FINAL
Publicatio n# 11560 Rev: GAmendment/+2
Issue Date: J a nua ry 1998
Am28F256
2 56 Kilobit (32 K x 8-Bit)
CMOS 12.0 Volt, Bulk Eras e Flash Memory
DISTINCTIVE CHARACTERISTICS
High performance
70 ns maximum access time
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
10,000 write/erase cycles minimum
Write and erase voltage 12.0 V ±5%
Latch-up protected to 100 mA
from –1 V to VCC +1 V
Flasherase Electrical B ulk Chip-Erase
One second typical chip-erase
Flashrite Programming
10 µs typical byte-program
0.5 second typical chip program
Command register architecture for
microprocessor/microcontro ller compatible
write interface
On-chip address and data latches
Advanced CMOS flash memory technology
Low cost single transistor memory cell
Automatic write/erase pulse stop tim er
GENERAL DESCRIPTION
The Am28F256 is a 256 K Flash memory organized as
32 Kbytes of 8 bits each. AMD’s Flash memories offer
the most cost-effective and reliable read/write non-
volatile random access memory. The Am28F256 is
pac kaged i n 32- pin PDIP, PLCC , and TSOP versions. It
is desi gned t o be repr ogram med and erased in- system
or in standard EPROM programmers. The Am28F256
is erased when shipped from the factory.
The standard Am28F256 off ers acce ss times as f ast as
70 ns, allowing operation of high-speed microproces-
sors without wait states. To eliminate bus contention,
the Am28F256 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
Am28F256 uses a command register to manage this
functionality, while maintaining a standard JEDEC
Flash Standard 32-pin pinout. The command register
allows for 100% TTL level control inputs and fixed
power supply levels during erase and programming.
AMD’s Flash technology reliably stores memory
contents even after 10,000 erase and program cycles.
The AMD cell is designed to optimize the erase and
programming mechanisms. In addition, the combina-
tion of advanced tunnel oxide processing and low
internal electric fields for erase and programming
operations produces reliable cycling. The Am28F256
uses a 12.0 V±5% VPP high voltage input to perfor m
the Flasherase and Flashrite algorithms.
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 milliamps
on address and data pins from –1 V to VCC +1 V.
The Am28F256 is byte programmable using 10 µs
programming pulses in accordance with AMD’s
Flashrite programming algorithm. The typical room
temperature programming time of the Am28F256 is a
half a second. The entire chip is bulk erased using
10 m s erase pulses according to AM D’s Flasherase
alrogithm. Typical erasure at room temperature is
accomplished in less than one second. The windowed
package and the 15-20 minutes required for EPROM
erasure using ultra-violet light are eliminated.
2 Am28F256
Commands are written to the command register using
standard microprocessor write timings. Register con-
t en ts se rve as inputs t o an internal s t at e-m ac hi ne wh ic h
controls the erase and programming circuitry. Dur ing
write cycles, the command register internally latches ad-
dress and data needed for the programming and erase
operations. For system design simplification, the
Am28F256 is designed to suppor t either WE# or CE#
controlled writes. During a system write cycle, ad-
dresses are latched on the falling edge of WE # or CE#
wh iche v er o ccur s las t. Da ta is latc hed on the ri sing edge
of WE# or CE# whiche v er occ urs first. To simplify the f ol-
lo wing 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.
AMD’s Flash technology combines years of EPROM
and EEPR OM ex perience to produce the hi ghest le vels
of quality, reliability, and cost effectiveness. The
Am28F256 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.
BLOCK DIAGRAM
PRODUCT SELECTOR GUIDE
Family Part N u mber Am28F256
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
Erase
Voltage
Switch
Command
Register Program
Voltage
Switch Chip Ena ble
Output Enable
Logic
Y-Decoder
X-Decoder
Y-Gating
262,144
Bit
Cell Matrix
11560F-1
A0–A14
OE#
CE#
WE#
VSS
VCC
To Array
DQ0–DQ7
Input/Output
Buffers
Data
Latch
VPP
Address
Latch
Low VCC
Detector
Program/Erase
Pulse Timer
State
Control
Am28F256 3
CONNECTION DIAGRAMS
VPP VCC
DQ0
A5
A12 A14
1
3
5
7
9
11
12
10
2
4
8
6
32
30
28
26
24
14
21
23
31
29
25
27
NC
A7
13
22
20
19
A6
15
16 18
17
A4
A3
A2
A1
A0
DQ1
DQ2
VSS
WE# (W#)
A13
A8
A9
A11
OE# (G#)
A10
CE# (E#)
DQ7
DQ6
DQ5
DQ4
DQ3
11560F-2
PDIP
NC
NC
Note: Pin 1 is marked for orientation.
DQ6
VPP
DQ5
DQ4
DQ3
131 30
2
3
4
5
6
7
8
9
10
11
12
13 17 18 19 2016
15
14
29
28
27
26
25
24
23
22
21
32
A7
A6
A5
A4
A3
A2
A1
A0
DQ0
A14
A13
A8
A9
A11
OE# (G#)
A10
CE# (E#)
DQ7
A12
NC
NC
VCC
WE# (W#)
NC
DQ1
DQ2
VSS
PLCC
11560F-3
4 Am28F256
CONNECTION DIAGRAMS (continued)
LOGIC SYMBOL
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
32-Pin TSOP—St a ndard Pinout
A11
A9
A8
A13
A14
NC
WE#
VCC
VPP
NC
NC
A12
A7
A6
A5
A4
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-Pin TSOP—Reverse Pinout
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
A11
A9
A8
A13
A14
NC
WE#
VCC
VPP
NC
NC
A12
A7
A6
A5
A4
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
11560G-4
15
8
DQ0
A0–A14
CE# (E#)
OE# (G#)
–DQ7
WE# (W#)
11560F-5
Am28F256 5
ORDERING INFORMATION
Standard Products
AMD standard products are available in several packages and operating ranges. The order number (V alid Combination) is formed
by a combination of:
Valid Comb inations
Valid Combinations list configurations planned to be sup-
ported in volume for this device. Consult the loca l AMD sales
office to confirm a vailability of specific v alid combi nations and
to check on newly released combinations.
DEVICE NUMBER/DESCRIPTION
Am28F256
256 Kilobit (32 K x 8-Bit) CMOS Flash Memory
AM28F256 -70 J C
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
B
Valid Combinations
AM28F256-70
PC, PI, PE,
JC, JI, JE,
EC, EI, EE,
FC, FI, FE
AM28F256-90
AM28F256-120
AM28F256-150
AM28F256-200
6 Am28F256
PIN DESCRIPTION
A0–A14
Address Inputs for memory locations. Inter nal latches
hold addresses during write cycles.
CE# (E#)
Chip Enab le activ e low i nput activates the chip’ s control
logic and input buffers. Chip Enable high will deselect
the device and operates the chip in stand-by mode.
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 lo w 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.
VCC
P o wer supply f or de vice operation. (5.0 V ± 5% or 10%)
VPP
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.
