020A is packaged in 32-pin PDIP, PLCC, and
TSOP versions. It is designed to be reprogrammed and
erased in-system or in standard EPROM programmers.
The Am28F020A is erased when shipped from
the factory.
The standard Am28F020A offers access times of as
fast as 70 ns, allowing high speed microprocessors to
operate without wait states. To eliminate bus contention, the device has separate chip enable (CE #) and
output enable (OE#) controls.
AMD’s Flash memories augment EPROM functionality
with in-circuit electrical erasure and programming. The
Am28F020A uses a command register to manage this
functionality. The command register allows for 100%
TTL level control inputs and fixed power supply levels
during erase and programming, while maintaining
maximum EPROM compatibility.
T he A m2 8F0 20A i s c omp ati bl e w i th th e AM D
Am28F256A, Am28F512A, and Am28F010A Flash
memories. All devices in the Am28Fxxx family follow
the JEDEC 32-pin pinout standard. In addition, all
Publication# 17502 Rev: D Amendment/+1
Issue Date: January 1998
AMD’s Flash technology reliably stores memory contents even after 100,000 erase and program cycles.
The AMD cell is designed to optimize the erase and
programming mechanisms. In addition, the combination of advanced tunnel oxide processing and low
internal electric fields for erase and programming operations produces reliable cycling. The Am28F020A uses
a 12.0±5% VPP supply input to perform the erase and
programming functions.
The highest degree of latch-up protection is achieved
with AMD’s proprietary non-epi process. Latch-up protection is provided for stresses up to 100 mA on
address and data pins from –1 V to VCC +1 V.
AMD’s Flash technology combines years of EPROM
and EEPROM experience to produce the highest levels
of quality, reliability, and cost effectiveness. The
Am28F020A electrically erases all bits simultaneously
using Fowler-Nordheim tunneling. The bytes are
programmed one byte at a time using the EPROM
programming mechanism of hot electron injection.
Embedded Program
Embedded Erase
The Am28F020A is byte programmable using the
Embedded Program algorithm, which does not require
the system to time-out or verify the data programmed.
The typical room temperature programming time of this
device is four seconds.
The entire device is bulk erased using the Embedded
Erase algorithm, which automatically programs the
entire array prior to electrical erase. The timing and verification of electrical erase are controlled internal to the
device. Typical erasure time at room temperature is five
seconds, including preprogramming.
Comparing Embedded Algorithms with Flasherase and Flashrite Algorithms
Embedded
Programming
Algorithm vs.
Flashrite
Programming
Algorithm
Embedded Erase
Algorithm vs.
Flasherase Erase
Algorithm
Am28F020A with
Embedded Algorithms
Am28F020 using AMD Flashrite
and Flasherase Algorithms
AMD’s Embedded Programming algorithm
requires the user to only write a program
set-up command and a program command
(program data and address). The device
automatically times the programming
pulse width, verifies the programming, and
counts the number of sequences. A status
bit, Data# Polling, provides the user with
the programming operation status.
The Flashrite Programming algorithm requires the
user to write a program set-up command, a program
command, (program data and address), and a
program verify command, followed by a read and
compare operation. The user is required to time the
programming pulse width in order to issue the
program verify command. An integrated stop timer
prevents any possibility of overprogramming.
AMD’s Embedded Erase algorithm
requires the user to only write an erase setup command and erase command. The
device automatically pre-programs and
verifies the entire array. The device then
automatically times the erase pulse width,
verifies the erase operation, and counts
the number of sequences. A status bit,
Data# Polling, provides the user with the
erase operation status.
The Flasherase Erase algorithm requires the device
to be completely programmed prior to executing an
erase command.
Commands are written to the command register using
standard microprocessor write timings. Register contents serve as input to an internal state-machine,
which controls the erase and programming circuitry.
During write cycles, the command register internally
latches addresses and data needed for the programming and erase operations. For system design
simplification, the Am28F010A is designed to support
2
Upon completion of this sequence, the data is read
back from the device and compared by the user with
the data intended to be written; if there is not a
match, the sequence is repeated until there is a
match or the sequence has been repeated 25 times.
To invoke the erase operation, the user writes an
erase set-up command, an erase command, and an
erase verify command. The user is required to time
the erase pulse width in order to issue the erase
verify command. An integrated stop timer prevents
any possibility of overerasure.
Upon completion of this sequence, the data is read
back from the device and compared by the user with
erased data. If there is not a match, the sequence is
repeated until there is a match or the sequence has
been repeated 1,000 times.
either WE# or CE # controlled writes. During a system
write cycle, addresses are latched on the falling edge
of WE # or CE #, whichever occurs last. Data is latched
on the rising edge of WE # or CE #, whichever occurs
first. To simplify the following discussion, the WE # pin
is used as the write cycle control pin throughout the
rest of this text. All setup and hold times are with
respect to the WE# signal.
Am28F020A
PRODUCT SELECTOR GUIDE
Am28F020A
Family Part Number
Speed Options (VCC = 5.0 V ± 10%)
-70
-90
-120
-150
-200
Max Access Time (ns)
70
90
120
150
200
CE# (E#) Access (ns)
70
90
120
150
200
OE# (G#) Access (ns)
35
35
50
55
55
BLOCK DIAGRAM
DQ0–DQ7
VCC
VSS
Erase
Voltage
Switch
VPP
Input/Output
Buffers
To Array
WE#
State
Control
Command
Register
Program
Voltage
Switch
CE#
OE#
Chip Enable
Output Enable
Logic
Data Latch
Program/Erase
Pulse Timer
Low VCC
Detector
Address Latch
Embedded
Algorithms
Y-Decoder
Y-Gating
X-Decoder
2,097,152
Bit
Cell Matrix
A0–A17
17502D-1
Am28F020A
3
CONNECTION DIAGRAMS
PDIP
32
VCC
A16
2
31
WE# (W#)
A15
3
30
A17
A12
4
29
A14
A7
5
28
A13
A6
6
27
A8
A5
7
26
A9
A4
8
25
A3
9
A2
A1
4 3 2
1 32 31 30
A7
5
29
A14
A6
6
28
A13
A5
27
A8
A4
7
8
26
A9
A3
9
25
A11
A11
A2
10
24
OE# (G#)
24
OE# (G#)
A1
11
23
A10
10
23
A10
A0
12
22
11
22
CE# (E#)
DQ0
13
21
CE# (E#)
DQ7
A0
12
21
DQ7
DQ0
13
20
DQ6
DQ1
14
19
DQ5
DQ2
15
18
DQ4
VSS
16
17
DQ3
Note: Pin 1 is marked for orientation.
