Analog-to-digital conversion with SAR

An analog digital converter (ADC abbreviation, or ADC A / D) is an electronic integrated circuit whose function is to generate numerical values from the continuous or analog inputs it receives. Our interest focuses on the new technique based on the successive approximation converters. This technique is none other than the segmentation method that is applied in the topology D-to-A or digital converter.

The segmentation method is to leave the value of components so that is proportional to the values per segment.

See also the section "Digitizing", "Sampling" and Quantification.

Description of these architectures

As in many applications that we use in everyday life, we need to move from the continuous mode to digital mode. A wide range of converter available on the market, not to mention the documents that come with it. Our task consists in presenting a new approach that is not currently used too. As far as architecture, you have the pleasure of discovering this in the following link.

* Analog-to-digital converter

Some examples

We will give a short summary of the two architectures that can be used for a converter Analog-to-digital in the chain of an ADC.

Reseau R-2R

As you can see the architecture of the network R-2R consists of a series of a resistance whose values vary between R and 2R. The problem with this architecture is that it consumes too much energy.

Switched-capacitor network

Here now, we have the same topology for a DAC but with capacitiesCommunitées. The difference between this architecture and aboveMore lies at the level of energy consumption. Element precious thanhuman beings are in a perpetual quest to decrease the value.

Principe of segmentation

The segmentation is to rearrange the value resistors or capabilities so that we have shared values and that repeat.

Take the example of a converter Analog-to-digital with network capacitif. In current architectures, the value of the capacity is exponentially. Admittedly, the consumption of capacity requires less energy but that the value of the capabilities vary widely. To overcome this problem of change and other aspects that come with, we propose the Segmentation. Suffice it to divide capacity in block of 4 or 3. Recall that the number of block depends on the application. Once the division made, we must assign values to the capabilities so that these same values have to be each block.

Example:Assume an ADC with nine switched-capacitorThe implementation of the Segmentation will be to allocate capacity 3 block. For a block, asking value. C0=2pF; C1=4pF; C2=6pF; For the rest of the blocks, we will have the same values in that order. C3=2pF; C4=4pF; C5=6pF; C6=2pF; C7=4pF; C8=8pF.

References

*http://www.analog.com/en/cat/0,2878,760,00.html
*http://ieeexplore.ieee.org/Xplore/dynhome.jsp
*http://cmosedu.com/cmos1/book.htm
*http://www.ansoft.com/