KBE00D002M-F407 [SAMSUNG]
Memory Circuit, Flash+SDRAM, CMOS, PBGA137;型号: | KBE00D002M-F407 |
厂家: | SAMSUNG |
描述: | Memory Circuit, Flash+SDRAM, CMOS, PBGA137 动态存储器 内存集成电路 |
文件: | 总73页 (文件大小:1306K) |
中文: | 中文翻译 | 下载: | 下载PDF数据表文档文件 |
Advance
MCP MEMORY
Preliminary
KBE00D002M-F407
Document Title
Multi-Chip Package MEMORY
256M Bit (16Mx16) Nand Flash*2 / 128M Bit (2Mx16x4Banks) Mobile SDRAM*2
Revision No. History
Draft Date
Remark
0.0
Initial issue.
June , 25, 2003
Preliminary
- 512M NAND DDP C-Die_ ver 1.0
- 256M MSDRAM DDP E-Die_ver 0.4
0.1
<Common>
August, 22 , 2003 Preliminary
- Changed errata for odering information : page 4
406 --> 407
<NAND Flash> .... ver 2.6
- Added the new definition of the number of invalid blocks : page 10
(Minimum 1004 valid blocks are guaranteed for each contiguous 128Mb
memory space.)
- Changed value of tR & tREA : page 11
tR : 12us --> 10us
tREA : 30ns --> 35ns
<M-SDR> .... ver 0.5
- Changing DC Current of ICC3P/3PS : page 32
10mA/2mA --> 8mA/8mA
- Added the word of "Internal: : page 38
Note : For more detailed features and specifications including FAQ, please refer to Samsung’s web site.
http://samsungelectronics.com/semiconductors/products/products_index.html
The attached datasheets are prepared and approved by SAMSUNG Electronics. SAMSUNG Electronics CO., LTD. reserve the right
to change the specifications. SAMSUNG Electronics will evaluate and reply to your requests and questions about device. If you have
any questions, please contact the SAMSUNG branch office near you.
Revision 0.1
August 2003
- 1 -
Advance
MCP MEMORY
Preliminary
KBE00D002M-F407
Multi-Chip Package MEMORY
256M Bit (16Mx16) Nand Flash*2 / 128M Bit (2Mx16x4Banks) Mobile SDRAM*2
FEATURES
<Common>
<SDRAM>
· Operating Temperature : -25°C ~ 85°C
· Package : 137-ball FBGA Type - 10.5x13mm, 0.8mm pitch
· Power Supply Voltage : 1.65~1.95V
• LVCMOS compatible with multiplexed address.
• Four banks operation.
<NAND>
· Power Supply Voltage : 1.7~1.95V
· Organization
• MRS cycle with address key programs.
-. CAS latency (1, 2 & 3).
-. Burst length (1, 2, 4, 8 & Full page).
-. Burst type (Sequential & Interleave).
• EMRS cycle with address key programs.
• All inputs are sampled at the positive going edge of the system
clock.
- Memory Cell Array : (32M + 1024K)bit x 16bit
- Data Register : (256 + 8)bit x16bit
· Automatic Program and Erase
- Page Program : (256 + 8)Word
- Block Erase : (8K + 256)Word
· Page Read Operation
• Burst read single-bit write operation.
• Special Function Support.
- Page Size : (256 + 8)Word
- Random Access : 10ms(Max.)
- Serial Page Access : 50ns(Min.)
· Fast Write Cycle Time
- Program time : 200ms(Typ.)
- Block Erase Time : 2ms(Typ.)
· Command/Address/Data Multiplexed I/O Port
· Hardware Data Protection
- Program/Erase Lockout During Power Transitions
· Reliable CMOS Floating-Gate Technology
- Endurance : 100K Program/Erase Cycles
- Data Retention : 10 Years
-. PASR (Partial Array Self Refresh).
-. Internal TCSR (Temperature Compensated Self Refresh)
-. DS (Driver Strength)
• DQM for masking.
• Auto refresh.
• 64ms refresh period (4K cycle).
• Commercial Temperature Operation (-25°C ~ 70°C).
Extended Temperature Operation (-25°C ~ 85°C).
· Command Register Operation
· Intelligent Copy-Back
· Unique ID for Copyright Protection
· Safe Lock Mechanism
GENERAL DESCRIPTION
The KBE00D002M is a Multi Chip Package Memory which combines 512Mbit Nand Flash Memory(organized with two pieces of
256Mbit Nand Flash Memory) and 256Mbit synchronous high data rate Dynamic RAM(organized with two pieces of 128Mbit Mobile
SDRAM)
512Mbit NAND Flash memory is organized as 32M x16 bits and 256Mbit SDRAM is organized as 2M x32 bits x4 banks.
In 256Mbit NAND Flash, a 264-word page program can be typically achieved within 200us and an 8K-word block erase can be typi-
cally achieved within 2ms. In serial read operation, a word can be read by 50ns. IO pins serve as the ports for address and data
input/output as well as command inputs. Even the write-intensive systems can take advantage of flash¢s extended reliability of 100K
program/erase cycles with real time mapping-out algorithm. These algorithms have been implemented in many mass storage appli-
cations.
256Mbit Mobile SDRAM is 268,435,456 bits synchronous high data rate Dynamic RAM organized as 4 x 2,097,152 words by 32 bits,
fabricated with SAMSUNG’s high performance CMOS technology. Synchronous design allows precise cycle control with the use of
system clock and I/O transactions are possible on every clock cycle. Range of operating frequencies, programmable burst lengths
and programmable latencies allow the same device to be useful for a variety of high bandwidth and high performance memory sys-
tem applications.
The KBE00D002M is suitable for use in data memory of mobile communication system to reduce not only mount area but also power
consumption. This device is available in 137-ball FBGA Type.
Revision 0.1
August 2003
- 2 -
Advance
MCP MEMORY
Preliminary
KBE00D002M-F407
PIN CONFIGURATION
9
10
1
2
3
4
5
6
7
8
A
B
C
D
E
F
DNU
NC
DNU
Vdd
A4
DNU
NC
NC
Vss
A2
A0
A11
BA1
Vdd
WEd
NC
Vss
CLK
CAS
CKE
DQ10
DQ25
DQ9
DQ24
DQ8
NC
RAS
CS
A3
A7
Vdd
A1
A10
Vss
IO15
IO6
BA0
DQM1
DQM2
DQ19
DQ3
A9
A6
IO7
NC
R/B
DQM0
WEn
DQ17
DQ1
DQM3
NC
A8
A5
IO14
IO13
IO3
NC
CE
IO4
IO12
IO10
IO1
IO5
G
H
J
DQ6
DQ23
DQ7
NC
Vcc
Vss
RE
CLE
ALE
WP
IO8
Vcc
Vss
IO11
NC
DQ4
DQ13
DQ28
DQ12
DQ11
DQ26
DQ27
Vss
IO2
DQ18
DQ20
DQ0
DQ21
DQ5
DQ14
DQ30
DQ31
DQ15
DQ29
Vddq
IO9
K
L
Vssq
Vss
Vddq
Vssq
NC
DQ16
Vdd
Vddq
Vssq
Vddq
DNU
IO0
DQ22
NC
NC
Vddq
Vssq
Vssq
Vssq
DNU
Vddq
Vss
Vdd
NC
M
DQ2
Vssq
Vddq
Vdd
NC
Vssq
Vddq
Vssq
Vddq
Vssq
Vddq
N
P
NAND
DNU
DNU
R
MSDRAM
137 FBGA: Top View (Ball Down)
Revision 0.1
August 2003
- 3 -
Advance
MCP MEMORY
Preliminary
KBE00D002M-F407
PIN DESCRIPTION
Pin Name
/CE
Pin Function(NAND Flash)
Pin Name
CLK
Pin Function(Mobile SDRAM)
Chip Enable
System Clock
/RE
Read Enable
CKE
Clock Enable
/WP
Write Protection
Write Enable
/CS
Chip Select
/WEn
ALE
/RAS
Row Address Strobe
Column Address Strobe
Write Enable
Address Latch Enable
Command Latch Enable
Ready/Busy Output
Data Input/Output
Power Supply
/CAS
CLE
/WEd
R/B
A0 ~ A11
BA0 ~ BA1
DQM0~DQM3
DQ0d ~ DQ31
Vdd
Address Input
IO0 ~ IO15
Vcc
Bank Address Input
Lower Input/Output Data Mask
Data Input/Output
Power Supply
Vss
Ground
Vddq
Data Out Power
Ground
Pin Name
NC
Pin Function
Vss
No Connection
Do Not Use
Vssq
DQ Ground
DNU
ORDERING INFORMATION
D
KB E 00
0 0 2 M - F 407
Samsung
MCP Memory(4chips)
Access Time
407 : NAND Flash 50ns
NAND Flash 50ns
Mobile SDRAM 15ns
Mobile SDRAM 15ns
Device Type
NAND + NAND + SDRAM+SDRAM
NOR Flash Density, Voltage,
Organization, Bank Size, Boot Block
00 = None
Package
F = FBGA
NAND Flash Density, Voltage, Organization
D = 256M+256M, 1.8V/1.8V, X16
Version
M = 1st Generation
UtRAM Density, Voltage, Organization
0 = None
DRAM Interface, Density,
SRAM Density, Voltage, Organization
0 = None
Voltage, Organization, Option
2 = SDR, 128M+128M, 1.8V/1.8V, X32
NOTE :
1. Samsung are not designed or manufactured for use in a device or system that is used under circumstance in which human life is potentially at stake.
Please contact to the memory marketing team in samsung electronics when considering the use of a product contained herein for any specific purpose,
such as medical, aerospace, nuclear, military, vehicular or undersea repeater use.
Revision 0.1
August 2003
- 4 -
Advance
MCP MEMORY
Preliminary
KBE00D002M-F407
FUNCTIONAL BLOCK DIAGRAM
CE
X-Buffers
512M+16M Bit
NAND flash ARRAY
RE
Latches
& Decoders
WP
(256Mb NAND * 2Chip)
(256 + 8)Word x 131072
Y-Buffers
Latches
& Decoders
WEn
ALE
page Register & S/A
Y-Gating
CLE
IO0 to IO15
R/B
Vcc
Command
Register
I/O Buffers & Latches
Vccq
Vss
Control Logic
& High Voltage
Generator
Output
Driver
Global Buffers
CLK
CKE
CS
Bank Select
Data Input Register
2M x 32
2M x 32
2M x 32
2M x 32
RAS
CAS
WEd
DQ0d to DQ31d
A0~A12
BA0~BA1
Column Decoder
DQM0~DQM3
UDQM
Latency & Burst Length
Vdd
Vddq
Programming Register
Vss
Vssq
Revision 0.1
August 2003
- 5 -
Advance
MCP MEMORY
Preliminary
KBE00D002M-F407
512Mb(32Mb x 16)
NAND Flash DDP C-Die
Revision 0.1
August 2003
- 6 -
Advance
MCP MEMORY
Preliminary
KBE00D002M-F407
Figure 1.NAND Flash ARRAY ORGANIZATION
1 Block =32 Pages
= (8K + 256) Word
1 Page = 264 Word
1 Block = 264 Word x 32 Pages
= (8K + 256) Word
1 Device = 264Words x 32Pages x 4096 Blocks
= 528 Mbits
128K Pages
(=4,096 Blocks)
Page Register
(=256 Words)
16 bit
256Word
8 Word
8 Word
I/O 0 ~ I/O 15
Page Register
256 Word
I/O 0
A0
I/O 1
A1
I/O 2
A2
I/O 3
I/O 4
A4
I/O 5
A5
I/O 6
A6
I/O 7
A7
I/O8 to 15
Column Address
1st Cycle
2nd Cycle
3rd Cycle
4rd Cycle
A3
A12
A20
L*
L*
L*
L*
L*
Row Address
(Page Address)
A9
A10
A18
L*
A11
A19
L*
A13
A21
L*
A14
A22
L*
A15
A23
L*
A16
A24
L*
A17
A25
NOTE : Column Address : Starting Address of the Register.
* L must be set to "Low".
Revision 0.1
August 2003
- 7 -
Advance
MCP MEMORY
Preliminary
KBE00D002M-F407
PRODUCT INTRODUCTION
The NAND Flash is a 528Mbit(553,648,218 bit) memory organized as 131,072 rows(pages) by 264 columns. Spare eight columns
are located from column address of 256~263. A 264-word data register is connected to memory cell arrays accommodating data
transfer between the IO buffers and memory during page read and page program operations. The memory array is made up of 16
cells that are serially connected to form a NAND structure. Each of the 16 cells resides in a different page. A block consists of two
NAND structured strings. A NAND structure consists of 16 cells. Total 135168 NAND cells reside in a block. The array organization
is shown in Figure 1. The program and read operations are executed on a page basis, while the erase operation is executed on a
block basis. The memory array consists of 4096 separately erasable 8K-Word blocks. It indicates that the bit by bit erase operation
is prohibited on the NAND Flash.
The NAND Flash has addresses multiplexed into lower 8 IO’s. The NAND Flash allows sixteen bit wide data transport into and out of
page registers. This scheme dramatically reduces pin counts while providing high performance and allows systems upgrades to
future densities by maintaining consistency in system board design. Command, address and data are all written through IO¢s by
bringing WE to low while CEn is low. Data is latched on the rising edge of WE. Command Latch Enable(CLE) and Address Latch
Enable(ALE) are used to multiplex command and address respectively, via the IO pins. Some commands require one bus cycle. For
example, Reset command, Read command, Status Read command, etc require just one cycle bus. Some other commands like
Page Program and Copy-back Program and Block Erase, require two cycles: one cycle for setup and the other cycle for execution.
The 32M-word physical space requires 25 addresses, thereby requiring four cycles for word-level addressing: column address, low
row address and high row address, in that order. Page Read and Page Program need the same four address cycles following the
required command input. In Block Erase operation, however, only the three row address cycles are used. Device operations are
selected by writing specific commands into the command register. Table 1 defines the specific commands of the NAND Flash.
Table 1. COMMAND SET
Function
1st. Cycle
00h
2nd. Cycle
Acceptable Command during Busy
Read 1
Read 2
Read ID
Reset
-
50h
-
-
90h
FFh
-
O
O
Page Program
Copy-Back Program
Block Erase
80h
10h
8Ah
D0h
-
00h
60h
Read Status
70h
Revision 0.1
August 2003
- 8 -
Advance
MCP MEMORY
Preliminary
KBE00D002M-F407
ABSOLUTE MAXIMUM RATINGS
Parameter
Symbol
VIN/OUT
VCC
Rating
Unit
-0.6 to + 2.45
-0.2 to + 2.45
-0.2 to + 2.45
-40 to +125
-65 to +150
5
Voltage on any pin relative to VSS
V
VCCQ
Temperature Under Bias
Storage Temperature
Short Circuit Current
NOTE :
TBIAS
°C
°C
TSTG
Ios
mA
1. Minimum DC voltage is -0.6V on input/output pins. During transitions, this level may undershoot to -2.0V for periods <30ns.
