PatentDe  


Dokumentenidentifikation EP0980076 11.12.2003
EP-Veröffentlichungsnummer 0980076
Titel Verfahren und Anordnungen zum Aktivieren eines Speichers
Anmelder Infineon Technologies AG, 81669 München, DE
Erfinder Sung, Chih-Ta Star, 85625 Glonn, DE;
Mattela, Venkat, San Jose, US;
Afsar, Muhammad, San Diego, US;
Singh, Balraj, San Jose, US;
Hung, Chic-Teng, San Jose, US
DE-Aktenzeichen 69912546
Vertragsstaaten DE, FR, GB, IE
Sprache des Dokument EN
EP-Anmeldetag 06.07.1999
EP-Aktenzeichen 991135351
EP-Offenlegungsdatum 16.02.2000
EP date of grant 05.11.2003
Veröffentlichungstag im Patentblatt 11.12.2003
IPC-Hauptklasse G11C 8/00

Beschreibung[en]
BACKGROUND OF THE INVENTION

The present invention relates generally to memory devices. More specifically, the present invention relates to method and apparatus for activating such memory devices.

Fig. 1A is a diagrammatic representation of a conventional memory device 100. As shown, the memory device 100 is configured to receive a plurality of address (ADR) signals, a clock signal (CLK), and a combined read and write enable (WR!/RD) signal. An "!" denotes that a write operation is enabled when the WR!/RD signal has a low value, while a high value enables a read operation. The memory device is also coupled with a databus (DB) that is used for receiving and outputting data into and out of the memory device 100.

Conventional memory devices typically include a memory array for storing data and a decoder for receiving the plurality of address signals and activating a particular memory location within the memory array that is associated with the address signals. The WR!/RD input signal indicates whether data is being retrieved from or input to the specified memory location of the memory array. Additionally, the memory device 100 uses the CLK signal to enable activation of the specified memory location.

Currently, memory devices typically are configured to follow one of two approaches for enabling the memory array: enabling during a rising edge of the CLK signal or enabling during a falling edge of the CLK signal. For both approaches, the memory device is configured to use the received CLK signal to enable activation of the specified memory location. These two conventional approaches are described below in reference to Figs. 1B.

Fig. 1B represents typical timing diagrams for CLK and ADR signals (ADR1 through ADRn), which are provided to the memory device 100 of Fig. 1A. As described above, the memory device 100 is enabled either during the rising edge 102 or the falling edge 104 of the CLK signal. When the memory device 100 is enabled (e.g., during transition 102 or 104), the specified memory location is then activated and data is read from or written to the specified memory location within the memory array.

Although these two conventional enabling approaches work well under certain conditions, both approaches have associated disadvantages. When the memory device is enabled at the CLK signal's rising edge 102, all or some of the ADR signals may have failed to completely transition to their new address values before the specified memory location is activated. The ADR signals typically begin to transition from a first value to a second value at the rising edge of the CLK signal. However, some of the ADR signals take a significantly longer time to transition than other address signals. As shown, the transition times for ADR1, ADR2, and ADRn, are d1, d2, and dn, respectively. Note, transition time dn is significantly longer than transition time d1. Thus, if the memory device is enabled at the rising edge 102 of the CLK signal, it is likely that some of the ADR signals have not transitioned when the memory device is then activated. Thus, data may then be read from or written to an erroneous memory location within the memory device.

Another conventional approach is to enable the memory device at the falling edge 104 of the CLK signal. Although this approach gives the ADR signals enough time to transition, this approach may not provide enough time to decode the ADR signals, activate the memory device and perform a read or a write operation. In other words, the ADR signals may begin to transition to a new value (e.g., at the CLK signal's next rising edge 106) before the read or write operation is finished, and the possibility of an erroneous read or write is thereby increased. For memory device designs that take longer than a half cycle of the CLK signal to perform a read or a write operation, this enabling approach may be inadequate.

EP 0 273 652 describes a memory device which includes a delay circuit for delaying as enable signal by a predetermined time corresponding to a worst case delay of the address signals US 5 208 783 describes a memory unit comprising a delaying means for delaying a clock signal by a preset time and thereby converting the clock signal into or enable signal.

In view of the foregoing, there is a need for an enabling mechanism and technique that provides adequate time for the ADR signals to transition to stable values, while providing enough time for the memory device to be activated and perform a read or a write operation after the ADR signals stabilize and before the ADR signals transition to new values. Especially, this times has to be optimized as regards transition times for the ADR signals.

SUMMARY OF THE INVENTION

Accordingly, the present invention provides an apparatus and method for activating a memory location within a memory device. In general terms, the present invention provides a system and method for generating a delayed enable signal within a memory device based on a received clock signal. That is, the memory device of the present invention receives the clock signal and delays the clock signal to create an enable signal for activating the memory device. The enable signal is delayed from the clock signal by at least about a worst case delay of a plurality of address signals that are also input into the memory device. As a result, a particular memory location of the memory device is not activated until all of the address signals have transitioned to stable values.

