Circuit Note

CN-0584

analog.com

Rev. 0 | 4 of 7

Figure 9 shows the frequency response of the output path, with

ideal waveforms generated in software, loaded into a cyclic buffer,

and written to the DAC. Slightly less attenuation is seen in this

plot due to the lower latency when compared to the full loopback

measurement, which must pass samples to the FPGA and back.

Figure 9. DAC Output Frequency Response

POWER ARCHITECTURE

All power for the CN0584 is provided by the CN0585 through the

AFE connector.

The CN0584 uses the +15 V and -15 V rails to provide the positive

and negative supply voltages for the ADG5421F input protection

switches. The +12 V and -12 V rails provide the positive and

negative supply voltages for the ADA4898-1 ADC buffer amplifiers.

The +3.3 V rail powers the EEPROM circuit.

SOFTWARE OVERVIEW

Python

The CN0584 can be interfaced to Python using the pyADI-IIO

library, which enables device configuration, capture of incoming

samples from the ADCs, and generation of waveforms to be trans-

mitted by the DACs. The output waveforms can then be pre- or

post-processed based on ADC data, allowing a user to quickly and

easily implement control loops or simulations of other hardware.

Figure 10 shows an arbitrary waveform generated using Python

and written to the DAC output. This example shows the generation

of a ±10 V peak-to-peak 5 kHz sine wave, but any waveform can

be implemented. For example, for an HIL setup one can generate a

damped sine wave, chirp, noise, or distorted waveform to simulate

defective hardware.

Figure 10. Example DAC Output Generation with Python

Figure 11 shows the measured input waveform for analysis, plotting

the sine wave generated in Figure 10. The samples read from

the ADCs can be processed and used as inputs to user-imple-

mented Python algorithms implementing control loops, hardware

emulations, or any other desired feedback systems.

Figure 11. ADC Samples Read and Plotted Using Python

MATLAB and Simulink

MATLAB® can also be used to interface with the CN0584 using

the High Speed Converter Toolbox, which enables largely the same

set of features as described for Python above. In addition to this,

Simulink® can be used to create custom HDL models which can

be loaded into the FPGA and run in real time using the CN0584.

Existing Simulink models or custom algorithms can be used to

implement control systems, perform digital signal processing (DSP)

operations on measured signals, or emulate hardware functions.

Figure 12 shows the placement of a Simulink model inserted in both

receive and transmit mode within the CN0584 block diagram. This

means that ADC samples are passed through the HDL model and

stored into memory and that DAC samples are passed through the

HDL model and output on the CN0584. HDL models could also be

inserted in receive-only or transmit-only modes, which gives access

exclusively to ADC inputs or DAC outputs, respectively.