1. Field of the Invention
The present invention relates to a selective negative voltage word line decoder, particularly for Flash memory.
2. Description of the Related Art
More particularly, the present invention aims to provide an improvement of a decoder of the type described in application WO 02/41322 (
The decoder WLDEC1 comprises a predecoder PREDEC and a postdecoder POSTDEC powered by a voltage Vcc, supplying signals SELi for selecting word lines. When a determined address ADR is supplied to the decoder, the signal SELi for selecting the word line WLi designated by this address is set to 1 (voltage Vcc) while all the other selection signals are on 0 (ground). The signals SELi are applied to voltage adaptor circuits ADi each delivering a voltage Vi to a word line WLi. Each voltage adaptor ADi also receives a signal ERASE, a voltage VPOS and a voltage VNEG. The voltage Vi can be positive, negative or zero according to the operation being executed, the value of the selection signal SELi and the value of the voltages VPOS, VNEG.
As described by table 1 below, the signal ERASE is on 1 in the erase mode and on 0 in the other operating modes of the memory. The voltage VPOS is equal to a read voltage VREAD in the read mode, to a programming voltage VPP in the programming mode and to an erase inhibit voltage VEINHIB in the erase mode. The voltage VNEG is equal to a non-read voltage VNREAD in the read mode, to a programming inhibit voltage VPINHIB in the programming mode and to a negative erase voltage VER in the erase mode.
The transfer function of the voltage adaptors ADi is described by table 1 below. In the page erase mode (ERASE=1) the voltage Vi applied to a word line WLi is equal to VER if the word line is selected (SELi=1) or to VEINHIB if the word line is not selected (SELi=0), the voltage VEINHIB here being equal to 4V. Outside periods of erasing, the voltage Vi applied to a selected word line WLi is equal to the voltage VPOS and the voltage Vi applied to a non-selected word line WLi (SELi=0) is zero.
This negative voltage decoder, although being fully satisfactory per se, has the disadvantage that the voltage adaptors ADi, which form the terminating elements of the decoder, are of a relatively complex structure.
Now, with the developments in manufacturing methods of Flash memories, the technological pitch of the memories, that is the minimum distance between two word lines, is increasingly reduced. It thus becomes apparent that the surface of silicon available opposite each word line WLi becomes too small to allow voltage adaptors having the structure described above to be integrated.
Thus, an embodiment of the present invention provides a selective negative voltage word line decoder structure that is compatible with the reduction in the technological pitch in non-volatile memories, particularly Flash memories.
More particularly, one embodiment of the present invention provides a selective negative voltage word line decoder structure that comprises compact terminating elements.
One embodiment of the present invention provides an address decoder for selectively applying to word lines of a memory array signals of variable polarity, negative or positive, the value of which varies according to a word line address applied to the decoder, comprising a group decoder delivering signals for selecting a group of word lines that are of variable polarity, at least one subgroup decoder delivering signals for selecting a subgroup of word lines that are also of variable polarity, one subgroup of word lines comprising a set of word lines belonging to different groups of word lines, and word line drivers with one word line driver per word line, each comprising means for multiplexing the group and subgroup selection signals, for selecting and selectively applying one of these signals to a word line.
According to one embodiment, a group of word lines comprises a set of word lines having address bits of identical determined significance, and a subgroup of word lines comprises a set of word lines having address bits of identical determined significance, while belonging to different groups of word lines.
According to one embodiment, a group of word lines comprises a set of word lines having identical most significant address bits, and a subgroup of word lines comprises a set of word lines having identical least significant address bits.
According to one embodiment, a word line driver comprises MOS-type switch transistors which are both driven on their gate and biased on their drain and their source by the group and subgroup selection signals, and are arranged for selecting one of these signals and for applying it to a word line.
According to one embodiment, a word line driver comprises switch transistors each having a terminal linked to a word line, a terminal receiving a group or subgroup selection signal, and a gate receiving a group or subgroup selection signal.