Am28F020A
DQ5
DQ6
DQ4
VSS
DQ3
DQ1
DQ2
14 15 16 17 18 19 20
17502D-2
4
VPP
VCC
WE# (W#)
A17
1
A16
VPP
A12
A15
PLCC
17502D-3
CONNECTION DIAGRAMS (Continued)
A11
A9
A8
A13
A14
A17
WE#
VCC
VPP
A16
A15
A12
A7
A6
A5
A4
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
32
31
30
29
28
27
26
25
24
23
22
21
20
19
18
17
OE#
A10
CE#
D7
D6
D5
D4
D3
VSS
D2
D1
D0
A0
A1
A2
A3
32
31
30
29
28
27
26
25
24
23
22
21
20
19
18
17
A11
A9
A8
A13
A14
A17
WE#
VCC
VPP
A16
A15
A12
A7
A6
A5
A4
32-Pin TSOP—Standard Pinout
OE#
A10
CE#
D7
D6
D5
D4
D3
VSS
D2
D1
D0
A0
A1
A2
A3
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
32-Pin TSOP—Reverse Pinout
17502D-4
LOGIC SYMBOL
18
8
A0–A17
DQ0–DQ7
CE# (E#)
OE# (G#)
WE# (W#)
17502D-5
Am28F020A
5
ORDERING INFORMATION
Standard Products
AMD standard products are available in several packages and operating ranges. The ordering number (Valid Combination) is
formed by a combination of the following:
AM28F020A
-70
J
C
B
OPTIONAL PROCESSING
Blank = Standard Processing
B
= Burn-In
Contact an AMD representative for more information.
TEMPERATURE RANGE
C = Commercial (0°C to +70°C)
I = Industrial (–40°C to +85°C)
E = Extended (–55°C to +125°C)
PACKAGE TYPE
P = 32-Pin Plastic DIP (PD 032)
J = 32-Pin Rectangular Plastic Leaded Chip
Carrier (PL 032)
E = 32-Pin Thin Small Outline Package (TSOP)
Standard Pinout (TS 032)
F = 32-Pin Thin Small Outline Package (TSOP)
Reverse Pinout (TSR032)
SPEED OPTION
See Product Selector Guide and Valid Combinations
DEVICE NUMBER/DESCRIPTION
Am28F020A
2 Megabit (256 K x 8-Bit) CMOS Flash Memory with Embedded Algorithms
Valid Combinations
Valid Combinations
AM28F020A-70
AM28F020A-90
AM28F020A-120
AM28F020A-150
PC, PI, PE,
JC, JI, JE,
EC, EI, EE,
FC, FI, FE
Valid Combinations list configurations planned to be supported in volume for this device. Consult the local AMD sales office to confirm availability of specific valid combinations and
to check on newly released combinations.
AM28F020A-200
6
Am28F020A
PIN DESCRIPTION
A0–A17
VPP
Address Inputs for memory locations. Internal latches
hold addresses during write cycles.
Program voltage input. VPP must be at high voltage in
order to write to the command register. The command
register controls all functions required to alter the memory array contents. Memory contents cannot be altered
when VPP ≤ VCC +2 V.
CE# (E#)
Chip Enable active low input activates the chip’s control
logic and input buffers. Chip Enable high will deselect
the device and operates the chip in stand-by mode.
VCC
Power supply for device operation. (5.0 V ± 5% or 10%)
DQ0–DQ7
Data Inputs during memory write cycles. Internal
latches hold data during write cycles. Data Outputs
during memory read cycles.
NC
No Connect-corresponding pin is not connected internally to the die.
OE# (G#)
Output Enable active low input gates the outputs of
the device through the data buffers during memory
read cycles. Output Enable is high during command
sequencing and program/erase operations.
VSS
Ground.
WE# (W#)
Write Enable active low input controls the write function
of the command register to the memory array. The target address is latched on the falling edge of the Write
Enable pulse and the appropriate data is latched on the
rising edge of the pulse. Write Enable high inhibits writing to the device.
Am28F020A
7
BASIC PRINCIPLES
The Am28FxxxA family uses 100% TTL-level control
inputs to manage the command register. Erase and
reprogramming operations use a fixed 12.0 V ± 5%
high voltage input.
Read Only Memory
Without high V PP voltage, the device functions as a
read only memory and operates like a standard
EPROM. The control inputs still manage traditional
read, standby, output disable, and Auto select modes.
Command Register
The command register is enabled only when high voltage is applied to the V PP pin. The erase and reprogramming operations are only accessed via the
register. In addition, two-cycle commands are required
for erase and reprogramming operations. The traditional read, standby, output disable, and Auto select
modes are available via the register.
The device’s command register is written using standard
microprocessor write timings. The register controls an
internal state machine that manages all device operations. For system design simplification, the device is designed to support either WE# or CE# controlled writes.
During a system write cycle, addresses are latched on
the falling edge of WE# or CE# whichever occurs last.
Data is latched on the rising edge of WE# or CE# whichever occur first. To simplify the following discussion, the
WE# pin is used as the write cycle control pin throughout
the rest of this text. All setup and hold times are with respect to the WE# signal.
quires the user to only write a program setup command
and a program command. The device automatically
times the programming pulse width, provides the program verify and counts the number of sequences. A
status bit, Data# Polling, provides feedback to the user
as to the status of the programming operation.
DATA PROTECTION
The device is designed to offer protection against accidental erasure or programming caused by spurious
system level signals that may exist during power transitions. The device powers up in its read only state. Also,
with its control register architecture, alteration of the
memory contents only occurs after successful completion of specific command sequences.
The device also incorporates several features to prevent inadver tent write cycles resulting from V CC
power-up and power-down transitions or system noise.
Low VCC Write Inhibit
To avoid initiation of a write cycle during VCC power-up
and power-down, the device locks out write cycles for
V CC < VLKO (see DC characteristics section for voltages). When VCC < V LKO, the command register is disabled, all internal program/erase circuits are disabled,
and the device resets to the read mode. The device ignores all writes until V CC > VLKO. The user must ensure
that the control pins are in the correct logic state when
VCC > VLKO to prevent unintentional writes.
Write Pulse “Glitch” Protection
Noise pulses of less than 10 ns (typical) on OE#, CE#
or WE# will not initiate a write cycle.
OVERVIEW OF ERASE/PROGRAM
OPERATIONS
Logical Inhibit
Embedded Erase Algorithm
AMD now makes erasure extremely simple and reliable. The Embedded Erase algorithm requires the user
to only write an erase setup command and erase command. The device will automatically pre-program and
verify the entire array. The device automatically times
the erase pulse width, provides the erase verify and
counts the number of sequences. A status bit, Data#
Polling, provides feedback to the user as to the status
of the erase operation.
Writing is inhibited by holding any one of OE# = V IL,
CE#=V IH or WE# = V IH. To initiate a write cycle CE#
and WE# must be a logical zero while OE# is a logical
one.
Power-Up Write Inhibit
Power-up of the device with WE# = CE# = VIL and
OE# = V IH will not accept commands on the rising
edge of WE#. The internal state machine is automatically reset to the read mode on power-up.
Embedded Programming Algorithm
AMD now makes programming extremely simple and
reliable. The Embedded Programming algorithm re-
8
Am28F020A
FUNCTIONAL DESCRIPTION
Description Of User Modes
Table 1.
