SDRAM Write-Read basics
Submitted by SDRAM Technology on Wed, 09/23/2009 - 10:24
Let us assume that we want to access the memory cell which is situated at the intersection of WL2 and BL3 in an SDRAM array-segment (see Figure 1 for illustrations). WL2 and BL3 are the word-line number two (2) and bit-line number three (3) correspondingly. See the article “What are bit-lines and word-lines on a SDRAM chip?” for more details about bit-lines and word-lines.
Figure 1: Schematic representation of the basic write/read process on a SDRAM chip
The read/write process of a target memory cell (in our case WL2/BL3) will take place as follows:
- The word-line-driver number two (WLD2) will be activated. “Activated” in this case means that the potential of the WL2 connected to the WLD2 will be increased (e.g. to 2.7 V depending on the technology). Thus, all cell-transistors connected to the WL2 will open and correspondingly the information contained in the cells will leave the cells (i.e. the capacitances of the SDRAM cells) and will reach the BLs. At a first glance it seems that the information which was in the cells before the activation of the WLD2 will be lost, due to the fact that the transistors are open, however this is where the so called sense-amplifiers (SAs) are coming into play.
- All SAs which are connected to the BLs of an array-segment will “sense” the information (“0” or “1”) coming from the “opened cells” and … ATTENTION will send the SAME information (“0” or “1”) back into the cell. So the SAs are sensing the information form the cells and immediately are writing the same information back into the cells. This process is known as information “refreshing”. For the read process, the SA which was specified by the read-address, in our case SA3/BL3, will send the sensed information not only back into the cell, but also to a higher level (e.g. to the computer). In this way the computer will obtain the information from the addressed cell (WL2-BL3) and use it for its purposes.
- The situation will change for the write-process compaired to the read-process. For a write- process the SA3 will not send back into the cell the original sensed information, but the new information which has to be written into the cell, i.e. the information received from a higher level, e.g. the computer.
- The read or write process will be closed by decreasing the potential of the WL2, e.g. to -0.5V, in order to close the transistors and in this way to keep the “new” or “new-old” information into the cells.