In an apparatus aspect of the invention, a memory device is disclosed. The memory device includes an enable unit arranged to receive a plurality of address signals and a clock signal and to output an activation signal. The address signals have an associated worst case delay, and the enable unit is further arranged to generate an enable signal that is delayed from the clock signal by an amount that is substantially the same as the worst case delay and a margin that is between 5 and 20 percent of a clock cycle of the clock signal. The memory device further includes a memory array arranged to receive the activation signal in response to which a corresponding memory location is activated.

In one embodiment, the enable unit is arranged to generate the enable signal to substantially directly control activation of the corresponding memory location. According to this embodiment, the enable unit includes (i) a decoder unit arranged to receive the address signals and to output a preactivation signal associated with the corresponding memory location, (ii) a delay unit arranged to receive the clock signal and generate the enable signal, and (iii) a plurality of multiplexers. A selected one of the multiplexers is arranged to receive the preactivation signal and the enable signal and to output an activation signal in response to the preactivation signal and the enable signal to the corresponding memory location.

In an alternative embodiment, the enable unit is arranged to generate the enable signal to control when the address signals are decoded and the corresponding memory location is activated. The enable unit includes (i) a delay unit arranged to receive the clock signal and to output the enable signal that is delayed from the clock signal by at least about the delay time and (ii) a decoder unit arranged to receive the address signals and the enable signal from the delay unit and to output the activation signal that corresponds to the memory location in response to the address signals and the enable signal.

In a method aspect, a method for selecting a memory location in a memory array is disclosed. In response to a first transition of a clock signal, a plurality of address signals corresponding to the memory location are transmitted from an address register to a decoder. A worst case delay is associated with the address signals. The clock signal is delayed by an amount that is substantially the same as the worst case delay and a margin that is between 5 and 20 percent of a clock cycle of the clock signal thereby generating an enable signal. The memory location is selected in response to a first transition of the enable signal. The first transition of the enable signal corresponds to the first transition of the clock signal.

The present invention has many advantages. For example, by generating an enable signal that is delayed from the clock signal by at least about a worst case delay of the address signals, the address signals that are input into the memory device will likely have stable values by the time the memory device is activated by the enable signal. That is, the address signals are less likely to have ambiguous or erroneous values as a result of the address signals failing to transition completely to new address values.

Additionally, the memory device of the present invention may utilize a significant portion of the clock cycle to decode the address signals, activate a memory location, and complete a read or a write operation since the memory device is enabled from a first transition of the enable signal that is delayed from a first rising edge of the clock signal to a second transition of the enable signal that is delayed from a second rising edge of the clock signal. Thus, the memory device may be designed with more relaxed timing specifications than a conventional memory device that utilizes the falling edge of the clock signal as an enable signal thereby allowing only a half cycle of the clock signal to decode the address signals and access the memory location.

Also, various fabrication processes may be used to manufacture the memory device without affecting the reliability of the memory device's enablement function when the enable signal includes a margin delay in addition to the worst case delay of the address signals. As a result of this increased reliability among various fabrication processes, a new simulation of the worst case ADR signal delay is not required for each new fabrication change. For example, when a 0.5 µm process is upgraded to a 0.4 µm process, a new simulation is not required for the new 0.4 µm process.

The features and advantages of the present invention will be presented in more detail in the following specification of the invention and the accompanying figures which illustrate by way of example the principles of the invention.

BRIEF DESCRIPTION OF THE DRAWINGS

The present invention will be readily understood by the following detailed description in conjunction with the accompanying drawings, wherein like reference numerals designate like structural elements, and in which:

  • Fig. 1A is a diagrammatic representation of a conventional memory device.
  • Fig. 1B represents typical timing diagrams for clock and address signals (ADR1 through ADRn), which are provided to the memory device of Fig. 1A.
  • Fig. 2A represents timing diagrams for address signals and clock signal that are input into a memory device of the present invention and a delayed enable signal that is generated from the clock signal by the memory device in accordance with one embodiment of the present invention.
  • Fig. 2B is a flowchart illustrating a process of selecting or activating a memory location in a memory device in accordance with one embodiment of the present invention.
  • Fig. 3 is a diagrammatic representation of a memory device that is suitable for implementing the process of Figure 2B of selecting a memory location of the memory device.
  • Fig. 4A is a diagrammatic representation of a first implementation of the enable unit of Fig. 3 in accordance with one embodiment of the current invention.
  • Fig. 4B is a diagrammatic representation of the word line decoder of the enable unit of Fig. 4A in accordance with one embodiment of the current invention.
  • Fig. 4C is a diagrammatic representation of the delay unit of the enable unit of Fig. 4A (and of the enable unit of Fig. 5, which is described below,) in accordance with one embodiment of the current invention.
  • Fig. 5 is a diagrammatic representation of a second implementation of the enable unit of Fig. 3 in accordance with an alternative embodiment of the current invention.

DETAILED DESCRIPTION OF SPECIFIC EMBODIMENTS

Reference will now be made in detail to a specific embodiment of the invention. An example of this embodiment is illustrated in the accompanying drawings. While the invention will be described in conjunction with this specific embodiment, it will be understood that it is not intended to limit the invention to one embodiment. On the contrary, it is intended to cover alternatives, modifications, and equivalents as may be included within the scope of the invention as defined by the appended claims. In the following description, numerous specific details are set forth in order to provide a thorough understanding of the present invention. The present invention may be practiced without some or all of these specific details. In other instances, well known process operations have not been described in detail in order not to unnecessarily obscure the present invention.

