Memory design
Yüklə 165,33 Kb.
|
|
DESIGN AND DC ANALYSIS OF SRAM 6T CELL INTERNSHIP REPORT Submitted By NISHA .P. JOSHI Roll no. - 09307201(temporary) Graduate Student University Of Illinois, Chicago njoshi7@uic.edu (UIN- 677316512) Under the guidance of Prof. Sachin B. Patkar 
Department of Electrical Engineering Indian Institute of Technology, Bombay Mumbai-400076 May 20 – July 20, 2010
This is to certify that Ms. Nisha Joshi has successfully completed her Internship at IIT Bombay, on Design and DC Analysis of SRAM 6T Cell, from 20th May-20th July. IIT Bombay Date: Prof. Sachin Ptakar ACKNOWLEDGEMENT I would like to thank Professor Sachin B. Patkar, Professor Jayanta Mukherji and Rewati Raman Raut for their reviews and invaluable comments and suggestions for the clarity of this report. I would like to thank the VLSI Lab staff Marshnil, Vineeth, Ani Xavier, Pallavi and Rajesh for helping me out with the Cadence 6.1 Tool issue and also for giving me few designing tip. I would also like to thank HPC lab colleagues for giving me ideas and helping me with my work. CONTENTS
Computer data storage, often called storage or memory, refers to computer components and recording media that retain digital data used for computing for some interval of time. The main purpose of storage is that without a significant amount of memory, a computer would merely be able to perform fixed operations and immediately output the result. It has to be reconfigured every time an operation needs to be performed. Hierarchy of storage is as shown in Figure 1.
Primary storage (or main memory or internal memory), often referred to simply as memory, is the only one directly accessible to the CPU. The CPU continuously reads instructions stored there and executes them as required. Any data actively operated on is also stored there in uniform manner.
Secondary storage (or external memory) differs from primary storage in that it is not directly accessible by the CPU. The computer usually uses its input/output channels to access secondary storage and transfers the desired data using intermediate area in primary storage. Secondary storage does not lose the data when the device is powered down—it is non-volatile. Per unit, it is typically also two orders of magnitude less expensive than primary storage. Consequently, modern computer systems typically have two orders of magnitude more secondary storage than primary storage and data is kept for a longer time there.
Tertiary storage or tertiary memory provides a third level of storage. Typically it involves a robotic mechanism which will mount (insert) and dismount removable mass storage media into a storage device according to the system's demands; this data is often copied to secondary storage before use.
Off-line storage is used to transfer information, since the detached medium can be easily physically transported. Additionally, in case a disaster, for example a fire, destroys the original data, a medium in a remote location will probably be unaffected, enabling disaster recovery. Off-line storage increases general information security. 
Figure1. Types of Computer data storage
The memory can be classified as volatile and non-volatile memories.
These are memories that require power to maintain the stored information. Random-access memory (RAM) is a form of computer data storage. It takes the form of integrated circuits that allow stored data to be accessed in any order. "Random" refers to the idea that any piece of data can be returned in a constant time, regardless of its physical location and whether or not it is related to the previous piece of data.
Dynamic random access memory (DRAM) is a type of random access memory that stores each bit of data in a separate capacitor within an integrated circuit. Since real capacitors leak charge, the information eventually fades unless the capacitor charge is refreshed periodically. Because of this refresh requirement, it is a dynamic memory as opposed to SRAM and other static memory. The advantage of DRAM is its structural simplicity only one transistor and a capacitor are required per bit, compared to six transistors in SRAM.
Static Random Access Memory (SRAM) is a type of semiconductor memory where the word static indicates that, unlike dynamic RAM (DRAM), it does not need to be periodically refreshed, as SRAM uses bistable latching circuitry to store each bit. SRAM exhibits data reminisce but is still volatile in the conventional sense that data is eventually lost when the memory is not powered.
Non-volatile memory, NVM or non-volatile storage, is computer memory that can retain the stored information even when not powered. Examples of non-volatile memory include read-only memory, flash memory and most types of magnetic computer storage devices.
Working of memory with processor: The processor communicates with the memory system memory interface. The processor sends the address over the address bus. For read, MemWrite is 0 and the memory returns the data on the ReadData bus. For write, MemWrite is 1 and the memory returns the data on the WriteData bus. The process is shown in the block diagram below. 