Maximum DC voltage on input/output pins is VCC,+0.3V which, during transitions, may overshoot to VCC+2.0V for periods <20ns.
2. Permanent device damage may occur if ABSOLUTE MAXIMUM RATINGS are exceeded. Functional operation should be restricted to the conditions
as detailed in the operational sections of this data sheet. Exposure to absolute maximum rating conditions for extended periods may affect reliability.
RECOMMENDED OPERATING CONDITIONS
(Voltage reference to GND, TA=-25 to 85°C)
Parameter
Supply Voltage
Supply Voltage
Supply Voltage
Symbol
VCC
Min
1.7
1.7
0
Typ.
1.8
1.8
0
Max
1.95
1.95
0
Unit
V
VCCQ
VSS
V
V
DC AND OPERATING CHARACTERISTICS(Recommended operating conditions otherwise noted.)
Parameter
Symbol
Test Conditions
tRC=50ns, CE=VIL
IOUT=0mA
Min
Typ
Max
Unit
Sequential Read
ICC1
-
8
20
Operating
Current
Program
Erase
ICC2
ICC3
ISB1
ISB2
ILI
-
-
-
-
-
-
-
-
8
8
-
20
20
mA
Stand-by Current(TTL)
Stand-by Current(CMOS)
Input Leakage Current
Output Leakage Current
CE=VIH, WP=0V/VCC
CE=VCC-0.2, WP=0V/VCC
VIN=0 to Vcc(max)
1
10
-
50
±10
±10
mA
ILO
VOUT=0 to Vcc(max)
-
VCCQ
I/O pins
VCCQ-0.4
VCC-0.4
-
-
+0.3
Input High Voltage
VIH
VCC
Except I/O pins
+0.3
V
Input Low Voltage, All inputs
Output High Voltage Level
Output Low Voltage Level
Output Low Current(R/B)
VIL
VOH
-
IOH=-100mA
IOL=100uA
-0.3
-
-
0.4
-
VCCQ-0.1
VOL
-
-
0.1
-
IOL(R/B)
VOL=0.1V
3
4
mA
Revision 0.1
August 2003
- 9 -
Advance
MCP MEMORY
Preliminary
KBE00D002M-F407
VALID BLOCK
Parameter
Symbol
Min
Typ.
Max
Unit
Valid Block Number
NVB
4026
-
4096
Blocks
NOTE :
1. The NAND Flash may include invalid blocks when first shipped. Additional invalid blocks may develop while being used. The number of valid blocks is
presented with both cases of invalid blocks considered. Invalid blocks are defined as blocks that contain one or more bad bits. Do not erase or
program factory-marked bad blocks. Refer to the attached technical notes for a appropriate management of invalid blocks.
2. The 1st block, which is placed on 00h block address, is fully guaranteed to be a valid block, does not require Error Correction.
3. The 2nd and 3rd blocks are good upon shipping.
4. Minimum 1004 valid blocks are guaranteed for each contiguous 128Mb memory space.
AC TEST CONDITION
( Vcc=1.7V~1.95V , TA=-40 to 85°C unless otherwise noted)
Parameter
Value
0V to VccQ
Input Pulse Levels
Input Rise and Fall Times
5ns
Input and Output Timing Levels
Output Load (VccQ:1.8V +/-10%)
VccQ/2
1 TTL GATE and CL=30pF
CAPACITANCE(TA=25°C, VCC=1.8V, f=1.0MHz)
Item
Symbol
Test Condition
Min
Max
20
Unit
pF
Input/Output Capacitance
Input Capacitance
CI/O
VIL=0V
-
-
CIN
VIN=0V
20
pF
NOTE : Capacitance is periodically sampled and not 100% tested.
MODE SELECTION
CLE
H
L
ALE
L
CE
L
WE
RE
H
WP
Mode
X
Command Input
Read Mode
Write Mode
H
L
H
X
Address Input(4clock)
Command Input
H
L
L
L
H
H
H
L
H
H
Address Input(4clock)
L
L
L
H
H
Data Input
L
L
L
H
H
X
X
X
X
X
Data Output
L
L
X
X
X
X
H
H
X
X
X
X
X
During Read(Busy)
X
X
H
During Program(Busy)
During Erase(Busy)
Write Protect
X
X
H
L
X(1)
X
X
(2)
X
Stand-by
0V/VCC
NOTE :
1. X can be VIL or VIH.
2. WP should be biased to CMOS high or CMOS low for standby.
Program/Erase Characteristics
Parameter
Symbol
Min
Typ
Max
500
2
Unit
Program Time
tPROG
-
-
-
-
200
ms
Main Array
-
-
cycles
cycles
ms
Number of Partial Program Cycles
in the Same Page
Nop
Spare Array
3
Block Erase Time
tBERS
2
3
Revision 0.1
August 2003
- 10 -
Advance
MCP MEMORY
Preliminary
KBE00D002M-F407
AC Timing Characteristics for Command / Address / Data Input
Parameter
Symbol
tCLS
tCLH
tCS
Min
Max
Unit
ns
ns
ns
ns
ns
ns
ns
ns
ns
ns
ns
CLE Set-up Time
CLE Hold Time
CE Setup Time
CE Hold Time
0
-
-
10
0
.-
-
tCH
10
25 (1)
0
WE Pulse Width
ALE Setup Time
ALE Hold Time
Data Setup Time
Data Hold Time
Write Cycle Time
tWP
-
tALS
tALH
tDS
-
10
20
10
45
15
-
-
tDH
-
tWC
-
WE High Hold Time
tWH
-
NOTE :
1. If tCS is set less than 10ns, tWP must be minimum 35ns, otherwise, tWP may be minimum 25ns.
AC Characteristics for Operation
Parameter
Symbol
Min
-
Max
Uni
ms
ns
ns
ns
ns
ns
ns
ns
ns
ns
ns
ns
ns
ns
ns
ms
Data Transfer from Cell to Register
tR
10
ALE to RE Delay
tAR
10
10
20
25
-
-
-
CLE to RE Delay
tCLR
tRR
Ready to RE Low
-
RE Pulse Width
tRP
-
WE High to Busy
tWB
tRC
100
-
Read Cycle Time
50
-
CE Access Time
tCEA
tREA
tRHZ
tCHZ
tOH
45
35
30
20
-
RE Access Time
-
RE High to Output Hi-Z
CE High to Output Hi-Z
RE or CE High to Output hold
RE High Hold Time
Output Hi-Z to RE Low
WE High to RE Low
Device Resetting Time(Read/Program/Erase)
-
-
15
15
0
tREH
tIR
-
-
tWHR
tRST
60
-
-
5/10/500(1)
NOTE :
1. If reset command(FFh) is written at Ready state, the device goes into Busy for maximum 5us.
Revision 0.1
August 2003
- 11 -
Advance
MCP MEMORY
Preliminary
KBE00D002M-F407
NAND Flash Technical Notes
Invalid Block(s)
Invalid blocks are defined as blocks that contain one or more invalid bits whose reliability is not guaranteed by Samsung. The infor-
mation regarding the invalid block(s) is so called as the invalid block information. Devices with invalid block(s) have the same quality
level as devices with all valid blocks and have the same AC and DC characteristics. An invalid block(s) does not affect the perfor-
mance of valid block(s) because it is isolated from the bit line and the common source line by a select transistor. The system design
must be able to mask out the invalid block(s) via address mapping. The 1st block, which is placed on 00h block address, is fully guar-
anteed to be a valid block, does not require Error Correction.
Identifying Invalid Block(s)
All device locations are erased(FFh) except locations where the invalid block(s) information is written prior to shipping. The invalid
block(s) status is defined by the 1st word in the spare area. Samsung makes sure that either the 1st or 2nd page sof every invalid
block has non-FFFFh data at the column address of 256 and 261. Since the invalid block information is also erasable in most cases,
it is impossible to recover the information once it has been erased. Therefore, the system must be able to recognize the invalid
block(s) based on the original invalid block information and create the invalid block table via the following suggested flow chart(Figure
2). Any intentional erasure of the original invalid block information is prohibited.
Start
Set Block Address = 0
Increment Block Address
Check "FFh" at the column address256 and 261
of the 1st and 2nd page in the block
*
No
Create (or update)
Check "FFh" ?
Invalid Block(s) Table
Yes
No
Last Block ?
Yes
End
Figure 2. Flow chart to create invalid block table.
Revision 0.1
August 2003
- 12 -
Advance
MCP MEMORY
Preliminary
KBE00D002M-F407
NAND Flash Technical Notes (Continued)
Error in write or read operation
Within its life time, the additional invalid blocks may develop with NAND Flash memory. Refer to the qualification report for the actual
data.The following possible failure modes should be considered to implement a highly reliable system. In the case of status read fail-
ure after erase or program, block replacement should be done. Because program status fail during a page program does not affect
the data of the other pages in the same block, block replacement can be executed with a page-sized buffer by finding an erased
empty block and reprogramming the current target data and copying the rest of the replaced block. To improve the efficiency of
memory space, it is recommended that the read or verification failure due to single bit error be reclaimed by ECC without any block
replacement. The said additional block failure rate does not include those reclaimed blocks.
Failure Mode
Detection and Countermeasure sequence
Erase Failure
Status Read after Erase --> Block Replacement
Status Read after Program --> Block Replacement
Read back ( Verify after Program) --> Block Replacement
or ECC Correction
Write
Read
Program Failure
Single Bit Failure
Verify ECC -> ECC Correction
: Error Correcting Code --> Hamming Code etc.
Example) 1bit correction & 2bit detection
ECC
Program Flow Chart
If ECC is used, this verification
operation is not needed.
Start
Write 80h
Write 00h
Write Address
Wait for tR Time
Write Address
Write Data
Write 10h
*
No
Program Error
Verify Data
Read Status Register
Yes
Program Completed
No
I/O 6 = 1 ?
or R/B = 1 ?
: If program operation results in an error, map out
the block including the page in error and copy the
target data to another block.
*
Yes
*
No
Program Error
I/O 0 = 0 ?
Yes
Revision 0.1
August 2003
- 13 -
Advance
MCP MEMORY
Preliminary
KBE00D002M-F407
NAND Flash Technical Notes (Continued)
Erase Flow Chart
Read Flow Chart
Start
Start
Write 60h
Write 00h
Write Block Address
Write Address
Read Data
Write D0h
Read Status Register
ECC Generation
No
No
I/O 6 = 1 ?
or R/B = 1 ?
Reclaim the Error
Verify ECC
Yes
Yes
*
No
Page Read Completed
Erase Error
I/O 0 = 0 ?
Yes
Erase Completed
: If erase operation results in an error, map out
the failing block and replace it with another block.
*
Block Replacement
Block A
1st
2
{
(n-1)th
nth
an error occurs.
(page)
Buffer memory of the controller.
Block B
1st
1
{
(n-1)th
nth
(page)
* Step1
When an error happens in the nth page of the Block ’A’ during erase or program operation.
* Step2
Copy the nth page data of the Block ’A’ in the buffer memory to the nth page of another free block. (Block ’B’)
* Step3
Then, copy the data in the 1st ~ (n-1)th page to the same location of the Block ’B’.
* Step4
Do not further erase Block ’A’ by creating an ’invalid Block’ table or other appropriate scheme.
Revision 0.1
August 2003
- 14 -
Advance
MCP MEMORY
Preliminary
KBE00D002M-F407
Pointer Operation
Samsung NAND Flash has two address pointer commands as a substitute for the most significant column address. ’00h’ command
sets the pointer to ’A’ area(0~255word), and ’50h’ command sets the pointer to ’B’ area(256~263word). With these commands, the
starting column address can be set to any of a whole page(0~263word). ’00h’ or ’50h’ is sustained until another address pointer com-
mand is inputted. To program data starting from ’A’ or ’B’ area, ’00h’ or ’50h’ command must be inputted before ’80h’ command is
written. A complete read operation prior to ’80h’ command is not necessary.
Table 2. Destination of the pointer
"A" area
"B" area
(00h plane)
(50h plane)
Command
Pointer position
Area
256 Word
8 Word
00h
50h
0 ~ 255 word
256 ~ 263 word
main array(A)
spare array(B)
"A"
"B"
Internal
Page Register
Pointer select
command
(00h, 50h)
Pointer
Figure 3. Block Diagram of Pointer Operation
(1) Command input sequence for programming ’A’ area
The address pointer is set to ’A’ area(0~255), and sustained
Address / Data input
Address / Data input
80h 10h
00h
80h
10h
00h
’A’,’B’ area can be programmed.
’00h’ command can be omitted.
It depends on how many data are inputted.
(2) Command input sequence for programming ’B’ area
The address pointer is set to ’B’ area(256~263), and sustained
Address / Data input
Address / Data input
80h 10h
50h
80h
10h
50h
Only ’B’ area can be programmed.
’50h’ command can be omitted.
Revision 0.1
August 2003
- 15 -
Advance
MCP MEMORY
Preliminary
KBE00D002M-F407
System Interface Using CE don’t-care.
For an easier system interface, CE may be inactive during the data-loading or sequential data-reading as shown below. The internal
264word page registers are utilized as seperate buffers for this operation and the system design gets more flexible. In addition, for
voice or audio applications which use slow cycle time on the order of u-seconds, de-activating CE during the data-loading and read-
ing would provide significant savings in power consumption.
Figure 4. Program Operation with CE don’t-care.
CLE
CE don’t-care
CE
WE
ALE
I/Ox
80h
Start Add.(4Cycle)
Data Input
Data Input
10h
tCS
tCH
tCEA
CE
CE
tREA
RE
tWP
tOH
WE
I/Ox
out
Figure 5. Read Operation with CE don’t-care.
CLE
CE
CE don’t-care
RE
ALE
tR
R/B
WE
I/Ox
Data Output(sequential)
00h
Start Add.(4Cycle)
Revision 0.1
August 2003
- 16 -
Advance
MCP MEMORY
Preliminary
KBE00D002M-F407
Command Latch Cycle
CLE
CE
tCLH
tCH
tCLS
tCS
tWP
WE
tALS
tALH
ALE
I/Ox
tDH
tDS
Command
Address Latch Cycle
tCLS
CLE
tCS
tWC
tWC
tWC
CE
tCH
tWP
tWP
tWP
tWP
WE
tWH
tALH tALS
tWH
tALH tALS
tWH
tALH tALS
tALH
tDH
tALS
ALE
I/OX
tDH
tDH
tDH
tDS
tDS
tDS
tDS
A0~A7
A9~A16
A17~A24
A25
Revision 0.1
August 2003
- 17 -
Advance
MCP MEMORY
Preliminary
KBE00D002M-F407
Input Data Latch Cycle
tCLH
CLE
CE
tCH
tWC
tALS
ALE
tWP
tWP
tWP
WE
tWH
tDH
tDH
tDH
tDS
tDS
tDS
I/Ox
DIN 0
DIN 1
DIN n
Sequential Out Cycle after Read(CLE=L, WE=H, ALE=L)
tRC
CE
tCHZ*
tOH
tREH
tREA
tREA
tREA
tRP
RE
tRHZ*
tRHZ*
tOH
I/Ox
Dout
Dout
Dout
tRR
R/B
NOTES :
1.Transition is measured ±200mV from steady state voltage with load.
