DAC Testbench

Measure DC and AC performance metrics of DAC output

Since R2020a

• 

×

Libraries:
Mixed-Signal Blockset / DAC / Measurements & Testbenches

Description

The DAC Testbench block measures both DC and AC performance metrics of a DAC (digital to analog converter). DC performance metrics include offset error and gain error. AC performance metrics include signal-to-noise ratio (SNR), signal to noise and distortion radio (SINAD), spurious-free dynamic range (SFDR), effective number of bits (ENOB), and noise floor.

The DAC Testbench block generates the stimulus to drive the device under test (DUT) from the Stimulus tab. The setup parameters for validating the DUT are defined on the Setup tab. The target validation metrics are defined on the Target Metric tab.

You can use the DAC Testbench block to validate the DAC architecture models provided in Mixed-Signal Blockset™, or you can validate a DAC of your own implementation.

Examples

Measure Offset and Gain Error of Binary Weighted DAC

Find the offset and gain errors of a binary weighted DAC block.

• Open Script

Measure AC Performance Metrics of Binary Weighted DAC

Find the AC performance metrics such as SNR, SINAD, SFDR, ENOB, noise floor and settling time of a binary weighted DAC block.

• Open Script

Ports

Input

from dac analog — Analog signal from DAC output

scalar

Analog input signal from the DAC output, specified as a scalar.

Data Types: double

Output

to dac digital — Digital stimulus signal for DAC input

scalar

Digital output stimulus signal for the DAC input, returned as a scalar.

Data Types: single | double | int8 | int16 | int32 | uint8 | uint16 | uint32 | fixed point

to dac start — External start conversion clock for DAC

scalar

External start conversion clock for DAC, returned as a scalar. The rising edge of this signal starts the conversion process in the DAC block.

Data Types: double

Parameters

Measurement — Select whether to measure DC or AC performance metrics

DC (default) | AC

Select whether to measure static (DC) or dynamic (AC) performance metrics:

• Select DC to measure offset error and gain error.

• Select AC to measure SNR, SINAD, SFDR, ENOB, and noise floor.

Recommended min. simulation stop time (s) — Minimum time simulation must run for meaningful result

2.048e-02 (default) | positive real scalar

Minimum time for which the simulation must run to obtain meaningful results, specified as a positive real scalar in seconds.

• To measure DC performance, the simulation must run so that the DAC can sample each digital code 20 times. Based on this assumption, the Recommended min. simulation stop time (s) T is given by:

  $$T=\frac{\text{Samples per bit}}{\left(StartFreq/2^{Nbits+1}\right)}+\text{Hold off time}$$,

  where StartFreq is the frequency of the conversion-start clock and Nbits is the resolution of the DAC.

  The number of samples per bit is calculated using the equation:

  $$\text{Samples per bit = }\max \left(\frac{1}{\text{Error tolerance}},10\right)$$.

• To measure AC performance, the simulation must run so that the DAC can generate six spectral updates of the DAC output. So, the Recommended min. simulation stop time (s) T is given by [1]:

  $$T=6\left(\frac{1.5}{RBW}+\text{Hold off time}\right)$$,

  where RBW is the resolution bandwidth of the spectrum estimator inside the DAC Testbench block and is given by the equation: $$RBW=\left[\min \left(\text{Input frequency}\right)0.1\right]$$.

This parameter is only reported by the testbench and is not editable.

Data Types: double

Set as model stop time — Automatically set recommended min. simulation stop time as model stop time

button

Click to automatically set the Recommended min. simulation stop time (s) as the stop time of the Simulink^® model.

Endpoint — Measure DNL, INL using endpoint method

on (default) | off

Measure the differential nonlinearity (DNL) error and integral nonlinearity (INL) error using the endpoint method. This method uses the endpoints of the actual transfer function to measure the DNL and INL errors.

Dependencies

To enable this parameter, set Measurement to DC.