VSS
Ground
WE# (W#)
Write Enable ac tive l ow 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.
Am28F256 7
BASIC PRINCIPLES
The device uses 100% TTL-level control inputs to
manage the command register. Erase and repro-
gramming operations use a fixed 12.0 V ± 5% high
voltage input.
Read Only Memory
Without high VPP 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 volt-
age is applied to the VPP pin. The erase and repro-
gramming operations are only accessed via the
register. In addition, two-cycle commands are required
for erase and reprogramming operations. The tradi-
tional read, standby, output disable, and Auto select
modes are available via the register.
The devices command register is written using stan-
dard microprocess or write timings. The register con-
trol s an internal state machine that manages all device
operations. For system design simplification, the de-
vice is designed to support either WE# or CE# con-
trolled writes. During a system write cycle, addresses
are latched on the falling edge of WE # or CE# which-
ever occurs last. Data is latched on the rising edge of
WE# or CE# whichever occur first. To simplify the fol-
lowing 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# sig-
nal.
Over v iew of Eras e/P rogr am Ope ra ti on s
Flasherase™ Sequ ence
A multiple step command sequence is required to
erase the Flash device (a two-cycle Erase command
and repeated one cycle ver ify commands).
Note: The Flash memory array must be completely
programmed to 0’s prior to erasure. Refer to the
Flashrite™ Programming Algorithm.
1. Erase Setup: Write the Setup Erase command to
the command register.
2. Erase: Wr ite the Erase command (same as Setup
Erase command) to the command register again.
The second command ini t iates the er ase operation.
The system software routines must now time-out
the erase pulse width (10 ms) prior to issuing the
Erase-verify command. An integrated stop timer
prevents any possibility of overerasure.
3. Erase-Verify: Write the Erase-verify command to
the command register. This command terminates
the erase operation. After the erase operation,
each byte of the arra y m ust be verified. Address in-
formation must be supplied with the Erase-verify
command. This command verifies the margin and
outputs the addressed byte in order to compare the
array data with FFh data (Byte erased).
After successful data verification the Erase-verify
command is written again with new address infor-
mation. E ach byte of the a rray is sequentially veri-
fied in this manner.
If data of the addressed location is not verified, the
Erase sequence is repeated until the entire array is
successfully verified or the sequence is repeated
1000 times.
Flashrite Programming Sequence
A three step c ommand sequence (a two-cy cle Progr am
command and one cycle Ver ify command) is required
to prog ram a b yte of the Flash arra y. Ref er to the Flash-
rite Algorithm.
1. Program Setup: Write the Setup Program com-
mand to the command register.
2. Program: Wri te the Prog ram c ommand to the com-
mand register with the appropriate Address and
Data. The sys tem softwar e routines m ust no w time-
out the program pulse width (10 µs) prior to issuing
the Program-verify command. An integrated stop
timer prevents any possibility of overprogramming.
3. Program-Verify: Write the Program-verify com-
mand to the command register. This command ter-
minates the programming operation. In addition,
this command verifies the margin and outputs the
by te just pr ogr ammed in or der to compare the arr a y
data with the original data programmed. After suc-
cessful data verification, the programming se-
quence is i nitiated again f or the ne xt b yte address to
be programmed.
If data is not verified successfully, the Program se-
quence is repeated until a successful comparison is
verified or the sequence is repeated 25 times.
Data Protection
The device i s designed to off er protection against acci-
dental erasure or programming caused by spurious
system le v el signals that ma y e xist during power transi-
tions . The device power s up i n its read only s tate. A lso,
with its control register architecture, alteration of the
memory contents only occurs after successful comple-
tion of specific command sequences.
The device also incorporates several features to pre-
v ent inadv ertent write cycles result ing fromVCC po wer-
up and power-down transitions or system noise.
Lo w VCC Write Inhibit
To avoid initiation of a wr ite cycle during VCC power-up
and power -down, the device locks out write cycles for
8 Am28F256
VCC < VLKO (see DC Characteristics section for
voltages). When VCC < VLKO, 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 VCC > VLKO. The user
must ensure that the control pins are in the correct logic
state w hen VCC > VLKO to pr event uni nitentional writes.
Write Pulse “Glitch” Protection
Noise pulses of less than 10 ns (typical) on OE# , CE#
or WE# will not initiate a write cycle.
Logical Inhibit
Writing i s inhibite d by holding any one of OE# = VIL, CE#
= VIH or WE# = VIH. To initiate a write cycle CE# and
WE# must be a logical zero whil e OE# is a logi cal one.
Power-Up Write Inhibit
Power-up of the device with WE# = CE# = VIL and
OE# = VIH will not accept commands on the rising
edge of W E# . T he in ter n al st ate ma chin e is a utoma t-
ically reset to the read mode on power-up.
FUNCTIONAL DESCRIPTION
Description Of User Modes
Table 1. Am28F256 Device Bus Operations (Notes 7 an d 8)
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, con nect e d with a resistor to gr ound, or < VCC + 2.0 V. VPPH is t he pro gramming voltag e spec if ied 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 < 1 3.0 V. Min i mum VID 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 D on’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.
Operation CE#
(E#)OE#
(G#)WE#
(W#)VPP
(Note 1) A0 A9 I/O
Read-Only
Read VIL VIL XV
PPL A0 A9 DOUT
Standby VIH XXV
PPL XXHIGH Z
Output Disable VIL VIH VIH VPPL XXHIGH Z
Auto-select Manufacturer
C ode (Note 2) VIL VIL VIH VPPL VIL VID
(Note 3) CODE
(01h)
Auto-select Device
C ode (Note 2) VIL VIL VIH VPPL VIH VID
(Note 3) CODE
(A1h)
Read/Write
Read VIL VIL VIH VPPH A0 A9 DOUT
(Note 4)
Standby (Note 5) VIH XXV
PPH XXHIGH Z
Output Disable VIL VIH VIH VPPH XXHIGH Z
Write VIL VIH VIL VPPH A0 A9 DIN
(Note 6)
Am28F256 9
READ ONLY MODE
When VPP is less than VCC + 2 V, the c ommand register
is inactive. The device can either read array or autose-
lect data, or be standby mode.
Read
The device functi ons as a read only memory when VPP
< VCC + 2 V. The de vice has two control func tions. Both
must be satisfied in order to output data. CE # controls
power to the device. This pin should be used for spe-
cific 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 dela y from stab le addresses and
stable CE # to valid data at the output pins. The output
enable access time is the del a y from the falling edge of
OE# to vali d data at the output pins (assuming the ad-
dresses have been stable at least tACC–tOE).
Standby Mode
The device has two standby modes. The CMOS
standby m ode (CE # input held a t VCC ± 0.5 V ), con-
sumes less than 100 µA of current. TTL standb y mode
(CE# is held at VIH) reduces the current requ irement s
to less than 1m A. Whe n in the sta ndby mode the out-
puts are in a high impedance state, independent of the
OE # input.