According to one embodiment, the subgroup decoder comprises a first subgroup decoder, delivering first subgroup selection signals the value of which varies according to the word line address applied to the decoder during phases of erasing memory cells, and is independent of the address applied to the decoder during phases of reading or programming memory cells; and a second subgroup decoder delivering second subgroup selection signals the value of which varies according to the word line address applied to the decoder during the phases of reading or programming, and is independent of the address applied to the decoder during the phases of erasing.
According to one embodiment, each of the subgroup decoders receives a first and a second reference voltage and supplies, in addition to a subgroup selection signal, a complementary subgroup selection signal equal to the second reference voltage when the subgroup selection signal is equal to the first reference voltage and equal to the first reference voltage when the subgroup selection signal is equal to the second reference voltage.
According to one embodiment, a word line driver comprises MOS transistors the gates of which are driven by one of the complementary signals, and MOS transistors the gates of which are driven by group selection signals.
According to one embodiment, a word line driver comprises a first MOS transistor having a drain or source terminal linked to a word line, receiving a group selection signal at its gate and receiving a first subgroup selection signal at a source or drain terminal, a second MOS transistor having a drain or source terminal linked to the word line, receiving the group selection signal at its gate and receiving a second subgroup selection signal at a source or drain terminal, a third MOS transistor having a drain or source terminal linked to the word line, receiving a first complementary subgroup selection signal at its gate and receiving a second subgroup selection signal at a source or drain terminal, and a fourth MOS transistor having a drain or source terminal linked to the word line, receiving a second complementary subgroup selection signal at its gate and receiving a first subgroup selection signal at a source or drain terminal.
According to one embodiment, the group and subgroup decoders receive two reference voltages that are respectively equal to a non-read voltage and a read voltage during the reading of memory cells.
According to one embodiment, the group and subgroup decoders receive two reference voltages that are respectively equal to a programming inhibit voltage and a programming voltage during the programming of memory cells.
According to one embodiment, the group and subgroup decoders receive two reference voltages that are respectively equal to an erase voltage and to an erase inhibit voltage during the erasing of memory cells.
According to one embodiment, the programming voltage is positive, the programming inhibit voltage and the erase voltage are negative, the non-read and erase inhibit voltages are zero.
According to one embodiment, the decoder comprises a predecode stage supplying predecoding signals to the group and subgroup decoders.
According to one embodiment, the group decoder and the subgroup decoder supply positive, negative or zero selection signals.
According to one embodiment, the group decoder comprises voltage elevator circuits for transforming a logic signal equal to 1 having a determined positive level of voltage into a logic signal having a higher positive level of voltage, equal to a reference voltage supplied to the voltage elevator circuits.
According to one embodiment, the group and subgroup decoders comprise voltage selector switches for transforming logic signals on 0 and on 1 into signals having different negative levels of voltage.
According to one embodiment, the group and subgroup decoders comprise voltage selector switches arranged for transforming a logic signal on 0 into a negative voltage signal and a logic signal on 1 into a positive voltage signal.
These features and advantages of the present invention will be explained in greater detail in the following description of an example of an embodiment of a decoder according to the present invention, given in relation with, but not limited to, the following figures:
Organization of the Memory Array
The memory array here comprises 8 sectors SCT0 to SCT7 designated by the 3 most significant address bits A0–AA2, and each sector comprises 256 word lines (only the sector SCT6 being represented partially). This breakdown into sectors is however optional and is only described here as a concrete example of implementation of the present invention.
The architecture of the decoder WLDEC2 is based on a breakdown of each sector (or of the entire memory array if a single sector is provided) into groups of word lines and into subgroups of word lines, such that the intersection of a group of word lines and of a subgroup of word lines corresponds to one and only one word line.
Therefore the breakdown is here as follows:
Naturally, the most significant address bits A0–A2 would be bits designating a group if the memory array did not comprise any sectors.