Fig. 2A represents timing diagrams for address (ADR) signals and clock (CLK) signal that are input into a memory device of the present invention and a delayed enable signal that is generated from the CLK signal by the memory device in accordance with one embodiment of the present invention. As shown, the ADR signals transition to a new value during time period 202. Time period 202 represents the worst case delay for a last ADR signal to transition to a new value. One of the reasons that the ADR signals have different transition times is that each ADR signal is transmitted along differently sized routing paths. In one embodiment of the present invention, which embodiment is described further below, this relationship between the routing and worst case delay of the ADR signals is utilized to generate the enable signal.

In the present invention, the enable signal is generated by delaying the CLK signal by at least about the worst case ADR signal delay (202). By delaying the enable signal by at least about the worst case ADR signal delay, the ADR signals will likely have enough time to transition to stable new values before the memory device is enabled. In other words, when data is read from or written into a particular memory location of the memory device, it is more likely that the memory location is correct than with conventional memory devices that use the CLK signal's rising edge to activate the memory location, wherein all of the ADR signals have likely not yet transitioned to new values.

To compensate for fabrication deviations and semiconductor process changes, the enable signal may be further delayed from the CLK signal by a margin 204. Any suitable value may be selected for the margin 204 to ensure that the memory device is not activated before the ADR signals transition to their respective new values. For example, the margin may be selected to be about 5 to 10 percent of the total CLK signal cycle. This range appears to provide enough time for ADR signals to transition within a wide range of memory devices that were fabricated under various semiconductor process conditions.

Fig. 2B is a flowchart illustrating a process 220 of selecting or activating a memory location in a memory device in accordance with one embodiment of the present invention. Initially, in operation 222, a first transition of a CLK signal is provided to the memory device. Any suitable transition of the CLK signal may be used for the first transition that corresponds to when the ADR signals begin to transition. Typically, the ADR signals will begin to transition during the rising edge of the CLK signal; thus, the first transition is usually in the form of a rising edge of the CLK signal.

After the first transition of the CLK signal is provided to the memory device, a plurality of ADR signals are then provided to the memory device in operation 224. Alternatively, the ADR signals may be provided prior to providing the CLK signals. The CLK signal is delayed by at least about a worst case delay of the transmitted ADR signals to generate the enable signal in operation 226. In operation 228, the memory location that corresponds to the transmitted ADR signals is then selected in response to a first transition of the generated enable signal that is delayed from the first transition of the CLK signal.

Any suitable system may be used to implement the process 220 of Fig. 2A. For example, an enable unit may be configured to generate the enable signal to control the decoding of the ADR signals. That is, in response to the enable signal, the ADR signals are decoded and a memory location is selected and activated. By way of another example, the enable unit may be configured to generate the enable signal to more substantially directly control activation of the memory location. That is, after the ADR signals are decoded based on the decoded ADR signals, the memory location is then selected or activated in response to the enable signal.

Fig. 3 is a diagrammatic representation of a memory device 300 that is suitable for implementing the process 220 of Fig. 2B of selecting a memory location of the memory device. Certain ones of the blocks of the memory device 300 may be implemented with discrete circuit elements, or as software routines executed by a suitable digital processor, such as a high speed signal processor. Alternatively, a combination of circuit elements and software routines may be employed. As such, the ensuing description is not intended to limit the application of this invention to any one particular technical embodiment

As shown, the memory device 300 includes an enable unit 302 and a memory array 304. The enable unit 302 is configured to receive a plurality of ADR signals, a CLK signal, and a WR!/RD signal. Each set of received ADR signals represent a selected memory location within the memory array. The CLK signal is used to generate an enable signal (not shown) that is, in turn, used to activate the selected memory location. A WR!/RD signal may also be utilized to indicate whether data is to be read from or written to the selected memory location. In response to the received signals, the enable unit 302 selects or activates the selected memory location by outputting an activation signal on a selected one of a plurality of activation lines (WL0 through WLn) that is coupled to the selected memory location.

The enable unit 302 is arranged to internally generate the enable signal (not shown) based on the CLK signal. As described above, the enable signal is delayed from the CLK signal by at least about a worst case delay of the ADR signals. In response to the enable signal, the enable unit is also arranged to activate the selected memory location that corresponds to the received ADR signals.

The enable unit 302 activates the selected memory location during a rising edge of the enable signal, which is delayed from a rising edge of the CLK signal by at least about the worst case delay of the ADR signals. The worst case delay may include the worst case delay for all of the ADR signals to reach the enable unit 302, as well as the time it takes for the ADR signals to be received and amplified by the enable unit 302.

The enable unit 302 may be configured to further delay the enable signal from the CLK signal by an extra margin so as to decrease the likelihood of activating the memory location when the enable signal's delay from the CLK signal does not exactly match the actual worst case delay of the ADR signals. Embodiments of the enable unit are further described below in reference to Figs. 4A, 4B, and 5.