Am28F020A Device Bus Operations (Notes 7 and 8)
CE#
(E#)
OE#
(G#)
WE#
(W#)
VPP
(Note 1)
A0
A9
I/O
Read
VIL
VIL
X
VPPL
A0
A9
DOUT
Standby
VIH
X
X
VPPL
X
X
HIGH Z
Output Disable
VIL
VIH
VIH
VPPL
X
X
HIGH Z
Auto-select Manufacturer
Code (Note 2)
VIL
VIL
VIH
VPPL
VIL
VID
(Note 3)
CODE
(01h)
Auto-select Device
Code (Note 2)
VIL
VIL
VIH
VPPL
VIH
VID
(Note 3)
CODE
(29h)
Read
VIL
VIL
VIH
VPPH
A0
A9
DOUT
(Note 4)
Standby (Note 5)
VIH
X
X
VPPH
X
X
HIGH Z
Output Disable
VIL
VIH
VIH
VPPH
X
X
HIGH Z
Write
VIL
VIH
VIL
VPPH
A0
A9
DIN
(Note 6)
Operation
Read-Only
Read/Write
Legend:
X = Don’t care, where Don’t Care is either VIL or VIH levels. VPPL = VPP < VCC + 2 V. See DC Characteristics for voltage levels
of VPPH. 0 V < An < VCC + 2 V, (normal TTL or CMOS input levels, where n = 0 or 9).
Notes:
1. VPPL may be grounded, connected with a resistor to ground, or < VCC + 2.0 V. VPPH is the programming voltage specified for
the device. Refer to the DC characteristics. When VPP = VPPL, memory contents can be read but not written or erased.
2. Manufacturer and device codes may also be accessed via a command register write sequence. Refer to Table 2.
3. 11.5 < VID < 13.0 V. Minimum V ID rise time and fall time (between 0 and VID voltages) is 500 ns.
4. Read operation with VPP = VPPH may access array data or the Auto select codes.
5. With VPP at high voltage, the standby current is ICC + IPP (standby).
6. Refer to Table 3 for valid DIN during a write operation.
7. All inputs are Don’t Care unless otherwise stated, where Don’t Care is either VIL or VIH levels. In the Auto select mode all
addresses except A9 and A0 must be held at VIL.
8. If VCC ≤ 1.0 Volt, the voltage difference between VPP and VCC should not exceed 10.0 volts. Also, the Am28F256 has a VPP
rise time and fall time specification of 500 ns minimum.
Am28F020A
9
READ-ONLY MODE
When V PP is less than VCC + 2 V, the command register
is inactive. The device can either read array or autoselect data, or be standby mode.
Read
The device functions as a read only memory when VPP
< VCC + 2 V. The device has two control functions. Both
must be satisfied in order to output data. CE# controls
power to the device. This pin should be used for specific device selection. OE# controls the device outputs
and should be used to gate data to the output pins if a
device is selected.
Address access time tACC is equal to the delay from
stable addresses to valid output data. The chip enable
access time tCE is the delay from stable addresses and
stable CE# to valid data at the output pins. The output
enable access time is the delay from the falling edge of
OE# to valid data at the output pins (assuming the addresses have been stable at least tACC - tOE).
Output Disable
Output from the device is disabled when OE# is at a
logic high level. When disabled, output pins are in a
high impedance state.
Auto Select
Flash memories can be programmed in-system or in a
standard PROM programmer. The device may be soldered to the circuit board upon receipt of shipment and
programmed in-system. Alternatively, the device may
initially be programmed in a PROM programmer prior
to soldering the device to the board.
The Auto select mode allows the reading out of a binary
code from the device that will identify its manufacturer and
type. This mode is intended for the purpose of automatically matching the device to be programmed with its corresponding programming algorithm. This mode is
functional over the entire temperature range of the device.
Programming In A PROM Programmer
Standby Mode
The device has two standby modes. The CMOS
standby mode (CE# input held at VCC ± 0.5 V), consumes less than 100 µA of current. TTL standby mode
(CE# is held at V IH) reduces the current requirements
to less than 1 mA. When in the standby mode the outputs are in a high impedance state, independent of the
OE# input.
If the device is deselected during erasure, programming, or program/erase verification, the device will
draw active current until the operation is terminated.
To activate this mode, the programming equipment
must force VID (11.5 V to 13.0 V) on address A9. Two
identifier bytes may then be sequenced from the device
outputs by toggling address A0 from V IL to VIH. All other
address lines must be held at VIL, and VPP must be
less than or equal to VCC + 2.0 V while using this Auto
select mode. Byte 0 (A0 = V IL) represents the manufacturer code and byte 1 (A0 = VIH) the device identifier
code. For the device the two bytes are given in the table
2 of the device data sheet. All identifiers for manufacturer and device codes will exhibit odd parity with the
MSB (DQ7) defined as the parity bit.
Table 2. (Am28F020A Auto Select Code
Type
A0
Code
(HEX)
Manufacturer Code
VIL
01
Device Code
VIH
29
10
Am28F020A
ERASE, PROGRAM, AND READ MODE
When V PP is equal to 12.0 V ± 5%, the command register is active. All functions are available. That is, the
device can program, erase, read array or autoselect
data, or be standby mode.
Write Operations
High voltage must be applied to the VPP pin in order to
activate the command register. Data written to the register serves as input to the internal state machine. The
output of the state machine determines the operational
function of the device.
The command register does not occupy an addressable memory location. The register is a latch that stores
the command, along with the address and data information needed to execute the command. The register
is written by bringing WE# and CE# to V IL, while OE#
is at VIH. Addresses are latched on the falling edge of
WE#, while data is latched on the rising edge of the
WE# pulse. Standard microprocessor write timings are
used.
The device requires the OE# pin to be V IH for write operations. This condition eliminates the possibility for
bus contention during programming operations. In
order to write, OE# must be V IH, and CE# and WE#
must be V IL. If any pin is not in the correct state a write
command will not be executed.
Refer to AC Write Characteristics and the Erase/Programming Waveforms for specific timing parameters.
Command Definitions
The contents of the command register default to 00h
(Read Mode) in the absence of high voltage applied to
the V PP pin. The device operates as a read only
memory. High voltage on the V PP pin enables the
command register. Device operations are selected by
writing specific data codes into the command register.
Table 3 in the device data sheet defines these register
commands.
Read Command
Memory contents can be accessed via the read command when V PP is high. To read from the device, write
00h into the command register. Standard microprocessor read cycles access data from the memory. The device will remain in the read mode until the command
register contents are altered.
The command register defaults to 00h (read mode)
upon VPP power-up. The 00h (Read Mode) register default helps ensure that inadvertent alteration of the
memory contents does not occur during the V PP power
transition. Refer to the AC Read Characteristics and
Waveforms for the specific timing parameters.
Table 3. Am28F020A Command Definitions
First Bus Cycle
Second Bus Cycle
Operation
(Note 1)
Address
(Note 2)
Data
(Note 3)
Operation
(Note 1)
Address
Note 2)
Data
(Note 3)
Read Memory (Note 4)
Write
X
00h/FFh
Read
RA
RD
Read Auto select
Write
X
80h or 90h
Read
00h/01h
01h/29h
Embedded Erase Set-up/
Embedded Erase
Write
X
30h
Write
X
30h
Embedded Program Set-up/
Embedded Program
Write
X
10h or 50h
Write
PA
PD
Reset (Note 4)
Write
X
00h/FFh
Write
X
00h/FFh
Command
Notes:
1. Bus operations are defined in Table 1.
2. RA = Address of the memory location to be read.
PA = Address of the memory location to be programmed.