Best fit — Measure DNL, INL using best fit method

on (default) | off

Measure the differential nonlinearity (DNL) error and integral nonlinearity (INL) error using the best fit method. This method uses a standard curve-fitting technique to find the best fit to measure the DNL and INL errors.

Dependencies

To enable this parameter, set Measurement to DC.

Plot DC analysis result — Plot DC analysis result

button

Click to plot DC analysis result for further analysis. To perform a complete DC analysis including integral nonlinearity (INL) and differential nonlinearity (DNL), use the DAC DC Measurement block.

Dependencies

To enable this parameter, set Measurement to DC.

Export measurement result — Store detailed test results to base workspace

button

Click to store detailed test results to a spreadsheet (XLS file) or as comma-separated values (CSV file) for further processing.

Stimulus

Digital input frequency (Hz) — Frequency of digital input signal to DAC

1e4 (default) | positive real scalar

Frequency of the digital input signal to the DAC block, specified as a positive real scalar in hertz. Digital input frequency (Hz) must match the input frequency of the DAC device under test.

Digital input frequency (Hz) needs to satisfy two requirements:

• All the output codes of the DAC must be activated.

• The Digital input frequency (Hz) must not share any common multiples other than 1 with the Start conversion frequency (Hz).

Dependencies

To enable this parameter, set Measurement to AC.

Programmatic Use

Block parameter: InputFrequency

Type: character vector

Values: positive real scalar

Default: 1e3

Data Types: double

Start conversion frequency (Hz) — Frequency of internal start-conversion clock

1e6 (default) | positive real scalar

Frequency of internal start-conversion clock, specified as a positive real scalar in Hz. Start conversion frequency (Hz) determines the rate of the DAC.

Programmatic Use

Block parameter: StartFreq

Type: character vector

Values: positive real scalar

Default: 1e6

Data Types: double

Error tolerance (LSB) — Maximum difference between successive samples of digital signal

0.1 (default) | positive scalar in the range (0, 1]

Maximum allowed difference in the amplitude of successive samples of the digital input signal, specified as a positive real scalar in least significant bit (LSB).

Dependencies

To enable this parameter, set Measurement to DC.

Programmatic Use

Block parameter: ErrorTolerance

Type: character vector

Values: positive scalar in the range (0, 1]

Default: 0.1

Data Types: double

Setup

Autofill setup parameters — Automatically propagate setup parameters from DAC

button

Click to automatically propagate setup parameters from the DAC.

Dependencies

The DAC must be a Binary Weighted DAC from the Mixed-Signal Blockset.

Number of bits — Number of bits in input word

10 (default) | positive real integer

Number of bits in the input word, specified as a unitless positive real integer. Number of bits determines the resolution of the DAC.

Programmatic Use

Block parameter: NBits

Type: character vector

Values: positive real integer

Default: 10

Data Types: double

Input polarity — Polarity of input signal to DAC

Bipolar (default) | Unipolar

Polarity of the input signal to the DAC.

Programmatic Use

Block parameter: Polarity

Type: character vector

Values: Bipolar|Unipolar

Default: Bipolar

Reference (V) — Reference voltage

1 (default) | real scalar

Reference voltage of the DAC, specified as a real scalar in volts. Reference (V) helps determine the output from the input digital code, Number of bits, and Bias (V) using the equation:

$$\text{DAC output = }\left(\left(\frac{\text{Digital input code}}{2^{\text{Number of bits}}}\right)\text{Reference}\right)+\text{Bias}$$.

Dependencies

To enable this parameter, set Measurement to DC.

Programmatic Use

Block parameter: Ref

Type: character vector

Values: real scalar

Default: 1

Data Types: double

Bias (V) — Bias voltage added to output

0 (default) | real scalar

Bias voltage added to the output of the DAC, specified as a real scalar in volts. Bias (V) helps determine the output from the input digital code, Number of bits, and Reference (V) using the equation:

$$\text{DAC output = }\left(\left(\frac{\text{Digital input code}}{2^{\text{Number of bits}}}\right)\text{Reference}\right)+\text{Bias}$$

Dependencies

To enable this parameter, set Measurement to DC.