2.This parameter is sampled and not 100% tested.
Revision 0.1
August 2003
- 18 -
Advance
MCP MEMORY
Preliminary
KBE00D002M-F407
Status Read Cycle
tCLR
CLE
CE
tCLS
tCS
tCLH
tCH
tWP
WE
RE
tCEA
tCHZ
tOH
tWHR
tRHZ
tOH
tDH
tREA
tDS
tIR
Status Output
I/Ox
70h
READ1 OPERATION(READ ONE PAGE)
CLE
CE
tCHZ
tOH
tWC
WE
tWB
tAR
ALE
tRHZ
tOH
tR
tRC
RE
N Address
tRR
Dout 264
A17 ~ A24
00h
A0 ~ A7
A9 ~ A16
Dout N
Dout N+1
Dout N+2
A25
I/Ox
R/B
Column
Address
Page(Row)
Address
Busy
X16 device : m = 264 , Read CMD = 00h
Revision 0.1
August 2003
- 19 -
Advance
MCP MEMORY
Preliminary
KBE00D002M-F407
READ1 OPERATION (INTERCEPTED BY CE)
CLE
CE
WE
ALE
RE
tWB
tCHZ
tOH
tAR
tR
tRC
tRR
A9 ~ A16 A17 ~ A24
Dout N+2
00h or 01h A0 ~ A7
Dout N
Dout N+1
A25
I/Ox
R/B
Page(Row)
Address
Column
Address
Busy
READ2 OPERATION (READ ONE PAGE)
CLE
CE
WE
ALE
RE
tR
tWB
tAR
tRR
Dout
n+m
n+M
50h
A0 ~ A7 A9 ~ A16 A17 ~ A24
A25
I/Ox
R/B
Selected
Row
M Address
A0~A3 : Valid Address
A4~A7 : Don¢t care
8
256
Start
address M
Revision 0.1
August 2003
- 20 -
Advance
MCP MEMORY
Preliminary
KBE00D002M-F407
PAGE PROGRAM OPERATION
CLE
CE
tWC
tWC
tWC
WE
ALE
RE
tPROG
tWB
Din
N
Din
255
A25
10h
80h
A0 ~ A7 A9 ~ A16 A17 ~ A24
70h
I/O0
I/Ox
R/B
Sequential Data Column
Input Command Address
Program
Command
1 up to 256 Word Data
Serial Input
Read Status
Command
Page(Row)
Address
I/O0=0 Successful Program
I/O0=1 Error in Program
COPY-BACK PROGRAM OPERATION
CLE
CE
tWC
WE
ALE
RE
tWB
tPROG
tWB
tR
8Ah
00h
A0~A7 A9~A16 A17~A24
A25
A0~A7 A9~A16 A17~A24 A25
10h
70h
I/O0
I/Ox
R/B
Column
Column
Read Status
Command
Page(Row)
Page(Row)
Address
Address
Address
Address
Busy
Busy
Copy-Back Data
Input Command
I/O0=0 Successful Program
I/O0=1 Error in Program
Revision 0.1
August 2003
- 21 -
Advance
MCP MEMORY
Preliminary
KBE00D002M-F407
BLOCK ERASE OPERATION (ERASE ONE BLOCK)
CLE
CE
tWC
WE
tBERS
tWB
ALE
RE
60h
A9 ~ A16 A17 ~ A24
DOh
70h
I/O 0
A25
I/Ox
R/B
Page(Row)
Address
Busy
I/O0=0 Successful Erase
Read Status I/O0=1 Error in Erase
Command
Auto Block Erase Setup Command
Erase Command
MANUFACTURE & DEVICE ID READ OPERATION
CLE
CE
WE
ALE
tAR
RE
tREA
I/Ox
90h
00h
XX46h
ECh
Read ID Command
Address. 1cycle
Maker Code
Device Code
Revision 0.1
August 2003
- 22 -
Advance
MCP MEMORY
Preliminary
KBE00D002M-F407
DEVICE OPERATION
PAGE READ
Upon initial device power up, the device defaults to Read1 mode. This operation is also initiated by writing 00h to the command reg-
ister along with three address cycles. Once the command is latched, it does not need to be written for the following page read opera-
tion. Two types of operations are available : random read, serial page read.
The random read mode is enabled when the page address is changed. The 264 words of data within the selected page are trans-
ferred to the data registers in less than 10ms(tR). The system controller can detect the completion of this data transfer(tR) by analyz-
ing the output of R/B pin. Once the data in a page is loaded into the registers, they may be read out in 50ns cycle time by sequentially
pulsing RE. High to low transitions of the RE clock output the data starting from the selected column address up to the last column
address[column 255 /263 depending on the state of GND input pin].
The way the Read1 and Read2 commands work is like a pointer set to either the main area or the spare area. The spare area of
256~263 words may be selectively accessed by writing the Read2 command with GND input pin low. Addresses A0~A2 set the start-
ing address of the spare area while addresses A3~A7 must be "L". The Read1 command is needed to move the pointer back to the
main area. Figures 6, 7 show typical sequence and timings for each read operation.
Figure 6. Read1 Operation
CLE
CE
WE
ALE
tR
R/B
RE
I/Ox
Start Add.(4Cycle)
A0 ~ A7 & A9 ~ A25
00h
Data Output(Sequential)
(00h Command)
Main array
Data Field
Spare Field
Revision 0.1
August 2003
- 23 -
Advance
MCP MEMORY
Preliminary
KBE00D002M-F407
Figure 7. Read2 Operation
CLE
CE
WE
ALE
R/B
tR
RE
Start Add.(4Cycle)
I/Ox
50h
Data Output(Sequential)
Spare Field
A0 ~ A2 & A9 ~ A25
Main array
A3 ~ A7 are "L"
Data Field
Spare Field
Revision 0.1
August 2003
- 24 -
Advance
MCP MEMORY
Preliminary
KBE00D002M-F407
PAGE PROGRAM
The device is programmed basically on a page basis, but it does allow multiple partial page programing of a word or consecutive
words up to 264, in a single page program cycle. The number of consecutive partial page programming operation within the same
page without an intervening erase operation should not exceed 2 for main array and 3 for spare array. The addressing may be done
in any random order in a block. A page program cycle consists of a serial data loading period in which up to 264 words of data may-
be loaded into the page register, followed by a non-volatile programming period where the loaded data is programmed into the
appropriate cell. About the pointer operation, please refer to the attached technical notes.
The serial data loading period begins by inputting the Serial Data Input command(80h), followed by the three cycle address input and
then serial data loading. The words other than those to be programmed do not need to be loaded.The Page Program confirm com-
mand(10h) initiates the programming process. Writing 10h alone without previously entering the serial data will not initiate the pro-
gramming process. The internal write controller automatically executes the algorithms and timings necessary for program and verify,
thereby freeing the system controller for other tasks. Once the program process starts, the Read Status Register command may be
entered, with RE and CE low, to read the status register. The system controller can detect the completion of a program cycle by mon-
itoring the R/B output, or the Status bit(I/O 6) of the Status Register. Only the Read Status command and Reset command are valid
while programming is in progress. When the Page Program is complete, the Write Status Bit(I/O 0) may be checked(Figure 8). The
internal write verify detects only errors for "1"s that are not successfully programmed to "0"s. The command register remains in Read
Status command mode until another valid command is written to the command register.
Figure 8. Program Operation
tPROG
R/B
I/Ox
Pass
80h
Address & Data Input
I/O0
Fail
10h
70h
COPY-BACK PROGRAM
The copy-back program is configured to quickly and efficiently rewrite data stored in one page within the array to another page within
the same array without utilizing an external memory. Since the time-consuming sequently-reading and its re-loading cycles are
removed, the system performance is improved. The benefit is especially obvious when a portion of a block is updated and the rest of
the block also need to be copied to the newly assigned free block. The operation for performing a copy-back is a sequential execu-
tion of page-read without burst-reading cycle and copying-program with the address of destination page. A normal read operation
with "00h" command with the address of the source page moves the whole 264words data into the internal buffer. As soon as the
Flash returns to Ready state, copy-back programming command "8Ah" may be given with three address cycles of target page fol-
lowed. The data stored in the internal buffer is then programmed directly into the memory cells of the destination page. Once the
Copy-Back Program is finished, any additional partial page programming into the copied pages is prohibited before erase. Since the
memory array is internally partitioned into two different planes, copy-back program is allowed only within the same memory plane.
Thus, A14 and A25, the plane address, of source and destination page address must be the same.
Figure 9. Copy-Back Program Operation
tR
tPROG
R/B
I/Ox
Add.(4Cycles)
Pass
00h
Add.(4Cycles)
I/O0
Fail
8Ah
70h
Source Address
Destination Address
Revision 0.1
August 2003
- 25 -
Advance
MCP MEMORY
Preliminary
KBE00D002M-F407
BLOCK ERASE
The Erase operation is done on a block basis. Block address loading is accomplished in three cycles initiated by an Erase Setup
command(60h). Only address A14 to A25 is valid while A9 to A13 is ignored. The Erase Confirm command(D0h) following the block
address loading initiates the internal erasing process. This two-step sequence of setup followed by execution command ensures that
memory contents are not accidentally erased due to external noise conditions.
At the rising edge of WE after the erase confirm command input, the internal write controller handles erase and erase-verify. When
the erase operation is completed, the Write Status Bit(I/O 0) may be checked. Figure 10 details the sequence.
Figure 10. Block Erase Operation
tBERS
R/B
Pass
60h
I/O0
Fail
I/Ox
70h
Address Input(3Cycle)
Block Add. : A9 ~ A25
D0h
READ STATUS
The device contains a Status Register which may be read to find out whether program or erase operation is completed, and whether
the program or erase operation is completed successfully. After writing 70h command to the command register, a read cycle outputs
the content of the Status Register to the I/O pins on the falling edge of CE or RE, whichever occurs last. This two line control allows
the system to poll the progress of each device in multiple memory connections even when R/B pins are common-wired. RE or CE
does not need to be toggled for updated status. Refer to table 3 for specific Status Register definitions. The command register
remains in Status Read mode until further commands are issued to it. Therefore, if the status register is read during a random read
cycle, a read command(00h or 50h) should be given before sequential page read cycle.
Table 3. Read Status Register Definition
I/O #
Status
Definition
"0" : Successful Program / Erase
I/O 0
Program / Erase
"1" : Error in Program / Erase
I/O 1
I/O 2
"0"
"0"
"0"
"0"
"0"
Reserved for Future
Use
I/O 3
I/O 4
I/O 5
I/O 6
Device Operation
Write Protect
Not use
"0" : Busy
"1" : Ready
"1" : Not Protected
I/O 7
"0" : Protected
Don’t care
I/O 8~15
Revision 0.1
August 2003
- 26 -
Advance
MCP MEMORY
Preliminary
KBE00D002M-F407
READ ID
The device contains a product identification mode, initiated by writing 90h to the command register, followed by an address input of
00h. Two read cycles sequentially output the manufacture code(ECh), and the device code respectively. The command register
remains in Read ID mode until further commands are issued to it. Figure 11 shows the operation sequence.
Figure 11. Read ID Operation
CLE
tCEA
CE
WE
tAR
ALE
RE
tWHR
tREA
I/Ox
ECh
00h
XX46h
90h
Address. 1cycle
Maker code
Device code
RESET
The device offers a reset feature, executed by writing FFh to the command register. When the device is in Busy state during random
read, program or erase mode, the reset operation will abort these operations. The contents of memory cells being altered are no
longer valid, as the data will be partially programmed or erased. The command register is cleared to wait for the next command, and
the Status Register is cleared to value C0h when WP is high. Refer to table 4 for device status after reset operation. If the device is
already in reset state a new reset command will not be accepted by the command register. The R/B pin transitions to low for tRST
after the Reset command is written. Refer to Figure 12 below.
Figure 12. RESET Operation
tRST
R/B
I/Ox
FFh
Table4. Device Status
After Power-up
After Reset
Operation Mode
Read 1
Waiting for next command
Revision 0.1
August 2003
- 27 -
Advance
MCP MEMORY
Preliminary
KBE00D002M-F407
READY/BUSY
The device has a R/B output that provides a hardware method of indicating the completion of a page program, erase and random
read completion. The R/B pin is normally high but transitions to low after program or erase command is written to the command reg-
ister or random read is started after address loading. It returns to high when the internal controller has finished the operation. The pin
is an open-drain driver thereby allowing two or more R/B outputs to be Or-tied. Because pull-up resistor value is related to tr(R/B) and
current drain during busy(ibusy) , an appropriate value can be obtained with the following reference chart. Its value can be deter-
mined by the following guidance.
Rp
ibusy
Vccqn
VOL : 0.1V, VOH : Vccq-0.1V
Ready Vccqn
R/Bn
open drain output
VOH
VOL
Busy
tf
tr
GND
Device
@ Vcc = 1.8V, Ta = 25°C , CL = 30pF
Ibusy
300n
3m
1.7
200n
100n
2m
1m
120
0.85
60
90
tr
30
0.57
1.7
0.43
1.7
1.7
1.7
2K
tf
4K
1K
3K
Rp(ohm)
Rp value guidance
Vccq(Max.) - VOL(Max.)
1.9V
3mA + SIL
Rp(min, 1.8V part) =
=
IOL + SIL
where IL is the sum of the input currents of all devices tied to the R/B pin.
Rp(max) is determined by maximum permissible limit of tr
Revision 0.1
August 2003
- 28 -
Advance
MCP MEMORY
Preliminary
KBE00D002M-F407
Data Protection & Power up sequence
The device is designed to offer protection from any involuntary program/erase during power-transitions. An internal voltage detector
disables all functions whenever Vcc is below about 1.1V. WP pin provides hardware protection and is recommended to be kept at VIL
during power-up and power-down and recovery time of minimum 10ms is required before internal circuit gets ready for any command
sequences as shown in Figure 13. The two step command sequence for program/erase provides additional software protection.
Figure 13. AC Waveforms for Power Transition
~ 1.5V
~ 1.5V
VCC
High
WP
WE
10ms
Revision 0.1
August 2003
- 29 -
Advance
MCP MEMORY
Preliminary
KBE00D002M-F407
256Mb(8Mb x 32)
Mobile SDRAM DDP E‘-Die
Revision 0.1
August 2003
- 30 -
Advance
MCP MEMORY
Preliminary
KBE00D002M-F407
ABSOLUTE MAXIMUM RATINGS
Parameter
Symbol
VIN, VOUT
VDD, VDDQ
TSTG
Value
Unit
V
Voltage on any pin relative to Vss
Voltage on VDD supply relative to Vss
Storage temperature
-1.0 ~ 2.6
-1.0 ~ 2.6
V
-55 ~ +150
°C
W
Power dissipation
PD
1.0
50
Short circuit current
IOS
mA
NOTES:
Permanent device damage may occur if ABSOLUTE MAXIMUM RATINGS are exceeded.