Addresses are latched on the falling edge of the WE# pulse.
X = Don’t care.
3. RD = Data read from location RA during read operation.
PD = Data to be programmed at location PA. Data latched on the rising edge of WE#.
4. Please reference Reset Command section.
Am28F020A
11
FLASH MEMORY PROGRAM/ERASE
OPERATIONS
Embedded Erase Algorithm
The automatic chip erase does not require the device
to be entirely pre-programmed prior to executing the
Embedded set-up erase command and Embedded
erase command. Upon executing the Embedded erase
command the device automatically will program and
verify the entire memory for an all zero data pattern.
The system is not required to provide any controls or
timing during these operations.
When the device is automatically verified to contain an
all zero pattern, a self-timed chip erase and verify begin. The erase and verify operation are complete when
the data on DQ7 is “1" (see Write Operation Status section) atwhich time the device returns to Read mode.
The system is not required to provide any control or
timing during these operations.
When using the Embedded Erase algorithm, the erase
automatically terminates when adequate erase margin
has been achieved for the memory array (no erase verify command is required). The margin voltages are internally generated in the same manner as when the
standard erase verify command is used.
The Embedded Erase Set-Up command is a command
only operation that stages the device for automatic
electrical erasure of all bytes in the array. Embedded
Erase Setup is performed by writing 30h to the command register.
To commence automatic chip erase, the command 30h
must be written again to the command register. The automatic erase begins on the rising edge of the WE and
terminates when the data on DQ7 is “1" (see Write Operation Status section) at which time the device returns
to Read mode.
Figure 1 and Table 4 illustrate the Embedded Erase algorithm, a typical command string and bus operation.
START
Apply VPPH
Write Embedded Erase Setup Command
Write Embedded Erase Command
Data# Poll from Device
Erasure Completed
17502D-6
Figure 1.
Embedded Erase Algorithm
Table 4. Embedded Erase Algorithm
Bus Operations
Command
Comments
Wait for VPP Ramp to VPPH (see Note)
Standby
Embedded Erase Setup Command
Data = 30h
Embedded Erase Command
Data = 30h
Write
Read
Data# Polling to Verify Erasure
Standby
Compare Output to FFh
Read
Available for Read Operations
Note: See AC and DC Characteristics for values of VPP parameters. The VPP power supply can be hard-wired to the device or
switchable. When VPP is switched, VPPL may be ground, no connect with a resistor tied to ground, or less than VCC + 2.0 V. Refer
to Functional Description.
12
Am28F020A
Embedded Programming Algorithm
of WE# also begins the programming operation. The
system is not required to provide further controls or
timings. The device will automatically provide an adequate internally generated program pulse and verify
margin. The automatic programming operation is
completed when the data on DQ7 is equivalent to data
written to this bit (see Write Operation Status section)
at which time the device returns to Read mode.
The Embedded Program Setup is a command only operation that stages the device for automatic programming. Embedded Program Setup is performed by
writing 10h or 50h to the command register.
Once the Embedded Setup Program operation is performed, the next WE# pulse causes a transition to an
active programming operation. Addresses are latched
on the falling edge of CE# or WE# pulse, whichever
happens later. Data is latched on the rising edge of
WE# or CE#, whichever happens first. The rising edge
Figure 2 and Table 5 illustrate the Embedded Program
algorithm, a typical command string, and bus operation.
START
Apply V PPH
Write Embedded Setup Program Command
Write Embedded Program Command (A/D)
Data# Poll Device
No
Increment Address
Last Address
Yes
Programming Completed
17502D-7
Figure 2. Embedded Programming Algorithm
Table 5. Embedded Programming Algorithm
Bus Operations
Command
Comments
Wait for VPP Ramp to VPPH (see Note)
Standby
Write
Embedded Program Setup Command
Data = 10h or 50h
Write
Embedded Program Command
Valid Address/Data
Read
Data# Polling to Verify Completion
Read
Available for Read Operations
Note: See AC and DC Characteristics for values of VPP parameters. The VPP power supply can be hard-wired to the device or
switchable. When VPP is switched, VPPL may be ground, no connect with a resistor tied to ground, or less than VCC + 2.0 V. Refer
to Functional Description. Device is either powered-down, erase inhibit or program inhibit.
Am28F020A
13
Write Operation Status
Data Polling—DQ7
The device features Data# Polling as a method to indicate to the host system that the Embedded algorithms
are either in progress or completed.
While the Embedded Programming algorithm is in operation, an attempt to read the device at a valid address
will produce the complement of expected Valid data on
DQ7. Upon completion of the Embedded Program algorithm an attempt to read the device at a valid address will
produce Valid data on DQ7. The Data# Polling feature is
valid after the rising edge of the second WE# pulse of
the two write pulse sequence.
While the Embedded Erase algorithm is in operation,
DQ7 will read “0" until the erase operation is completed. Upon completion of the erase operation, the
data on DQ7 will read “1.” The Data# Polling feature is
valid after the rising edge of the second WE# pulse of
the two Write pulse sequence.
The Data# Polling feature is only active during Embedded Programming or erase algorithms.
See Figures 3 and 4 for the Data# Polling timing specifications and diagrams. Data# Polling is the standard
method to check the write operation status, however,
an alternative method is available using Toggle Bit.
START
Read Byte
(DQ0–DQ7)
Addr = VA
DQ7 = Data
?
VA = Byte address for programming
= XXXXh during chip erase
Yes
No
No
DQ5 = 1
?
Yes
Read Byte
(DQ0–DQ7)
Addr = VA
DQ7 = Data
?
Yes
No
Fail
Pass
17502D-8
Note:
DQ7 is rechecked even if DQ5 = “1” because DQ7 may change simultaneously with DQ5 or after DQ5.
Figure 3.
14
Data# Polling Algorithm
Am28F020A
tCH
CE#
tDF
tOE
OE#
tOEH
WE#
tCE
tOH
*
DQ7
DQ7#
DQ7 =
Valid Data
High Z
tWHWH 3 or 4
DQ0–DQ6
DQ0–DQ6 = Invalid
DQ0–DQ7
Valid Data
17502D-9
*DQ7 = Valid Data (The device has completed the Embedded operation.)
Figure 4.
AC Waveforms for Data# Polling during Embedded Algorithm Operations
Am28F020A
15
Toggle Bit—DQ6
The device also features a “Toggle Bit” as a method to
indicate to the host system that the Embedded algorithms are either in progress or completed.
Successive attempts to read data from the device at a
valid address, while the Embedded Program algorithm
is in progress, or at any address while the Embedded
Erase algorithm is in progress, will result in DQ6 toggling between one and zero. Once the Embedded Program or Erase algorithm is completed, DQ6 will stop
toggling to indicate the completion of either Embedded
operation. Only on the next read cycle will valid data be
obtained. The toggle bit is valid after the rising edge of
the first WE# pulse of the two write pulse sequence, unlike Data# Polling which is valid after the rising edge of
the second WE# pulse. This feature allows the user to
determine if the device is partially through the two write
pulse sequence.
