U.S. patent application number 14/831016 was filed with the patent office on 2015-12-10 for stack bank type semiconductor memory apparatus capable of improving alignment margin.
The applicant listed for this patent is SK hynix Inc.. Invention is credited to Seung Wook KWAK, Sang Hoon SHIN, Keun Soo SONG.
Application Number | 20150357003 14/831016 |
Document ID | / |
Family ID | 40623583 |
Filed Date | 2015-12-10 |
United States Patent
Application |
20150357003 |
Kind Code |
A1 |
KWAK; Seung Wook ; et
al. |
December 10, 2015 |
STACK BANK TYPE SEMICONDUCTOR MEMORY APPARATUS CAPABLE OF IMPROVING
ALIGNMENT MARGIN
Abstract
A semiconductor memory apparatus is capable of improving the
alignment margin for a bank and sufficiently ensuring a space for
forming a global input/output line. The semiconductor memory
apparatus includes a stack bank structure having at least two
sub-banks continuously stacked without disconnection of data signal
lines, and a control block arranged at one side of the stack bank
structure to simultaneously control column-related signals of the
sub-banks.
Inventors: |
KWAK; Seung Wook; (Icheon-si
Gyeonggi-do, KR) ; SHIN; Sang Hoon; (Icheon-si
Gyeonggi-do, KR) ; SONG; Keun Soo; (Icheon-si
Gyeonggi-do, KR) |
|
Applicant: |
Name |
City |
State |
Country |
Type |
SK hynix Inc. |
Icheon-si Gyeonggi-do |
|
KR |
|
|
Family ID: |
40623583 |
Appl. No.: |
14/831016 |
Filed: |
August 20, 2015 |
Related U.S. Patent Documents
|
|
|
|
|
|
Application
Number |
Filing Date |
Patent Number |
|
|
12169595 |
Jul 8, 2008 |
9123395 |
|
|
14831016 |
|
|
|
|
Current U.S.
Class: |
365/51 |
Current CPC
Class: |
G11C 5/02 20130101; G11C
8/10 20130101; G11C 7/10 20130101; G11C 5/063 20130101; G11C 5/025
20130101 |
International
Class: |
G11C 5/02 20060101
G11C005/02; G11C 7/10 20060101 G11C007/10; G11C 8/10 20060101
G11C008/10 |
Foreign Application Data
Date |
Code |
Application Number |
Nov 9, 2007 |
KR |
10-2007-0114146 |
Nov 12, 2007 |
KR |
10-2007-0114944 |
Nov 13, 2007 |
KR |
10-2007-0115462 |
Claims
1. A semiconductor memory apparatus comprising: a plurality of
stack bank structures having at least two sub-banks continuously
arranged without disconnection of data signal lines; a share block
having a predecoder and a fuse set interposed between sub-banks in
the stack bank structure; and a main decoder interposed between the
sub-bank and the share block.
2. The semiconductor memory apparatus of claim 1, wherein the
sub-bank includes a mat array having a plurality of mat rows and a
plurality of mat columns, and wherein the predecoder and the fuse
set are alternately arranged in the share block based on an
arrangement of the mat rows.
3. The semiconductor memory apparatus of claim 2, wherein the
predecoder is configured to predecode column-related signals of the
sub-banks, and wherein the fuse set includes a signal line
configured to repair a column-related signal line.
4. The semiconductor memory apparatus of claim 3, wherein the fuse
set is configured to receive information on the column-related
signal line, and provide a control signal for the predecoder.
5. The semiconductor memory apparatus of claim 1, further
comprising a plurality of global input/output lines arranged
between the stack bank structures.
6. The semiconductor memory apparatus of claim 1, further
comprising a plurality of Input/Output control block arranged
between the stack bank structure and the global input/output line.
Description
CROSS-REFERENCES TO RELATED APPLICATION
[0001] The present application claims priority under 35 U.S.C.
119(a) to Korean application numbers 10-2007-0114146 filed on Nov.
9, 2007, 10-2007-0114944 filed on Nov. 12, 2007 and 10-2007-0115462
filed on Nov. 13, 2007 in the Korean Intellectual Property office,
each of which is incorporated by reference in its entirety as if
set forth in full.