Fig. 4A is a diagrammatic representation of a first implementation 302 of the enable unit of Fig. 3 in accordance with one embodiment of the current invention. As shown, the enable unit 302 includes a word line decoder 402 configured to decode the received ADR signals and a delay unit 404 configured to generate the enable signal based on the received CLK signal.

The delay unit 404 is arranged to receive and delay the CLK signal by at least about the worst case delay of the ADR signals. The delayed CLK signal is output to the word line decoder 402 as the enable signal. The word line decoder 402 is arranged to receive the enable signal from the delay unit 404 and the ADR signals and to activate one of a plurality of activation lines (WL0 through WLn) based on the received ADR signals and enable signaL The selected activation line is associated with a row of the memory array. Thus, the word line decoder is arranged to activate a selected row of the memory array.

Additionally, a bit line decoder (not shown) may be employed to activate a pair of complementary bit lines (not shown) that are associated with a particular column of memory locations within the memory array. The word line decoder and bitline decoder work together to activate the selected memory location, which is located at the intersection of the activated row and column of the memory array. Although the present invention is described as being implemented within a word line decoder, it may also be implemented with any suitable ADR decoder, such as a bitline decoder.

As described above, the word line decoder 402 is configured to activate a memory location of the memory array 304 (see Fig. 3) through one of the activation lines (WL0 through WLn) in response to the delayed enable signal. Fig. 4B is a diagrammatic representation of the word line decoder 402 of the enable unit 302 of Fig. 4A in accordance with one embodiment of the current invention.

As shown, the decoder 402 includes a plurality of AND gates 406 that are configured to receive selected ADR signals and output a corresponding activation signal on one of activation lines WL0 through WLn that is associated with the ADR signals. Of course, each AND gate may be replaced by an equivalent multiplexer arrangement, such as a NAND gate and an inverter. The associated activation line is not selected or activated until the enable signal transitions to an enable state (e.g., a logical "1" or high voltage state). In other words, a selected one of the AND gates outputs the activation signal on one of the activation lines WL0 through WLn when the selected AND gate's selected ADR signals and enable signal input are equal to "1".

Fig. 4C is a diagrammatic representation of the delay unit 404 of the enable unit 302 of Fig. 4A (and of the enable unit 302' of Fig. 5, which is described below,) in accordance with one embodiment of the current invention. As shown, the delay unit 404 includes a worst case delay unit 408 for delaying the CLK signal by at least about a worst case delay of the ADR signals.

The delay unit 404 may be implemented in any suitable manner so as to delay the CLK by at least about the worst case delay of the ADR signals. For example, the routing path that is used to transmit the worst case ADR signal may be duplicated within the delay unit 404 to mimic the worst case delay. That is, the CLK signal is transmitted through a duplicated routing path in the delay unit to produce the enable signal. By way of another example, the RC delay for the worst case delay ADR signal may be calculated and a corresponding capacitor and/or resistor structure may be implemented that has the same RC delay. The CLK signal is then transmitted through the capacitor and/or resistor structure to generate the enable signal. Additionally, the delay unit may be formed from a single device or a plurality of devices (such as an even number of chained inverters), wherein the CLK signal is transmitted through device(s) having the same RC delay as the worst case delay of the ADR signals.

The delay unit 404 may also include a margin unit 410 for providing an extra delay time and increasing the reliability of memory devices that are manufactured under various fabrication conditions, for example. By way of a more specific example, a memory device that is manufactured with a 0.5 µm process will function differently than a memory device manufactured with a 0.4 µm process. Specifically, the delay of the enable signal may not precisely match the worst case delay of the ADR signals. Thus, the margin unit 410 may be implemented in any suitable configuration that leaves enough time to compensate for this mismatch. Any suitable value may be used for the margin so as to increase reliability of the memory device. For example, a margin value of about 5 to 20 percent of the CLK cycle may work well, and, preferably about 5 to 10 percent.

The delay unit 404 may also include a buffer matching unit 412 for matching the ADR signal delay through the enable unit's input buffers. That is, the time taken by the enable units input buffers to receive and amplify the ADR signals is added to the delay of the enable signal relative to the CLK signal. The buffer matching unit 412 may be implemented by duplicating the input buffer's sizing, shape, and routing such that the buffer matching unit 412 has the same performance as the actual input buffers, even under variable conditions, such as changes in process and temperature.

Fig. 5 is a diagrammatic representation of a second implementation 302' of the enable unit of Fig. 3 in accordance with an alternative embodiment of the current invention. As shown in this second implementation, the enable unit 302' includes a word line decoder 502 configured to receive ADR signals and output a preactivation signal on one of a plurality of preactivation lines (XWL0 through XWLn) based on the ADR signals. More specifically, one of the preactivation lines that is associated with the ADR signals is activated and output to one of a plurality of AND gates 504.

The enable unit 302' also includes the delay unit 404 that was described above in reference to Fig. 4C. The delay unit 404 is arranged to receive the CLK signal and generate the enable signal by delaying the CLK signal by at least about the worst case delay of the ADR signals. The delay unit 404 is further arranged to output the enable signal to each of the plurality of AND gates 504.