See Figures 5 and 6 for the Toggle Bit timing specifications and diagrams.
START
VA = Byte address for programming
= XXXXh during chip erase
Read Byte
(DQ0–DQ7)
Addr = VA
DQ6 = Toggle
?
No
Yes
No
DQ5 = 1
?
Yes
Read Byte
(DQ0–DQ7)
Addr = VA
DQ6 = Toggle
?
No
Yes
Fail
Pass
17502D-10
Note:
DQ6 is rechecked even if DQ5 = “1” because DQ6 may stop toggling at the same time as DQ5 changing to “1”.
Figure 5.
16
Toggle Bit Algorithm
Am28F020A
CE#
tOEH
WE#
OE#
*
Data
DQ0–DQ7
DQ6 =
DQ6
Stop Toggling
DQ6 =
DQ0–DQ7
Valid
tOE
17502D-11
Note:
*DQ6 stops toggling (The device has completed the Embedded operation.)
Figure 6. AC Waveforms for Toggle Bit during Embedded Algorithm Operations
DQ5
Power-Up/Power-Down Sequence
Exceeded Timing Limits
The device powers-up in the Read only mode. Power
supply sequencing is not required. Note that if VCC ≤
1.0 Volt, the voltage difference between V PP and VCC
should not exceed 10.0 Volts. Also, the device has a
rise VPP rise time and fall time specification of 500 ns
minimum.
DQ5 will indicate if the program or erase time has
exceeded the specified limits. This is a failure condition and the device may not be used again (internal
pulse count exceeded). Under these conditions DQ5
will produce a “1.” The program or erase cycle was not
successfully completed. Data# Polling is the only operating function of the device under this condition. The
CE# circuit will partially power down the device under
these conditions (to approximately 2 mA). The OE#
and WE# pins will control the output disable functions
as described in the Command Definitions table in the
corresponding device data sheet.
Parallel Device Erasure
The Embedded Erase algorithm greatly simplifies parallel device erasure. Since the erase process is internal
to the device, a single erase command can be given to
multiple devices concurrently. By implementing a parallel erase algorithm, total erase time may be minimized.
Note that the Flash memories may erase at different
rates. If this is the case, when a device is completely
erased, use a masking code to prevent further erasure
(over-erasure). The other devices will continue to erase
until verified. The masking code applied could be the
read command (00h).
Reset Command
The Reset command initializes the Flash memory device to the Read mode. In addition, it also provides the
user with a safe method to abort any device operation
(including program or erase).
The Reset must be written two consecutive times after
the Setup Program command (10h or 50h). This will
reset the device to the Read mode.
Following any other Flash command, write the Reset
command once to the device. This will safely abort any
previous operation and initialize the device to the Read
mode.
The Setup Program command (10h or 50h) is the only
command that requires a two-sequence reset cycle. The
first Reset command is interpreted as program data.
However, FFh data is considered as null data during programming operations (memory cells are only programmed from a logical “1" to “0"). The second Reset
command safely aborts the programming operation and
resets the device to the Read mode.
Memory contents are not altered in any case.
Am28F020A
17
This detailed information is for your reference. It may
prove easier to always issue the Reset command two
consecutive times. This eliminates the need to determine if you are in the Setup Program state or not.
In-System Programming Considerations
Flash memories can be programmed in-system or in a
standard PROM programmer. The device may be soldered to the circuit board upon receipt of shipment and
programmed in-system. Alternatively, the device may
initially be programmed in a PROM programmer prior
to soldering the device to the circuit board.
Auto Select Command
AMD’s Flash memories are designed for use in applications where the local CPU alters memory contents.
In order to correctly program any Flash memories
18
in-system, manufacturer and device codes must be
accessible while the device resides in the target
system. PROM programmers typically access the signature codes by raising A9 to a high voltage. However,
multiplexing high voltage onto address lines is not a
generally desired system design practice.
The device contains an Auto Select operation to supplement traditional PROM programming methodologies.
The operation is initiated by writing 80h or 90h into the
command register. Following this command, a read
cycle address 0000h retrieves the manufacturer code of
01h (AMD). A read cycle from address 0001h returns
the device code (see the Auto Select Code table of the
corresponding device data sheet). To terminate the operation, it is necessary to write another valid command,
such as Reset (00h or FFh), into the register.
Am28F020A
ABSOLUTE MAXIMUM RATINGS
OPERATING RANGES
Storage Temperature . . . . . . . . . . . . –65°C to +125°C
Commercial (C) Devices
Ambient Temperature
with Power Applied. . . . . . . . . . . . . . –55°C to +125°C
Ambient Temperature (TA). . . . . . . . . . . .0°C to +70°C
Voltage with Respect to Ground
All pins except A9 and VPP (Note 1) .–2.0 V to +7.0 V
Ambient Temperature (TA). . . . . . . . . .–40°C to +85°C
VCC (Note 1). . . . . . . . . . . . . . . . . . . .–2.0 V to +7.0 V
A9, VPP (Note 2) . . . . . . . . . . . . . . . –2.0 V to +14.0 V
Output Short Circuit Current (Note 3) . . . . . . 200 mA
Notes:
1. Minimum DC voltage on input or I/O pins is –0.5 V. During
voltage transitions, input or I/O pins may overshoot VSS to
–2.0 V for periods of up to 20 ns. Maximum DC voltage on
input or I/O pins is VCC +0.5 V. During voltage transitions,
input or I/O pins may overshoot to VCC +2.0 V for periods
up to 20 ns.
Industrial (I) Devices
Extended (E) Devices
Ambient Temperature (TA). . . . . . . . .–55°C to +125°C
VCC Supply Voltages
VCC . . . . . . . . . . . . . . . . . . . . . . . . +4.50 V to +5.50 V
VPP Voltages
Read . . . . . . . . . . . . . . . . . . . . . . . . –0.5 V to +12.6 V
Program, Erase, and Verify . . . . . . +11.4 V to +12.6 V
Operating ranges define those limits between which the
functionality of the device is guaranteed.
2. Minimum DC input voltage on pins A9 and VPP is –0.5 V.
During voltage transitions, A9 and VPP may overshoot
VSS to –2.0 V for periods of up to 20 ns. Maximum DC
input voltage on pin A9 and VPP is +13.0 V, which may
overshoot to 14.0 V for periods up to 20 ns.
3. No more than one output may be shorted to ground at a
time. Duration of the short circuit should not be greater
than one second.
4. Stresses above those listed under “Absolute Maximum
Ratings” may cause permanent damage to the device. This
is a stress rating only; functional operation of the device at
these or any other conditions above those indicated in the
operational sections of this data sheet is not implied.
Exposure of the device to absolute maximum rating
conditions for extended periods may affect device reliability.