Functional operation should be restricted to recommended operating condition.
Exposure to higher than recommended voltage for extended periods of time could affect device reliability.
DC OPERATING CONDITIONS
Recommended operating conditions (Voltage referenced to VSS = 0V, TA = -25 to 85°C for Extended, -25 to 70°C for Commercial)
Parameter
Symbol
VDD
VDDQ
VIH
Min
1.65
Typ
Max
Unit
V
Note
1.8
1.95
Supply voltage
1.65
1.8
1.95
V
Input logic high voltage
Input logic low voltage
Output logic high voltage
Output logic low voltage
Input leakage current
NOTES :
0.8 x VDDQ
-0.3
-
0
-
VDDQ + 0.3
V
1
VIL
0.3
-
V
2
VOH
VOL
VDDQ -0.2
-
V
IOH = -0.1mA
IOL = 0.1mA
3
-
0.2
10
V
ILI
-10
-
uA
1. VIH (max) = 2.2V AC.The overshoot voltage duration is £ 3ns.
2. VIL (min) = -1.0V AC. The undershoot voltage duration is £ 3ns.
3. Any input 0V £ VIN £ VDDQ.
Input leakage currents include Hi-Z output leakage for all bi-directional buffers with tri-state outputs.
4. Dout is disabled, 0V £ VOUT £ VDDQ.
CAPACITANCE (VDD = 2.5V, TA = 23°C, f = 1MHz, VREF =0.9V ± 50 mV)
Pin
Symbol
CCLK
CIN
Min
3.0
3.0
1.5
3.0
3.0
Max
8.0
8.0
4.0
8.0
6.5
Unit
Note
Clock
pF
pF
pF
pF
pF
RAS, CAS, WE, CS, CKE
DQM
CIN
Address
CADD
COUT
DQ0 ~ DQ31
Revision 0.1
August 2003
- 31 -
Advance
MCP MEMORY
Preliminary
KBE00D002M-F407
DC CHARACTERISTICS
Recommended operating conditions (Voltage referenced to VSS = 0V, TA = -25 to 85°C for Extended, -25 to 70°C for Commercial)
Version
Parameter
Symbol
Test Condition
Unit Note
-1L
-15
Burst length = 1
tRC ³ tRC(min)
IO = 0 mA
Operating Current
(One Bank Active)
ICC1
70
60
mA
mA
1
ICC2P CKE £ VIL(max), tCC = 10ns
0.6
0.6
Precharge Standby Current in
power-down mode
ICC2PS CKE & CLK £ VIL(max), tCC = ¥
CKE ³ VIH(min), CS ³ VIH(min), tCC = 10ns
ICC2N
20
2
Input signals are changed one time during 20ns
Precharge Standby Current
in non power-down mode
mA
mA
CKE ³ VIH(min), CLK £ VIL(max), tCC = ¥
ICC2NS
Input signals are stable
ICC3P CKE £ VIL(max), tCC = 10ns
8
8
Active Standby Current
in power-down mode
ICC3PS CKE & CLK £ VIL(max), tCC = ¥
CKE ³ VIH(min), CS ³ VIH(min), tCC = 10ns
ICC3N
40
10
mA
mA
Active Standby Current
in non power-down mode
(One Bank Active)
Input signals are changed one time during 20ns
CKE ³ VIH(min), CLK £ VIL(max), tCC = ¥
ICC3NS
Input signals are stable
IO = 0 mA
Operating Current
(Burst Mode)
Page burst
4Banks Activated
ICC4
100
80
mA
1
tCCD = 2CLKs
Refresh Current
ICC5
ICC6
tARFC ³ tARFC(min)
170
Max 40
320
150
mA
2
3
Internal TCSR
4Banks
Max 85/70
600
°C
Self Refresh Current
CKE £ 0.2V
-G/F
uA
2Banks
1Bank
260
400
230
300
NOTES:
1. Measured with outputs open.
2. Refresh period is 64ms.
3. Internal TCSR can be supported.
In commercial Temp : Max 40°C/Max 70°C, In extended Temp : Max 40°C/Max 85°C
4. Unless otherwise noted, input swing IeveI is CMOS(VIH /VIL=VDDQ/VSSQ).
Revision 0.1
August 2003
- 32 -
Advance
MCP MEMORY
Preliminary
KBE00D002M-F407
AC OPERATING TEST CONDITIONS(VDD = 2.5V ± 0.2V, TA = -25 to 85°C for Extended, -25 to 70°C for Commercial)
Parameter
AC input levels (Vih/Vil)
Value
0.9 x VDDQ / 0.2
0.5 x VDDQ
tr/tf = 1/1
Unit
V
Input timing measurement reference level
Input rise and fall time
V
ns
V
Output timing measurement reference level
Output load condition
0.5 x VDDQ
See Figure 2
VDDQ
13.9KW
Vtt=0.5 x VDDQ
VOH (DC) = VDDQ - 0.2V, IOH = -0.1mA
VOL (DC) = 0.2V, IOL = 0.1mA
30pF
Output
50W
10.6KW
Output
Z0=50W
30pF
Figure 1. DC Output Load Circuit
Figure 2. AC Output Load Circuit
Revision 0.1
August 2003
- 33 -
Advance
MCP MEMORY
Preliminary
KBE00D002M-F407
OPERATING AC PARAMETER
(AC operating conditions unless otherwise noted)
Version
Parameter
Symbol
Unit
Note
-1L
19
-15
30
30
30
60
Row active to row active delay
RAS to CAS delay
tRRD(min)
tRCD(min)
ns
ns
1
1
1
1
28.5
28.5
60
Row precharge time
tRP(min)
ns
tRAS(min)
ns
Row active time
tRAS(max)
tRC(min)
100
us
Row cycle time
88.5
90
ns
1
2
3
2
2
Last data in to row precharge
Last data in to Active delay
Last data in to new col. address delay
Last data in to burst stop
tRDL(min)
2
CLK
-
tDAL(min)
tRDL + tRP
1
tCDL(min)
CLK
CLK
ns
tBDL(min)
1
105
120
1
Auto refresh cycle time
tARFC(min)
tSRFX(min)
tCCD(min)
Exit self refresh to active command
Col. address to col. address delay
Number of valid output data
Number of valid output data
Number of valid output data
ns
CLK
4
5
CAS latency=3
CAS latency=2
CAS latency=1
2
ea
1
0
NOTES:
1. The minimum number of clock cycles is determined by dividing the minimum time required with clock cycle time
and then rounding off to the next higher integer.
2. Minimum delay is required to complete write.
3. Minimum tRDL=2CLK and tDAL(= tRDL + tRP) is required to complete both of last data write command(tRDL) and precharge command(tRP).
4. All parts allow every cycle column address change.
5. In case of row precharge interrupt, auto precharge and read burst stop.
Revision 0.1
August 2003
- 34 -
Advance
MCP MEMORY
Preliminary
KBE00D002M-F407
AC CHARACTERISTICS(AC operating conditions unless otherwise noted)
-1L
-15
Parameter
Symbol
Unit
Note
Min
9.5
15
Max
Min
15
Max
CLK cycle time
CLK cycle time
CLK cycle time
CAS latency=3
CAS latency=2
CAS latency=1
CAS latency=3
CAS latency=2
CAS latency=1
CAS latency=3
CAS latency=2
CAS latency=1
tCC
tCC
tCC
tSAC
tSAC
tSAC
tOH
tOH
tOH
tCH
tCL
1000
1000
ns
1
15
25
30
CLK to valid output delay
CLK to valid output delay
CLK to valid output delay
Output data hold time
Output data hold time
Output data hold time
CLK high pulse width
CLK low pulse width
Input setup time
7
8
9
9
ns
ns
1,2
2
20
24
2.5
2.5
2.5
3.5
3.5
3.0
1.5
1
2.5
2.5
2.5
3.5
3.5
4.0
2.0
1
ns
ns
ns
ns
ns
3
3
3
3
2
tSS
Input hold time
tSH
CLK to output in Low-Z
tSLZ
CAS latency=3
CAS latency=2
CAS latency=1
7
8
9
9
CLK to output in Hi-Z
tSHZ
ns
20
24
NOTES :
1. Parameters depend on programmed CAS latency.
2. If clock rising time is longer than 1ns, (tr/2-0.5)ns should be added to the parameter.
3. Assumed input rise and fall time (tr & tf) = 1ns.
If tr & tf is longer than 1ns, transient time compensation should be considered,
i.e., [(tr + tf)/2-1]ns should be added to the parameter.
Revision 0.1
August 2003
- 35 -
Advance
MCP MEMORY
Preliminary
KBE00D002M-F407
SIMPLIFIED TRUTH TABLE
A11,
A9 ~ A0
COMMAND
CKEn-1 CKEn CS RAS CAS WE DQM BA0,1 A10/AP
Note
Register
Refresh
Mode Register Set
Auto Refresh
H
H
X
H
L
L
L
L
L
L
L
L
X
X
OP CODE
1, 2
3
H
X
Entry
3
Self
L
H
L
H
X
L
H
X
H
H
X
H
3
Refresh
Exit
L
H
H
H
X
X
X
X
X
X
3
Bank Active & Row Addr.
V
V
Row Address
Read &
Column Address
Auto Precharge Disable
Auto Precharge Enable
L
Column
4
L
H
L
H
Address
(A0~A8)
H
4, 5
Write &
Column Address
Auto Precharge Disable
Auto Precharge Enable
L
Column
Address
(A0~A8)
4
H
H
H
X
X
X
L
L
L
H
H
L
L
H
H
L
L
L
X
X
X
V
H
4, 5
Burst Stop
Precharge
X
6
Bank Selection
All Banks
V
X
L
X
H
H
L
X
V
X
X
H
X
V
X
X
H
X
V
X
X
H
X
V
X
V
X
X
H
X
V
Entry
H
L
L
H
L
X
X
X
Clock Suspend or
Active Power Down
X
X
Exit
X
H
L
Entry
H
Precharge Power Down
Mode
H
L
Exit
L
H
H
H
X
X
V
X
DQM
X
X
7
H
L
X
H
X
H
No Operation Command
(V=Valid, X=Don¢t Care, H=Logic High, L=Logic Low)
NOTES :
1. OP Code : Operand Code
A0 ~ A11 & BA0 ~ BA1 : Program keys. (@MRS)
2. MRS can be issued only at all banks precharge state.
A new command can be issued after 2 CLK cycles of MRS.
3. Auto refresh functions are the same as CBR refresh of DRAM.
The automatical precharge without row precharge command is meant by "Auto".
Auto/self refresh can be issued only at all banks precharge state.
Partial self refresh can be issued only after setting partial self refresh mode of EMRS.
4. BA0 ~ BA1 : Bank select addresses.
5. During burst read or write with auto precharge, new read/write command can not be issued.
Another bank read/write command can be issued after the end of burst.
New row active of the associated bank can be issued at tRP after the end of burst.
6. Burst stop command is valid at every burst length.
7. DQM sampled at the positive going edge of CLK masks the data-in at that same CLK in write operation (Write DQM latency
is 0), but in read operation, it makes the data-out Hi-Z state after 2 CLK cycles. (Read DQM latency is 2).
Revision 0.1
August 2003
- 36 -
Advance
MCP MEMORY
Preliminary
KBE00D002M-F407
A. MODE REGISTER FIELD TABLE TO PROGRAM MODES
Register Programmed with Normal MRS
*2
Address
A11 ~ A10/AP
A8
A7
A6
A5
A4
A3
A2
A1
A0
BA0 ~ BA1
A9
"0" Setting for
Normal MRS
*1
Function
W.B.L
Test Mode
CAS Latency
BT
Burst Length
RFU
Normal MRS Mode
Test Mode
CAS Latency
Burst Type
Type
Burst Length
A8 A7
Type
Mode Register Set
Reserved
A6 A5
A4
0
Latency
Reserved
1
A3
0
A2
A1
0
A0
0
BT=0
BT=1
0
0
1
1
0
1
0
1
0
0
0
0
1
1
1
1
0
0
1
1
0
0
1
1
Sequential
Interleave
0
0
0
0
1
1
1
1
1
2
4
8
1
2
4
8
1
1
0
1
Reserved
0
2
Mode Select
1
0
Reserved
1
3
BA1 BA0
Mode
1
1
Write Burst Length
Length
0
Reserved
Reserved
Reserved
Reserved
0
0
Reserved Reserved
Reserved Reserved
Setting
for Nor-
mal MRS
A9
0
1
0
1
0
0
Burst
0
1
0
Reserved Reserved
*3
1
Single Bit
1
1
1
Reserved
Full Page
Register Programmed with Extended MRS
Address
BA1
BA0
A11 ~ A10/AP
A9
A8
A7
A6
A5
A4
RFU
A3
A2
A1
A0
*1
*1
Function
Mode Select
DS
PASR
RFU
EMRS for PASR(Partial Array Self Ref.) & DS(Driver Strength)
*4
Mode Select
Driver Strength
A5 Driver Strength
Full
PASR
A0
BA1
BA0
Mode
A6
0
A2
A1
0
# of Banks
4 Banks
0
0
1
1
0
1
0
1
Normal MRS
Reserved
0
0
0
0
0
1
1
1
1
0
1
0
1
0
1
0
1
0
1
0
1
1/2
1/4
1/8
0
2 Banks
EMRS for Mobile SDRAM
Reserved
1
1
1 Bank
1
1
Reserved
Reserved
Reserved
Reserved
Reserved
Reserved Address
0
A11~A10/AP
0
A9
0
A8
0
A7
0
A4
0
A3
0
0
1
1
NOTES:
1. RFU(Reserved for future use) should stay "0" during MRS cycle.
2. If A9 is high during MRS cycle, "Burst Read Single Bit Write" function will be enabled.
3. Full Page Length : x32 : 64Mb(256) , 128Mb (256), 256Mb (512), 512Mb (512)
4. Mobile SDRAM supports PASR of all banks, 1/2 of all banks and 1/4 of all banks.
Revision 0.1
August 2003
- 37 -
Advance
MCP MEMORY
Preliminary
KBE00D002M-F407
Partial Array Self Refresh
1. In order to save power consumption, Mobile SDRAM has PASR option.
2. Mobile SDRAM supports 3 kinds of PASR in self refresh mode : 4 Banks(256Mb), 2 Banks(128Mb) and 1 Bank(64Mb).
BA1=0
BA0=0
BA1=0
BA0=1
BA1=0
BA0=0
BA1=0
BA0=1
BA1=0
BA0=0
BA1=0
BA0=1
BA1=1
BA0=0
BA1=1
BA0=1
BA1=1
BA0=0
BA1=1
BA0=1
BA1=1
BA0=0
BA1=1
BA0=1
- 1 Bank
- 4 Banks
- 2 Banks
Partial Self Refresh Area
Internal Temperature Compensated Self Refresh
1. In order to save power consumption, Mobile-DRAM includes the internal temperature sonsor and control units to control the self
refresh cycle automatically according to the two temperature range : Max 40 °C and Max 85 °C(for Extended), Max 70 °C(for
Commercial).