Programmatic Use

Block parameter: Bias

Type: character vector

Values: real scalar

Default: 0

Data Types: double

Settling time (s) — Time required for output to settle

0.25/1e-6 (default) | nonnegative real scalar

The time required for the output of the DAC to settle to within some fraction of its final value, specified as a nonnegative real scalar in seconds.

Dependencies

To enable this parameter, set Measurement to DC.

Programmatic Use

Block parameter: SettlingTime

Type: character vector

Values: real scalar

Default: 0.25/1e-6

Data Types: double

Settling time tolerance (LSB) — Tolerance for calculating settling time

0.5 (default) | positive real scalar

The tolerance allowed for calculating settling time, specified as a positive real scalar in LSB. The output of the DAC must settle within the Settling time tolerance (LSB) by Settling time (s).

Dependencies

To enable this parameter, set Measurement to AC.

Programmatic Use

Block parameter: SettlingTimeTolerance

Type: character vector

Values: positive real scalar

Default: 0.5

Data Types: double

Hold off time (s) — Delay before measurement analysis

0 (default) | nonnegative real scalar

Delay before measurement analysis to avoid corruption by transients, specified as a nonnegative real scalar in seconds.

Programmatic Use

• Use get_param(gcb,'HoldOffTime') to view the current value of Hold off time (s).

• Use set_param(gcb,'HoldOffTime',value) to set Hold off time (s) to a specific value.

Data Types: double

Show spectrum analyzer during simulation — Displays Spectrum Analyzer during simulation

off (default) | on

Select this parameter to display the Spectrum Analyzer window during simulation. By default, this parameter is deselected.

Dependencies

To enable this parameter, set Measurement to AC.

Target Metric

Autofill target metric — Automatically propagate target metrics from DAC

button

Click to automatically propagate target metrics from the DAC.

Dependencies

• To enable this parameter, set Measurement to DC.

• The DAC must be a Binary Weighted DAC from the Mixed-Signal Blockset.

Offset error — Shifts quantization steps by specific value

0 LSB (default) | real scalar

Shifts quantization steps by a specific value, specified as a real scalar in %FS (percentage of full scale), FS (full scale), or LSB (least significant bit).

Note

The full scale range of the converter is defined as the difference between the last and first code on the +0.5 LSB compensated transfer curve. In a +0.5 LSB compensated transfer curve, first code is 0.5 LSB wide while the last code is 1.5 LSB wide. The input values must be considered within the full scale range of the converter.

Note

LSB is calculated by the equation $$\text{LSB = }\frac{\text{Full scale range of converter}}{2^{NBits}}$$.

Dependencies

To enable this parameter, set Measurement to DC.

Programmatic Use

Block parameter: TargetOffsetError

Type: character vector

Values: real scalar

Default: 0 LSB

Data Types: double

Gain error — Error in slope of DAC transfer curve

0 LSB (default) | real scalar

Error in the slope of the straight line interpolating the DAC transfer curve, specified as a real scalar in %FS (percentage of full scale), FS (full scale), or LSB (least significant bit).

Note

The full scale range of the converter is defined as the difference between the last and first code on the +0.5 LSB compensated transfer curve. In a +0.5 LSB compensated transfer curve, first code is 0.5 LSB wide while the last code is 1.5 LSB wide. The input values must be considered within the full scale range of the converter.

Note

LSB is calculated by the equation $$\text{LSB = }\frac{\text{Full scale range of converter}}{2^{NBits}}$$.

Dependencies

To enable this parameter, set Measurement to DC.

Programmatic Use

Programmatic Use

Block parameter: TargetGainError

Type: character vector

Values: real scalar

Default: 0 LSB

Data Types: double

References

[1] Spectrum Analyzer

Version History

Introduced in R2020a