Each of the plurality of AND gates 504 are configured to receive one of the preactivation lines from the word line decoder 502, the enable signal from the delay unit 404, as well as a RD! enable signal (the RD! signal is set high for a write). As shown, each of the AND gates 504 is configured to output an activation signal to an associated one of the activation lines (WL0 through WLn) when the received preactivation signal, enable signal, and RD! signal indicate that the corresponding memory location is to be activated. For example, the AND gate 504a transmits the activation signal along WL0 when the preactivation signal XWL0, the enable signal, and the RD! each have a logical "1" value.

Although the present invention is described in terms of enabling a write operation to the memory array, it may also be implemented for a read operation. Additionally, any suitable multiplexing devices may be used in place of the plurality of AND gates 504.

Although the foregoing invention has been described in some detail for purposes of clarity of understanding, it will be apparent that certain changes and modifications may be practiced within the scope of the appended claims. It should be noted that there are many alternative ways of implementing both the process and apparatus of the present invention. For example, the enable signal may be delayed from the CLK signal by an amount that is significantly lower or higher than the worst case delay of the ADR signals. If the amount is lower, then the enable signal may also be delayed by a margin delay that compensates for the difference between the worst case delay and the enable delay. Accordingly, the present embodiments are to be considered as illustrative and not restrictive, and the invention is not to be limited to the details given herein, but may be modified within the scope and equivalents of the appended claims.