2. If the EMRS for external TCSR is issued by the controller, this EMRS code for TCSR is ginored.
Self Refresh Current (Icc6)
Temperature Range
Unit
4 Banks
600
2 Banks
400
1 Bank
300
Max 85/70 °C
Max 40 °C
uA
320
260
230
B. POWER UP SEQUENCE
1. Apply power and attempt to maintain CKE at a high state and all other inputs may be undefined.
- Apply VDD before or at the same time as VDDQ.
2. Maintain stable power, stable clock and NOP input condition for a minimum of 200us.
3. Issue precharge commands for all banks of the devices.
4. Issue 2 or more auto-refresh commands.
5. Issue a mode register set command to initialize the mode register.
6. Issue a extended mode register set command to define DS or PASR operating type of the device after normal MRS.
EMRS cycle is not mandatory and the EMRS command needs to be issued only when DS or PASR is used.
The default state without EMRS command issued is the half driver strength and all 4 banks refreshed.
The device is now ready for the operation selected by EMRS.
For operating with DS or PASR, set DS or PASR mode in EMRS setting stage.
In order to adjust another mode in the state of DS or PASR mode, additional EMRS set is required but power up sequence is not
needed again at this time. In that case, all banks have to be in idle state prior to adjusting EMRS set.
Revision 0.1
August 2003
- 38 -
Advance
MCP MEMORY
Preliminary
KBE00D002M-F407
C. BURST SEQUENCE
1. BURST LENGTH = 4
Initial Address
Sequential
Interleave
A1
0
A0
0
0
1
2
3
1
2
3
0
2
3
0
1
3
0
1
2
0
1
2
3
1
0
3
2
2
3
0
1
3
2
1
0
0
1
1
0
1
1
2. BURST LENGTH = 8
Initial Address
Sequential
Interleave
A2
0
A1
0
A0
0
0
1
2
3
4
5
6
7
1
2
3
4
5
6
7
0
2
3
4
5
6
7
0
1
3
4
5
6
7
0
1
2
4
5
6
7
0
1
2
3
5
6
7
0
1
2
3
4
5
7
0
1
2
3
4
5
6
0
1
2
3
4
5
6
7
1
2
3
0
1
6
7
4
5
3
2
1
0
7
6
5
4
4
5
6
7
0
1
2
3
5
4
7
6
1
0
3
2
6
7
4
5
2
3
0
1
7
6
5
4
3
2
1
0
0
0
1
6
7
0
1
2
3
4
0
3
2
5
4
7
6
0
1
0
0
1
1
1
0
0
1
0
1
1
1
0
1
1
1
Revision 0.1
August 2003
- 39 -
Advance
MCP MEMORY
Preliminary
KBE00D002M-F407
D. DEVICE OPERATIONS
ADDRESSES of 256Mb
ADDRESSES of 512Mb
BANK ADDRESSES (BA0 ~ BA1)
BANK ADDRESSES (BA0 ~ BA1)
: In case x 16
: In case x 16
This SDRAM is organized as four independent banks of
8,388,608 words x 16 bits memory arrays. The BA0 ~ BA1 inputs
are latched at the time of assertion of RAS and CAS to select the
bank to be used for the operation. The bank addresses BA0 ~
BA1 are latched at bank active, read, write, mode register set
and precharge operations.
This SDRAM is organized as four independent banks of
4,194,304 words x 16 bits memory arrays. The BA0 ~ BA1 inputs
are latched at the time of assertion of RAS and CAS to select the
bank to be used for the operation. The bank addresses BA0 ~
BA1 are latched at bank active, read, write, mode register set
and precharge operations.
: In case x 32
: In case x 32
This SDRAM is organized as four independent banks of
4,194,304 words x 32 bits memory arrays. The BA0 ~ BA1 inputs
are latched at the time of assertion of RAS and CAS to select the
bank to be used for the operation. The bank addresses BA0 ~
BA1 are latched at bank active, read, write, mode register set
and precharge operations.
This SDRAM is organized as four independent banks of
2,097,152 words x 32 bits memory arrays. The BA0 ~ BA1 inputs
are latched at the time of assertion of RAS and CAS to select the
bank to be used for the operation. The bank addresses BA0 ~
BA1 are latched at bank active, read, write, mode register set
and precharge operations.
ADDRESS INPUTS (A0 ~ A12)
ADDRESS INPUTS (A0 ~ A12)
: In case x 16
: In case x 16
The 22 address bits are required to decode the 4,194,304 word
locations are multiplexed into 13 address input pins (A0 ~ A12).
The 13 bit row addresses are latched along with RAS and BA0 ~
BA1 during bank activate command. The 9 bit column addresses
are latched along with CAS, WE and BA0 ~ BA1 during read or
write command.
The 23 address bits are required to decode the 8,388,608 word
locations are multiplexed into 13 address input pins (A0 ~ A12).
The 13 bit row addresses are latched along with RAS and BA0 ~
BA1 during bank activate command. The 10 bit column
addresses are latched along with CAS, WE and BA0 ~ BA1 dur-
ing read or write command.
: In case x 32
: In case x 32
The 21 address bits are required to decode the 2,097,152 word
locations are multiplexed into 12 address input pins (A0 ~ A11).
The 12 bit row addresses are latched along with RAS and BA0 ~
BA1 during bank activate command. The 9 bit column addresses
are latched along with CAS, WE and BA0 ~ BA1 during read or
write command.
The 22 address bits are required to decode the 4,194,304 word
locations are multiplexed into 13 address input pins (A0 ~ A12).
The 13 bit row addresses are latched along with RAS and BA0 ~
BA1 during bank activate command. The 9 bit column addresses
are latched along with CAS, WE and BA0 ~ BA1 during read or
write command.
Revision 0.1
August 2003
- 40 -
Advance
MCP MEMORY
Preliminary
KBE00D002M-F407
D. DEVICE OPERATIONS (continued)
CLOCK (CLK)
The clock input is used as the reference for all SDRAM opera-
DQM OPERATION
The DQM is used to mask input and output operations. It works
similar to OE during read operation and inhibits writing during
write operation. The read latency is two cycles from DQM and
zero cycle for write, which means DQM masking occurs two
cycles later in read cycle and occurs in the same cycle during
write cycle. DQM operation is synchronous with the clock. The
DQM signal is important during burst interruptions of write with
read or precharge in the SDRAM. Due to asynchronous nature of
the internal write, the DQM operation is critical to avoid unwanted
or incomplete writes when the complete burst write is not
required. Please refer to DQM timing diagram also.
tions. All operations are synchronized to the positive going edge
of the clock. The clock transitions must be monotonic between
VIL and VIH. During operation with CKE high all inputs are
assumed to be in a valid state (low or high) for the duration of
set-up and hold time around positive edge of the clock in order to
function well Q perform and ICC specifications.
CLOCK ENABLE (CKE)
The clock enable(CKE) gates the clock onto SDRAM. If CKE
goes low synchronously with clock (set-up and hold time are the
same as other inputs), the internal clock is suspended from the
next clock cycle and the state of output and burst address is fro-
zen as long as the CKE remains low. All other inputs are ignored
from the next clock cycle after CKE goes low. When all banks are
in the idle state and CKE goes low synchronously with clock, the
SDRAM enters the power down mode from the next clock cycle.
The SDRAM remains in the power down mode ignoring the other
inputs as long as CKE remains low. The power down exit is syn-
chronous as the internal clock is suspended. When CKE goes
high at least "1CLK + tSS" before the high going edge of the
clock, then the SDRAM becomes active from the same clock
edge accepting all the input commands.
MODE REGISTER SET (MRS)
The mode register stores the data for controlling the various
operating modes of SDRAM. It programs the CAS latency, burst
type, burst length, test mode and various vendor specific options
to make SDRAM useful for variety of different applications. The
default value of the mode register is not defined, therefore the
mode register must be written after power up to operate the
SDRAM. The mode register is written by asserting low on CS,
RAS, CAS and WE (The SDRAM should be in active mode with
CKE already high prior to writing the mode register). The state of
address pins A0 ~ An and BA0 ~ BA1 in the same cycle as CS,
RAS, CAS and WE going low is the data written in the mode reg-
ister. Two clock cycles is required to complete the write in the
mode register. The mode register contents can be changed
using the same command and clock cycle requirements during
operation as long as all banks are in the idle state. The mode
register is divided into various fields depending on the fields of
functions. The burst length field uses A0 ~ A2, burst type uses
A3, CAS latency (read latency from column address) use A4 ~
A6, vendor specific options or test mode use A7 ~ A8, A10/AP ~
An and BA0 ~ BA1. The write burst length is programmed using
A9. A7 ~ A8, A10/AP ~ An and BA0 ~ BA1 must be set to low for
normal SDRAM operation. Refer to the table for specific codes
for various burst length, burst type and CAS latencies.
NOP and DEVICE DESELECT
When RAS, CAS and WE are high, the SDRAM performs no
operation (NOP). NOP does not initiate any new operation, but is
needed to complete operations which require more than single
clock cycle like bank activate, burst read, auto refresh, etc. The
device deselect is also a NOP and is entered by asserting CS
high. CS high disables the command decoder so that RAS, CAS,
WE and all the address inputs are ignored.
Revision 0.1
August 2003
- 41 -
Advance
MCP MEMORY
Preliminary
KBE00D002M-F407
D. DEVICE OPERATIONS (continued)
The SDRAM has four internal banks in the same chip and shares
part of the internal circuitry to reduce chip area, therefore it
restricts the activation of four banks simultaneously. Also the
noise generated during sensing of each bank of SDRAM is high,
requiring some time for power supplies to recover before another
bank can be sensed reliably. tRRD(min) specifies the minimum
time required between activating different bank. The number of
clock cycles required between different bank activation must be
calculated similar to tRCD specification. The minimum time
required for the bank to be active to initiate sensing and restoring
the complete row of dynamic cells is determined by tRAS(min).
Every SDRAM bank activate command must satisfy tRAS(min)
specification before a precharge command to that active bank
can be asserted. The maximum time any bank can be in the
active state is determined by tRAS(max). The number of cycles for
both tRAS(min) and tRAS(max) can be calculated similar to tRCD
specification.
EXTENDED MODE REGISTER SET (EMRS)
The extended mode register stores the data for selecting driver
strength, partial self refresh or temperature compensated self
refresh. EMRS cycle is not mandatory and the EMRS command
needs to be issued only when DS or PASR is used. The default
state without EMRS command issued is half driver strength, and
all 4 banks refreshed. The extended mode register is written by
asserting low on CS, RAS, CAS, WE and high on BA1 ,low on
BA0(The SDRAM should be in all bank precharge with CKE
already high prior to writing into the extended mode register).
The state of address pins A0 ~ A11 in the same cycle as CS,
RAS, CAS and WE going low is written in the extended mode
register. Two clock cycles are required to complete the write
operation in the extended mode register. The mode register con-
tents can be changed using the same command and clock cycle
requirements during operation as long as all banks are in the idle
state. A0 - A2 are used for partial self refresh , A5 - A6 are used
for Driver strength, "Low" on BA1 and "High" on BA0 are used
for EMRS. All the other address pins except A0-A2, A5-A6 and
BA1, BA0 must be set to low for proper EMRS operation. Refer to
the table for specific codes.
BURST READ
The burst read command is used to access burst of data on con-
secutive clock cycles from an active row in an active bank. The
burst read command is issued by asserting low on CS and CAS
with WE being high on the positive edge of the clock. The bank
must be active for at least tRCD(min) before the burst read com-
mand is issued. The first output appears in CAS latency number
of clock cycles after the issue of burst read command. The burst
length, burst sequence and latency from the burst read command
is determined by the mode register which is already pro-
grammed. The burst read can be initiated on any column address
of the active row. The address wraps around if the initial address
does not start from a boundary such that number of outputs from
each I/O are equal to the burst length programmed in the mode
register. The output goes into high-impedance at the end of the
burst, unless a new burst read was initiated to keep the data out-
put gapless. The burst read can be terminated by issuing another
burst read or burst write in the same bank or the other active
bank or a precharge command to the same bank. The burst stop
command is valid at every page burst length.
BANK ACTIVATE.
The bank activate command is used to select a random row in an
idle bank. By asserting low on RAS and CS with desired row and
bank address, a row access is initiated. The read or write opera-
tion can occur after a time delay of tRCD(min) from the time of
bank activation. tRCD is an internal timing parameter of SDRAM,
therefore it is dependent on operating clock frequency. The mini-
mum number of clock cycles required between bank activate and
read or write command should be calculated by dividing
tRCD(min) with cycle time of the clock and then rounding off the
result to the next higher integer.
Revision 0.1
August 2003
- 42 -
Advance
MCP MEMORY
Preliminary
KBE00D002M-F407
D. DEVICE OPERATIONS (continued)
BURST WRITE
AUTO PRECHARGE
The burst write command is similar to burst read command and The precharge operation can also be performed by using auto
is used to write data into the SDRAM on consecutive clock precharge. The SDRAM internally generates the timing to satisfy
cycles in adjacent addresses depending on burst length and tRAS(min) and "tRP" for the programmed burst length and CAS
burst sequence. By asserting low on CS, CAS and WE with valid latency. The auto precharge command is issued at the same
column address, a write burst is initiated. The data inputs are time as burst read or burst write by asserting high on A10/AP. If
provided for the initial address in the same clock cycle as the burst read or burst write by asserting high on A10/AP, the bank is
burst write command. The input buffer is deselected at the end of left active until a new command is asserted. Once auto pre-
the burst length, even though the internal writing can be com- charge command is given, no new commands are possible to
pleted yet. The writing can be completed by issuing a burst read that particular bank until the bank achieves idle state.
and DQM for blocking data inputs or burst write in the same or
another active bank. The burst stop command is valid at every
burst length. The write burst can also be terminated by using
AUTO REFRESH
DQM for blocking data and procreating the bank tRDL after the The storage cells of 64Mb, 128Mb and 256Mb SDRAM need to
last data input to be written into the active row. See DQM be refreshed every 64ms to maintain data. An auto refresh cycle
OPERATION also.
accomplishes refresh of a single row of storage cells. The inter-
nal counter increments automatically on every auto refresh cycle
to refresh all the rows. An auto refresh command is issued by
asserting low on CS, RAS and CAS with high on CKE and WE.