Each sector thus comprises 32 groups of 8 word lines and 8 subgroups of 8 word lines, and each group comprises 8 word lines that each belong to one of the 8 subgroups, a subgroup only comprising word lines belonging to different groups. In the description below, a group is designated GRPi and the rank of a group designated by the index “i”, the rank of a subgroup being designated by the index “j”. A word line belonging to a group of rank i and to a subgroup of rank j is designated WLi,j. As an example,
Architecture of the Decoder WLDEC2
The decoder WLDEC2 has an architecture corresponding to the breakdown of the memory array into groups and into subgroups, and comprises:
The predecoder PREDEC1 supplies group selection signals Lx(a), Ly(b), LBS to the group decoder GPGEN.
The signal LBS varies according to the 3 most significant address bits A0–A2 and is identical for all the groups of word lines of a same sector. If, for example, the address bits A0–A2 designate the sector SECT6, the signal LBS(S6) supplied by the predecoder for all the groups of word lines of this sector is on 1.
The signals Lx(a), Ly(b) vary according to the 5 most significant address bits A3–A7 and here comprise 8 signals Lx(0) to Lx(7) and 4 signals Ly(0) to Ly(3), allowing 32 different combinations of signals to be obtained, starting with the combination Lx(0)/Ly(0) and ending with the combination Lx(7)/Ly(3). Therefore, each combination of signals Lx(a), Ly(b) corresponds to one and only one group out of the 32 groups of the sector.
The predecoder PREDEC1 also supplies subgroup selection signals PDj that are applied to the subgroup decoder DECGEN, here 8 selection signals PD0 to PD7, and subgroup selection signals PSj that are applied to the subgroup decoder SPGEN, here 8 selection signals PS0 to PS7. These signals vary according to the 3 least significant address bits A8–A10.
The predecoder PREDEC1 comprises 8 predecode blocks PREDEC1(S0) to PREDEC1(S7), with one block per sector, each decode block being linked to a decode block of the decoder GPGEN described below. The signals supplied by the block PREDEC(S6) are partially represented on
Group Decoder GPGEN
The group decoder GPGEN comprises one decode block per sector, i.e. here 8 decode blocks GPGEN(S0) to GPGEN(S7). Each decode block comprises decode elements GPGENi in equal number to the number of groups of word lines per sector, i.e. here 32 decode elements GPGEN0 to GPGEN31. The decode elements of the block GPGEN(S6) are partially represented in
Each decode element GPGENi receives a combination of three selection signals Lx(a), Ly(b), LBS. For example, the decode element GPGEN0 of the block GPGEN(S6) receives the combination Lx(0)/Ly(0)/LBS(S6) and the decode element GPGEN31 receives the combination Lx(7)/Ly(3)/LBS(S6), the signal LBS being identical for all the word lines of the same sector. As the signals Lx(a), Ly(b) vary according to the 5 most significant address bits A3–A7, a single group decode element GPGENi in each decode block receives a combination of signals Lx(a)/Ly(b) equal to “11” (Vcc), and a single decode element GPGENi out of all the decode blocks receives a combination of signals Lx(a)/Ly(b)/LBS equal to “111”.
Subgroup Decoder DECGEN
The subgroup decoder DECGEN comprises one decode block per sector, i.e. here 8 decode blocks DECGEN(S0) to DECGEN(S7). Each decode block comprises decode elements DECGENj in equal number to the number of subgroups of word lines per sector, i.e. here 8 decode elements DECGEN0 to DECGEN7. The decode elements of the block SPGEN(S6) are partially represented on
Each decode element DECGENj of rank j receives a selection signal PDj of corresponding rank and the signal LBS for selecting the corresponding sector. For example, the decode elements DECGEN0 and DECGEN7 of the block DECGEN(S6) respectively receive selection signals PD0 and PD7 and the signal LBS(S6). The selection signals PD0 to PD7 are common to all the blocks DECGEN(S0) to DECGEN(S7).