BACKGROUND
[0002] 1. Technical Field
[0003] The embodiments described herein relate to a stack bank type
semiconductor memory apparatus, and more particularly, to a
semiconductor memory including a plurality of banks having a
plurality of sub-banks.
[0004] 2. Related Art
[0005] A conventional semiconductor memory apparatus includes a
plurality of memory cells and a circuit for controlling the memory
cells. At present, a bank concept has been introduced to control a
plurality of memory cells by classifying the memory cells into
groups. A bank represents an area that includes a plurality of
memory cells. The memory cells are grouped into banks and
controlled to improve the signal transmission characteristics of
the semiconductor memory apparatus.
[0006] Recently, as the number of memory cells included in
semiconductor memory apparatus has increased significantly, a
multi-bank scheme has been proposed to control banks by dividing
the banks into sub-banks.
[0007] FIG. 1 is a plan view illustrating a conventional
semiconductor memory apparatus employing such a multi-bank scheme.
Referring to FIG. 1, a semiconductor chip 10 may be divided into
four banks 12a to 12d. The banks 12a to 12d are spaced apart from
each other and a peripheral area 14 is interposed therebetween. For
example, each of the banks 12a to 12d may be divided into an up
bank UP and a down bank DOWN about a half line HL. Each up bank may
be divided into four sub-banks 15 and each down bank may also be
divided into the four sub-banks 15.
[0008] Each sub-bank 15 includes a plurality of word lines, a
plurality of bit lines crossing the word lines, and a plurality of
memory cells defined by the word and bit lines. The word and bit
lines may extend in the y and x directions of FIG. 1.
[0009] At the present time, the semiconductor memory apparatus
performs hierarchical data input/output. To this end, the
semiconductor memory apparatus employs a plurality of data bus
lines. A conventional semiconductor memory apparatus hierarchically
transfers data, which is loaded on a bit line, to a
sub-input/output (SIO, not shown) line, a local input/output (LIO,
not shown) line, and a global input/output (GIO, not shown) line.
The global input/output line is arranged between the sub-banks 15
perpendicular to the extension direction of the bit line.
[0010] Further, Y-control blocks 20 are arranged between the
sub-banks 15 adjacent to the global input/output line to control a
`Yi` signal of a corresponding sub-bank 15, respectively.
Furthermore, an X-hole 25, which includes circuits used for driving
the word line, is arranged between the sub-banks 15 perpendicular
to the Y-control block 20.
[0011] FIG. 2 is an enlarged plan view illustrating one bank in
FIG. 1. AS can be seen, one fuse set 23 is installed in each
sub-bank 15 to repair a memory cell defect occurring between the
Y-control block 20 and the global input/output line. Further,
circuits (not shown) are arranged in the peripheral area 14 to
control the banks 12.
[0012] Each bank 12a of the semiconductor chip 10 receives commands
and signals from the control circuits arranged in the peripheral
area 14. However, the number of memory cells integrated in the
banks 12 is increased due to the increase in the integration degree
of conventional semiconductor memory apparatus. The increased
number of memory cells increases the area of the sub-banks 15 and
the area of the banks 12 including the sub-banks 15. In addition,
the area of the blocks 20 and 25, which control the banks 12, must
also increase.
[0013] Therefore, the alignment margin between the banks 12 is
insufficient, and a sufficient gap between the sub-banks 15 may not
be sufficiently ensured. The reduction in the gap between the
sub-banks 15 may cause a reduction in the line width and spacing
related to the global input/output line. The reduction in the
spacing related to the global input/output line causes crosstalk,
and the reduction in the line width of the global input/output line
causes a signal delay.
SUMMARY
[0014] A semiconductor memory apparatus capable of improving the
alignment margin for a bank is described herein. A semiconductor
memory apparatus capable of sufficiently ensuring a space for
forming a global input/output line is also described herein.
[0015] According to one aspect, a semiconductor memory apparatus
includes a stack bank structure having at least two sub-banks
continuously stacked, and a control block arranged at one side of
the stack bank structure to simultaneously control column-related
signals of the sub-banks. Wherein the data signal lines of the
sub-banks for constructing the stack bank structure, which receive
the same signal, are continuously connected each other.
