摘要:
A description is given of a method for operating an integrated memory which has memory cells each having a selection transistor and a storage capacitor with a ferroelectric storage effect. The memory contains a plate line, which is connected to one of the column lines via a series circuit containing a selection transistor and a storage capacitor of respective memory cells. A memory access is carried out according to the “pulsed plate concept”. In this case, the temporal sequence is controlled in such a way that, in an access cycle, the storage capacitor of the memory cell to be selected is in each case charged and discharged by the same amount. An attenuation or destruction of the information stored in the memory cells, which is caused by source-drain leakage currents of unactivated selection transistors, is thus avoided.
摘要:
An integrated semiconductor memory has memory cells with a ferroelectric memory property. The memory cells are in each case connected between a column line and a charge line. The column line is connected to a read amplifier which supplies an output signal. The charge line is connected to a driver circuit which provides the charge line with a given potential. In an inactive mode, the column line and the charge line are jointly connected to a connection for a common supply potential in the read amplifier or in the driver circuit. As a result, a relatively quick equalization of a potential between the lines is possible. Thus, unintended changes in the memory cell content due to interfering voltages are avoided.
摘要:
A circuit configuration for controlling write operations and read operations in an MRAM memory configuration includes selection transistors grouped in sections of equal numbers of the selection transistors. The selection transistors of each of the sections are jointly connected, at the ends of the bit lines, to a respective interacting pair of read/write amplifiers via those electrode terminals of the selection transistors that are not connected to the bit lines. The read/write amplifiers are controlled such that if a write signal is fed thereto, write currents for writing a logic “1” or “0” flow in a first direction or a second direction in all of the bit lines selected by a corresponding column select signal and, if a read signal is fed in, a logic state stored in one of the magnetoresistive memory cells can be read out.
摘要:
An integrated memory has a memory cell array with memory cells which are connected to word lines and bit lines. For the purpose of reading from or writing to one of the memory cells, a first word line can be connected to a supply circuit via a controllable first switching device and a second word line can be connected to the supply circuit via a controllable second switching device. A control circuit can drive the first switching device in dependence of an activation state of the second word line and the second switching device in dependence of an activation state of the first word line. Consequently, existing word lines that are not currently being used can be used for addressing one of the memory cells. As a result, only one wiring plane is required for the word lines.
摘要:
A magnetic memory configuration stores data and avoids ageing effects. The memory configuration contains a cell array containing magnetic memory cells disposed along a first direction and a second direction crossing the former, a multiplicity of electrical lines along the first direction, and a multiplicity of electrical lines along the second direction. The magnetic memory cells in each case are disposed at crossover points of the electrical lines. A first current supply device supplies respectively selected electrical lines along the first direction with current. A second current supply device supplies respectively selected electrical lines along the second direction with current. The second current supply device is configured for setting the direction of the current in accordance with an information item to be written. The first current supply device is suitable for changing over the direction of the current as desired.
摘要:
The invention relates to a configuration for implementing redundancy for a memory chip, in which a fuse bank is connected to a comparator via a redundancy predecoder so that predecoded addresses can be compared with one another in the comparator and undecoded addresses can be stored in the fuse bank. This provides for a low-power and space-saving design.
摘要:
Embodiments of the invention provide a memory arrangement having an even number k=4 of physically spaced RAM chips, on each of which it is possible for m data items to be simultaneously written or read via an m-bit data bus, also having a register for buffer-storing and transmitting n respective parallel data bits as a packet between an n-bit parallel port and the data buses, and having a selection device which responds to selection bits in order to select a respective separate cell group within the plurality of the chips for each of the disjoint m-bit groups (d) of the n-bit packet. The k chips are classified into q=2 disjoint chip groups, each of which comprises k/q chips which differ as little as possible from one another in terms of their distance from the register. The number m is chosen to be equal to q*n/k, and the selection device is designed to select a respective separate chip from the same chip group and a cell group in this chip for each m-bit group of the same n-bit packet.
摘要:
A configuration and method for low-loss writing of an MRAM includes setting voltages at bit lines and word lines such that the voltage across the memory cells between a selected word/bit line and the individual bit line/word lines is minimal. A voltage drop occurs on a selected word/bit line connected to a particular memory cell when writing into the memory cell and voltages at the bit/word lines are set to minimize a cell voltage across the memory cells between a selected word/bit line and individual bit/word lines. A voltage drop occurs on a selected word/bit line connected to a particular memory cell when writing into the particular cell, and, when a voltage V1 and a voltage V2
摘要:
Embodiments of the invention provide a memory arrangement having an even number k=4 of physically spaced RAM chips, on each of which it is possible for m data items to be simultaneously written or read via an m-bit data bus, also having a register for buffer-storing and transmitting n respective parallel data bits as a packet between an n-bit parallel port and the data buses, and having a selection device which responds to selection bits in order to select a respective separate cell group within the plurality of the chips for each of the disjoint m-bit groups (d) of the n-bit packet. The k chips are classified into q=2 disjoint chip groups, each of which comprises k/q chips which differ as little as possible from one another in terms of their distance from the register. The number m is chosen to be equal to q*n/k, and the selection device is designed to select a respective separate chip from the same chip group and a cell group in this chip for each m-bit group of the same n-bit packet.
摘要:
One embodiment of the invention provides a hub chip comprising: an address bus input for receiving a plurality of successively sent portions of address and/or command data, a shift register which has register elements and is connected to the address bus input to receive the plurality of portions of the address and/or command data, the shift register being connected to the address bus input so that, when the address and/or command data are received, the portions of the address and/or command data are successively written to the register elements, an address bus output for outputting the received address and/or command data, a memory module interface for connecting one or more memory modules, where the hub chip addresses none, one or a plurality of the connected memory modules, depending on the address and/or command data transferred, and a driver element provided to output the received portion of the address and/or command data to the address bus output before all of the portions of the address and/or command data have been received in full.