If the device is deselected dur ing erasure, program-
ming, or program/erase verification, the device will
draw active current until the operation is terminated.
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 sol-
dered to the circuit board upon r ecei pt of shipment and
programmed in-system. Alternatively, the device may
initially be programmed in a PROM programmer pr io r
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 manufactur er
and type. This mode is intended for the purpose
of automatically matching the device to be pro-
grammed with its corresponding programming algo-
rithm. This mode is functional over the entire
temperature range of the device.
Programming In A PROM Programmer
To activate this mode, the programming equipment
must fo rce VID (11.5 V to 13.0 V) on address A9. Two
identifier bytes may then be sequenced from the de vice
outputs b y toggling addr ess A0 from VIL to VIH. All other
address lines must be held at VIL, and VPP must be
less than or equal to VCC + 2.0 V w hile using this Auto
select mode. Byte 0 (A0 = VIL) represents the man ufac-
turer code and byte 1 (A0 = V IH) the device identifier
code. For the device these two bytes ar e giv en in Tab le
2 below. All identifiers for manufacturer and device
codes will exhibit odd parity with the MSB (DQ7) de-
fined as the parity bit.
Table 2. Am28F256 Auto Select Code
Type A0 Code
(HEX)
M anufact ur er Code V IL 01
Device Code VIH A1
10 Am28F256
ERASE, PROGRAM, AND READ MODE
When VPP is equal to 12.0 V ± 5%, the command reg-
ister 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 reg-
ister serves as input to the internal state machine. The
output of the state mac hine determines the operational
function of the device.
T he co mma nd r eg is ter do es no t o cc upy an addr es s ab l e
memor y location. The register is a latch that stores the
com m an d, al on g w ith th e ad dr es s a nd data in for m atio n
needed to ex ecute the command. The register is written
by br inging WE # and CE# to VIL, while OE# is at VIH.
Ad dres ses ar e latc hed on th e f a lli ng edge of WE#, wh il e
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 VIH for write op-
erations. This condition eliminates the possibility for
bus contention during programming operations. In
order to write, OE# must be VIH, and CE# and WE#
must be VIL. If any pin is not in the correct state a write
command will not be executed.
Refer to AC Write Characteristics and the Erase/Pro-
gramming 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 VPP pin. The device operates as a read only mem-
ory. High vol tage on the VPP pin enables the command
regis ter. De vic e oper ati ons are s elected b y writing spe-
cific data codes into the command register. Table 3 de-
fines these register commands.
Read Command
Memory contents can be accessed via the read com-
mand when V PP is high. To read from the device, write
00h in to the command register. Standard microproces-
sor read cycles access data from the memory. The de-
vice 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 de-
fault helps ensure that inadvertent alteration of the
memory contents does not oc cur during the VPP po wer
transition. Refer to the AC Read Characteristics and
Waveforms for the specific timing parameters.
Table 3. Am28F256 Command Definitions
Notes:
1. Bus operations are defined in Table 1.
2. RA = Address of the memory location to be read.
EA = Address of the memory location to be read during erase-verify.
PA = Address of the memory location to be programmed.
X = Don t care.
Addresses are latched on the falling edge of the WE# pulse.
3. RD = Data read from location RA during read operation.
EVD = D ata read from location EA during erase-verify.
PD = Data to be programmed at location PA. Data latched on the rising edge of WE#.
PVD = D ata read from location PA during program-verify. PA is latched on the Program comm and.
4. Refer to the appropriate section for algorithms and timing diagrams.
Command (Note 4)
First Bus Cycle Second Bus Cycle
Operation
(No te 1) Address
(Note 2) Data
(Note 3) Operation
(Note 1) Address
(Note 2) Data
( Note 3)
Read Memory Write X 00h/FFh Read RA RD
Read Auto select Write X 80h or 90h Read 00h/01h 01h/A1h
Erase Set-up/Erase Write Write X 20h Write X 20h
Erase-Verify Write EA A0h Read X EVD
Program Setup/Program Write X 40h Write PA PD
Program-Verify Write X C0h Read X PVD
Reset Write X FFh Write X FFh
Am28F256 11
FLASHERASE E RASE SEQ UENCE
Erase Setup
Eras e Setup is the fi rst of a two-cycle erase command.
It is a command-only operation that stages the device
for bulk chip erase. The array contents are not altered
with this command. 20h i s written to the command reg-
ister in order to perform the Erase Setup operation.
Erase
The second two-cycle erase command initiates the
bulk erase operation. You must write the Erase com-
mand (20h) again to the register. The erase operation
begins with the rising edge of the WE# pulse. The
erase operation must be terminated by writing a new
command (Erase-verify) to the register.
This two step sequence of the Setup and Erase com-
mands helps to ensure that memory contents are not
accidentally erased. Also, chip erasure can only occur
when hi gh v ol tage is applied to the VPP pi n and al l c on-
trol pins are in their proper state . In absence of this high
voltage, memory contents cannot be altered. Refer to
AC Erase Characteristics and Waveforms for specific
timing parameters.
Note: The Flash memory device must be fully
programmed to 00h data prior to erasure. This
equalizes the charge on all memory cells ensuring
reliable erasure.
Erase-Verify Command
The erase operation erases all bytes of the array
in parallel. A fter the erase operation, all bytes must be
sequentially v erified. The E rase- verify operation is initi -
ated by w riting A0h to the regis ter. The byte address to
be verified mus t be supplied with the command. Ad-
dresses are latched on the falling edge of the WE#
pulse or C E# pulse, whichever occurs later. The rising
edge of the WE# pulse terminates the erase oper ation.
Margin Verif y
During the Erase-verify operation, the device applies
an internally generated margin voltage to the
address ed b yte . Reading FFh from the addressed b yte
indicates that all bits in the byte are properly erased.
Verify Next Address
You must w rite the Erase-verify command with the ap-
propriate address to the register pr ior to verification of
each address . Each ne w address is latched on the fall-
ing edge of WE# or CE# pulse, whi che ver occurs later .
The process continues for each byte in the memory
array until a byte does not return FFh data or all the
bytes in the array are accessed and verified.
If an address is not v erified to FFh data, the entire chip
is erased again (refer to Erase Setup/Erase). Erase
verification then resumes at the address that failed to
verify. Erase is complete when all bytes in the array
have been verified. The device is now ready to be pro-
gr ammed. At this point, the verification operation is ter-
minated by writing a valid command (e.g. Program
Setup) to the command register. Figure 1 and Table 4,
the Flasherase electrical eras e algorithm, illust rate how
commands and bus operations are combined to per-
form electrical eras ure . Refer to A C Er ase Char acteris-
tics and Waveforms for specific timing parameters.