[0016] According to another aspect, a semiconductor memory
apparatus includes a plurality of banks divided based on a
peripheral area, a plurality of stack bank structures arranged in
the banks, being spaced apart from each other at a predetermined
interval, wherein the stack bank structure includes a plurality of
sub-banks, control blocks arranged at one side of the stack bank
structure to control all column-related signals of the sub-banks
constituting the stack bank structure, and a plurality of global
input/output lines arranged between the stack bank structures.
[0017] According to still another aspect, a semiconductor memory
apparatus includes a plurality of banks having a plurality of word
lines, and a plurality of bit lines crossing the word lines, and a
peripheral area dividing the banks and providing each bank with a
control signal. One bank is divided into an up bank and a down
bank, and each of the up bank and the down bank has a plurality of
sub-banks arranged in a form of a matrix. Within the up and down
banks, the sub-banks located at rows or columns parallel to an
extension direction of the bit line are arranged while interposing
a decoding block therebetween without disconnection of signal
lines, thereby forming stack bank structures. Control blocks are
provided at opposite sides of the stack bank structures to control
bit line selection of all sub-banks constituting the stack bank
structure, and a global input/output line is arranged between the
control blocks to be used for data input/output of the sub-banks
constituting the stack bank structure.
[0018] According to still another aspect, a semiconductor memory
apparatus includes a plurality of stack bank structures having at
least two sub-banks continuously arranged without disconnection of
data signal lines, a plurality of global input/output lines
arranged between the stack bank structures, a share block having a
predecoder and a fuse set interposed between sub-banks in the stack
bank structure, and a main decoder interposed between the sub-bank
and the share block.
[0019] According to still another aspect, a semiconductor memory
apparatus includes a pair of sub-banks constituting one bank and
having a plurality of mat rows and a plurality of mat columns
formed by a plurality of mats, a pair of data bus lines arranged in
a space between the mat rows of the sub-bank and a plurality of
precharge units arranged between the sub-banks and in a space
between the mat columns to be electrically connected with the data
bus lines of the sub-banks.
[0020] According to still another aspect, a semiconductor memory
apparatus comprises a plurality of stack bank structures having at
least two sub-banks continuously arranged without disconnection of
data signal lines, a plurality of global input/output lines
arranged between the stack bank structures, a share block having a
predecoder and a fuse set interposed between sub-banks in the stack
bank structure, a main decoder interposed between the sub-bank and
the share block, and a precharge unit interposed between the
predecoder and the fuse set in the share block.
[0021] These and other features, aspects, and embodiments are
described below in the section entitled "Detailed Description."
BRIEF DESCRIPTION OF THE DRAWINGS
[0022] The above and other aspects, features and other advantages
of the subject matter of the present disclosure will be more
clearly understood from the following detailed description taken in
conjunction with the accompanying drawings, in which:
[0023] FIG. 1 is a plan view illustrating a conventional
semiconductor memory apparatus employing a multi-bank scheme;
[0024] FIG. 2 is an enlarged plan view illustrating one bank of the
semiconductor memory apparatus shown in FIG. 1;
[0025] FIG. 3 is a plan view illustrating a stack bank type
semiconductor memory apparatus according to one embodiment;
[0026] FIG. 4 is an enlarged plan view illustrating one stack bank
of the semiconductor memory apparatus shown in FIG. 3 according to
one embodiment;
[0027] FIG. 5 is an enlarged plan view illustrating one stack bank
of the semiconductor memory apparatus shown in FIG. 3 according to
another embodiment;
[0028] FIG. 6 is a plan view illustrating a stack bank type
semiconductor memory apparatus according to another embodiment;
[0029] FIG. 7 is an enlarged plan view illustrating one stack bank
of the semiconductor memory apparatus shown in FIG. 6;
[0030] FIG. 8 is an enlarged plan view illustrating an "A" part
that can be included in the one stack bank shown in FIG. 7;
[0031] FIG. 9 is a plan view illustrating a stack bank type
semiconductor memory apparatus according to still another
embodiment;
[0032] FIG. 10 is a plan view illustrating a general bank having a
precharge block;
[0033] FIG. 11 is a timing diagram illustrating a reset interval of
a local input/output line in a conventional semiconductor memory
apparatus;
[0034] FIG. 12 is a plan view illustrating a stack bank type
semiconductor memory apparatus according to still another
embodiment;
[0035] FIG. 13 is an enlarged plan view illustrating one stack bank
of the semiconductor memory apparatus shown in FIG. 12;
[0036] FIG. 14 is a circuit diagram illustrating a precharge unit
according to an embodiment; and
[0037] FIG. 15 is a timing diagram illustrating a reset interval of
a local input/output line in a semiconductor memory apparatus
according to an embodiment.
