Tuesday, December 14, 2010

Design Guide series: Q & A DAC

DAC's function

The real world of analog signals, such as temperature, pressure, sound or image, etc., are continually converted to easier to store, process and transmit in digital form.

But in many systems, digital information also must be converted to analog signals for some real-world functionality. Digital-to-analog converter (DAC) can do this, but their output can also be used to drive a variety of devices such as microphones, engine, radio frequency transmitters and temperature controller.

DAC generally are placed in a digital system.

In digital systems, some real-world signals from analog to digital converter (ADC), digital and processing and then needs to be converted into analog signals of the system. These systems required DAC performance will be affected by the rest of the system performance and requirements.

The basic principle

DAC will produce a quantization (discrete phase) analog output to respond to a binary input encoding.

This digital input is probably the ECL and TTL, CMOS or LVDS, and analog output is a voltage or current output.

To produce output, a reference quantity (voltage or current) is divided into binary segment or linear segments.

Next, the digital input will drive switch, the switch will be a certain number of segments connected to the output. Paragraph number reflects the possible number of digital input encoding, this is a feature of the converter resolution, or you can reflect the number of digits in the input encoding (N). With n digits, the number of possible coding to 2N.

DAC output amplitude can be expressed in the following ways:

Analog output = digital input encoding/(2N-1) × reference input

Analog signal is continuity with infinite resolution time domain signal.

However, the DAC output is one built on discrete values (quantitative) of the signal, this discrete values of unity but also has an infinite interval (sampling). In other words, the DAC output is trying to show an infinite resolution and bandwidth analog signal.

Quantization and sampling performance of DAC and a basic but it can be expected to limit.

Quantitative decision a converter, the maximum dynamic range and noise quantization error and produced. According to the Nyquist criterion, sampling the DAC output signal to determine the maximum bandwidth.

In an ideal DAC, analog signal is a least significant bit (LSB), and a LSB is a divided into 2N segment full analog output amplitude, N is a digit representation of DAC resolution.

However, the DAC in the real world operations is affected by some of the effects of non-ideal and unable to meet those requirements by quantifying and sampling the ideal effect.

Some of these errors to AC and DC performance specifications for features, it is precisely these specifications determine the converter of static and dynamic performance.

Offset errors

The offset error is DAC output and when gain error zero ideal transfer function of the deviation between the output.

For all the input encoding to offset error is constant.

Gain error

Gain error is a converter of the transfer function of the slope and the ideal of the transfer function of the deviation between the adoption of compensation to the zero offset error to measure.

Differential nonlinearity (DNL)

Differential nonlinearity is a real step size and an ideal deviation between LSB step.

DNL error will produce more than quantitative effects outside the extra noise and burrs.

Integral nonlinearity (INL)

Integral nonlinearity is ideal for the actual output voltage and output voltage at the transfer function of the straight line between the endpoints of the error.

INL is offset errors and gain error elimination before. INL error will cause extra harmonic and burrs.

Monotonicity

If a DAC output as digital input encoding grows or remain unchanged, the DAC has the monotony.

Conversely, if a DAC output as digital encoding of increases and decreases, the DAC has non-monotonic.

Establish time

Establishment time is the analog output response digital input on a step change in qualified error identified a number of times.

False signals

False signals are input and output when the converter changes the State from input to output loading.

Digital feed through

Digital feed through refers to when the DAC is no startup, high-frequency signal penetrates to the converter logic output produced by the output noise.

Noise has more characteristics parameters such as parasitic free dynamic range (SFDR), the total harmonic distortion (THD) and signal to noise ratio (SNR).

Another parameter THD + N is the total harmonic distortion plus noise and fundamental amplitude ratio.

The following diagram shows the structure of DAC.

No comments:

Post a Comment