12 Am28F256
Figure 1. Flasherase Electrical Erase Algorit hm
Start
Program All Bytes to 00h
Appl y VPPH
Address = 00h
PLSCNT = 0
Write Erase Setup Command
Write Erase Command
Time out 10 ms
Write Erase Verify
Time out 6 µs
Read Data from Device
Data = FFh
Last Addre ss
Write Reset Command
Apply VPPL
Erasure Completed
PLSCNT =
1000
Increment Address
Apply V PPL
Erase Error
No
Yes
No
11559G-6
Yes
Yes Yes
No
No
Increment
PLSCNT
Data = 00h
Am28F256 13
FLASHERASE ELECTRICAL ERASE ALGORITHM
This Flash memory device erases the entire array in
parallel. The erase time depends on VPP
, temperature,
and number of erase/program cycles on the device. In
general, reprogramming time increases as the number
of erase/program cycles increases.
The Flasherase electrical erase algorithm employs an
interactive closed loop flow to simultaneously erase all
bits in the arra y . Erasure begins with a read of the mem-
ory contents. The device is erased when shipped from
the factory. Reading FFh data from the device would
immediately be f ollo wed by ex ecuting the Flashrite pro-
gramming algorithm with the appropriate data pattern.
Should the dev ice be currentl y programmed, data other
than FFh will be returned from address locations.
Follow the Flasherase algorithm. Uniform and reliable
erasure is ensured by first programming all bits in the
device to their charged state (Data = 00h). This is
accomplished using the Flashrite Programming
algorithm. Erasure then continues with an initial erase
operation. Erase verification (Data = FFh) begins at
address 0000h and continues through the array to the
last address, or until data other than FFh is
encountered. If a byte fails to verify, the device is
erased again. With each erase operation, an
increasing number of bytes verify to the erased state.
Typically, devices are erased in less than 100 pulses
(one second). Erase efficiency may be improved by
storing the address of the last byte that fails to verify in
a register. Following the next erase operation,
ver ification may star t at the stored address location. A
total of 1000 erase pulses are allowed per reprogram
cycle, which corresponds to appr o ximatel y 10 seconds
of cum ulative er ase ti me. The entire sequence of er ase
and byte verification is performed with high voltage
applied to the VPP pin. Figure 1 i llustrates the elect rical
erase algorithm.
Table 4. Flasherase Electrical Erase Algorithm
Notes:
1. 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.
2. Era se V erify is performed only after chip erasure. A final read compare may be perf ormed (optional) after the register is written
with the read command.
3. The erase algorithm Must Be Followed to ensure proper and reliable operation of the device.
Bus Operations Command Comments
Entire memory must = 00h before erasure (Note 3)
Note: Use Flashrite programming algorithm (Figure 3) for
programming.
Standby
Wait for VPP Ra mp to VPPH (Note 1)
Initialize:
Addresses
PLSCNT (Pulse count)
Write Erase Setup Data = 20h
Erase Data = 20h
Standby Duration of Erase Operation (tWHWH2)
Write Erase-Verify (Note 2) Address = Byte to Verify
Data = A0h
Stops Erase Operation
Standby Write Recovery Time before Read = 6 µs
Read Read byte to ver ify erasure
Standby Compare output to FFh
Increm ent pulse count
Write Reset Data = FFh, reset the register for read operations
Standby Wait for VPP Ramp to VPPL (Note 1)
14 Am28F256
Figure 2. AC Waveforms For Erase Operat ions
ANALYSIS OF ERASE TIMING WAVEFORM
Note: This analysis does not include the requirement
to progr am the entire arra y to 00h data prior to er asure.
Refer to the Flashrite
Programming algorithm.
Erase Setup /Erase
This analysis illustrates the use of two-cycle erase
commands (section A and B). The first erase com-
mand (20h) is a Setup command and does not affect
the array data (section A). The second erase com-
mand (20h) initiates the erase operation (section B)
on th e r isi ng ed ge o f thi s W E # pu lse. A ll bytes of t he
memory array are erased in parallel. No address infor-
mation is required.
The erase pulse occurs in section C.
Time-Out
A software timing routine (10 ms duration) must be ini-
tiated on the rising edge of the WE# pulse of section B.
Note: An integrated stop timer prev ents any possibil-
ity of overerasure by limiting each time-out period of
10 ms.
Erase-Verify
Upon completion of the erase software timing routine,
the microprocessor must write the Erase-verify com-
mand (A0h). This c ommand terminates the eras e oper-
ation on the rising edge of the WE # pulse (section D).
The Erase-verify command also stages the device for
data verification (section F).
After each erase operation each byte must be verified.
The byte address to be ver ified must be supplied with
Addresses
CE#
OE#
WE#
Data
VPP
VCC
11559G-7
20h
20h
Section
A0h Data
Out
Bus Cycle Write Write Time-out Write Time-out Read Standby
Command 20h 20h N/A A0h N/A Compare
Data N/A
Function Erase
Setup Erase Erase
(10 ms) Erase-
Verify Transition
(6 µs) Erase
Verification
Proceed per
Erase
Algorithm
AB DEFCG
AB DEFCG
Am28F256 15
the Erase-verify command (section D). Addresses ar e
latched on the falling edge of the WE# pulse.
Another softw are timing routine (6 µs duration) mus t be
exec uted to allow for generation of internal voltages for
margin checking and read operation (section E).
During Eras e-v erification (section F) eac h address that
returns FFh data is successfully er ased. Each address
of the arra y is sequentially v erified in this manner by re-
peating sections D thru F until the entir e array is veri-
fied or an address fails to verify. Should an address
location fail to verify to FFh data, erase the device
again. Repeat sections A thru F. Resume verification
(section D) with the failed address.
Each data change sequence allow s the device to use
up to 1,000 eras e pulses to completel y erase . Typic ally
100 erase pulses are required.
Note: All address locations must be programmed to
00h prior to erase. This equalizes the charge on all
memory cells and ensures reliable erasure.
FLASHRITE PROGRAMMING SEQUENCE
Program Setup
The device is programmed byte by byte. Bytes may be
programmed sequentially or at random. Program Setup
is the first of a two-cycle program command. It stages
the device for byte programming. The Program Setup
operation is perfor med by writing 40h to the command
register.
Program
Only after the program Setup operation is completed
will the next WE# pulse initiate the active programming
operation. The appropriate address and data for pro-
g ramming must be av ailab le on the s econd WE# pulse.
Addresses and data are internall y latched on the f alling
and rising edge of the WE# pulse respectiv ely. The ris-
ing edge of WE # also begins the programming opera-
tion. You must write the Program-verify command to
terminate the programming operation. This two step
sequence of the Setup and Program commands helps
to ensure that memory contents are not accidentally
written. Also, programming can only occur when high
v oltage is applied to the VPP pin and all control pins are
in their proper state. In absence of this high voltage,
memory contents cannot be programmed.
Ref er to AC Characteristi cs and W av ef orms for specific
timing parameters.
Program Verify Command
Following each programming operation, the byte just
programmed must be verified.
Write C0h into the command r egister i n or der to initiate
the Program-verify operation. The rising edge of this
WE pulse ter minates the programming operation. The
Program-verify operation stages the de vice for verifica-
tion of the last byte programmed. Addresses were pre-
viously latched. No new information is required.