DETAILED DESCRIPTION
[0038] FIG. 3 is a diagram illustrating a semiconductor chip 100,
e.g., a 512M DRAM, in accordance with one embodiment. Referring to
FIG. 3, the semiconductor chip 100 can be divided into four banks
110. Each bank 110 can include a plurality of word lines, a
plurality of bit lines crossing the word lines, and a plurality of
memory cells defined by the word and bit lines. In this example, it
is assumed that the word and bit lines WL and BL extend in the y
and x directions of FIG. 3, respectively. A peripheral area can be
made up of a first peripheral area 120a, which isolates the banks
110 in the x direction, and a second peripheral area 120b that
isolates the banks 110 in the y direction.
[0039] Further, the second peripheral area 120b can be divided into
a center area CPERI and a peripheral area DPERIL. The center area
CPERI can be located at the center of the semiconductor chip 100
and can include circuits configured to receive commands for driving
the DRAM. The peripheral area DPERIL can be located between the
banks 110 to allow data pads (not shown) to be arranged
therein.
[0040] Each bank 110 can be divided into half banks, i.e., an up
bank 110u and a down bank 110d, about a virtual half line HL. The
half line HL can be parallel to the word lines. Further, each of
the up bank 110u and the down bank 110d can include four sub-banks
130 (B0-B3 and B4-B7), i.e. quarters. Thus, one bank 110 can
include eight sub-banks 130 (B0-B7).
[0041] As shown in FIGS. 3 and 4, the semiconductor memory
apparatus 100 can be designed such that two sub-banks 130 share one
Y-control block (Y_CTRL) 140. In detail, the semiconductor memory
apparatus 100 can be designed such that the sub-banks 130 are
continuously arranged in the same rows (r1 and r2 or the same
columns) within one half bank 110u or 110d, and the continuously
arranged sub-banks 130 share one Y-control block 140 without
uncoupling of signal lines.
[0042] When the sub-banks 130 are continuously arranged as
described above, it will be referred to as "the sub-banks are
stacked", in which the "stack" means the status in which sub-banks
different from each other have been continuously arranged without
uncoupling or disconnection of the signal lines (e.g. bit lines).
Further, the row or column can be parallel to the bit line, and the
same row or column represents that the row or column exists on the
extension line of the same bit line. In addition, the stacked
sub-banks 130 will be referred to as a stack bank structure
135.
[0043] According to the present embodiment, a stack bank structure
135 can include two sub-banks 130. However, it will be understood
that more than two sub-banks 130 can be included in a stack bank
structure 135.
[0044] The Y-control block 140 can be located at an edge of one
side of the stack bank structure 135. For example, since the
Y-control block 140 includes circuits for controlling a `Yi` signal
used for driving the bit line BL, the Y-control block 140 can be
located at the edge of the stack bank structure 135 that is
perpendicular to the bit line BL. The Y-control block 140 can
include an address control block (not shown) and an Input/output
control block (not shown). Here, the address control block can be a
decoding block and the address control block can be a controller of
input/output sense amplifier, a read driver and a write driver.
[0045] A plurality of global input/output lines can be arranged
between the Y-control blocks 140. For example, the global
input/output (GIO) lines can be located adjacently to the half line
that separate the up bank 110u from the down bank 110d.
[0046] A predetermined number of global input/output (GIO) lines
can be allocated to each stack bank structure 135, and the
allocated global input/output (GIO) lines can be used for
transferring data to and from the sub-banks 130 constituting the
stack bank structure 135. Thus, as compared with a conventional
circuit in which the global input/output lines are allowed to each
sub-bank 130, in the embodiment of FIGS. 3 and 4, the global
input/output (GIO) lines can be allowed to each stack bank
structure 135, so that the total number of the global input/output
(GIO) lines can be reduced.