Figure 2.1 Memory Interface Computers store most commonly used instructions and data in faster but smaller memory called cache. Cache is usually built out of SRAM on the same chip as the processor. If the processor requests data that is available in the cache, it is returned quickly. This is called cache hit. Otherwise, the processor retrieves the data from main memory (DRAM). This is called cache miss. 
Figure 2.2 Cache The performance analysis i.e. the miss rate, the hit rate and average memory access time (AMAT) where AMAT is the average time a processor must wait for memory per load or store instruction. 
Random access memories which can retain their data content as long as electric power is supplied to the memory device, and do not need any rewrite or refresh operation, are called static random access memories. The following elements are required to build a SRAM,
Figure3.1 Typical Memory Architecture
SRAM adopts a multi-divided memory cell array structure to achieve high speed word decoding and reduce column power dissipation. The multi-stage decoder circuit technique is adopted as it has advantage over the one-stage decoder in reducing the number transistors and fan-in. For column decoders, which select the desired bit pairs out of the set of the sets of bit pairs in the selected row a typical dynamic and gate could be used. 
Figure 3.2 Static CMOS AND Gate 
Figure 3.3 Dynamic CMOS AND Gate Digital Analysis of Row Decoder: The address lines i.e. the word line is generated using this logic. For example, if we have 1024*32 bit memory, we need a 10:1024 (2^10) decoder. Since to obtain such a large decoder logic with and gates makes it slower, cascading of multiple smaller decoders is done. Here, there are 2 approaches that were considered:
Figure 3.4 Decoders Table 3.1 Truth table for 5:32 decoder
The basic cell for static RAM consists of 6 transistors in which there are two pass gates. We are using 2 inverters, which are cross coupled and they store the data being read or written whereas the pass transistors purpose is to select the cell which the word line activates and pass the data inputs to be read or written to the cross coupled inverters. The circuit diagram of 6T bit cell is shown below. 
Figure 3.5 6T SRAM Cell Digital Analysis 6T SRAM Cell:
Sense amplifiers, in association with memory cells, are key elements in defining the performance and environmental tolerance of CMOS memories. In an integrated memory circuit sensing means the detection and determination of the data content of a selected memory cell. CMOS memories are required to increase speed, improve capacity and maintain low power dissipation. These objectives are somewhat conflicting when it comes to memory sense amplifier design. With increased memory capacity, usually comes an increased bit-line parasitic capacitance. This increased bit line capacitance in turn slows down voltage sensing and makes bit line voltage swings expensive. 
Figure3.6 Typical Sense Amplifier The main function of sense amplifier is to sense or detect stored data from a read selected memory cell. The memory cell being read produces a current Idata that removes some charge (dQ) stored on the pre-charged bit lines. Since bit lines are very long and are shared by other similar cells, the parasitic resistance Rbl and capacitance Cbl are large. Thus the resulting bit line voltage swing (dVbl) caused by removal of dQ from the bit line is very small dVbl=dQ/Cbl. Sense amplifiers are used to translate this small voltage signal to a full signal that can be further used by digital logic. Positive Feedback Differential Voltage Sense Amplifier: 
Figure3.7 Positive feedback voltage sense amplifier Digital analysis of Positive Feedback Voltage Sense Amplifier: The positive feedback amplifier has 2 data nodes Vin/out1 and Vin/out2 and 3 control nodes SANen, SAPen and PRE. The data nodes act as input and output to the sense amplifier and its operation is as follows:
Due to its positive feedback this voltage sensing amplifier achieves a very high differential gain. This high gain minimizes sensing time by being able to sense small voltage swings on the bit line.
The write-in buffers are enabled in such a way as to have them control the bit lines. More precisely, by setting WE and OE the data terminal D is made to drive one BL and its conjugate BLB in direct and complemented form respectively. Digital Analysis of Data Read: This can be achieved by setting WE=0 which disables the write-in buffers, thereby permitting the memory cell selected to exert control over bit lines BL and BLB. Further by setting OE=1 propagates the state stored in that cell to the data terminal D via the enabled output buffer. Digital Analysis of Data Write: This is achieved by setting WE=1 and OE=0 and the data terminal D drives the bit lines. The word line is activated causing the cross-connected inverters in the selected cell to assume the value of bit lines imposed externally. After the cell is deselected, the positive feedback mechanism maintains that state indefinitely until the cell’s content is overwritten or the supply voltage gets turned off. 
Figure3.8 Read/Write Control Circuit
Tristate buffers are the buffers which are used to control the read and write operations. The write enable and output enable signals are generated through these buffers. 