Margin Verif y
During the Program-verify operation, the device applies
an internally generated margin voltage to the ad-
dress ed b yte . A normal mi croprocessor read c ycle out-
puts the data. A successful comparison between the
programmed byte and the true data indicates that the
byte was successfully programmed. The original pro-
grammed data should be stored for comparison. Pro-
gramming then proceeds to the next desired byte
location. Should the byte fail to verify, reprogram (refer
to Program Setup/Program). Figure 3 and Table 5 indi-
cate how instructions are combined with the bus oper-
ations to perform byte programming. Refer to AC
Programming Characteristics and Waveforms for spe-
cific timing parameters.
Flashrite Programming Algorithm
The device Flashrite P rogram ming algorithm employs
an interactive closed loop flow to program data byte by
b y te. Bytes ma y be pr ogrammed sequentiall y or at ran-
dom. The Flashrite Programming algorithm uses 10 µs
programming pulses. Each operation is followed by a
by te v eri fication to determi ne when the addr essed b yte
has been successfully programmed. The program al-
gorithm al lows f or up to 25 progr amming oper ations per
byte per r eprogramming cycle. Most bytes verify after
the first or second pulse. The entire sequence of pr o-
gramming and byte ver ification is performed w ith high
voltage applied to the VPP pin. Figure 3 and Table 5 il-
lustrate the programming algorithm.
16 Am28F256
Figure 3. Flashrite Programming Algorithm
Start
Apply VPPH
PLSCNT = 0
Write Program Setup Com mand
Write Pr og ram C ommand (A/D)
Tim e out 10 µs
Write Prog ra m Verif y Command
Time out 6 µs
Read Data from Device
Last Address
Write Reset Command
Apply V PPL
Programming Completed
PLSCNT =
25?
Increment Address
Apply VPPL
De vi ce Failed
No
11559G-8
Yes
Yes
No No
Verify Byte Incre ment PLSCNT
Yes
Am28F256 17
Table 5. Flashrite Programming Algorithm
Notes:
1. 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.
2. Pro gr am Verif y is pe rf ormed only after b y te prog rammin g. A fi nal r ead/c ompare may be pe rformed (opti onal) af ter th e r egister
is w ritten with the read comm and.
Bus Operations Command Comments
Standby Wait for VPP Ramp to VPPH (Note 1)
Initialize Pulse counter
Write Program Setup D ata = 40 h
Program Valid Address/Data
Standby Duration of Programming O peration (tWHWH1)
Write Program-Verify (Note 2) Data = C0h Stops Program Operation
Standby Write Recover y Time before Read = 6 µs
Read Read Byte to Verify Programming
Standby Compa re Data Output to Data Expected
Write Reset Data = FFh, resets the register for read operations.
Standby Wait for VPP Ramp to VPPL (Note 1)
18 Am28F256
Figure 4. AC Waveforms for Programming Operations
ANALYSIS OF PROGRAM TIMING WAVEFORMS
Program Setup/Program
Two-cycle write commands are required for program
operations (section A and B). The first program com-
mand (40h) is a Setup command and does not affect
the array data (section A). The second program com-
mand latches address and data required for program-
ming on the falling and rising edge of WE# respectiv ely
(s ection B). The rising edge of this WE# pulse (section
B) also initiates the pr ogramming pulse. The device is
prog rammed on a by te by b yte basis either sequenti ally
or randomly.
The program pulse occurs in section C.
Time-Out
A softwar e timing routine (10 µs dur ati on) must be initi-
ated on the rising edge of the WE# pulse of section B.
Note: An integrated stop timer prev ents any possibi lity
of overprogramming by limiting each time-out period of
10 µs.
Program-Verify
Upon completion of the progr am timing routine , the mi-
croprocessor must write the program-verify command
(C0h) . This command terminates the progr ammi ng op-
er ation on the rising edge of the WE# pulse (secti on D).
The program-verify command also stages the device
f or data verification (section F). Another software timing
routine (6 µs duration) must be executed to allow for
Addresses
CE#
OE#
WE#
Data
VPP
VCC
11559G-9
Data
In
20h
Section
A0h Data
Out
Bus Cycle Write Write Time-out Write Time-out Read Standby
Command 40h Program
Address,
Program Data N/A C0h
(Stops
Program) N/A Compare
Data N/A
Function Program
Setup
Program
Command
Latch
Address and
Data
Program
(10 µs) Program
Verify Transition
(6 µs) Program
Verification
Proceed per
Programming
Algorithm
AB DEFCG
A
B
DE FCG
Am28F256 19
gener ation of internal v ol t ages for margin chec k ing and
read operations (section E).
During program-verification (section F) each byte just
programmed is read to compare array data with original
prog ram data. When succes sfully veri fied, the next de-
sir ed address is programmed. Should a b yte f ail to v er -
ify, reprogram the byte (repeat section A thru F). E ach
data change sequence allows the device to use up to
25 program pulses per byte . Typical ly, bytes ar e v erified
within one or two pulses.
Algorithm Timing Delays
There are four different timing delays associated with
the Flasherase and Flashrite algorithms:
1. The first delay is associated with the VPP ri se-time
when VPP first turns on. The capacitors on the VPP
bus cause an RC ramp. A fter switching on the V PP
,
the delay required is propor tional to the number of
devices being erased and the 0.1 mF/device. VPP
must reach its final value 100 ns before commands
are executed.
2. The second dela y time is the erase time pulse width
(10 ms). A software timing routine should be run by
the local microprocessor to time out the delay. The
erase operation must be terminated at the conclu-
sion of the timing routine or prior to executing any
system interrupts that may occur during the erase
operation. To ensure proper device operation, write
the Erase-verify operation after each pulse.
3. A third delay time is required for each programming
pulse width (10 ms). The programming algorithm is
interactive and verifies each byte after a program
pulse . The program oper ation must be terminated at
the conclusion of the timing routine or prior to exe-
cuting any system interrupts that may occur during
the programming operation.
4. A fourth timing delay associated with both the
Flasherase and Flashrite algorithms is the wr ite re-
cov ery time (6 ms). During this time internal circuitry
is changing voltage levels from the erase/ program
level to those used for margin ver ify and read oper-
ations . A n attempt to read the device during this pe-
riod will result in possible false data (it may appear
the device is not properly erased or programmed).
Note: Software timing routines should be written in
machine language for each of the dela ys . Code written
in machine language requires kno wledge of the appro-
priate microprocessor clock speed in order to accu-
rately time each delay.
Parallel Device Erasure
Many applications will use more than one Flash
memory device. Total erase time may be minimized by
implementing a parallel erase algorithm. Flash
memories may erase at different rates. Therefore each
device must be verified separately. When a device is
completely erased and ver ified use a masking code to
pre v ent further er asure. The other de vices will continue
to eras e until verified. The masking code applied could
be the read command ( 00h).
Power-Up/Power-Down Sequence
The device powers-up in the Read only mode. Power
supply sequencing is not required. Note that if VCC
1.0 Volt, the voltag e difference betw een V PP and V CC
should not exceed 10.0 Volts. Also , the de vice has VPP
rise time and fall time specification of 500 ns mini mum.