Each decode element DECGENj of rank j supplies selection signals DECj, DECNJ. These selection signals are applied to the word line drivers Di,j of the word lines belonging to the same subgroup of rank j. As the word lines of a same group GRPi each belong to a different subgroup, the word line drivers Di,j linked to these word lines each receive an individual combination of selection signals DECj, DECNj.
Subgroup Decoder SPGEN
The subgroup decoder SPGEN, of a general architecture similar to that of the subgroup decoder DECGEN, comprises one decode block per sector, i.e. here 8 decode blocks SPGEN(S0) to SPGEN(S7). Each decode block comprises decode elements SPGENj in equal number to the number of subgroups of word lines per sector, i.e. here 8 decode elements SPGEN0 to SPGEN7. The decode elements of the block SPGEN(S6) are partially represented on
Each decode element SPGENj of rank j receives a selection signal PSj of corresponding rank and the signal LBS for selecting the corresponding sector. For example, the decode elements SPGEN0 and SPGEN7 of the block SPGEN(S6) respectively receive selection signals PS0 and PS7 and the signal LBS(S6). The selection signals PS0 to PS7 are common to all the blocks SPGEN(S0) to SPGEN(S7).
Each decode element SPGENj of rank j supplies selection signals SPj, SPNj. These selection signals are applied to the word line drivers Di,j of the word lines belonging to the same subgroup of rank j. As the word lines of a same group GRPi each belong to a different subgroup, the word line drivers Di,j linked to these word lines each receive an individual combination of selection signals SPj, SPNj.
Word Line Driver Di,j
Contrary to the classical decoder described above, the word line drivers Di,j according to the present invention do not convert a selection signal of the type 0/Vcc into a negative or positive voltage depending on the operation performed in the memory array (read, erase, programming). The word line drivers are here simple multiplexers of signals the function of which is to combine the signals of variable polarity supplied by the group decode elements GPGENi and the subgroup decode elements DECGENj, SPGENj, i.e. the signals GPi, SPj and SPNj, DECj and DECNj, to obtain voltages Vi,j of variable polarity applied to the word lines WLi,j. The word line drivers can therefore be of a very simple structure.
The transistors TP1, TP2 receive the signal SPj at their sources and the transistors TN1, TN2 receive the signal DECj at their sources. The gates of the transistors TP1, TN1 receive the signal GPi. The gate of the transistor TP2 receives the signal DECNj and the gate of the transistor TN2 receives the signal SPNj. The drains of the transistors TP1, TP2, TN1, TN2 are connected to an output node of the word line driver, which supplies the voltage Vi,j to the corresponding word line WLi,j.
The word line drivers Di,j are therefore compact and are adapted to the reduction of the technological pitch in the memories in integrated circuit. The negative or positive voltages are supplied by the group GPGEN and subgroup DECGEN, SPGEN decoders. As a result, the part of the decoder according to the present invention performing the postdecoding of the predecoding signals, here comprising the decoders GPGEN, DECGEN and SPGEN, is of a more complex structure than that of a positive voltage postdecoder, as it will be seen below. However, the surface of silicon available to produce the group and subgroup decoders is decidedly more extensive than that imposed by the technological pitch between the word lines, since each group or subgroup covers a set of word lines. The present invention is therefore based on a complexification of the part upstream from the decoder, to the benefit of a simplification of its part downstream, that is formed by the word line drivers.
To supply signals of variable polarity, the decoders GPGEN, SPGEN, DECGEN receive voltages VPOS and VNEG supplied by a voltage generator PWGEN controlled by a sequencer SEQ.
Voltage Generator PWGEN and Sequencer SEQ
The generator PWGEN and the sequencer SEQ are represented in greater detail in
The control circuit CONTCT is driven by a mode signal MDS that is supplied by the sequencer SEQ in response to a command CMD for reading or writing the memory array applied to the sequencer. The hard-wired logic or microprocessor sequencer SEQ, also supplies signals SES (“Sector Erase Signal”) and INVSEL (“Inverse Selection”). The signals MDS, SES are applied to the predecoder PREDEC1. The signal INVSEL is applied to the group decoder GPGEN.
