<mosaic.cnfolio.com>

Coursework report


Introduction
Semiconductor memories where created when scientists and engineers where able to create logic gates out of transistors and they configured them to store data in ones and zeros. Current known storage mediums are RAM, ROM, EEPROM and Flash. where RAM is Random Access Memory which is a class of storage medium that can store data till the computer/microcontroller gets restarted and the data stored will be lost, and it is used to store value of certain variables. ROM is Read Only Memory, it represents how a computer/microcontroller executes the program and it is set by the manufacturer and can't be modified, the last bit is EEPROM which means Electrically Erasable Programmable Read Only Memory, where you can store information in it and this information or data aren't volatile that when the computer/microcontroller gets restarted the information and/or data will remain safely stored, and can only erased "electrically", i.e when the user or programmer gives a command to erase some or all the data from the chip. Flash is kindEE of similar to EEPROM but erasing the data in it is done faster than that of EEPROM, EEPROM you erase the data bit by bit, while in Flash you can erase the data all at once.

EEPROM/Flash relationship with Microcontrollers
Microcontrollers is a Microprossors with programmable input/output peripherals that can communicate with LEDs, Sensors, Digital to Analogue Converters, etc. Microcontrollers was , and that was the begining of setting in motion an engineering feat that dramatically altered the course of electronics. it was Intel 4004, with a ROM for storing the application and RAM for storing data that is being processed, and it could be used to be used for a variety of products not just for calculations. when the memory technology improved so did the Microprocessors/Microcontrollers.

I have chose 2 Microcontrollers intended for Mixed Signal Processing in this coursework, one of them is C8051F120 is a mixed signal processing chip created by Silicon Labs and have a flash memory in it, while the other is an MSP430G2955 a MSP chip created by Texas Instrumentation

Priorities and differences in Design

C8051F120

MSP430G2955

I have put the some of the features of the 2 microcontrollers to show the difference in Design and goals for each microcontroller, and as you can see the MSP430G2955 prides its self by the low power consumption and its fast speed wake up from sleep mode and this is seen in the fact that lots of its features are related to power saving , while on the other hand the C8051F120 have more features than the MSP430G2955 like the temperature sensor, more ADC with different resolution, more timers, where you can modify them as counters, much more RAM but in terms of power consumption, the only feature that relates to power consumption is that it have a stand by, power off and active mode, which is standard in most microcontrollers. Hence you can see that C8051 is intended for more versatile roles and have more tools so the User can make much more complex programs relating to Instrumentation, while on the other hand the Texas Instrumentation realised that most sensors and microcontrollers are put in remote places (e.g a sensor is put in an oil pipeline in the middle of the desert) and figured that the top priority for a microcontroller is to conserve as much energy as possible and the MSP430G2955 is intended to do just that.

The power consumption is related to the Memory they use, the C8051F120 uses the classic semiconductor Flash memory while the MSP430G2955 uses the FRAM Flash memory

Flash Technology

A single cell or unit of flash consists of an npn layer, with gate or worldline consisting of two layers a control gate and beneath it is a floating gate, with the floating gate surrounded by a an oxide insulator. Without any Current or Voltage, electrons in the Source can't pass through the p layer and into the drain, because of the buffer zone, but if there is a Voltage potential difference between the Ground, the Wordline and the Bitline, where Wordline and Bitline have a positive charge, the buffer zone will decrease allowing the electrons to go from the source to the drain and in doing that some electrons will go and pass through the floating gate to the Wordline, and when the voltage is removed, the charge will stay in the floating gate because it is sandwiched between two thin oxide layers that act as insulators and prevents the charge inside from going anywhere, the only way to remove the charge is to apply a negative charge through the wordline which would repel the electrons inside.

FRAM Technology

Property of a Ferroelectric material, when the ferroroelectic is formed (by applying a strong electric field to the ferro polymer, making the crystal polarized in a macroscopic level) it will remain polarized even when the strong electric field is removed, and you can change the polarity of the polymer/crystal by applying an electric field, this property allows it to store information, being that the polymer at one polarity (-+) means TRUE/ONE and at the other polarity (+-) means FALSE/ZERO.

Because of how the structure is made that when an electric field is removed, the polarity remains (non-volatile) and power consumption to the polarity is very low, while for EEPROM and Flash and other medias, they use semiconductor Integrated Circuits (where they store data in semiconductor diodes).

Energy Consumption

if we compare FRAM to Flash and EEPROMs we will find that FRAM power consumption is 250 times less than the Flash when microcontrollers run at similar speeds, also it operates at a much lower current of 250 uA to 9 uA to 0.1 uA compared to that of the Flash which is 2.3 to 2.5mA. FRAM writes and reads data at a minimum of 1.8 V compared to that of Flash which is at a minimum of 3.6V. We will also find that the high speed write time of the FRAM is 30000 times less than that of the Flash.

The difference between the FRAM and other memory products we will find that there isn't much difference except for the fact that it consumes less energy/ have more writes in it and write data faster than other memory products because of the properties of the ferroelectric (low energy consumption, don't have to spend lots of energy in changing from 1 to 0 and vice versa), it is faster (changing the electric polarity is quicker than changing the magnetic polarity or sending data through and IC)

This all because the FRAM's ones and zeros is caused by changing the polarity of the FRAM crystal/polymer, which consumes less energy and is much faster than the classical semiconductor memories which keeps the ones and zeros (electrical chargers) in a loop made out of logic gates and transistors.

Reference
http://www.designspark.com/nodes/view/type:knowledge-item/slug:fram-vs-flash
http://www.ti.com/general/docs/lit/getliterature.tsp?genericPartNumber=msp430g2955&fileType=pdf
http://www.keil.com/dd/docs/datashts/silabs/c8051f12x.pdf
http://www.intel.com/content/www/us/en/history/museum-story-of-intel-4004.html


Attachment Timestamp Size