Reset Command
The Reset command initializes the Flash memor y de-
vice to the R ead mode. In addition, it also provides the
user with a safe method to abort any device operation
(including program or erase).
The Reset command must be w ritten two consecutive
times after the setup Program command (40h). This will
reset the device to the Read mode.
Following any other Flash command write the Reset
command once to the de vice. This will saf ely abort any
previous operation and initialize the device to the
Read mode.
The Setup Program command (40h) is the only com-
mand that requires a two sequence reset cycle. The
first Reset command is inter preted as program data.
How ever, FFh data is considered null data during pr o-
gramming operations (memory cells are only pro-
grammed from a logical “1” to “0”). The second Reset
command safely abor ts the programming operation
and resets the device to the Read mode.
Memory contents are not altered in any case.
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 deter-
mine if you are in the setup Program state or not.
Programming In-System
Flash memories can be programmed in-system or in a
standard PROM programmer. The device may be sol-
dered to the circuit board upon r ecei pt of shipment and
programmed in-system. Alternatively, the device may
initially be programmed in a PROM programmer pr io r
to soldering the device to the board.
20 Am28F256
Auto Select Command
AMD’ s Flash memories are designed for use in applica-
tions w here the local CPU alters memory contents. Ac-
cordingly, manufacturer and device codes must be
accessible while the device resides in the target sys-
tem. PROM programmers typically access the signa-
ture 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 sup-
plement traditional PROM programming methodology.
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. A read cycle from address 0001h returns the
de vice code. To terminate the operati on, it is necessary
to write another valid command, such as Reset (FFh),
into the register.
Am28F256 21
ABSOLUTE MAXIMUM RATINGS
Storage Temperature . . . . . . . . . . . . –65°C to +150°C
Plastic Packages . . . . . . . . . . . . . . . –65 °C to +125°C
Ambient Temperature
with Powe r Applied. . . . . . . . . . . . . . –55°C to + 125°C
Voltage with Respect To Ground
All pins except A 9 and VPP (N ote 1) .–2.0 V to +7.0 V
VCC (Note 1). . . . . . . . . . . . . . . . . . . .–2.0 V to +7.0 V
A9 (Note 2). . . . . . . . . . . . . . . . . . . .–2.0 V to +14.0 V
VPP (Note 2). . . . . . . . . . . . . . . . . . .–2.0 V to +14.0 V
Output Short Circuit Current (Note 3) . . . . . . 200 mA
Notes:
1. Mini mum DC voltage on input or I/ O p ins i s – 0.5 V. During
v oltage transitions, inputs may o vershoot VSS to –2.0 V for
peri ods of up to 20 ns. Maxim um DC v oltage on i nput and
I/O pins is V CC + 0.5 V. During voltage transitions, input
and I/O pins may overshoot to VCC + 2.0V for periods up
to 20n s.
2. Minimum DC input voltage on A9 and VPP pins is –0.5 V.
During voltage transitions, A9 and V PP may overshoot
VSS t o –2.0 V for periods of up to 20 ns. Maximum DC
input voltage on A9 and VPP is +13.0 V which may
overshoot to 14.0 V for periods up to 20 ns.
3. No m ore than one output s horted t o g round at a t ime . Du-
ration of the short circuit should not be greater than one
second.
Stresses above those listed under “Absolute Maximum
Ratings” may cause permanent damage to th e device. This is
a stress rating only; functional operation of the de vice at these
or any other conditions above those indicated in the opera-
tional s ections of this specifi c a tion is no t implied. Exposure o f
the device to absolute maximum rating conditions for extended
pe rio ds ma y af f e c t devi c e reliabil it y.
OPERATING RANGES
Commercial (C) Devices
Ambient Temperature (TA) . . . . . . . . . . . .0°C to +70°C
Industrial (I) Devices
Ambient Temperature (TA) . . . . . . . . . .–40°C to +85°C
Extended (E) Device s
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.
22 Am28F256
MAXIMUM O VERSHO OT
Maximum Negative Input Overshoot
Maximum Positive Input Overshoot
Maxim um V PP Overshoot
11560F-10
20 ns
20 ns
+0.8 V
–0.5 V
20 ns
–2.0 V
11560F-11
20 ns
VCC + 0.5 V
2.0 V
20 ns 20 ns
VCC + 2.0 V
11560F-12
20 ns
13.5 V
VCC + 0.5 V
20 ns 20 ns
14.0 V
Am28F256 23
DC CHARACTERISTICS over operating range unless otherw ise specified
TTL/NMOS Compatible
Notes:
1. Caution: The Am28 F256 m ust not b e remov ed fr om (or inserted into ) a sock et when VCC or VPP is applie d. 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 fa ll
tim e specification of 500 ns minimum.
2. ICC1 is tested with OE
#
= V IH to simulate open outputs.
3. Maximum active power usage is the sum of ICC and IPP
..
4. Not 100% tested.
Parameter
Symbol Parameter Description Test Conditions Min Typ Max Unit
ILI Input Leakage Current VCC = VCC Max, VIN = V CC 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 Ma x, 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 VCC Programming Current CE# = VIL
Programming in Progress (Note 4) 20 30 mA
ICC3 VCC Erase Current CE# = VIL
Erasure in Progress (Note 4) 20 30 mA
IPPS VPP Standby Current VPP = VPPL ±1.0 µA
IPP1 VPP Read Current VPP = VPPH 70 200 µA
VPP = VPPL ±1.0
IPP2 VPP Programming Current VPP = VPPH
Programming in Progress (Note 4) 10 30 mA
IPP3 VPP Erase Current VPP = VPPH
Erasure in Progress (Note 4) 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 0.4 5 V
VOH1 Output High Voltage IOH = –2.5 mA, VCC = VCC Min 2.4 V
VID A9 Auto S e le ct Voltage A9 = VID 11.5 13.0 V
IID A9 Auto S e le ct Current A9 = V ID Max , VCC = VCC Max 5 50 µA
VPPL VPP duri ng Re ad-Only
Operations
Note: Erase/Program are inhibited
when VPP = V PPL
0.0 VCC +2 .0 V
VPPH VPP during Read/W rite
Operations 11.4 12.6 V
VLKO Low VCC Lock-out Voltage 3.2 3.7 V
24 Am28F256
DC CHARACTERISTICS
CMOS Compatible
Notes:
1. Caution: The Am 28F 256 m us t not b e rem o v ed from (or inserted into) a soc k et w hen VCC or VPP i s applied. I f VCC
1.0 volt,
the voltage difference between VPP and VCC shoul d not ex ceed 10.0 v olts. Also, the Am28F256 has a VPP rise time and fall
tim e specification of 500 ns minimum.
2. ICC1 is tested with OE
#
= V IH to simulate open outputs.
3. Maximum active power usage is the sum of ICC and IPP
.