Operation of the Decoder WLDEC2
The operation of the decoder WLDEC2 is described by table 2 below. In this table, the following columns can be distinguished:
This column describes the operating modes of the memory. A mode READ corresponding to the reading of memory cells, a mode PROG corresponding to the programming of memory cells, a mode PERASE (“Page Erase”) corresponding to the erasing of a page (word line) of the memory array, and a mode SERASE (“Sector Erase”) corresponding to the erasing of an entire sector can be distinguished. The last erase mode is that of classical Flash memories that do not have a selective negative voltage decoder.
This column describes the value of the signal SES (“Sector Erase Signal”). This signal determines whether an erase operation must be applied to an entire sector (mode SERASE) or only to one page of the memory array (mode PERASE). The signal SES is set to 1 in response to a sector erase command (mode SERASE) and to 0 in response to a page erase command (mode PERASE). Its default value is 0 in the modes PROG and READ.
This column describes the value of the mode signal MDS, that here comprises two bits B1, B2. The bit B1 is set to 1 when the voltage VNEG must be taken to a negative value and is set to 0 when the voltage VNEG must be set to 0 (ground). The bit B2 is set to 0 when the voltage VPOS must be taken to a positive value and is set to 1 when the voltage VPOS must be set to 0 (ground).
This column describes the value of the signal INVSEL (“Inverse Selection”), that is on 0 in the modes READ and PROG and on 1 in the modes PERASE and SERASE. The signal INVSEL allows the signals PDj to be inhibited in the modes READ and PROG and the signals PSj to be inhibited in the modes PERASE and SERASE. More particularly, the signals PDj are forced to 0 in the mode READ and are forced to 1 in the mode PROG, while the signals PSj are forced to 1 in the modes PERASE and SERASE. That allows the subgroup decoder DECGEN to be inhibited in the modes READ and PROG so as to take the signals DECj, DECNj to predetermined and fixed values described by table 2. That also allows the subgroup decoder SPGEN to be inhibited in the modes PERASE and SERASE so as to take the signals SPj, SPNj to predetermined and fixed values described by table 2.
This column describes the value of the voltage VNEG in the four operating modes of the decoder. In the mode READ, the voltage VNEG is equal to a non-read voltage VNREAD that is here equal to 0V. In the mode PROG, the voltage VNEG is equal to a programming inhibit voltage VPINHIB that is here equal to −1.5V. In the mode PERASE and SERASE, the voltage VNEG is equal to a negative erase voltage VER that is here equal to −9V.
This column describes the value of the voltage VPOS in the four operating modes of the decoder. In the mode READ, the voltage VPOS is equal to a read voltage VREAD that is here equal to 4.5V. In the mode PROG, the voltage VPOS is equal to a programming voltage VPP that is here equal to 7.5V. In the mode PERASE and SERASE, the voltage VPOS is equal to an erase inhibit voltage VEINHIB that is here equal to 0V.
This column describes the selected (SEL) or non-selected (UNS) state of a word line, according to the predecoding signals Lx(a), Ly(b), PSj, PDj supplied by the predecoder.
These columns describe, in the four operating modes of the decoder, examples of values of predecoding signals for selected or non-selected word lines. As indicated above, the signal PDj is forced to 0 in the mode READ and is forced to 1 in the mode PROG, while the signal PSj is forced to 1 in the modes PERASE and SERASE. A word line WLi,j is in the selected state when the corresponding predecoding signals Lx(a), Ly(b), PSj (modes READ, PROG) or Lx(a), Ly(b), PDj (modes PERASE, SERASE) are simultaneously on 1. The selection signal LBS is not described out of concern to remain simple, and it is considered that table 2 describes the operations occurring in a selected sector (LBS=1).