Figure 3.9 CMOS Implementation of Tristate Buffer
The figure given below shows a typical static memory cell. The circuit consists of a flipflop comprising two cross-coupled inverters and two access transistors Q5 and Q6. 
Figure 4.1 Typical Static Memory Cell Prior to initiating a read operation, the bit lines are precharged to Vdd. The read operation is initiated by enabling the word line and connecting bit lines, BL and BLB to the internal nodes of the cell. Assume that the cell is storing a 1. In this case Q will be at Vdd and Qbar will be at 0 V. When the word line is selected and access transistors are turned Q5 and Q6 are turned on. 
Figure 4.2 Operation performed during Read 1 From the above given figure, we see that current flows from Bbar line through Q5. This current charges CQbar. Equilibrium is reached when CQbar is charged to VQbar at which I1 equals I5. To provide a non destructive read operation, the following design constraints are to be taken care of:
 …… for the above circuit
 …………… general form
b). The aspect ratio for read 0 is same as above except for the fact that the access transistors are different i.e. ratio if Q6/Q2 = Q5/Q1.
Figure 4.3 Bit line voltage during Read 1 operation Whereas in read 0 operation, VBLB is maintained at precharge voltage and VBL is discharged to ∆V.
 …… time delay to reach ∆V
For economic reasons, minimizing the die area becomes important in memory design as the basic storage cell gets repeated over and over again. The circuit consists of latch and pass-transistor switches. Depending on the preserved state of the inverters, data is stored in cell. To access the cell, we need switches which are connected between cell and bit line, which is controlled by read/write circuitry. Design Strategy: The most important and basic requirement for efficient 6T cell design is the aspect ratio of the transistors which is calculated depending on the following requirements:
6T Cell Design & Analysis using Cadence Tools:
From the above figure we see that the value initially stored is 0. Output Enable (OE) is turned on and the word line is activated. The voltage then developed between BL and BLbar with VBbar lower than VB when the stored value is 0 which is given by V21. RESULTS: VBL=747.7 mV and VBLB=913.7mV, Vtn =0.677 (process defined) Step1: VQbar = 606.1mV, we see that is VQbar is less than Vtn. Step2: ∆V= VBL (pre) –VBL (new) = 1.38-747.7mV=0.6323V, we see that ∆V<= Vtn Step3: Aspect Ratio (W/L) m26/ (W/L) m4 = 5 
Figure current in access transistor Step4: Time delay, ∆t = (CBL * ∆V)/I26 = (110pF * 0.6323V)/138.4uA = 50.255usec.
The aspect ratio of all the transistors used in this design is tabulated in the table shown below:
Table1. Aspect Ratio Read Zero (W/L)
From the above figure we see that the value initially stored is 1. Output Enable (OE) is turned on and the word line is activated. The voltage then developed between BL and BLbar with VB higher than VBbar when the stored value is 1 as given by V21. RESULTS: VB=1.208V and VBbar=827mV Vtn= 0.6777 (process dependant) Step1: VQbar = 468.6mV, we see that is VQbar is less than Vtn. Step2: ∆V= VBLB (pre) –VBLB (new) = 1.38-827mV=0.553V, ∆V<=Vtn Step3: Aspect Ratio (W/L) m25/ (W/L) m0 = 730 
Figure current in access transistor Step4: Time delay, ∆t = (CBL * ∆V)/I25 = (110pF * 0.553V)/19.05uA = 3.1931usec
The aspect ratio of all the transistors used in this design is tabulated in the table shown below
Table2. Aspect Ratio Read One (W/L)
Write Operation: The figure given below is the circuit diagram of a basic 6T cell. We wish to write a zero, the objective now is to pull down the node Q (as VQ= Vdd and VQbar= 0V) and node Qbar up and have the voltages of at least one of these two nodes pass.