4. Not 100% tested.
Parameter
Symbol Parameter Description Test Conditions Min Typ Max Unit
ILI Input Leakage Current VCC = VCC Ma x, 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 µA
ICC1 VCC Active Read Current VCC = VCC Max, CE# = V IL , OE # = VIH
IOUT = 0 mA, at 6 MHz 20 30 mA
ICC2 VCC Programming Current CE # = VIL
Programming in Progress (N ote 4) 20 30 mA
ICC3 VCC Erase Current CE# = VIL
Erasure in Progress (Note 4) 20 30 mA
IPPS VPP Standby Current VPP = VPPL ±1.0 µA
IPP1 VPP Read Current VPP = VPPH 70 200 µA
IPP2 VPP Programming Current VPP = VPPH
Programming in Progress (N ote 4) 10 30 mA
IPP3 VPP Erase Current VPP = VPPH
Erasure in Progress (Note 4) 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 0.4 5 V
VOH1 O utput High Voltage IOH = –2.5 mA, VCC = VCC Min 0.85 VCC V
VOH2 IOH = –100 µA, VCC = V CC Min VCC –0.4
VID A9 Auto S e le ct Voltage A9 = VID 11.5 13.0 V
IID A9 Auto S e le ct Current A9 = V ID Max , VCC = VCC Max 5 50 µA
VPPL VPPL dur ing Read-Only
Operations
Note: Erase/Program are inhibited
when VPP = VPPL
0.0 VCC + 2.0 V
VPPH VPP dur ing Re ad/Write
Operations 11.4 12.6 V
VLKO Low VCC Lock-out Voltage 3.2 3.7 V
Am28F256 25
Figure 5. Am28F256—Average ICC Active vs. Frequency
VCC = 5.5 V, A ddressing Pa ttern = Minmax
Data Pattern = Checkerboard
TEST CONDITIONS Table 6. Test Specifications
ICC Active
in mA
25
20
15
10
5
0
01 23456789101112
Frequency in MHz
11560F-13
55°C
0°C
25°C
70°C
125°C
2.7 k
CL6.2 k
5.0 V
Device
Under
Test
11560G-14
Figure 6. Test Setup
Note: Diodes are IN3064 or equivalent
Test Condition -70 All others Unit
Output Load 1 TTL gate
Output Load Cap acitance, CL
(including jig capacitance) 30 100 pF
Input Rise and Fall Times 10 ns
Input Pulse Levels 0.0–3.0 0. 45–2 .4 V
Input timing measurem ent
refere nc e levels 1.5 0.8, 2.0 V
Output timing measurem ent
refere nc e levels 1.5 0.8, 2.0 V
26 Am28F256
SWITCHING TEST WAVEFORMS
SWITCHING CHARACTERISTICS over operating range unless otherwise specified
AC Characteristics—Read Only Operation
Notes:
1. Guaranteed by design not tested.
2. Not 100% tested.
Parameter Symbols
Parameter Descrip tion
Am28F 256 Speed Opti ons
UnitJEDEC Standard -70 -90 -120 -150 -200
tAVAV tRC Read Cycle Time (Note 2) Min 70 90 120 150 200 ns
tELQV tCE Chip Enable AccessTime 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
(Note 2) Min00000ns
t
EHQZ tDF Ch ip Disable to Output in High Z
(Note 1) Max2020303535ns
t
GLQX tOLZ Ou tput Enable to Output in Low Z
(Note 2) Min00000ns
t
GHQZ tDF Ou tput Disable to Output in High Z
(Note 2) Max2020303535ns
t
AXQX tOH Output Hold from first of Address,
CE#, or OE # Change (Note 2) Min00000ns
t
WHGL Write Reco v ery Time bef o re Read Min 6 6 6 6 6 µs
tVCS VCC S etup Time to Valid Read
(Note 2) Min5050505050µs
11560G-15
3 V
0 V Input Output
1.5 V 1.5 V
Test Points
AC Testing for -70 devices: Inputs are driven at 3.0 V for a
lo gic “1 ” and 0 V for a logic “0”. I nput pulse rise and fall ti me
s
are
10 ns.
2.4 V
0.45 V Input Output
Test Points
2.0 V 2. 0 V
0.8 V0.8 V
A C Testing (all speed opt ions except -70): Inputs are driven at
2. 4 V for a logic “1” and 0.45 V for a logic “0”. Input puls e rise
and fall times are
10 ns.
Am28F256 27
AC Characteristics—Write/Erase/Program Operations
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. Maximum pulse widths not required beca use the on-chip prog ram/erase stop timer will te rminate the pulse widths intern ally
on th e devi ce.
3. Chip-Enable Controlled W rites: Write operations are driven by the valid combination of Chip-Enable and Write-Enable. In
syst ems wher e Chip-Ena ble def ines t he Write Pul se Widt h (withi n a longer W rite -Enab le ti ming w av ef orm) all set up , hol d and
inactive W rite-Enable times should be measured relative to the Chip-Enable waveform.