The auto refresh command can only be asserted with all banks
being in idle state and the device is not in power down mode
(CKE is high in the previous cycle). The time required to com-
plete the auto refresh operation is specified by tRC(min). The min-
imum number of clock cycles required can be calculated by
driving tRC with clock cycle time and them rounding up to the next
higher integer. The auto refresh command must be followed by
NOP's until the auto refresh operation is completed. All banks will
ALL BANKS PRECHARGE
All banks can be precharged at the same time by using Pre-
charge all command. Asserting low on CS, RAS, and WE with
high on A10/AP after all banks have satisfied tRAS(min) require-
ment, performs precharge on all banks. At the end of tRP after
performing precharge to all the banks, all banks are in idle state.
PRECHARGE
The precharge operation is performed on an active bank by be in the idle state at the end of auto refresh operation. The auto
asserting low on CS, RAS, WE and A10/AP with valid BA0 ~ BA1 refresh is the preferred refresh mode when the SDRAM is being
of the bank to be precharged. The precharge command can be used for normal data transactions. The 64Mb and 128Mb
asserted anytime after tRAS(min) is satisfied from the bank active SDRAM’s auto refresh cycle can be performed once in 15.6us or
command in the desired bank. tRP is defined as the minimum a burst of 4096 auto refresh cycles once in 64ms. The 256Mb
number of clock cycles required to complete row precharge is and 512Mb SDRAM’s auto refresh cycle can be performed once
calculated by dividing tRP with clock cycle time and rounding up in 7.8us or a burst of 8192 auto refresh cycles once in 64ms.
to the next higher integer. Care should be taken to make sure
that burst write is completed or DQM is used to inhibit writing
before precharge command is asserted. The maximum time any
bank can be active is specified by tRAS(max). Therefore, each
bank activate command. At the end of precharge, the bank
enters the idle state and is ready to be activated again. Entry to
Power down, Auto refresh, Self refresh and Mode register set
etc. is possible only when all banks are in idle state.
Revision 0.1
August 2003
- 43 -
Advance
MCP MEMORY
Preliminary
KBE00D002M-F407
D. DEVICE OPERATIONS(continued)
SELF REFRESH
The self refresh is another refresh mode available in the
SDRAM. The self refresh is the preferred refresh mode for data
retention and low power operation of SDRAM. In self refresh
mode, the SDRAM disables the internal clock and all the input
buffers except CKE. The refresh addressing and timing are inter-
nally generated to reduce power consumption.
The self refresh mode is entered from all banks idle state by
asserting low on CS, RAS, CAS and CKE with high on WE. Once
the self refresh mode is entered, only CKE state being low mat-
ters, all the other inputs including the clock are ignored in order
to remain in the self refresh mode.
The self refresh is exited by restarting the external clock and then
asserting high on CKE. This must be followed by NOP's for a
minimum time of tSRFX before the SDRAM reaches idle state to
begin normal operation. In case that the system uses burst auto
refresh during normal operation, it is recommended to use burst
8192 auto refresh cycles for 256Mb and 512Mb, and burst 4096
auto refresh cycles for 128Mb and 64Mb immediately before
entering self refresh mode and after exiting in self refresh mode.
On the other hand, if the system uses the distributed auto
refresh, the system only has to keep the refresh duty cycle.
Revision 0.1
August 2003
- 44 -
Advance
MCP MEMORY
Preliminary
KBE00D002M-F407
E. BASIC FEATURE AND FUNCTION DESCRIPTIONS
1. CLOCK Suspend
1) Clock Suspended During Write
CLK
2) Clock Suspended During Read (BL=4)
CLK
CMD
CKE
WR
CMD
CKE
RD
Masked by CKE
Masked by CKE
Internal
CLK
Internal
CLK
D0
D0
D1
D1
D2
D2
D3
D3
DQ(CL2)
DQ(CL3)
DQ(CL2)
DQ(CL3)
Q0
Q1
Q2
Q1
Q3
Q2
Q0
Q3
Not Written
Suspended Dout
2. DQM Operation
1) Write Mask (BL=4)
CLK
2) Read Mask (BL=4)
CLK
CMD
DQM
CMD
DQM
WR
RD
Masked by CKE
D3
Masked by CKE
Hi-Z
DQ(CL2)
DQ(CL3)
DQ(CL2)
DQ(CL3)
Q0
Q2
Q1
Q3
Q2
D0
D0
D1
D1
Hi-Z
D3
Q3
DQM to Data-in Mask = 0
DQM to Data-out Mask = 2
3) DQM with Clock Suspended (Full Page Read) *2
CLK
CMD
RD
CKE
DQM
Hi-Z
Hi-Z
Hi-Z
Hi-Z
Hi-Z
Q0
Q2
Q1
Q4
Q3
Q6
Q5
Q7
Q6
Q8
Q7
DQ(CL2)
DQ(CL3)
Hi-Z
*NOTES :
1. CKE to CLK disable/enable = 1CLK.
2. DQM makes data out Hi-Z after 2CLKs which should masked by CKE " L"
3. DQM masks both data-in and data-out.
Revision 0.1
August 2003
- 45 -
Advance
MCP MEMORY
Preliminary
KBE00D002M-F407
3. CAS Interrupt (I)
1) Read interrupted by Read (BL=4) *1
CLK
RD RD
CMD
ADD
A
B
DQ(CL2)
DQ(CL3)
QA0 QB0 QB1 QB1 QB3
QA0 QB0 QB1 QB1 QB3
*2
tCCD
2) Write interrupted by Write (BL=2)
CLK
3) Write interrupted by Read (BL=2)
CLK
WR RD
WR WR
CMD
CMD
*2
*2
tCCD
tCCD
ADD
A
B
A
B
ADD
DQ
DQ(CL2)
DQ(CL3)
DA0
DA0
QB0 QB1
QB0 QB1
DA0 DB0 DB1
*3
tCDL
*3
tCDL
*NOTES:
1. By " Interrupt", It is meant to stop burst read/write by external command before the end of burst.
By "CAS Interrupt", to stop burst read/write by CAS access ; read and write.
2. tCCD : CAS to CAS delay. (=1CLK)
3. tCDL : Last data in to new column address delay. (=1CLK)
Revision 0.1
August 2003
- 46 -
Advance
MCP MEMORY
Preliminary
KBE00D002M-F407
4. CAS Interrupt (II) : Read Interrupted by Write & DQM
(a) CL=2, BL=4
CLK
i) CMD
RD WR
DQM
DQ
D0
D1
D2
D3
D2
ii) CMD
RD
WR
DQM
DQ
Hi-Z
D0
D1
D3
iii) CMD
RD
WR
DQM
DQ
Hi-Z
D0
D1
D2
D1
D3
D2
iv) CMD
RD
WR
DQM
DQ
Hi-Z
*1
Q0
D0
D3
(b) CL=3, BL=4
CLK
RD WR
i) CMD
DQM
DQ
D0
D1
D2
D3
D2
ii) CMD
RD
WR
DQM
DQ
D0
D1
D3
D2
iii) CMD
RD
RD
WR
DQM
DQ
D0
D1
D3
D2
D1
iv) CMD
WR
DQM
DQ
Hi-Z
D0
D1
D3
D2
v) CMD
RD
WR
DQM
DQ
Hi-Z
*1
Q0
D0
D3
*NOTE:
1. To prevent bus contention, there should be at least one gap between data in and data out.
Revision 0.1
August 2003
- 47 -
Advance
MCP MEMORY
Preliminary
KBE00D002M-F407
5. Write Interrupted by Precharge & DQM
1) tRDL = 2CLK
CLK
*3
CMD
DQM
WR
D0
PRE
*2
D1
D2
DQ
Masked by DQM
*NOTES:
1. To prevent bus contention, DQM should be issued which makes at least one gap between data in and data out.
2. To inhibit invalid write, DQM should be issued.
3. This precharge command and burst write command should be of the same bank, otherwise it is not precharge interrupt but only another bank pre-
charge of four banks operation.
6. Precharge
1) Normal Write
BL=4 & tRDL=2CLK
CLK
CMD
DQ
WR
D0
PRE
D1
D2
D3
*1
tRDL
2) Normal Read (BL=4)
CLK
*2
RD
PRE
Q2
CMD
1
Q0
Q1
Q0
Q3
Q2
DQ(CL2)
2
Q1
Q3
DQ(CL3)
7. Auto Precharge
1) Normal Write (BL=4)
CLK
2) Normal Read (BL=4)
CLK
CMD
DQ
WR
D0
ACT
CMD
RD
DQ(CL2)
DQ(CL3)
Q0
Q1
Q0
Q2
Q1
Q3
Q2
D1
D2
D3
tRDL =2CLK
tDAL =tRDL + tRP
Q3
*4
*3
Auto Precharge Starts
*3
Auto Precharge Starts@tRDL=2CLK
*NOTES:
1. SAMSUNG can support tRDL=1CLK and tRDL=2CLK for all memory devices. SAMSUNG recommends tRDL=2 CLK.
2. Number of valid output data after row precharge : 1, 2 for CAS Latency = 2, 3 respectively.
3. The row active command of the precharge bank can be issued after tRP from this point.
The new read/write command of other activated bank can be issued from this point.
At burst read/write with auto precharge, CAS interrupt of the same bank is illegal
4. tDAL defined Last data in to Active delay. SAMSUNG can support tDAL=tRDL+ tRP .
Revision 0.1
August 2003
- 48 -
Advance
MCP MEMORY
Preliminary
KBE00D002M-F407
8. Burst Stop & Interrupted by Precharge
1) Normal Write
BL=4 & tRDL=2CLK
CLK
WR
D0
PRE
CMD
DQM
D1
D2
tRDL
DQ
*1
2) Write Burst Stop (BL=8)
CLK
3) Read Interrupted by Precharge (BL=4)
CLK
WR
STOP
RD
PRE
Q0
CMD
CMD
DQ(CL2)
DQ(CL3)
1
Q1
Q0
DQM
DQ
2
D0
D1
D2
D3
Q1
*2
tBDL
4) Read Burst Stop (BL=4)
CLK
RD
STOP
Q0
CMD
1
Q1
Q0
DQ(CL2)
DQ(CL3)
2
Q1
9. MRS
1) Mode Register Set
CLK
*4
CMD
PRE
MRS
ACT
tRP
2CLK
*NOTES:
1. SAMSUNG can support tRDL=1CLK and tRDL=2CLK for all memory devices. SAMSUNG recommends tRDL=2 CLK.
2. tBDL : 1 CLK ; Last data in to burst stop delay.
Read or write burst stop command is valid at every burst length.
3. Number of valid output data after row precharge or burst stop : 1, 2 for CAS latency= 2, 3 respectively.
4. PRE : All banks precharge is necessary.
MRS can be issued only at all banks precharge state.
Revision 0.1
August 2003
- 49 -
Advance
MCP MEMORY
Preliminary
KBE00D002M-F407
10. Clock Suspend Exit & Power Down Exit
1) Clock Suspend (=Active Power Down) Exit
CLK
2) Power Down (=Precharge Power Down) Exit
CLK
CKE
CKE
tSS
tSS
*2
*1
Internal
Internal
CLK
CLK
RD
NOP ACT
CMD
CMD
11. Auto Refresh & Self Refresh
Auto Refresh
An auto refresh command is issued by having CS, RAS and CAS held low with CKE and WE high at the rising edge of the
clock(CLK). All banks must be precharged and idle for tRP(min) before the auto refresh command is applied. No control of the external
address pins is required once this cycle has started because of the internal address counter. When the refresh cycle has completed,
all banks will be in the idle state. A delay between the auto refresh command and the next activate command or subsequent auto
refresh command must be greater than or equal to the tARFC(min).
CLK
Auto
Refresh
PRE
CMD
Command
CKE = High
tRP
tARFC(min) = 105ns
Self Refresh
A Self Refresh command is defined by having CS, RAS, CAS and CKE held low with WE high at the rising edge of the clock. Once
the self Refresh command is initiated, CKE must be held low to keep the device in Self Refresh mode. After 1 clock cycle from the self
refresh command, all of the external control signals including system clock(CLK) can be disabled except CKE. The clock is internally
disabled during Self Refresh operation to reduce power. To exit the Self Refresh mode, supply stable clock input before returning
CKE high, assert deselect or NOP command and then assert CKE high. In case that the system uses burst auto refresh during normal
opreation, it is recommended to use burst 4096 auto refresh cycle immediately before entering self refresh mode and after exiting in
self refresh mode. On the other hand, if the system uses the distributed auto refresh, the system only has to keep the refresh duty
cycle.
CLK
Stable Clock
Self
Refresh
ACT
NOP
Command
tSRFX(min) = 120ns
CKE
tSS
tSS
Revision 0.1
August 2003
- 50 -
Advance
MCP MEMORY
Preliminary
KBE00D002M-F407
12. About Burst Type Control
At MRS A3 = "0". See the BURST SEQUENCE TABLE. (BL=4, 8)
BL=1, 2, 4, 8 and full page.
Sequential Counting
Basic
MODE
At MRS A3 = "1". See the BURST SEQUENCE TABLE. (BL=4, 8)
BL=4, 8. At BL=1, 2 Interleave Counting = Sequential Counting.
Interleave Counting
Every cycle Read/Write Command with random column address can realize Random
Column Access.
That is similar to Extended Data Out (EDO) Operation of conventional DRAM.
Random
MODE
Random column Access
tCCD = 1 CLK
13. About Burst Length Control
At MRS A2,1,0 = "000".
At auto precharge, tRAS should not be violated.
1
At MRS A2,1,0 = "001".
At auto precharge, tRAS should not be violated.
2
Basic
MODE
4
8
At MRS A2,1,0 = "010".
At MRS A2,1,0 = "011".
At MRS A2,1,0 = "111".
Full Page
Wrap around mode(infinite burst length) should be stopped by burst stop.
RAS interrupt or CAS interrupt.
At MRS A9 = "1".
Read burst =1, 2, 4, 8, full page write Burst =1.
At auto precharge of write, tRAS should not be violated.
Special
MODE
BRSW
Random
MODE
tBDL= 1, Valid DQ after burst stop is 1, 2 for CAS latency 2, 3 respectively
Using burst stop command, any burst length control is possible.
Burst Stop
Before the end of burst, Row precharge command of the same bank stops read/write
burst with Row precharge.
RAS Interrupt
(Interrupted by Precharge) tRDL= 2 with DQM, valid DQ after burst stop is 1, 2 for CAS latency 2, 3 respectively.
During read/write burst with auto precharge, RAS interrupt can not be issued.
Interrupt
MODE
Before the end of burst, new read/write stops read/write burst and starts new
CAS Interrupt
read/write burst.
During read/write burst with auto precharge, CAS interrupt can not be issued.