Anspruch[de]
  1. Speicherbaustein (300), umfassend:
    • eine Freigabeeinheit (302), die so ausgelegt ist, daß sie mehrere Adressensignale (ADR) und ein Taktsignal (CLK) empfängt und ein Aktivierungssignal (WL) ausgibt, wobei den Adressensignalen (ADR) eine Verzögerung im ungünstigsten Fall (202) zugeordnet ist, wobei die Freigabeeinheit (302) weiterhin so ausgelegt ist, daß sie ein Freigabesignal erzeugt, das um einen Betrag von dem Taktsignal (CLK) verzögert ist, der im wesentlichen mit der Verzögerung im ungünstigsten Fall (202) und einer Reserve (204), die zwischen etwa 5 und 20 Prozent eines Taktzyklus des Taktsignals (CLK) liegt, übereinstimmt, und ein Speicherarray (304), das so ausgelegt ist, daß es das Aktivierungssignal (WL) empfängt, woraufhin eine entsprechende Speicherstelle aktiviert wird.
  2. Speicherbaustein nach Anspruch 1, wobei die Freigabeeinheit (302) so ausgelegt ist, daß sie das Freigabesignal erzeugt, um im wesentlichen direkt die Aktivierung der entsprechenden Speicherstelle zu steuern.
  3. Speicherbaustein nach Anspruch 2, wobei die Freigabeeinheit (302) folgendes umfaßt:
    • eine Decodierereinheit (502), die so ausgelegt ist, daß sie die Adressensignale (ADR) empfängt und ein der entsprechenden Speicherstelle zugeordnetes Voraktivierungssignal (XWL) ausgibt;
    • eine Verzögerungseinheit (404), die so ausgelegt ist, daß sie das Taktsignal (CLK) empfängt und das Freigabesignal erzeugt; und
    • mehrere Multiplexer (504), wobei ein ausgewählter der Multiplexer (504a, b, c) so ausgelegt ist, daß er das Voraktivierungssignal (XWL) und das Freigabesignal empfängt und als Reaktion auf das Voraktivierungssignal (XWL) und das Freigabesignal das Aktivierungssignal (WL) an die entsprechende Speicherstelle ausgibt.
  4. Speicherbaustein nach Anspruch 1, wobei die Freigabeeinheit (302) so ausgelegt ist, daß sie das Freigabesignal erzeugt, um zu steuern, wann die Adressensignale (ADR) decodiert und die entsprechende Speicherstelle aktiviert wird.
  5. Speicherbaustein nach Anspruch 4, wobei die Freigabeeinheit (302) folgendes umfaßt:
    • eine Verzögerungseinheit (404), die so ausgelegt ist, daß sie das Taktsignal (CLK) empfängt und das Freigabesignal ausgibt, das um mindestens etwa die Verzögerungszeit von dem Taktsignal (CLK) verzögert ist; und
    • eine Decodierereinheit (402), die so ausgelegt ist, daß sie die Adressensignale (ADR) und das Freigabesignal aus der Verzögerungseinheit (404) empfängt und als Reaktion auf die Adressensignale (ADR) und das Freigabesignal das Aktivierungssignal (WL) ausgibt, das der Speicherstelle entspricht.
  6. Speicherbaustein nach Anspruch 1, wobei die Freigabeeinheit (302) weiterhin so konfiguriert ist, daß sie das Freigabesignal um eine Reserve von dem Taktsignal (CLK) verzögert, die so gewählt ist, daß die Zuverlässigkeit des Speicherbausteins erhöht wird.
  7. Speicherbaustein nach Anspruch 1, wobei die Reserve (204) zwischen etwa 5 und 10 Prozent eines Zyklus eines Takts des Taktsignals (CLK) beträgt.
  8. Speicherbaustein nach Anspruch 1, wobei die. Freigabeeinheit (302) mehrere Eingangspuffer (306) enthält, wobei jeder Eingangspuffer (306) so ausgelegt ist, daß er ein zugeordnetes Adressensignal (ADR) verstärkt, und die Freigabeeinheit (302) so ausgelegt ist, daß sie das Freigabesignal noch weiter um eine Pufferangleichzeit von dem Taktsignal (CLK) verzögert, die eine Zeitdauer im wesentlichen gleich einer Verstärkungszeit eines der Eingangspuffer (306) aufweist.
  9. Speicherbaustein nach Anspruch 8, wobei die Freigabeeinheit (302) weiterhin eine Pufferangleicheinheit (412) enthält, die die Bemessung, die Form und das Routing des Eingangspuffers (306) dupliziert, so daß die Pufferangleicheinheit (412) auch unter variablen Bedingungen, wie zum Beispiel Prozeß- und Temperaturänderungen, eine selbe Leistung wie die Eingangspuffer (306) aufweist.
  10. Speicherbaustein nach Anspruch 1, wobei der erste Übergang des Taktsignals (CLK) in Form einer ansteigenden Flanke (102) vorliegt.
  11. Speicherbaustein nach Anspruch 3, wobei die Verzögerungseinheit (302) Schaltkreise umfaßt, die einen Routing-Weg eines Adressensignals des ungünstigsten Falls der mehreren Adressensignale (ADR), die der Verzögerung im ungünstigsten Fall (202) zugeordnet sind, duplizieren.
  12. Speicherbaustein nach Anspruch 3, wobei die Verzögerungseinheit (302) folgendes umfaßt:
    • Schaltkreise, einschließlich einer Struktur mit RC-Verzögerung ähnlich der Verzögerung im ungünstigsten Fall (202), wobei die Struktur aus der folgenden Gruppe ausgewählt wird: eine. Kondensatorstruktur, eine Widerstandsstruktur, mehrere diskrete Bauelemente und ein einziges Bauelement.
  13. Verfahren zum Auswählen einer Speicherstelle in einem Speicherarray (304), mit den folgenden Schritten:
    • als Reaktion auf einen ersten Übergang eines Taktsignals (CLK), Senden mehrerer Adressensignale (ADR) entsprechend der Speicherstelle aus einem Adressenregister zu einem Decodierer, wobei den Adressensignalen (ADR) eine Verzögerung im ungünstigsten Fall (202) zugeordnet ist;
    • Verzögern des Taktsignals (CLK) um einen Betrag, der im wesentlichen gleich der Verzögerung im ungünstigsten Fall (202) und einer Reserve (204), die zwischen etwa 5 und 20 Prozent eines Taktzyklus des Taktsignals (CLK) liegt, ist, wodurch ein Freigabesignal erzeugt wird; und
    • Auswählen der Speicherstelle als Reaktion auf einen ersten Übergang des Freigabesignals, wobei
    • der erste Übergang des Freigabesignals dem ersten Übergang des Taktsignals (CLK) entspricht.
  14. Verfahren nach Anspruch 13, wobei die Adressensignale (ADR) an dem ersten Übergang des Taktsignals (CLK) beginnen, überzugehen.
  15. Verfahren nach Anspruch 14, wobei der erste Übergang des Taktsignals (CLK) eine ansteigende Flanke (102) ist.
  16. Verfahren nach Anspruch 13, wobei die Reserve (204) zwischen etwa 5 und 10 Prozent eines Zyklus eines Takts des Taktsignals (CLK) beträgt.
  17. Verfahren nach Anspruch 13, bei dem weiterhin die Adressensignale (ADR) als Reaktion auf das Freigabesignal decodiert werden.
  18. Verfahren nach Anspruch 15, wobei die Speicherstelle bis zu einem nächsten Übergang des Freigabesignals, entsprechend einer nächsten ansteigenden Flanke (106) des Taktsignals (CLK), ausgewählt wird.
Anspruch[en]
  1. A memory device (300) comprising:
    • an enable unit (302) arranged to receive a plurality of address signals (ADR) and a clock signal (CLK) and to output an activation signal (WL), the address signals (ADR) having a worst case delay (202) associated therewith, the enable unit (302) being further arranged to generate an enable signal that is delayed from the clock signal (CLK) by an amount that is substantially the same as the worst case delay (202) and a margin (204) that is between about 5 and 20 percent of a clock cycle of the clock signal (CLK), and a memory array (304) arranged to receive the activation signal (WL) in response to which a corresponding memory location is activated.
  2. A memory device as recited in claim 1, wherein the enable unit (302) is arranged to generate the enable signal to substantially directly control activation of the corresponding memory location.
  3. A memory device as recited in claim 2, wherein the enable unit (302) comprises:
    • a decoder unit (502) arranged to receive the address signals (ADR) and to output a preactivation signal (XWL) associated with the corresponding memory location;
    • a delay unit (404) arranged to receive the clock signal (CLK) and generate the 'enable signal; and
    • a plurality of multiplexers (504), wherein a selected one of the multiplexers (504a,b,c) is arranged to receive the preactivation signal (XWL) and the enable signal and to output said activation signal (WL) in response to the preactivation signal (XWL) and the enable signal to the corresponding memory location.
  4. A memory device as recited in claim 1, wherein the enable unit (302) is arranged to generate the enable signal to control when the address signals (ADR) are decoded and the corresponding memory location is activated.
  5. A memory device as recited in claim 4, wherein the enable unit (302) comprises:
    • a delay unit (404) arranged to receive the clock signal (CLK) and to output the enable signal that is delayed from the clock signal (CLK) by at least about the delay time; and
    • a decoder unit (402) arranged to receive the address signals (ADR) and the enable signal from the delay unit (404) and to output the activation signal (WL) that corresponds to the memory location in response to the address signals (ADR) and the enable signal.
  6. A memory device as recited in claim 1, wherein the enable unit (302) is further configured to delay the enable signal form the clock signal (CLK) by a margin that is selected to increase reliability of the memory device.
  7. A memory device as recited in claim 1, wherein the margin (204) is between about 5 and 10 percent of a cycle of the clock signal (CLK).
  8. A memory device as recited in claim 1, wherein the enable unit (302) includes a plurality of input buffers (306), each input buffer (306) being arranged to amplify an associated address signal (ADR), and the enable unit (302) is configured to further delay the enable signal from the clock signal (CLK) by a buffer matching time that has a duration time substantially equal to an amplification time of one of the input buffers (306).
  9. A memory device as recited in claim 8, the enable unit (302) further includes a buffer matching unit (412) duplicating the input buffer's (306) sizing, shape, and routing such that the buffer matching unit (412) has a same performance as the input buffers (306), even under variable conditions, such as changes in process and temperature.
  10. A memory device as recited in claim 1, wherein the first transition of the clock signal (CLK) is in the form of a rising edge (102).
  11. A memory device as recited in claim 3, wherein the delay unit (302) comprises circuitry duplicating a routing path of a worst case address signal of the plurality of address signals (ADR) associated with the worst case delay (202).
  12. A memory device as recited in claim 3, wherein the delay unit (302) comprises circuitry including a structure having an RC delay similar to the worst case delay (202), the structure being selected from the group consisting of a capacitor structure, a resistor structure, a plurality of discrete devices, and a single device.
  13. A method for selecting a memory location in a memory array (304), comprising:
    • in response to a first transition of a clock signal (CLK), transmitting a plurality of address signals (ADR) corresponding to the memory location from an address register to a decoder, a worst case delay (202) being associated with the address signals (ADR);
    • delaying the clock signal (CLK) by an amount that is substantially the same as the worst case delay (202) and a margin (204) that is between about 5 and 20 percent of a clock cycle of the clock signal (CLK) thereby generating an enable signal; and
    • selecting the memory location in response to a first transition of the enable signal, the first transition of the enable signal corresponding to the first transition of the clock signal (CLK).
  14. A method as recited in claim 13, wherein the address signals (ADR) begin to transition at the first transition of the clock signal (CLK).
  15. A method as recited in claim 14, wherein the first transition of the clock signal (CLK) is a rising edge (102).
  16. A method as recited in claim 13 wherein the margin (204) is between about 5 and 10 percent of a clock cycle of the clock signal (CLK).
  17. A method as recited in claim 13, further comprising decoding the address signals (ADR) in response to the enable signal.
  18. A method as recited in claim 15, wherein the memory location is selected until a next transition of the enable signal corresponding to a next rising edge (106) of the clock signal (CLK).
Anspruch[fr]
  1. Dispositif mémoire (300) comprenant :
    • une unité d'activation (302) disposée pour recevoir une pluralité de signaux d'adresse (ADR) et un signal d'horloge (CLK), et pour produire un signal d'activation (WL), les signaux d'adresse (ADR) ayant un retard le plus pessimiste (202) qui leur est associé, l'unité d'activation (302) étant en outre disposée pour générer un signal d'activation qui est retardé par rapport au signal d'horloge (CLK) d'une quantité qui est pour l'essentiel la même que le retard le plus pessimiste (202) et une marge (204) qui est d'environ 5 à 20 pour cent d'un cycle d'horloge du signal d'horloge (CLK), et un tableau mémoire (304) disposé pour recevoir le signal d'activation (WL) en réponse auquel un emplacement mémoire correspondant peut être activé.
  2. Dispositif mémoire selon la revendication 1,