[0047] In a conventional circuit, two global input/output lines
have been necessary for each sub-bank 130. In general, eight global
input/output lines at minimum have been necessary for the sub-banks
130. However, in the embodiment of FIGS. 4 and 5, only two global
input/output (GIO) lines are necessary for each stack bank
structure 135. Thus, the number of the global input/output lines
can be reduced per each bank, so that the alignment margin for the
global input/output lines can be improved. Further, since there is
no need to arrange the global input/output line between the
sub-banks 130, the size of the sub-banks 130 can be increased
without increasing the overall area of the circuit 100.
[0048] In the embodiment of FIGS. 3 and 4, two sub-banks 130 are
stacked to share one Y-control block 140. Further, the global
input/output (GIO) line arranged between the sub-banks 130 is
shifted and arranged between the stack bank structures 135. Thus a
gap between the sub-banks 130 can be decrease.
[0049] When two sub-banks 130 are stacked in such a manner, an area
corresponding to one Y-control block 140 can be reduced, so that
the area of the sub-bank 130 can be increased by the area of one
Y-control block 140. Further, the area needed for a Y-control block
140 can be provided between the stack bank structures 135 that
ensures enough space for forming the global input/output (GIO)
line.
[0050] Further, since the number of global input/output (GIO) lines
is reduced, the alignment margin for the global input/output (GIO)
lines can be improved. Consequently, since the alignment margin (or
arrangement area) for the global input/output (GIO) lines can be
ensured, the line width of the global input/output (GIO) lines not
an issue, which can prevent signal delay. Further, the spacing
related to the global input/output (GIO) lines can also be
sufficiently ensured, thereby preventing crosstalk from
occurring.
[0051] Meanwhile, referring to FIG. 5, in the stack bank structure
135, a part of the Y-control block 140 can be interposed between
the sub-banks 130. For example, the address control block of the
Y-control block 140 can be interposed between the sub-banks 130 and
the input/output control block 142 of the Y-control block can be
located at the edge of one side of the stack bank structure 135.
The address control block can be a decoding block 200. The
input/output control block can include a controller of an
input/output sense amplifier (not shown), a read driver (not shown)
and a write driver (not shown). The decoding block 200 can include
a main decoder 210 and a predecorder 220 and can decode `Yi`
signals of the sub-banks 130.
[0052] In general, the stacked sub-banks 130 can share one decoding
block 200, similarly to manner in which one Y-control block 140 is
shared. Thus, sufficient area for each decoding block 200 can be
ensured. Further, since each decoding block 200 is arranged between
the sub-banks 130, the efficiency of the address signal can be
improved.
[0053] Further, as shown in FIGS. 6 and 7, the sub-banks 130
constituting the stack bank structure 135 can share the predecoder
220 and a fuse set 230. Hereinafter, the predecoder 220 and the
fuse set 230 will be referred to as a share block 240. In detail,
the main decoders 210 can be provided between opposite edges of the
sub-banks 130 constituting the stack bank structure 135, and the
share block 240 can be provided between the main decoders 210.
[0054] As is generally known, the predecoder 220 can predecode the
`Yi` signal for selecting the bit line BL, and the fuse set 230 can
include a repair Yi signal transmission line for repairing a defect
of a Yi signal transmission line, which occurs in the memory cells
constituting the sub-banks 130.
[0055] As shown in FIG. 8, the predecoder 220 and the fuse set 230
can be alternately arranged in the share block 240, e.g., in the
vertical direction (e.g. word line extension direction). The
predecoder 220 and the fuse set 230 can be alternately arranged
corresponding to mat rows Mr1 to Mr3, respectively. Such an
arrangement can facilitate transmission of the `Yi` signal between
the fuse set 230 and the predecoder 220 and reduce the length of a
line for transmitting information on a defective `Yi` signal.
[0056] Conventional fuse sets are configured such that they
correspond to entire mat rows. In the embodiments described herein,
however, since the fuse sets 230 are selectively arranged
corresponding to odd mat rows or even mat rows, the number of the
fuse sets can be reduced by a half as compared with the number of
conventional fuse sets 230. In general, since a defect of the `Yi`
signal does not occur relative to entire mats 132 in the sub-bank
130, problems may not occur in a repair operation even if the
number of the fuse sets 230 is reduced by 1/2.