Revision 0.1
August 2003
- 51 -
Advance
MCP MEMORY
Preliminary
KBE00D002M-F407
FUNCTION TRUTH TABLE (TABLE 1)
Current
State
CS
RAS
CAS
WE
BA
Address
Action
Note
H
L
L
L
L
L
L
L
H
L
L
L
L
L
L
L
H
L
L
L
L
L
L
L
H
L
L
L
L
L
L
L
H
L
L
L
L
L
H
L
L
L
L
L
X
H
H
H
L
X
H
H
L
X
H
L
X
X
X
X
X
NOP
NOP
X
ILLEGAL
2
2
X
H
L
BA
BA
BA
X
CA, A10/AP ILLEGAL
IDLE
H
H
L
RA
A10/AP
X
Row (& Bank) Active ; Latch RA
L
NOP
4
5
5
L
H
L
Auto Refresh or Self Refresh
L
L
OP code
X
OP code Mode Register Access
X
H
H
H
H
L
X
H
H
L
X
H
L
X
X
X
NOP
X
NOP
X
ILLEGAL
2
2
Row
Active
H
L
BA
BA
BA
BA
X
CA, A10/AP Begin Read ; latch CA ; determine AP
CA, A10/AP Begin Read ; latch CA ; determine AP
L
H
H
L
H
L
RA
ILLEGAL
L
A10/AP
Precharge
L
X
X
H
L
X
X
X
X
ILLEGAL
X
H
H
H
H
L
X
H
H
L
X
NOP (Continue Burst to End --> Row Active)
NOP (Continue Burst to End --> Row Active)
Term burst --> Row active
X
X
H
L
BA
BA
BA
BA
X
CA, A10/AP Term burst, New Read, Determine AP
CA, A10/AP Term burst, New Write, Determine AP
Read
L
3
2
H
H
L
H
L
RA
ILLEGAL
L
A10/AP
Term burst, Precharge timing for Reads
ILLEGAL
L
X
X
H
L
X
X
X
X
X
H
H
H
H
L
X
H
H
L
X
NOP (Continue Burst to End --> Row Active)
NOP (Continue Burst to End --> Row Active)
Term burst --> Row active
X
X
H
L
BA
BA
BA
BA
X
CA, A10/AP Term burst, New read, Determine AP
CA, A10/AP Term burst, New Write, Determine AP
3
3
2
3
Write
L
H
H
L
H
L
RA
ILLEGAL
L
A10/AP
Term burst, precharge timing for Writes
ILLEGAL
L
X
X
H
L
X
X
X
X
X
H
H
H
L
X
H
H
L
X
NOP (Continue Burst to End --> Precharge)
NOP (Continue Burst to End --> Precharge)
ILLEGAL
X
Read with
Auto
Precharge
X
X
X
X
X
H
L
BA
BA
X
CA, A10/AP ILLEGAL
RA, RA10 ILLEGAL
H
L
2
2
L
X
X
X
X
ILLEGAL
X
H
H
H
L
X
H
H
L
X
NOP (Continue Burst to End --> Precharge)
NOP (Continue Burst to End --> Precharge)
ILLEGAL
X
Write with
Auto
Precharge
X
X
X
X
BA
BA
X
CA, A10/AP ILLEGAL
RA, RA10 ILLEGAL
H
L
L
X
ILLEGAL
Revision 0.1
August 2003
- 52 -
Advance
MCP MEMORY
Preliminary
KBE00D002M-F407
FUNCTION TRUTH TABLE (TABLE 1)
Current
CS
H
L
RAS
X
H
H
H
L
CAS
X
H
H
L
WE
X
H
L
BA
X
Address
Action
NOP --> Idle after tRP
Note
X
X
X
NOP --> Idle after tRP
ILLEGAL
L
X
X
2
2
2
4
Precharging
L
X
H
L
BA
BA
BA
X
CA
ILLEGAL
L
H
H
L
RA
ILLEGAL
L
L
A10/AP
NOP --> Idle after tRP
ILLEGAL
L
L
X
X
H
L
X
H
L
X
H
H
H
L
X
H
H
L
X
X
NOP --> Row Active after tRCD
NOP --> Row Active after tRCD
ILLEGAL
X
X
Row
Activating
L
X
X
2
2
2
2
L
X
H
L
BA
BA
BA
X
CA
ILLEGAL
L
H
H
L
RA
ILLEGAL
L
L
A10/AP
ILLEGAL
L
L
X
X
X
X
X
X
X
H
L
X
X
X
X
X
X
X
X
X
X
X
ILLEGAL
H
L
X
H
H
L
X
H
L
X
NOP --> Idle after tRC
NOP --> Idle after tRC
ILLEGAL
X
Refreshing
L
X
L
H
L
X
ILLEGAL
L
L
X
ILLEGAL
H
L
X
H
H
H
L
X
H
H
L
X
NOP --> Idle after 2 clocks
NOP --> Idle after 2 clocks
ILLEGAL
Mode
Register
Accessing
X
L
X
L
X
X
X
ILLEGAL
L
X
X
ILLEGAL
Abbreviations : RA = Row Address
NOP = No Operation Command
BA = Bank Address
CA = Column Address
AP = Auto Precharge
*NOTES:
1. All entries assume the CKE was active (High) during the precharge clock and the current clock cycle.
2. Illegal to bank in specified state ; Function may be Iegal in the bank indicated by BA, depending on the
state of that bank.
3. Must satisfy bus contention, bus turn around, and/or write recovery requirements.
4. NOP to bank precharging or in idle state. May precharge bank indicated by BA (and A10/AP).
5. Illegal if any bank is not idle.
Revision 0.1
August 2003
- 53 -
Advance
MCP MEMORY
Preliminary
KBE00D002M-F407
FUNCTION TRUTH TABLE (TABLE 2)
Current
State
CKE
(n-1)
CKE
n
CS
RAS
CAS
WE
Address
Action
Note
H
L
X
H
H
H
H
H
L
X
H
L
X
X
H
H
H
L
X
X
H
H
L
X
X
H
L
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
RA
X
Exit Self Refresh --> Idle after tsRFX(ABI)
Exit Self Refresh --> Idle after tsRFX (ABI)
Exit Self Refresh --> Idle after tsRFX (ABI)
ILLEGAL
6
6
L
Self
Refresh
L
L
L
L
X
X
X
X
X
H
L
ILLEGAL
L
L
X
X
X
X
H
H
L
ILLEGAL
L
X
X
H
L
X
X
X
H
H
H
L
NOP (Maintain Self Refresh)
INVALID
H
L
X
H
H
H
H
H
L
Exit Power Down --> ABI
Exit Power Down --> ABI
ILLEGAL
All
Banks
Precharge
Power
L
7
7
L
L
L
L
X
X
X
X
X
H
L
ILLEGAL
Down
L
L
X
X
X
X
H
H
L
ILLEGAL
L
X
X
H
L
X
X
X
H
H
H
L
NOP (Maintain Low Power Mode)
Refer to Table 1
H
H
H
H
H
H
H
H
L
H
L
Enter Power Down
Enter Power Down
ILLEGAL
L
8
8
All
Banks
Idle
L
L
L
L
X
H
H
L
ILLEGAL
L
L
H
L
Row (& Bank) Active
Enter Self Refresh
L
L
L
8
L
L
L
L
OP Code Mode Register Access
L
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
NOP
H
H
L
H
L
Refer to Operations in Table 1
Begin Clock Suspend next cycle
Exit Clock Suspend next cycle
Maintain Clock Suspend
Any State
other than
Listed
9
9
H
L
above
L
Abbreviations : ABI = All Banks Idle, RA = Row Address
*NOTES:
6. CKE low to high transition is asynchronous.
7. CKE low to high transition is asynchronous if restarts internal clock.
A minimum setup time 1CLK + tSS must be satisfied before any command other than exit.
8. Power down and self refresh can be entered only from the all banks idle state.
9. Must be a legal command.
Revision 0.1
August 2003
- 54 -
Advance
MCP MEMORY
Preliminary
KBE00D002M-F407
Power Up Sequence
Single Bit Read - Write - Read Cycle(Same Page) @CAS Latency=3, Burst Length=1
Read & Write Cycle at Same Bank @Burst Length=4, tRDL=2CLK
Page Read & Write Cycle at Same Bank @Burst Length=4, tRDL=2CLK
Page Read Cycle at Different Bank @Burst Length=4
Page Write Cycle at Different Bank @Burst Length=4, tRDL=2CLK
Read & Write Cycle at Different Bank @Burst Length=4
Read & Write Cycle With Auto Precharge l @Burst Length=4
Read & Write Cycle With Auto Precharge ll @Burst Length=4
Clock Suspension & DQM Operation Cycle @CAS Letency=2, Burst Length=4
Read Interrupted by Precharge Command & Read Burst Stop Cycle @ Full Page Burst
Write Interrupted by Precharge Command & Write Burst Stop Cycle @ Full Page Burst, tRDL=2CLK
Burst Read Single bit Write Cycle @Burst Length =2
Active/precharge Power Dower Down Mode @CAS Latency=2 Burst Length=4
Self Refresh Entry & Exit Cycle & Exit Cycle
Mode Register Set Cycle and Auto Refresh Cycle
Extended Mode Register Set Cycle
Revision 0.1
August 2003
- 55 -
Advance
MCP MEMORY
Preliminary
KBE00D002M-F407
Power Up Sequence for Mobile SDRAM
0
1
2
3
4
5
6
7
8
9
10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25
CLOCK
CKE
Hi
CS
RAS
CAS
ADDR
BA0
BA1
Key
Key
RAa
A10/AP
DQ
RAa
Hi-Z
Hi-Z
WE
High level is necessary
tRP
DQM
tARFC
tARFC
Precharge
(All Bank)
Auto
Refresh
Auto
Refresh
Normal
MRS
Extended
MRS
Row Active
(A-Bank)
: Don’t care
*NOTES:
1. Apply power and attempt to maintain CKE at a high state and all other inputs may be undefined.
- Apply VDD before or at the same time as VDDQ.
2. Maintain stable power, stable clock and NOP input condition for a minimum of 200us.
3. Issue precharge commands for all banks of the devices.
4. Issue 2 or more auto-refresh commands.
5. Issue a mode register set command to initialize the mode register.
6. Issue a extended mode register set command to define DS or PASR operating type of the device after normal MRS.
EMRS cycle is not mandatory and the EMRS command needs to be issued only when DS or PASR is used.
The default state without EMRS command issued is half driver strength, all 4 banks refreshed.
The device is now ready for the operation selected by EMRS.
For operating with DS or PASR, set DS or PASR mode in EMRS setting stage.
In order to adjust another mode in the state of DS or PASR mode, additional EMRS set is required but power up sequence is not needed again
at this time. In that case, all banks have to be in idle state prior to adjusting EMRS set.
Revision 0.1
August 2003
- 56 -
Advance
MCP MEMORY
Preliminary
KBE00D002M-F407
Single Bit Read-Write-Read Cycle(Same Page) @CAS Latency=3, Burst Length=1
tCH
0
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
CLOCK
CKE
CS
tCC
tCL
HIGH
tSH
tRP
tRAS
tRC
*Note 1
tRCD
tSS
tSH
tSH
RAS
CAS
tSS
tSH
tSS
Ca
ADDR
BA0,BA1
A10/AP
Ra
Cb
Cc
Rb
tSS
*Note 2
BS
*Note 2,3
BS
*Note 2,3
BS
*Note 2,3 *Note 4
BS BS
*Note 2
BS
*Note 3
*Note 3
*Note 3 *Note 4
Ra
Rb
tSH
Db
tSAC
Qa
DQ
WE
Qc
tSS
tSS tSH
tSLZ
tOH
tSS tSH
DQM
Row Active
Read
Write
Read
Row Active
Precharge
: Don’t care
*NOTES:
1. All input except CKE & DQM can be don't care when CS is high at the CLK high going edge.
2. Bank active & read/write are controlled by BA0,BA1.
Revision 0.1
August 2003
- 57 -
Advance
MCP MEMORY
Preliminary
KBE00D002M-F407
Read & Write Cycle at Same Bank @Burst Length=4, tRDL=2CLK
0
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
CLOCK
CKE
HIGH
*Note 1
tRC
CS
RAS
*Note 2
CAS
ADDR
BA0
Ra
Ca
Rb
Cb
BA1
A10/AP
CL=2
Ra
Rb
tOH
Qa0 Qa1 Qa2 Qa3
Db0 Db1 Db2 Db3
Db0 Db1 Db2 Db3
tRDL
tSHZ
tRCD
tSAC
tOH
*Note 4
tSHZ
DQ
{
CL=3
Qa0 Qa1 Qa2 Qa3
tSAC
tRDL
*Note 4
WE
DQM
Row Active
(A-Bank)
Read
(A-Bank)
Precharge
(A-Bank)
Row Active
(A-Bank)
Write
(A-Bank)
Precharge
(A-Bank)
: Don’t care
*NOTES:
1. Minimum row cycle times is required to complete internal DRAM operation.
2. Row precharge can interrupt burst on any cycle. [CAS Latency - 1] number of valid output data
is available after Row precharge. Last valid output will be Hi-Z(tSHZ) after the clcok.
3. Ouput will be Hi-Z after the end of burst. (1, 2, 4, 8 & Full page bit burst)
Revision 0.1
August 2003
- 58 -
Advance
MCP MEMORY
Preliminary
KBE00D002M-F407
Page Read & Write Cycle at Same Bank @Burst Length=4, tRDL=2CLK
0
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
CLOCK
CKE
HIGH
CS
RAS
*Note 2
CAS
ADDR
BA0
Ra
Ca
Cb
Cc
Cd
Rb
BA1
A10/AP
CL=2
Ra
Rb
tRDL
Qa0 Qa1 Qb0 Qb1 Qb2
Dc0 Dc1 Dd0 Dd1
tRCD
tDAL
*Note 4
DQ
{
CL=3
Qa0 Qa1 Qb0 Qb1
Dc0 Dc1 Dd0 Dd1
tCDL
WE
*Note 1
*Note 3
DQM
Row Active
(A-Bank)
Read
(A-Bank)
Read
(A-Bank)
Write
(A-Bank)
Write
(A-Bank)
Precharge
(A-Bank)
Row Active
(A-Bank)
: Don’t care
*NOTES:
1. To write data before burst read ends, DQM should be asserted three cycle prior to write
command to avoid bus contention.
2. Row precharge will interrupt writing. Last data input, tRDL before Row precharge, will be written.
3. DQM should mask invalid input data on precharge command cycle when asserting precharge
before end of burst. Input data after Row precharge cycle will be masked internally.
4. tDAL ,last data in to active delay, is 2CLK + tRP.