 … General form
From the above figure we see that the value stored is 1. Write Enable (WE) are turned on and the word line is activated. The voltage then developed between BL and BLB with VBL higher than VBLB when the value to be written is 1 as given by V21. RESULTS: VBL=1.001V and VBLB=836.6mV, Vtn= 0.6777 (process dependant) Step1: VQ = 1 V, we see that is VQ is more than Vtn. Step2: ∆V= VBL (pre) –VBL (new) = 1.38-1.001V=0.379V Step3: Aspect Ratio (W/L) m26/ (W/L) m5 = 10 
Figure current in access transistor during write 1 Step4: Time delay, ∆t = (CBL * ∆V)/I26 = (110pF * 0.379V)/13.4fA = 3111sec
The aspect ratio of all the transistors used in this design is tabulated in the table shown below:
Table3. Aspect Ratio Write One (W/L)
From the above figure we see that the value stored is 0. Write Enable (WE) are turned on and the word line is activated. The voltage then developed between BL and BLB with VBL lower than VBLB when the value to be written is 0. RESULTS: VBL=218.1mV and VBLB=927.8mV, Vtn= 0.6777 (process dependant) Step1: VQ = 592.7m V, we see that is VQ is less than Vtn. Step2: ∆V= VBL (pre) –VBL (new) = 1.38- 218.1mV=1.1619V Step3: Aspect Ratio (W/L) m26/ (W/L) m5 = 10 
Figure current in access transistor during write Step4: Time delay, ∆t = (CBL * ∆V)/I26 = (110pF * 1.1619V)/13.15mA = 9.719nsec
The aspect ratio of all the transistors used in this design is tabulated in the table shown below
Table4. Aspect Ratio Write Zero (W/L)
Cadence is an electronic design automation (EDA) software and engineering services company.
Cadence can be opened from any operating system. From windows it can be opened on Xmanager. 
The Xterm window opens up as shown below, 
After giving the icfb& command, the following window opens up 
After the virtuoso window opens up click on File NewLibrary 
As the library window opens up, Namedo not need process information 
Then go back to the virtuoso window FileNewCell. The following window opens up. In the cell tag put in the name of the cell you want. Once the name is given, the virtuoso schematic editor window opens up. 
On your Xterm window type vi cds.lib as shown in the figure below then you have to make changes in the library path to add it in your home directory. 
The figure shown below is the Virtuoso Schematic Editor which opens when we make a new library and cell view. 
LibraryBrowse the window below opens up. Select the component you want to use. 
The figures given below shows how to use wire, instance properties, move, copy, delete, pin names, wire names. 
Once the schematic is complete, you do check and save and no errors should be there on the CIW window. 
Fast keys which could be used during designing. 
Once the design is checked and saved with no errors, we have to perform the simulation. That can be achieved by clicking on LaunchADE L. it would look like the following two window as shown in the figure. 
Then we have to set the model libraries by selecting Setup model libraries as shown in the figure given below. 
As the model library setup window opens as shown in the figure given below, click on the grey tab to browse the libraries to be added. 
The browser window opens up to choose the model file as shown in the figure given below. 
The steps are followed as FreePDK45ncsu_basekitmodelshspicetran_modelsmodels_nom 
Once the model libraries are setup, we can choose the analyses we want to run by clicking on Analyses. 
We choose the type of analyses whether it is transient or dc or ac. 
Once the analyses are chosen we can plot the outputs by selecting the Outputto be plottedSelect on Schematic. The parameters to be plotted are shown in the figure given below 
Once all these things are taken care of then we can simulate it by clicking on Netlist and Run 
The following result window opens up . 
The CIW window showing that simulation is completed successfully. 
Once the simulation is successful, the outputs to be plotted are displayed. To read the output in more detail, few steps have to be followed. The figure given below is the merged graph of the outputs to be plotted. 
To read in detail click on AxisStrips 
The graph will show individual readings of each output. To read the value on a particular point on graph, click on MarkerPlaceTrace Marker as shown below. 
The point at which you place the marker will show the x-axis and y-axis reading at that particular point. 
[1] Sedra/ Smith, Microelectronic Circuit, chapter 15, 6th edition [2] Wai-Kai-Chen, Memory, Microprocessor an ASIC, 2003 [3] Tegze P. Haraszti, CMOS Memory Circuits, Kluwer Publication, 2002 [4] David Money Harris & Sarah L. Harris, Digital Design and Computer Architecture [5] Mohammed Sharifkhani, Design and Analysis of low power Sram, thesis report, Waterloo University, Canada [6] Prof. Etienna Sicard and Asst Prof. Sonia Delmas Bendhia, INSA Electronic Engineering School of Toulouse, France, Advanced CMOS Cell Design [7] Randall L. Geiger, Phillip E. Allen and Noel R. Strider, VLSI Design Techniques for Analog and Digital Circuits [8] R. Jacob Baker, Harry W. Li and David E. Boyce, CMOS Circuit Design, Layout and Simulation [9] Neil Weste, CMOS VLSI Design: A Circuit and System Perspective, International Edition Yüklə 165,33 Kb. Dostları ilə paylaş:
|