[0057] Further, as shown in FIG. 9, the sub-banks 130 constituting
the stack bank structure 135 can simultaneously share the share
block 240, which can include the predecoder 220 and the fuse set
230, and the Input/Output controller 142 of the Y-control block.
The Input/Output controller 142 of Y-control block can be arranged
at the edge of the stack bank structure 135 perpendicular to the
bit lines BL.
[0058] According to the embodiments described herein, the sub-banks
135 located on the extension line of the same bit lines can be
stacked in the half banks 110u and 110d while sharing the
predecoder 220 and the fuse set 230. Thus, the area needed for the
corresponding to the predecoder 220 and the fuse set 230 can be
ensured in the banks as well as the half banks, even in the face of
increased integration.
[0059] Further, since the predecoder 220 and the fuse set 230 can
be alternately arranged corresponding to the mat rows constituting
the sub-bank in the share block space, mutual control signals
(repair `Yi` signals) can be easily transmitted and thus the length
of the signal line can be reduced.
[0060] Furthermore, the sub-banks can be stacked so that the global
input/output (GIO) lines can be collected between the stack bank
structures. Thus, there is no need for ensuring a space between the
sub-banks to accommodate the global input/output (GIO) lines. In
addition, the global input/output (GIO) lines arranged
corresponding to the sub-banks can be arranged corresponding to the
stack bank structures, so that the number of the global
input/output (GIO) lines can be reduced.
[0061] Moreover, as shown in FIG. 10, a conventional semiconductor
memory apparatus is designed such that one precharge controller 50
and one precharge block 60 are provided per sub-bank 15. In
addition, the precharge controller 50 and the precharge block 60
are positioned on either side of the sub-bank 15, and are
electrically interconnected through a precharge interconnection 70.
Thus, since the length of the precharge interconnection 70 is
greater than that of the sub-bank 15, serious signal skew and delay
occur. In FIG. 10, reference LIO denotes a local input/output
signal line and reference LIOb denotes an inverted local
input/output signal line.
[0062] FIG. 11 is a timing diagram illustrating a reset interval of
an input/output line when the precharge controller and the
precharge block are positioned as illustrated in FIG. 10. Referring
to FIGS. 10 and 11, when the precharge controller 50 and the
precharge block 60 are spaced apart from each other by the length
of the sub-bank 15, since a signal transmission path is increased,
a skew may occur in a precharge control signal. Therefore, a reset
interval (a') of the local input/output line LIO is narrower than a
desired interval (a).
[0063] In the embodiments described herein, as shown in FIG. 12, a
precharge block 300 can be provided between the stacked sub-banks
130. Thus, the sub-banks 130 constituting the stack bank structure
135 share one precharge block 300. The "stack arrangement" of the
sub-banks means that the sub-banks 130 are continuously arranged
without electrical uncoupling or disconnection of data lines, e.g.
the local input/output lines LIO and LIOb. Further, if there is no
disconnection of data lines between sub-banks, the sub-banks can be
referred to as being in a "stack arrangement" as described herein
even if the sub-banks are spaced apart from each other at a
predetermined interval. That is, if the data lines are connected
with each other through a connection medium, the sub-banks can be
referred to as being in a "stack arrangement".
[0064] The precharge block 300 can include a plurality of precharge
units 310 connected with the local input/output lines LIO and LIOb
of the sub-bank 130, respectively. A precharge controller 320 can
be provided in at least one of both sides of the precharge block
300 to receive a precharge command from a peripheral circuit area
(not shown) and then provide the precharge block 300 with control
signals `LIOpcg_UP` and `LIOpcg_DN`. Since the precharge controller
320 is provided in at least one of both sides of the precharge
block 300, a path for receiving the control signals can be
significantly shortened. For example, the prechare controller 320
can be located at the peripheral area. Consequently, the signal
delay and signal skew can be reduced or eliminated.