Revision 0.1
August 2003
- 59 -
Advance
MCP MEMORY
Preliminary
KBE00D002M-F407
Page Read Cycle at Different Bank @Burst Length=4
0
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
CLOCK
CKE
HIGH
*Note 1
CS
RAS
*Note 2
CAS
ADDR
BA0
RAa
RBb CAa
RCc CBb
RDd CCc
CDd
BA1
A10/AP
CL=2
RAa
RBb
RCc
RDd
QAa0 QAa1 QAa2 QBb0 QBb1 QBb2 QCc0 QCc1 QCc2
QDd0 QDd1 QDd2
DQ
{
CL=3
QAa0 QAa1 QAa2 QBb0 QBb1 QBb2 QCc0 QCc1 QCc2 QDd0 QDd1 QDd2
WE
DQM
Row Active
(A-Bank)
Read
(A-Bank)
Read
(B-Bank)
Read
(C-Bank)
Read
(D-Bank)
Precharge
(D-Bank)
Row Active
(B-Bank)
Row Active
(C-Bank)
Row Active
(D-Bank)
Precharge
(C-Bank)
Precharge
(A-Bank)
Precharge
(B-Bank)
: Don’t care
*NOTES:
1. CS can be don't cared when RAS, CAS and WE are high at the clock high going dege.
2. To interrupt a burst read by row precharge, both the read and the precharge banks must be the same.
Revision 0.1
August 2003
- 60 -
Advance
MCP MEMORY
Preliminary
KBE00D002M-F407
Page Write Cycle at Different Bank @Burst Length=4, tRDL=2CLK
0
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
CLOCK
CKE
CS
HIGH
RAS
CAS
ADDR
BA0
*Note 2
RAa
RAb CAa
CBb RCc
RDd CCc
CDd
BA1
A10/AP
DQ
RAa
RBb
RCc
RDd
DAa0 DAa1 DAa2 DAa3 DBb0 DBb1 DBb2 DBb3 DCc0 DCc1 DDd0 DDd1 DDd2
tCDL
tRDL
WE
*Note 1
DQM
Row Active
(A-Bank)
Write
(A-Bank)
Write
(B-Bank)
Row Active
(D-Bank)
Write
(D-Bank)
Precharge
(All Banks)
Row Active
(B-Bank)
Row Active
(C-Bank)
Write
(C-Bank)
: Don’t care
*NOTES:
1. To interrupt burst write by Row precharge, DQM should be asserted to mask invalid input data.
2. To interrupt burst write by Row precharge, both the write and the precharge banks must be the same.
Revision 0.1
August 2003
- 61 -
Advance
MCP MEMORY
Preliminary
KBE00D002M-F407
Read & Write Cycle at Different Bank @Burst Length=4
0
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
CLOCK
CKE
HIGH
CS
RAS
CAS
ADDR
BA0
RAa
CAa
RDb
CDb RBc
CBc
BA1
A10/AP
CL=2
RAa
RDb
RBc
tCDL
*Note 1
QAa0 QAa1 QAa2 QAa3
DDb0 DDb1 DDb2 DDb3
DDb0 DDb1 DDb2 DDb3
QBc0 QBc1 QBc2
QBc0 QBc1
DQ
{
CL=3
QAa0 QAa1 QAa2 QAa3
WE
DQM
Row Active
(A-Bank)
Read
(A-Bank)
Precharge
(A-Bank)
Write
(D-Bank)
Read
(B-Bank)
Row Active
(D-Bank)
Row Active
(B-Bank)
: Don’t care
*NOTE:
1. tCDL should be met to complete write.
Revision 0.1
August 2003
- 62 -
Advance
MCP MEMORY
Preliminary
KBE00D002M-F407
Read & Write Cycle with Auto Precharge I @Burst Length=4
0
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
CLOCK
CKE
HIGH
CS
RAS
CAS
ADDR
BA0
RAa
RBb CAa
CBb
RAc
CAc
BA1
A10/AP
DQ CL=2
RAa
RBb
RAc
QAa0 QAa1 QBb0 QBb1 QBb2 DBb3
DAc0 DAc1
DAc0 DAc1
CL=3
QAa0 QAa1 QBb0 QBb1 QBb2 DBb3
WE
DQM
Row Active
(A-Bank)
Read with Read without Auto
Auto Pre Precharge(B-Bank)
Precharge
(B-Bank)
Row Active
(A-Bank)
Write with
Auto Precharge
(A-Bank)
charge
Auto Precharge
Start Point
(A-Bank)
(A-Bank) *Note1
Row Active
(B-Bank)
: Don’t care
*NOTE:
1. When Read(Write) command with auto precharge is issued at A-Bank after A and B Bank activation.
- if Read(Write) command without auto precharge is issued at B-Bank before A-Bank auto precharge starts, A-Bank
auto precharge will start at B-Bank read command input point .
- any command can not be issued at A-Bank during tRP after A-Bank auto precharge starts.
Revision 0.1
August 2003
- 63 -
Advance
MCP MEMORY
Preliminary
KBE00D002M-F407
Read & Write Cycle with Auto Precharge II @Burst Length=4
0
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
CLOCK
CKE
HIGH
CS
RAS
CAS
ADDR
BA0
Ra
Ca
Rb
Cb
BA1
A10/AP
DQ CL=2
Ra
Rb
Qa0 Qa1 Qa2 Qa3
Qb0 Qb1 Qb2 Qb3
CL=3
Qa0 Qa1 Qa2 Qa3
Qb0 Qb1 Qb2 Qb3
WE
DQM
*Note1
Auto Precharge
Start Point
Row Active
(A-Bank)
Read with
Auto Precharge
(A-Bank)
Read with
Auto Precharge
(B-Bank)
Auto Precharge
Start Point
(A-Bank)
(B-Bank)
Row Active
(B-Bank)
: Don’t care
*NOTE:
1. Any command to A-bank is not allowed in this period.
tRP is determined from at auto precharge start point
Revision 0.1
August 2003
- 64 -
Advance
MCP MEMORY
Preliminary
KBE00D002M-F407
Clock Suspension & DQM Operation Cycle @CAS Latency=2, Burst Length=4
0
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
CLOCK
CKE
CS
RAS
CAS
ADDR
BA0
Ra
Ca
Cb
Cc
BA1
A10/AP
DQ
Ra
Qa0 Qa1
Qa2
Qa3
Qb0 Qb1
tSHZ
Dc0
Dc2
tSHZ
WE
*Note 1
DQM
Row Active
Read
Clock
Suspension
Read
Read DQM
Write
DQM
Write
DQM
Write
Clock
Suspension
: Don’t care
*NOTE:
1. DQM is needed to prevent bus contention.
Revision 0.1
August 2003
- 65 -
Advance
MCP MEMORY
Preliminary
KBE00D002M-F407
Read Interrupted by Precharge Command & Read Burst Stop Cycle @Full Page Burst
0
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
CLOCK
CKE
HIGH
CS
RAS
CAS
ADDR
BA0
RAa
CAa
CAb
BA1
A10/AP
CL=2
RAa
1
1
QAb0 QAb1 QAb2 QAb3 QAb4 QAb5
QAa0 QAa1 QAa2 QAa3 QAa4
DQ
2
2
{
CL=3
QAa0 QAa1 QAa2 QAa3 QAa4
QAb0 QAb1 QAb2 QAb3 QAb4 QAb5
WE
DQM
Precharge
(A-Bank)
Row Active
(A-Bank)
Read
(A-Bank)
Burst Stop
Read
(A-Bank)
: Don’t care
*NOTES:
1. At full page mode, burst is finished by burst stop or precharge.
2. About the valid DQs after burst stop, it is same as the case of RAS interrupt.
Both cases are illustrated above timing diagram. See the label 1, 2 on them.
But at burst write, Burst stop and RAS interrupt should be compared carefully.
Refer the timing diagram of "Full page write burst stop cycle".
3. Burst stop is valid at every burst length.
Revision 0.1
August 2003
- 66 -
Advance
MCP MEMORY
Preliminary
KBE00D002M-F407
Write Interrupted by Precharge Command & Write Burst Stop Cycle @ Full Page Burst,
tRDL=2CLK
0
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
CLOCK
CKE
CS
HIGH
RAS
CAS
ADDR
BA0
RAa
CAa
CAb
BA1
A10/AP
DQ
RAa
tBDL
tRDL
*Note 1
*Note 1,2
DAa0 DAa1 DAa2 DAa3 DAa4
DAb0 DAb1 DAb2 DAb3 DAb4 DAb5
WE
DQM
Row Active
(A-Bank)
Write
(A-Bank)
Burst Stop
Write
(A-Bank)
Precharge
(A-Bank)
: Don’t care
*NOTES:
1. At full page mode, burst is finished by burst stop or precharge.
2. Data-in at the cycle of interrupted by precharge can not be written into the corresponding
memory cell. It is defined by AC parameter of tRDL.
DQM at write interrupted by precharge command is needed to prevent invalid write.
DQM should mask invalid input data on precharge command cycle when asserting precharge
before end of burst. Input data after Row precharge cycle will be masked internally.
3. Burst stop is valid at every burst length.
Revision 0.1
August 2003
- 67 -
Advance
MCP MEMORY
Preliminary
KBE00D002M-F407
Burst Read Single bit Write Cycle @Burst Length=2
0
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
CLOCK
CKE
HIGH
CS
RAS
*Note 2
CAS
ADDR
BA0
RAa
CAa RBb CAb
RCc
CBc
CCd
BA1
A10/AP
CL=2
RAa
RBb
RCc
DAa0
QAb0 QAb1
DBc0
DBc0
QCd0 QCd1
DQ
{
CL=3
DAa0
QAb0 QAb1
QCd0 QCd1
WE
DQM
Precharge
(C-Bank)
Row Active
(A-Bank)
Row Active
(B-Bank)
Row Active
(C-Bank)
Read
(C-Bank)
Write
Read with
Write with
Auto Precharge
(B-Bank)
(A-Bank) Auto Precharge
(A-Bank)
: Don’t care
*NOTES:
1. BRSW modes is enabled by setting A9 "High" at MRS (Mode Register Set).
At the BRSW Mode, the burst length at write is fixed to "1" regardless of programmed burst length.
2. When BRSW write command with auto precharge is executed, keep it in mind that tRAS should not be violated.
Auto precharge is executed at the burst-end cycle, so in the case of BRSW write command,
the next cycle starts the precharge.
Revision 0.1
August 2003
- 68 -
Advance
MCP MEMORY
Preliminary
KBE00D002M-F407
Active/Precharge Power Down Mode @CAS Latency=2, Burst Length=4
0
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
CLOCK
CKE
CS
tSS
*Note 1
tSS
tSS
*Note 2
*Note 2
*Note 3
RAS
CAS
ADDR
BA
Ra
Ca
Ra
A10/AP
DQ
Qa0 Qa1 Qa2
tSHZ
WE
DQM
Precharge
Power-down
Entry
Row Active
Read
Precharge
Precharge
Power-down
Exit
Active
Power-down
Entry
Active
Power-down
Exit
: Don’t care
*NOTES:
1. All banks should be in idle state prior to entering precharge power down mode.
2. CKE should be set high at least 1CLK + tSS prior to Row active command.
3. Can not violate minimum refresh specification. (64ms)
Revision 0.1
August 2003
- 69 -
Advance
MCP MEMORY
Preliminary
KBE00D002M-F407
Self Refresh Entry & Exit Cycle
0
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
CLOCK
CKE
*Note 2
*Note 4
tSRFX
*Note 1
*Note 6
*Note 3
tSS
CS
RAS
CAS
ADDR
BA0,BA1
A10/AP
DQ
Hi-Z
Hi-Z
WE
DQM
Self Refresh Entry
Self Refresh Exit
Auto Refresh
: Don’t care
*NOTES:
TO ENTER SELF REFRESH MODE
1. CS, RAS & CAS with CKE should be low at the same clcok cycle.
2. After 1 clock cycle, all the inputs including the system clock can be don't care except for CKE.
3. The device remains in self refresh mode as long as CKE stays "Low".
cf.) Once the device enters self refresh mode, minimum tRAS is required before exit from self refresh.
TO EXIT SELF REFRESH MODE
4. System clock restart and be stable before returning CKE high.
5. CS starts from high.
6. Minimum tRC is required after CKE going high to complete self refresh exit.
7. 4K cycle(64Mb ,128Mb) or 8K cycle(256Mb, 512Mb) of burst auto refresh is required before self refresh entry and
after self refresh exit if the system uses burst refresh.
Revision 0.1
August 2003
- 70 -
Advance
MCP MEMORY
Preliminary
KBE00D002M-F407
Mode Register Set Cycle
Auto Refresh Cycle
0
1
2
3
4
5
6
0
1
2
3
4
5
6
7
8
9
10
CLOCK
CKE
CS
HIGH
HIGH
*Note 2
tARFC
RAS
CAS
ADDR
BA0
*Note 1
*Note 3
Key
Ra
BA1
DQ
Hi-Z
Hi-Z
WE
DQM
MRS
New Command
Auto Refresh
New Command
: Don’t care
* All banks precharge should be completed before Mode Register Set cycle and auto refresh cycle.
*NOTES:
MODE REGISTER SET CYCLE
1. CS, RAS, CAS, BA0, BA1 & WE activation at the same clock cycle with address key will set internal mode register.
2. Minimum 2 clock cycles should be met before new RAS activation.
3. Please refer to Mode Register Set table.
Revision 0.1
August 2003
- 71 -
Advance
MCP MEMORY
Preliminary
KBE00D002M-F407
Extended Mode Register Set Cycle
0
1
2
3
4
5
6
CLOCK
CKE
CS
HIGH
*Note 2
RAS
CAS
ADDR
BA0
*Note 1
*Note 3
Key
Ra
BA1
Hi-Z
DQ
WE
DQM
EMRS New Command
: Don’t care
*NOTES:
EXTENDED MODE REGISTER SET CYCLE
1. CS, RAS, CAS, BA0, BA1 & WE activation at the same clock cycle with address key will set internal mode register.
2. Minimum 2 clock cycles should be met before new RAS activation.
3. Please refer to Mode Register Set table.
Revision 0.1
August 2003
- 72 -
Advance
MCP MEMORY
Preliminary
KBE00D002M-F407
PACKAGE DIMENSION
137-Ball Fine pitch Ball Grid Array Package (measured in millimeters)
Units:millimeters
#A1 INDEX MARK
10.50±0.10
A
0.10 MAX
10.50±0.10
0.80x9=7.20
(Datum A)
B
10 9
8 7 6 5 4 3 2 1
0.80
A
B
C
D
#A1
(Datum B)
E
F
G
H
J
K
L
M
N
P
R
3.60
0.32±0.05
1.30±0.10
137-Æ0.45±0.05
BOTTOM VIEW
TOP VIEW
Æ
0.20
M A B
Revision 0.1
August 2003
- 73 -
相关型号:
KBE00D002M-F4070
Memory Circuit, 16MX16, CMOS, PBGA137, 10.50 X 13 MM, 0.80 MM PITCH, FBGA-137
SAMSUNG
KBE00F005A-D4110
Memory Circuit, Flash+SDRAM, 8MX32, CMOS, PBGA137, 10.50 X 13 MM, 0.80 MM PITCH, LEAD FREE, FBGA-137
SAMSUNG
KBE00S009M-D4110
Memory Circuit, 16MX16, CMOS, PBGA137, 12 X 14 MM, 0.80 MM PITCH, LEAD FREE, FBGA-137
SAMSUNG
©2020 ICPDF网 联系我们和版权申明