These columns describe, in the four operating modes of the decoder, the values of the signals GPi, SPj, SPNj, DECj, DECNJ corresponding to the values of the predecoding signals Lx(a), Ly(b), PSj, PDj and to the values of the voltages VNEG and VPOS. These signals are expressed in Volts and correspond to the values allocated to the voltages VNEG and VPOS, i.e. VNREAD and VREAD in the mode READ, VPINHIB and VPP in the mode PROG, VER and VEINHIB in the modes PERASE and SERASE, that is here 0V and 4.5 V, −1.5V and 7.5V, −9V and 0V, respectively. The signals DECj and DECNj are forced to 0V and 4.5V in the mode READ as the predecoding signal PDj is forced to 0, and are forced to −1.5V and 7.5V in the mode PROG as the predecoding signal PDj is forced to 1. Similarly, the signals SPj and SPNj are forced to 0 and −9V in the modes PERASE and SERASE as the predecoding signal PSj is forced to 1.
It will be noted that the signal DECNj is the opposite of the signal DECj and that the signal SPNj is the opposite of the signal SPj, in a system of logic signals in which the voltages VNEG and VPOS respectively define the 0 and the logic 1.
This column describes the value of the voltage Vi,j applied to a selected or non-selected word line, in each of the operating modes. In the mode READ, a selected word line receives the read voltage VREAD (here 4.5V) while a non-selected line receives the non-read voltage VNREAD (here 0V). In the mode PROG, a selected word line receives the programming or refresh voltage VPP (here 7.5V) while a non-selected word line receives the programming inhibit voltage VPINHIB (here −1.5V) that protects the memory cells against a drain stress that can lead to spurious programming. In the mode PERASE, a selected word line (forming the page to be erased) receives the negative erase voltage VER (here −9V) while a non-selected word line receives the erase inhibit voltage VEINHIB (here 0V). This voltage protects the memory cells against a spurious erasure by reducing the source-gate voltage difference of the floating-gate transistors, as a positive voltage is applied to the sources of all the memory cells of the sector, as described in application WO 02/41322. In the mode PERASE, all the word lines of the selected sector receive the negative erase voltage VER while the word lines of the other sectors receive the erase inhibit voltage VEINHIB.
“FIGS. 5A–5L” Column
In summary, the decoder according to the present invention allows word line drivers Di,j to be provided that are simple in structure and compatible with memories with a high density of word lines, for selectively applying negative or positive voltages to these word lines.
Some examples will now be described of embodiments of a group decode element GPGENi, of a subgroup decode element SPGENj and of a subgroup decode element DECGENj. In the following description, the NMOS type transistors are designated by reference signs starting with “TN” and the PMOS type transistors by reference signs starting with “TP”.
Group Decode Element GPGENi
The element GPGEN also comprises a voltage elevator branch comprising a transistor TP10, a transistor TP11, and a transistor TN10 in series. The voltage VPOS is applied to the source of the transistor TP10 the drain of which is connected to the source of the transistor TP11. The drain of the transistor TP11 is connected to the drain of the transistor TN10 the source of which receives the voltage VNEG. The transistor TP11 is a cascode transistor the gate of which is for example linked to the ground.
The signal GPi is supplied by the drain node of the transistors TP11 and TN10, forming the output of the element GPGENi. This drain node is further linked to the ground by two transistors TN11, TN12 in series. The gate of the transistor TN12 is driven by the signal INVSEL.
The element GPGENi also comprises a voltage elevator circuit ELVCT powered by the voltage VPOS. The circuit ELVCT comprises an input IN1 receiving the signal GPATVPOS, an input IN2 receiving the signal GPATVNEG and an output OUTN that drives the gate of the transistor TP10. The output OUTN, which is an inverting output relative to the input IN1, delivers a signal equal to VPOS when the input IN1 is on 0, and a signal equal to 0 when the input IN1 is on 1 (Vcc) (refer to table 7 described below).