Am28F020A
19
MAXIMUM OVERSHOOT
Maximum Negative Input Overshoot
20 ns
20 ns
+0.8 V
–0.5 V
–2.0 V
20 ns
17502D-12
Maximum Positive Input Overshoot
20 ns
VCC + 2.0 V
VCC + 0.5 V
2.0 V
20 ns
20 ns
17502D-13
Maximum VPP Overshoot
20 ns
14.0 V
13.5 V
VCC + 0.5 V
20 ns
20 ns
17502D-14
20
Am28F020A
DC CHARACTERISTICS over operating range unless otherwise specified
TTL/NMOS Compatible
Parameter
Symbol
Parameter Description
Test Conditions
Min
Typ
Max
Unit
ILI
Input Leakage Current
VCC = VCC Max,
VIN = VCC or VSS
±1.0
µA
ILO
Output Leakage Current
VCC = VCC Max,
VOUT = VCC or VSS
±1.0
µA
ICCS
VCC Standby Current
VCC = VCC Max
CE# = VIH
0.2
1.0
mA
ICC1
VCC Active Read Current
VCC = VCC Max, CE# = VIL, OE# = VIH
IOUT = 0 mA, at 6 MHz
20
30
mA
ICC2
(Note 4)
VCC Programming Current CE# = VIL
Programming in Progress
20
30
mA
ICC3
(Note 4)
VCC Erase Current
CE# = VIL
Erasure in Progress
20
30
mA
IPPS
VPP Standby Current
VPP = VPPL
±1.0
µA
IPP1
VPP Read Current
VPP = VPPH
70
200
±1.0
VPP = VPPL
µA
IPP2
(Note 4)
VPP Programming Current VPP = VPPH
Programming in Progress
10
30
mA
IPP3
(Note 4)
VPP Erase Current
10
30
mA
VIL
Input Low Voltage
–0.5
0.8
V
VIH
Input High Voltage
2.0
VCC + 0.5
V
VOL
Output Low Voltage
IOL = 5.8 mA
VCC = VCC Min
0.45
V
VOH1
Output High Voltage
IOH = –2.5 mA
VCC = VCC Min
2.4
VID
A9 Auto Select Voltage
A9 = VID
11.5
IID
A9 Auto Select Current
A9 = VID Max
VCC = VCC Max
VPPL
VPP during Read-Only
Operations
Note: Erase/Program are inhibited when
VPP = VPPL
VPPH
VLKO
VPP = VPPH
Erasure in Progress
V
13.0
V
50
µA
0.0
VCC +2.0
V
VPP during Read/Write
Operations
11.4
12.6
V
Low VCC Lock-out Voltage
3.2
5
3.7
V
Notes:
1. Caution: the Am28F020A must not be removed from (or inserted into) a socket when V CC or VPP is applied. If VCC ≤ 1.0
Volt, the voltage difference between VPP and VCC should not exceed 10.0 Volts. Also, the Am28F020A has a VPP rise time
and fall time specification of 500 ns minimum.
2. ICC1 is tested with OE# = VIH to simulate open outputs.
3. Maximum active power usage is the sum of ICC and IPP .
4. Not 100% tested.
Am28F020A
21
DC CHARACTERISTICS
CMOS Compatible
Parameter
Symbol
Parameter Description
Test Conditions
Min
Typ
Max
Unit
ILI
Input Leakage Current
VCC = VCC Max,
VIN = VCC or VSS
±1.0
µA
ILO
Output Leakage Current
VCC = VCC Max,
VOUT = VCC or VSS
±1.0
µA
ICCS
VCC Standby Current
VCC = VCC Max
CE# = VCC + 0.5 V
15
100
mA
ICC1
VCC Active Read Current
VCC = VCC Max, CE# = VIL, OE# = VIH
IOUT = 0 mA, at 6 MHz
20
30
mA
ICC2
(Note 4)
VCC Programming Current CE# = VIL
Programming in Progress
20
30
mA
ICC3
(Note 4)
VCC Erase Current
CE# = VIL
Erasure in Progress
20
30
mA
IPPS
VPP Standby Current
VPP = VPPL
±1.0
µA
IPP1
VPP Read Current
VPP = VPPH
70
200
µA
IPP2
(Note 4)
VPP Programming Current VPP = VPPH
Programming in Progress
10
30
mA
IPP3
(Note 4)
VPP Erase Current
10
30
mA
VIL
Input Low Voltage
–0.5
0.8
V
VIH
Input High Voltage
0.7
VCC
VCC + 0.5
V
VOL
Output Low Voltage
IOL = 5.8 mA
VCC = VCC Min
0.45
V
VOH1
Output High Voltage
IOH = –2.5 mA, VCC = VCC Min
0.85
VCC
IOH = –100 µA, V CC = VCC Min
VCC
–0.4
11.5
VOH2
VPP = VPPH
Erasure in Progress
VID
A9 Auto Select Voltage
A9 = VID
IID
A9 Auto Select Current
A9 = VID Max
VCC = VCC Max
VPPL
VPP during Read-Only
Operations
Note: Erase/Program are inhibited when
VPP = VPPL
VPPH
VLKO
V
13.0
V
50
µA
0.0
VCC +2.0
V
VPP during Read/Write
Operations
11.4
12.6
V
Low VCC Lock-out Voltage
3.2
5
3.7
V
Notes:
1. Caution: the Am28F020A must not be removed from (or inserted into) a socket when V CC or VPP is applied. If VCC ≤ 1.0
Volt, the voltage difference between VPP and VCC should not exceed 10.0 Volts. Also, the Am28F020A has a VPP rise time
and fall time specification of 500 ns minimum.
2. ICC1 is tested with OE# = VIH to simulate open outputs.
3. Maximum active power usage is the sum of ICC and IPP .
4. Not 100% tested.
22
Am28F020A
30
25
20
ICC Active
in mA
15
-55 °C
0° C
25° C
70° C
125°C
10
5
0
0
1
2
3
4
5
6
7
8
9
10
11
12
Frequency in MHz
17502D-15
Figure 7.
Am28F020A - Average ICC Active vs. Frequency
VCC = 5.5 V, Addressing Pattern = Minmax
Data Pattern = Checkerboard
TEST CONDITIONS
Table 6.
5.0 V
Test Condition
2.7 kΩ
Device
Under
Test
Test Specifications
Output Load
Output Load Capacitance, CL
(including jig capacitance)
CL
-70
Input Pulse Levels
Note: Diodes are IN3064 or equivalent
Unit
1 TTL gate
30
100
≤10
Input Rise and Fall Times
6.2 kΩ
All others
pF
ns
0.0–3.0
0.45–2.4
V
Input timing measurement
reference levels
1.5
0.8, 2.0
V
Output timing measurement
reference levels
1.5
0.8, 2.0
V
17502D-16
Figure 8. Test Setup
Am28F020A
23
SWITCHING TEST WAVEFORMS
3V
2.4 V
2.0 V
2.0 V
Test Points
0.8 V
Test Points
1.5 V
1.5 V
0.8 V
0V
0.45 V
Input
Input
Output
AC Testing (all speed options except -70): Inputs are driven at
2.4 V for a logic “1” and 0.45 V for a logic “0”. Input pulse rise
and fall times are ≤10 ns.
Output
AC Testing for -70 devices: Inputs are driven at 3.0 V for a
logic “1” and 0 V for a logic “0”. Input pulse rise and fall times
are ≤10 ns.
