In my recent series of blogs, we have been looking at driving the power supply inputs of an ADC. This led to a discussion of the ADIsimPower design tool. We looked at how to use it with the ADP2114 and the ADP2442 to derive a circuit solution to provide power supply inputs for an ADC. The tool is quite handy for this purpose and allows us to generate power supply circuits for these two DC/DC converters based on different input parameters and criteria.
This discussion reminded me of the various design tools that Analog Devices (ADI) offers for many different types of components. We will begin this next series of blogs by taking a look at ADIsimADC. This is a quite relevant design tool for me, since I am an applications engineer in the high-speed ADC group at ADI. It is in the process of moving to MOTIF (Modeling Operation Through the Implementation of Features). The purpose of the simulator is the same, but the implementation is getting an update.
Before we begin looking at MOTIF, let's look at VisualAnalog, the software that ADI provides users with the ability to evaluate high-speed ADCs. MOTIF resides within the VisualAnalog software. Let's take a look at an example with one of ADI's latest high-speed ADCs, the AD9680. In Figure 1, the main selection screen is shown for the AD9680. Various measurements can be made with the AD9680 evaluation board and the ADS7-V1EBZ data capture board. The measurements are taken in different canvases, including FFT, Average FFT, Two-Tone, Average Two-Tone, Samples, and Logic. The FFT, Average FFT, Two-Tone, and Average Two-Tone canvases are all for reporting the FFT of the AD9680. The Samples canvas can be used to see a time-domain representation of the ADC data. The logic canvas can be used to see the logic bits as the transition.
In addition to the measurement canvases, notice the MOTIF canvases. There are the same canvases available for MOTIF as there are for actual measurements of real hardware. The intent of MOTIF is to provide the same look and feel of actual measurements, as well as to provide simulated data on the ADC that is very close to the actual hardware performance. The beauty of this is that the user can accurately predict the ADC performance under specific input frequencies and sample rates without the need for the actual hardware. All this is done while operating in a software environment that mimics the measurement environment on the actual hardware.
To get an idea of a typical measurement, Figure 2 shows the main screen in VisualAnalog for capturing an FFT with the AD9680 evaluation board and the ADS7-V1EBZ data capture board. This screen shows the blocks necessary to generate the FFT. As shown, the FFT is windowed using Blackman-Harris windowing. Various settings need to be set up. We will not dive too deeply into those items, since they are covered in more detail on the Wiki page for the AD9680 evaluation board. In addition, this user manual for VisualAnalog provides a more in-depth view of the software's capabilities.
Now let's take a look at the MOTIF FFT canvas. To open this canvas from VisualAnalog, select File -> New, and select the MOTIF FFT canvas, as shown in Figure 3. This will open the screen shown in Figure 4. Notice that the MOTIF FFT canvas looks remarkably similar to the FFT canvas used for measurements on the AD9680 evaluation board.
Let's now take a quick look at the measurement results of the AD9680 evaluation board compared to the MOTIF simulation. I have included measurement results for the AD9680 evaluation board with fS = 1.0 GSPS and fIN = 170.3 MHz. I set up the same parameters for the MOTIF simulation. Figures 5 and 6 show the results of the measurement and the simulation, respectively. As you can see, the results are very close. The measured performance of the AD9680-1000 gives SNRFS = 66.04 dBFS and SFDR = 84.3 dBc, while the MOFIT simulation gives SNRFS = 66.23 dBFS and SFDR = 86.4 dBc. This shows quite good correlation between the measured and simulated data.
We have taken a high-level look at VisualAnalog and MOTIF to see that the measured and simulated results are very similar. Next time, we will look more closely at the various settings that can be configured. Certain common settings that can be adjusted are independent of the ADC, while other settings are included in the model files specific to the particular converter that the user is simulating. Stay tuned as we dive a little deeper in our look at MOTIF.