The element GPGENi also comprises a negative voltage selector switch circuit NEGSW powered by the voltages Vcc and VNEG. The circuit NEGSW comprises an input IN to which the signal GPATVNEG is applied, a non-inverting output OUT of low level that drives the gate of the transistor TN10, an inverting output OUTHN of high level that drives the gate of the transistor TN11. The circuit NEGSW also comprises an inverting output OUTN of low level and a non-inverting output OUTH of high level, not used here. The voltages delivered by these various outputs according to the signal applied to the input IN are described by table 8 below, some varying according to the operating mode of the memory.
The value of the selection signal GPi, equal to VPOS, VNEG or 0, varies according to the signals Lx(a), Ly(b), LBS, INVSEL, as described by table 2 and tables 3 and 4 below, in which it is considered that LBS=1. In the mode PERASE or SERASE (VPOS=0), the setting of the signal GPi to 0 is performed by the transistor TN11 when the output OUTHN of the circuit NEGSW changes to Vcc, as the signal INVSEL is on 1 and the transistor TN12 is on (refer to the last line of table 4).
Subgroup Decode Element DECGENi
Subgroup Decode Element SPGENi
In the mode READ or PROG, the signal SPj is therefore equal to 0 or to VPOS (with VPOS=VREAD or VPP) and the signal DECNj is equal to VPOS or VNEG (with VNEG=VNREAD or VPINHIB), according to the value of the signal PSj, as described by table 2 and table 6 below, in which it is assumed that LBS=1.
Voltage Elevator Circuit ELVCT
Negative Voltage Selector Switch NEGSW
The low level output OUT of the negative voltage selector switch is connected to the drain of the transistor TN51 and the low level inverting output OUTN is connected to the drain of the transistor TN53. The high level output OUTH is connected to the drain of the transistor TP51 and the high level inverting output OUTHN is connected to the drain of the transistor TP51. When the voltage Vcc is applied to the input IN the transistor TN51 is on and the transistor TN53 is off. When the input IN is set to 0 the transistor TN51 is off, the output of the gate 13 supplies the voltage Vcc and the transistor TN53 is on. The selector switch NEGSW therefore operates like a sort of flip-flop that switches in one direction when the input IN receives the voltage Vcc, and in the other direction when the input IN receives a zero voltage. The voltages delivered by the outputs OUT, OUTN, OUTH, OUTHN are described by table 8 below. It will be noted that the voltages delivered by the low level outputs OUT, OUTN depend on the value of the voltage VCASC and that the latter has different values according to the operating mode of the memory. The voltage VCASC is therefore equal to VPOS in the operating modes READ and PROG (i.e. when the voltage VNEG is equal to 0 or to −1.5V) and is equal to 0 in the operating modes PERASE, SERASE (i.e. when the voltage VNEG is equal to VERASE).
It will be understood by those skilled in the art that various alternative embodiments of the word line decoder that has just been described may be made, particularly as far as the production of the word line drivers, the group decoders and the subgroup decoders are concerned, without departing from the scope of the present invention.
1 (Vcc)
1 (Vcc)
All of the above U.S. patents, U.S. patent application publications, U.S. patent applications, foreign patents, foreign patent applications and non-patent publications referred to in this specification and/or listed in the Application Data Sheet, are incorporated herein by reference, in their entirety.
From the foregoing it will be appreciated that, although specific embodiments of the invention have been described herein for purposes of illustration, various modifications may be made without deviating from the spirit and scope of the invention. Accordingly, the invention is not limited except as by the appended claims.
Number | Date | Country | Kind |
---|---|---|---|
03 00615 | Jan 2003 | FR | national |
Number | Name | Date | Kind |
---|---|---|---|
5825205 | Ohtsuka | Oct 1998 | A |
6426908 | Hidaka | Jul 2002 | B1 |
6741515 | Lazar et al. | May 2004 | B1 |
6747903 | Pan et al. | Jun 2004 | B1 |
20020067639 | Byeon et al. | Jun 2002 | A1 |
Number | Date | Country |
---|---|---|
1 255 254 | Nov 2002 | EP |
WO 0241322 | May 2002 | WO |
Number | Date | Country | |
---|---|---|---|
20040230736 A1 | Nov 2004 | US |