17502D-17
SWITCHING CHARACTERISTICS over operating range, unless otherwise specified
AC Characteristics—Read-Only Operations
Parameter
Symbols
Am28F020A Speed Options
JEDEC
Standard Parameter Description
-70
-90
-120
-150
-200
Unit
tAVAV
tRC
Read Cycle Time (Note 3)
Min
70
90
120
150
200
ns
tELQV
tCE
Chip Enable Access Time
Max
70
90
120
150
200
ns
tAVQV
tACC
Address Access Time
Max
70
90
120
150
200
ns
tGLQV
tOE
Output Enable Access Time
Max
35
35
50
55
55
ns
tELQX
tLZ
Chip Enable to Output in Low Z
(2)
Min
0
0
0
0
0
ns
tEHQZ
tDF
Chip Disable to Output in High Z
(1)
Max
20
20
30
35
35
ns
tGLQX
tOLZ
Output Enable to Output in Low Z (2)
Min
0
0
0
0
0
ns
tGHQZ
tDF
Output Disable to Output in High Z (2)
Max
20
20
30
35
35
ns
tAXQX
tOH
Output Hold Time From First Address, CE#,
or OE# change (2)
Min
0
0
0
0
0
ns
VCC Set-up Time to Valid Read (2)
Min
50
50
50
50
50
ns
tVCS
Notes:
1. Guaranteed by design; not tested.
2. Not 100% tested
24
Am28F020A
AC Characteristics—Write (Erase/Program) Operations
Parameter Symbols
Am28F020A Speed Options
JEDEC
Standard
Description
-70
-90
-120
-150
-200
Unit
tAVAV
tWC
Write Cycle Time (4)
Min
70
90
120
150
200
ns
tAVWL
tAS
Address Setup Time
Min
0
0
0
0
0
ns
tWLAX
tAH
Address Hold Time
Min
45
45
50
60
75
ns
tDVWH
tDS
Data Setup Time
Min
45
45
50
50
50
ns
tWHDX
tDH
Data Hold Time
Min
10
10
10
10
10
ns
tOEH
Output Enable Hold Time for Embedded
Algorithm only
Min
10
10
10
10
10
ns
tGHWL
Read Recovery Time Before Write
Min
0
0
0
0
0
ns
tELWLE
tCSE
CE# Embedded Algorithm Setup Time
Min
20
20
20
20
20
ns
tWHEH
tCH
CE# Hold Time
Min
0
0
0
0
0
ns
tWLWH
tWP
Write Pulse Width
Min
45
45
50
60
60
ns
tWHWL
tWPH
Write Pulse Width High
Min
20
20
20
20
20
ns
tWHWH3
Embedded Program Operation (2)
Min
14
14
14
14
14
µs
tWHWH4
Embedded Erase Operation (3)
Typ
5
5
5
5
5
sec
tVPEL
VPP Setup Time to Chip Enable Low (4)
Min
100
100
100
100
100
ns
tVCS
VCC Setup Time (4)
Min
50
50
50
50
50
µs
tVPPR
VPP Rise Time (4) 90% VPPH
Min
500
500
500
500
500
ns
tVPPF
VPP Fall Time (4) 90% VPPL
Min
500
500
500
500
500
ns
tLKO
VCC < VLKO to Reset (4)
Min
100
100
100
100
100
ns
Notes:
1. Read timing characteristics during read/write operations are the same as during read-only operations. Refer to AC
Characteristics for Read Only operations.
2. Embedded Program Operation of 14 µs consists of 10 µs program pulse and 4 µs write recovery before read. This is the
minimum time for one pass through the programming algorithm.
3. Embedded Erase operation of 5 sec consists of 4 sec array pre-programming time and one sec array erase time. This is a
typical time for one embedded erase operation.
4. Not 100% tested.
Am28F020A
25
KEY TO SWITCHING WAVEFORMS
WAVEFORM
INPUTS
OUTPUTS
Steady
Changing from H to L
Changing from L to H
Don’t Care, Any Change Permitted
Changing, State Unknown
Does Not Apply
Center Line is High Impedance State (High Z)
SWITCHING WAVEFORMS
Power-up, Standby
Device and
Address Selection
Addresses
Outputs
Enabled
Data
Valid
Standby, Power-down
Addresses Stable
tAVAV (tRC)
CE# (E#)
tEHQZ
(tDF)
OE# (G#)
tWHGL
tGHQZ
(tDF)
WE# (W#)
tGLQV (tOE)
tELQV (tCE )
tGLQX (tOLZ )
tVCS
Data (DQ)
tAXQX (tOH)
tELQX (tLZ)
High Z
tVCS
Output Valid
High Z
tAVQV (tACC)
5.0 V
VCC
0V
17502D-18
Figure 9.
26
AC Waveforms for Read Operations
Am28F020A
SWITCHING WAVEFORMS
Embedded
Erase Setup
Embedded
Erase
tWC
tAS
Erase
Data# Polling
Read
Standby
Addresses
CE#
tRC
tAH
tGHWL
OE#
tWHWH3 OR 4
tWP
WE#
tWPH
tCSE
Data
30h
tVCS
tDF
tDH
tOE
30h
DQ7#
DQ7#
tDS
tOH
VCC
tCE
VPP
tVPEL
17502D-19
Note:
DQ7# is the complement of the data written to the device.
Figure 10.
AC Waveforms for Embedded Erase Operation
Am28F020A
27
SWITCHING WAVEFORMS
Embedded
Program Setup
Embedded
Program
Addresses
Data# Polling
PA
tWC
Read
PA
tRC
tAS
CE#
tAH
tGHWL
OE#
tWHWH3 OR 4
tWP
WE#
tCSE
tDH
Data
tWPH
tDF
tOE
DIN
50h
DQ7#
DQ7#
DOUT
tVCS
tOH
tDS
VCC
tCE
VPP
tVPEL
17502D-20
Notes:
1. DIN is data input to the device.
2. DQ7# is the complement of the data written to the device.
3. DOUT is the data written to the device.
Figure 11.
28
AC Waveforms for Embedded Programming Operation
Am28F020A
AC CHARACTERISTICS—WRITE (ERASE/PROGRAM) OPERATIONS
Alternate CE# Controlled Writes
Parameter Symbols
Am28F020A Speed Options
JEDEC
Standard
Description
-70
-90
-120
-150
-200
Unit
tAVAV
tWC
Write Cycle Time (4)
Min
70
90
120
150
200
ns
tAVEL
tAS
Address Setup Time
Min
0
0
0
0
0
ns
tELAX
tAH
Address Hold Time
Min
45
45
50
60
75
ns
tDVEH
tDS
Data Setup Time
Min
45
45
50
50
50
ns
tEHDX
tDH
Data Hold Time
Min
10
10
10
10
10
ns
tOEH
Output Enable Hold Time for Embedded
Algorithm only
Min
10
10
10
10
10
ns
tGHEL
Read Recovery Time Before Write
Min
0
0
0
0
0
ns
tWLEL
tWS
WE# Setup Time by CE#
Min
0
0
0
0
0
ns
tEHWK
tWH
WE# Hold Time
Min
0
0
0
0
0
ns
tELEH
tCP
Write Pulse Width
Min
65
65
70
80
80
ns
tEHEL
tCPH
Write Pulse Width High
Min
20
20
20
20
20
ns
tEHEH3
Embedded Program Operation (2)
Min
14
14
14
14
14
µs
tEHEH4
Embedded Erase Operation (3)
Typ
5
5
5
5
5
sec
tVPEL
VPP Setup Time to Chip Enable Low (4)
Min
100
100
100
100
100
ns
tVCS
VCC Setup Time (4)
Min
50
50
50
50
50
µs
tVPPR
VPP Rise Time (4) 90% VPPH
Min
500
500
500
500
500
ns
tVPPF
VPP Fall Time (4) 90% VPPL
Min
500
500
500
500
500
ns
tLKO
VCC < VLKO to Reset (4)
Min
100
100
100
100
100
ns
Notes:
1. Read timing characteristics during read/write operations are the same as during read-only operations. Refer to AC
Characteristics for Read Only operations.