    dans lequel

    l'unité d'activation (302) est disposée pour générer le signal d'activation pour contrôler directement pour l'essentiel l'activation de l'emplacement mémoire correspondant.
  3. Dispositif mémoire selon la revendication 2,

    dans lequel

    l'unité d'activation (302) comprend :
    • une unité de décodage (502) disposée pour recevoir les signaux d'adresse (ADR) et pour produire un signal de préactivation (XWL) associé à l'emplacement mémoire correspondant ;
    • une unité de retard (404) disposée pour recevoir le signal d'horloge (CLK) et générer le signal d'activation ; et
    • une pluralité de multiplexeurs (504) dans lesquels l'un des multiplexeurs (504a, b, c) est disposé pour recevoir le signal de préactivation (XWL) et le signal d'activation et pour produire ledit signal d'activation (WL) en réponse au signal de préactivation (XWL) et le signal d'activation dans l'emplacement mémoire correspondant.
  4. Dispositif mémoire selon la revendication 1,

    dans lequel

    l'unité d'activation (302) est disposée pour générer le signal d'activation pour contrôler le moment où les signaux d'adresse (ADR) sont décodés et le moment où l'emplacement mémoire correspondant est activé.
  5. Dispositif mémoire selon la revendication 4,

    dans lequel

    l'unité d'activation (302) comprend:
    • une unité de retard (404) disposée pour recevoir le signal d'horloge (CLK) et pour produire le signal d'activation qui est retardé par rapport au signal d'horloge (CLK) d'au moins le retard ; et
    • une unité de décodage (402) disposée pour recevoir les signaux d'adresse (ADR) et le signal d'activation de l'unité de retard (404) et pour produire le signal d'activation (WL) qui correspond à l'emplacement mémoire en réponse aux signaux d'adresse (ADR) et au signal d'activation.
  6. Dispositif mémoire selon la revendication 1,

    dans lequel

    l'unité d'activation (302) est en outre configurée pour retarder le signal d'activation par rapport au signal d'horloge (CLK) d'une marge qui est sélectionnée pour augmenter la fiabilité du dispositif mémoire.
  7. Dispositif mémoire selon la revendication 1,

    dans lequel

    la marge (204) se situe entre 5 et 10 pour cent d'un cycle du signal d'horloge (CLK).
  8. Dispositif mémoire selon la revendication 1,

    dans lequel

    l'unité d'activation (302) inclut une pluralité de tampons d'entrée (306), chaque tampon d'entrée (306) étant disposé pour amplifier un signal d'adresse associé (ADR), et dans lequel l'unité d'activation (302) est configurée pour retarder encore le signal d'activation par rapport au signal d'horloge (CLK) d'un temps d'appariement de tampon d'une durée pour l'essentiel égale à une durée d'amplification de l'un des tampons d'entrée (306).
  9. Dispositif mémoire selon la revendication 8,

    dans lequel

    l'unité d'activation (302) inclut en outre une unité d'appariement de tampon (412) dupliquant la taille, la forme et le routage du tampon d'entrée (306) de façon que l'unité d'appariement de tampon (412) ait les mêmes performances que les tampons d'entrée (306), même dans des conditions variables telles que changements de procédé et de température.
  10. Dispositif mémoire selon la revendication 1,

    dans lequel

    la première transition du signal d'horloge (CLK) a la forme d'un front montant (102).
  11. Dispositif mémoire selon la revendication 3,

    dans lequel

    l'unité de retard (302) comprend un circuit dupliquant un chemin de routage d'un signal d'adresse le plus pessimiste de la pluralité de signaux d'adresse (ADR) associés au retard le plus pessimiste (202).
  12. Dispositif mémoire selon la revendication 3,

    dans lequel

    l'unité de retard (302) comprend un circuit incluant une structure ayant un retard RC similaire au retard le plus pessimiste (202), la structure étant sélectionnée dans le groupe composé d'une structure de capacitance, une structure de résistance, une pluralité de dispositifs discrets et un dispositif unique.
  13. Procédé pour sélectionner un emplacement mémoire dans un tableau mémoire (304), comprenant les étapes consistant à :
    • en réponse à une première transition d'un signal d'horloge (CLK), transmettre une pluralité de signaux d'adresse (ADR) correspondant à l'emplacement mémoire d'un registre d'adresse vers un décodeur, un retard le plus pessimiste (202) étant associé aux signaux d'adresse (ADR),
    • retarder le signal d'horloge (CLK) d'une quantité qui soit pour l'essentiel la même que le retard le plus pessimiste (202) et une marge (204) qui se situe entre 5 et 20 pour cent d'un cycle d'horloge du signal d'horloge (CLK), générant ainsi un signal d'activation ; et
    • sélectionner l'emplacement mémoire en réponse à une première transition du signal d'activation, la première transition du signal d'activation correspondant à la première transition du signal d'horloge (CLK).
  14. Procédé selon la revendication 13,

    dans lequel

    les signaux d'adresse (ADR) commencent à transitionner à la première transition du signal d'horloge (CLK).
  15. Procédé selon la revendication 14,

    dans lequel

    la première transition du signal d'horloge (CLK) est un front montant (102).
  16. Procédé selon la revendication 13,

    dans lequel

    la marge (204) est d'environ 5 à 10 pour cent d'un cycle d'horloge du signal d'horloge (CLK).
  17. Procédé selon la revendication 13,

    comprenant en outre

    le décodage des signaux d'adresse (ADR) en réponse au signal d'activation.
  18. Procédé selon la revendication 15,

    dans lequel

    l'emplacement mémoire est sélectionné jusqu'à une prochaine transition du signal d'activation correspondant au front montant suivant (106) du signal d'horloge (CLK).






IPC
A Täglicher Lebensbedarf
B Arbeitsverfahren; Transportieren
C Chemie; Hüttenwesen
D Textilien; Papier
E Bauwesen; Erdbohren; Bergbau
F Maschinenbau; Beleuchtung; Heizung; Waffen; Sprengen
G Physik
H Elektrotechnik

Anmelder
Datum

Patentrecherche

Patent Zeichnungen (PDF)

Copyright © 2008 Patent-De Alle Rechte vorbehalten. eMail: info@patent-de.com