4. Not 100% tested.
Parameter Symbols
Parameter Description
Am28F256 Speed Options
UnitJEDEC Standard -70 -90 -120 -150 -200
tAVAV tWC Write Cycle Time (Note 4) M in 70 90 120 150 200 ns
tAVWL tAS Address Set-up Time Min 0 0 0 0 0 ns
tWLAX tAH Address Hold Time Min45455060 75ns
t
DVWH tDS Data Setup Time M in 45 45 50 50 50 ns
tWHDX tDH Data Hold Tim e Min 10 10 10 1 0 10 ns
tWHGL tWR Writ e Recovery T ime
before Read Min66666µs
t
GHWL Read Recovery Time
before Write Min00000µs
t
ELWL tCS Chip Enable Set-up Time Min 0 0 0 0 0 ns
tWHEH tCH Chip E nable 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
Widt h H I GH Min2020202020ns
t
WHWH1 Duration of Programming
Operation (Note 2) Min1010101010µs
t
WHWH2 D uration of
Erase Operation (Note 2) Min 9.5 9.5 9.5 9.5 9.5 ms
tVPEL VPP Setup Time to
Chip Enable LOW (Note 4) Min 100 100 100 100 100 ns
tVCS VCC Set-up Time to
Chip Enable LOW (Note 4) Min5050505050µs
t
VPPR VPP Rise Time
90% VPPH (Note 4) Min 500 500 500 500 500 ns
tVPPF VPP Fall Time
10% VPPL (Note 4) Min 500 500 500 500 500 ns
tLKO VCC < VLKO
to Reset (Note 4) Min 100 100 100 100 100 ns
28 Am28F256
KEY TO SWITCHING WAVEFORMS
SWITCHING WAVEFORMS
Figure 7. AC Waveforms for Read Operation s
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)
Addresses
CE# (E#)
OE# (G#)
WE# (W#)
Data (DQ)
5.0 V
VCC
0 V
Power-up, Standby Devi ce and
Addr ess Selection Outputs
Enabled Data
Valid Standby, Power-Down
Addresses Stable
High Z High Z
tWHGL
tAVQV (tACC)
tEHQZ
(tDF)
tGHQZ
(tDF)
tELQX (tLZ)
tGLQX (tOLZ)
tELQV (tCE)
tGLQV (tOE)
tAXQX (tOH)
Output Valid
tAVAV (tRC)
tVCS
11560G-16
Am28F256 29
SWITCHING WAVEFORMS
Figure 8. AC Waveforms for Erase Operations
DA TA IN =
A0h VALID
DATA
OUT
Erase-Verify
Command Erase
Verification Standby,
Power-down
tWLAX (tAH)
tEHQZ (tDF)
tWHGL
tGHQZ (tDF)
tGLQX (tOLZ)
tGLQV (tOE)
tELQV (tCE)
11560F-17
tELQX (tLZ)
tAVAV (tRC)
tAXQX (tOH)
DA T A IN
= 20h DA T A IN
= 2 0h
Setup Erase
Command E ras e
Command
P o w er-up,
Standby
tAVWL (tAS)
tAVAV (tWC)
tELWL (tCS)
tGHWL (tOES)
tWHEH (tCH)
tWHWH2
tWHWL (tWPH)
tWHDX (tDH)
tWL WH (tWP)
tDVWH (tDS)
tVCS
tVPEL
Addresses
HIGH Z
CE# (E#)
OE# (G#)
WE# (W#)
Data (DQ)
5.0 V
VCC
0 V
VPPH
VPP
VPPL
Erasure
30 Am28F256
SWITCHING WAVEFORMS
Figure 9. AC Waveforms for Pro gramming Operatio ns
DA TA IN
= C0h VALID
DA T A
OUT
Verify
Command Programming
Verification Standby,
Power-down
tWLAX (tAH)
tGHQZ (tDF)
tWHGL
tGHQZ (tDF)
tGLQX (tOLZ)
tGLQV (tOE)
tELQV (t CE)
11560F-18
tELQX (tLZ)
tAVAV (tRC)
tAXQX (tOH)
DA TA IN
= 40h DATA IN
Setup Program
Command
Program
Command
Latch Address
and Data
Power-up,
Standby
tAVWL (tAS)
tAVAV (tWC)
tELWL (tCS)
tGHWL (tOES)
tWHEH (tCH)
tWHWH1
tWHWL (tWPH)
tW HDX (tDH)
tWLWH (tWP)
tDVWH (tDS)
tVCS
tVPEL
Addresses
HIGH Z
CE# (E#)
OE# (G#)
WE# (W#)
Data (DQ)
5.0 V
VCC
0 V
VPPH
VPP
VPPL
Programming
Am28F256 31
ERASE AND PROGRAMMING PERFORMANCE
Notes:
1. 25
°
C, 12 V VPP
.
2. Maximum time specified is lower than worst case. Worst case is derived from the Flasherase/Flashrite pulse count
(Flashera se = 1000 max and Flashrite = 25 max). Typical w orst case for program and erase is significantly less than the actual
dev ice limit.
LATCHUP CHARACTERISTICS
PIN CAPACITANCE
Note: Sampled, not 100% tested. Test conditions TA = 25°C, f = 1.0 MHz.
DATA RETENTIO N
Parameter
Limits
CommentsMin Typ
(Note 1) Max
(Note 2) Unit
Chip Erase Time 1 1 0 sec Excludes 00h programming prior to erasure
Chip Programming Time 0.5 3 sec Excludes system-level overhead
Write/Erase Cycles 10,000 Cycles
Min Max
Input Voltage with respect to VSS on all pins except I/O pins (Including A9 and VPP) –1.0 V 1 3.5 V
Input Voltage with respect to VSS on all pins I/O pins –1.0 V VCC + 1.0 V
Current –100 m A +100 mA
Includes all pins except VCC. Test conditions: VCC = 5.0 V, one pin at a time.
Parameter
Symbol Parameter Description Test Conditions Typ Max Unit
CIN Input Capa cit anc e VIN = 0 8 10 pF
COUT Ou tp ut C apacitanc e VOUT = 0 8 1 2 pF
CIN2 VPP Input Capacitance VPP = 0 8 12 pF
Parameter Test Conditions Min Unit
Minimum Pa tte rn Data Retention Time 150°C10Years
125°C20Years
32 Am28F256
PHY S ICAL DIMEN SION S
PD032—32-Pin P last ic DIP (measured in inches)
PL032—32-Pin Plastic Leaded Chip Carrier (measured in inches)
Pin 1 I.D.
1.640
1.670
.530
.580
.005 MIN
.045
.065
.090
.110
.140
.225
.120
.160 .016
.022
SEATING PLANE
.015
.060
16-038-S_AG
PD 032
EC75
5-28-97 lv
32 17
16 .630
.700
0°
10°
.600
.625
.009
.015
.050 REF.
.026
.032 TOP VIEW
Pin 1 I.D.
.485
.495
.447
.453
.585
.595
.547
.553
16-038FPO-5
PL 032
DA79
6-28-94 ae
SIDE VIEW
SEATING
PLANE
.125
.140
.009
.015
.080
.095
.042
.056
.013
.021
.400
REF. .490
.530
Am28F256 33
PHY S ICAL DIMEN SION S
TS032—32-Pin Standard Thin Small Outline Package (measured in millimeters)
Pin 1 I.D.
1
18.30
18.50
7.90
8.10
0.50
B
0.05
0.15
0.95
1.05
16-038-TSOP-2
TS 032
DA95
3-25-97 lv
19.80
20.20
1.20
MAX
0.50
0.70
0.10
0.21
0°
5°
0.08
0.20
34 Am28F256
PHY S ICAL DIMEN SION S
TSR032—32-Pin Reversed Thin Small Outline Package (measured in millimeters)
1
18.30
18.50
19.80
20.20
7.90
8.10
0.50 BSC
0.05
0.15
0.95
1.05
16-038-TSOP-2
TSR032
DA95
3-25-97 lv
Pin 1 I.D.
1.20
MAX
0.50
0.70
0.10
0.21
0°
5°
0.08
0.20
Am28F256 35
DATA SHEET REVISION SUMMARY FOR
AM28F256
Revision G
Deleted -75, -95, and -250 speed options. Matched f or-
matting to other current data sheets.
Revision G+1
Figure 3, Fl ashrite Programm ing Algorithm:
Mo ved end
of arrow originating from Increment Address box so
that it points to the P LSCNT = 0 bo x, not the Write Pro-
gram Verify Command box. This is a corr ection to the
diagram on page 6-189 of the 1998 Flash Memory
Data Book.
Revision G+2
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 de-
vice in-system).”
Trademarks
Copyright © 1998 Advanced Micro Devices, Inc. All rights reser ved.
ExpressFlash is a trademark of Advanced Micro Devices, Inc.
AMD, the AMD logo, and combinations thereof are registered trademarks of Advanced Mi cro Devices, Inc.
Product names used in this publication are for identification purposes only and may be trademarks of their respective companies.