2. Embedded Program Operation of 14 µs consists of 10 µs program pulse and 4 µs write recovery before read. This is the
minimum time for one pass through the programming algorithm.
3. Embedded erase operation of 5 sec consists of 4 sec array pre-programming time and one sec array erase time. This is a
typical time for one embedded erase operation.
4. Not 100% tested.
Am28F020A
29
SWITCHING WAVEFORMS
Embedded
Program Setup
Embedded
Program
Data# Polling
PA
PA
Addresses
tWC
tAS
tAH
WE#
tGHEL
OE#
tCPH
tEHEH3 OR 4
tCP
CE#
tDH
tWS
Data
50h
DIN
DQ7#
DQ7#
DOUT
tDS
VCC
VPP
tVPEL
17502D-21
Notes:
1. DIN is data input to the device.
2. DQ7# is the complement of the data written to the device.
3. DOUT is the data written to the device.
Figure 12. AC Waveforms for Embedded Programming Operation Using CE# Controlled Writes
30
Am28F020A
ERASE AND PROGRAMMING PERFORMANCE
Limits
Typ
(Note 1)
Max
(Note 2)
Unit
Chip Erase Time
1
10
sec
Excludes 00h programming prior to erasure
Chip Programming Time
4
25
sec
Excludes system-level overhead
Parameter
Min
Write/Erase Cycles
100,000
Byte Programming Time
Comments
Cycles
14
µs
96
(Note 3)
ms
Notes:
1. 25°C, 12 V VPP .
2. Maximum time specified is lower than worst case. Worst case is derived from the Embedded Algorithm internal counter which
allows for a maximum 6000 pulses for both program and erase operations. Typical worst case for program and erase is
significantly less than the actual device limit.
3. Typical worst case = 84 µs. DQ5 = “1” only after a byte takes longer than 96 ms to program.
LATCHUP CHARACTERISTICS
Parameter
Min
Max
Input Voltage with respect to VSS on all pins except I/O pins (Including A9 and VPP)
–1.0 V
13.5 V
Input Voltage with respect to VSS on all pins I/O pins
–1.0 V
VCC + 1.0 V
–100 mA
+100 mA
Current
Includes all pins except VCC. Test conditions: VCC = 5.0 V, one pin at a time.
PIN CAPACITANCE
Parameter
Symbol
Parameter Description
Test Conditions
Typ
Max
Unit
Input Capacitance
VIN = 0
8
10
pF
COUT
Output Capacitance
VOUT = 0
8
12
pF
CIN2
VPP Input Capacitance
VPP = 0
8
12
pF
CIN
Note: Sampled, not 100% tested. Test conditions TA = 25°C, f = 1.0 MHz.
DATA RETENTION
Parameter
Test Conditions
Min
Unit
150°C
10
Years
125°C
20
Years
Minimum Pattern Data Retention Time
Am28F020A
31
PHYSICAL DIMENSIONS
PD032—32-Pin Plastic DIP (measured in inches)
1.640
1.670
.600
.625
17
32
.009
.015
.530
.580
Pin 1 I.D.
.630
.700
16
.045
.065
0°
10°
.005 MIN
.140
.225
16-038-S_AG
PD 032
EC75
5-28-97 lv
SEATING PLANE
.090
.110
.120
.160
.016
.022
.015
.060
PL032—32-Pin Plastic Leaded Chip Carrier (measured in inches)
.447
.453
.485
.495
.009
.015
.585
.595
.042
.056
.125
.140
Pin 1 I.D.
.080
.095
.547
.553
SEATING
PLANE
.400
REF.
.490
.530
.013
.021
.050 REF.
.026
.032
TOP VIEW
32
SIDE VIEW
Am28F020A
16-038FPO-5
PL 032
DA79
6-28-94 ae
PHYSICAL DIMENSIONS
TS032—32-Pin Standard Thin Small Outline Package (measured in millimeters)
0.95
1.05
Pin 1 I.D.
1
7.90
8.10
0.50 BSC
0.05
0.15
18.30
18.50
19.80
20.20
0.08
0.20
0.10
0.21
1.20
MAX
0°
5°
16-038-TSOP-2
TS 032
DA95
3-25-97 lv
0.50
0.70
Am28F020A
33
PHYSICAL DIMENSIONS
TSR032—32-Pin Reversed Thin Small Outline Package (measured in millimeters)
0.95
1.05
Pin 1 I.D.
1
7.90
8.10
0.50 BSC
0.05
0.15
18.30
18.50
19.80
20.20
0.08
0.20
0.10
0.21
1.20
MAX
0°
5°
0.50
0.70
34
Am28F020A
16-038-TSOP-2
TSR032
DA95
3-25-97 lv
DATA SHEET REVISION SUMMARY FOR
AM28F020A
Revision C+1
AC Characteristics:
Read Only Operations Characteristics: Added the -70
column and test conditions.
Distinctive Characteristics:
Deleted -95 and -250 speed options.
High Performance: The fastest speed option available
is now 70 ns.
AC Characteristics:
General Description:
Paragraph 2: Changed fastest speed option to 70 ns.
Write/Erase/Program Operations, Alternate CE # Controlled Writes: Added the -70 column. Deleted -95 and
-250 speed options.
Switching Test Waveforms:
Product Selector Guide:
In the 3.0 V waveform caption, changed -95 to -70.
Added -70, deleted -95 and -250 speed options.
Revision D
Ordering Information, Standard Products:
Matched formatting to other current data sheets.
The -70 speed option is now listed in the example.
Valid Combinations: Added -70, deleted -95 and -250
combinations.
Erase, Program, and Read Mode:
Corrected the subheading to read “VPP = 12.0 V ± 5%”.
Operating Ranges:
Revision D+1
Programming In A PROM Programmer:
Deleted the paragraph “(Refer to the AUTO SELECT
paragraph in the ERASE, PROGRAM, and READ
MODE section for programming the Flash memory device in-system).”
VCC Supply Voltages: Added -70, deleted -95 and -250
speed options.
Trademarks
Copyright © 1998 Advanced Micro Devices, Inc. All rights reserved.
ExpressFlash is a trademark of Advanced Micro Devices, Inc.
AMD, the AMD logo, and combinations thereof are registered trademarks of Advanced Micro Devices, Inc.
Product names used in this publication are for identification purposes only and may be trademarks of their respective companies.
Am28F020A
35