[0065] In more detail, as shown in FIG. 13, each sub-bank 130 can
have mats 115 arranged in the form of a matrix including a
plurality of mat rows Mr1 to Mr3 and a plurality of mat columns. As
will be understood, the mats 115 can include a set of a plurality
of word lines (not shown), a plurality of bit lines (not shown),
and a plurality of memory cells. The mats 115 can be spaced apart
from each other at a predetermined interval, and a pair of the
local input/output lines LIO and LIOb can be arranged in the space
between the mats 115, respectively, i.e., in the space
corresponding to the direction parallel to bit lines (not shown)
constituting the mats 115.
[0066] Meanwhile, the main decoders 210 can be arranged at opposite
sides of the sub-banks 130 to generate the column control signals
('Yi' signals).
[0067] Further, the predecoder 220, the fuse set 230 and the
precharge unit 310 can be arranged in the space between the
opposite main decoders 210. The predecoder 220 and the fuse set 230
can be alternately arranged corresponding to the mat rows Mr1 to
Mr3, and the precharge unit 310 can be arranged between the
predecoder 220 and the fuse set 230 corresponding to the local
input/output lines LIO and LIOb.
[0068] The fuse set 230 can be used for repairing a defect, which
occurs in a column line of a corresponding mat column, and can
include a first fuse (Fu) 230a, which controls a defect of a
corresponding mat row of a first sub-bank 130u, and a second fuse
(Fd) 230b that controls a defect of a corresponding mat row of a
second sub-bank 130d. A fuse controller (not shown) can be
interposed between the first and second fuses 230a and 230b.
[0069] As described above, the precharge units 310 can be arranged
between the fuses 230a and 230b and the predecoder 220,
respectively. In detail, the precharge units 310 can be arranged in
the spaces among mat rows Mr1 to Mr3 in which the local
input/output lines LIO and LIOb are arranged. Thus, the precharge
units 310 can be connected with the local input/output lines LIO
and LIOb so that the first and second sub-banks 130u and 130d can
be stacked without substantial disconnection of the local
input/output lines LIO and LIOb.
[0070] When the control signals `LIOpcg_UP` and `LIOpcg_DN` are
enabled, the precharge unit 310 can precharge the local
input/output lines LIO and LIOb by using predetermined voltage,
e.g., bit line precharge voltage VBLP.
[0071] For example, as shown in FIG. 14, the precharge unit 310 can
include first to third MOS transistors N1 to N3. The first MOS
transistor N1 can have a gate for receiving the precharge control
signal `LIOpcg_UP` or `LIOpcg_DN`, a source connected with the
local input/output line LIOb, and a drain for receiving the bit
line precharge voltage VBLP. The second MOS transistor N2 can have
a gate for receiving the precharge control signal `LIOpcg_UP` or
`LIOpcg_DN`, a source connected with the local input/output line
LIO, and a drain for receiving the bit line precharge voltage VBLP.
The third MOS transistor N3 can have a gate for receiving the
precharge control signal `LIOpcg_UP` or `LIOpcg_DN`, a source
connected with the local input/output line LIOb, and a drain
connected with the local input/output line LIO.
[0072] As described above, the Input/Output controller 140 of the
Y-control block can be located at one edge of the stack bank
structure 135.
[0073] The precharge block 300 including the precharge units 310
having the afore-described construction can be arranged in the
share block 240 between the stacked sub-banks, so that an area
corresponding to one precharge block 300 can be saved. Further, the
precharge controller 320 can be provided in at least one side of
the precharge block 300, so that the transmission path for the
precharge control signal can be reduced.
[0074] FIG. 15 is a timing chart illustrating a reset interval of
the local input/output line for the embodiments described
herein.
[0075] Referring to FIG. 15, since the signal skew is significantly
reduced as the transmission path for the precharge control signal
is reduced, the reset interval LIO_rst/ of the local input/output
lines LIO and LIOb occupies the entire space limited to a precharge
interval PCG. Thus, the reset interval of the local input/output
lines LIO and LIOb can be sufficiently ensured.
[0076] While certain embodiments have been described above, it will
be understood that the embodiments described are by way of example
only. Accordingly, the systems and methods described herein should
not be limited based on the described embodiments. Rather, the
systems and methods described herein should only be limited in
light of the claims that follow when taken in conjunction with the
above description and accompanying drawings.
* * * * *