I’ve now discussed the basics of the frequency folding tool from Analog Devices in two previous blog posts this year, Analog Devices Design Tools: ADISimADC Frequency Folding Tool and Analog Devices Design Tools: Frequency Folding Tool Revisited. I discussed where to find the tool on the website and the basics of the tool. We looked at single tone and wideband signal inputs. We also explored some basic examples of aliasing in an ADC and, in my most recent blog on the subject, we looked at how certain input frequencies and sample rates can result in a situation where tones alias to the same locations. Let’s now take a look at a real example with the AD9680-500. We’ll see how this simple yet powerful tool can be used to aid in understanding the aliasing effects of an ADC as well as help with understanding effects of some digital processing blocks in the AD9680.
In this example we’ll look at the AD9680-500 operating with an input clock of 368.64 MHz and an analog input frequency of 270 MHz. First it is important to understand the setup for the digital processing blocks in the AD9680. The AD9680 will be set to use the digital downconverter (DDC) where the input is real, the output is complex, the numerically controlled oscillator (NCO) tuning frequency is set to 98 MHz, half band filter 1 (HB1) is enabled, and the 6dB gain is enabled. Since the output is complex the complex to real conversion block is disabled. The basic diagram for the DDC is shown below. In order to understand how the input tones are processed, it is important to understand that the signal first passes through the NCO which shifts the input tones in frequency, then passes through the decimation, optionally through the gain block, and then optionally through the complex to real conversion.
It is important to understand the macro view of the signal flow through the AD9680 as well. The signal enters through the analog inputs, passes through the ADC core, into the DDC, through the JESD204B serializer, and then out through the JESD204B serial output lanes. This is illustrated by the block diagram of the AD9680.
Let’s now look at how the signal appears as it passes through the various processing blocks in the AD9680. We will use the frequency folding tool as well as VisualAnalog (which we’ve also discussed in a couple of blogs now: Analog Devices Design Tools: VisualAnalog and Analog Devices Design Tools: VisualAnalog Part 2). With an input sample clock of 368.64 MHz and an analog input frequency of 270 MHz, the input signal will alias into the first Nyquist zone at 98.64 MHz. The second harmonic of the input frequency will alias into the first Nyquist zone at 171.36 MHz while the third harmonic aliases to 72.72 MHz. This is illustrated by the plot of the Frequency Folding Tool below.
The Frequency Folding Tool plot shown above gives the state of the signal at the output of the ADC core BEFORE it passes through the DDC in the AD9680. The first processing block that the signal passes through in the AD9680 is the NCO which will shift the spectrum to the left in the frequency domain by 98 MHz (recall our tuning frequency is 98 MHz). This will shift the analog input from 98.64 MHz down to 0.64 MHz, the second harmonic will shift down to 73.36 MHz, and the third harmonic will shift down to -25.28 MHz (recall we are looking at a complex output). This is shown in the FFT plot in VisualAnalog below.
From the FFT plot we can see clearly how the NCO has shifted the frequencies that we observed in the Frequency Folding Tool. What is interesting is that we see an ‘unexplained’ tone in the FFT; or is it unexplained? The NCO is not subjective and shifts all frequencies. It has shifted the fundamental input tone alias at 98 MHz down to 0.64 MHz and shifted the second harmonic to 73.36 MHz and the third harmonic to -25.28 MHz. It appears also that another tone has been shifted as well and appears at 86.32 MHz.
The frequency folding tool does not include the DC offset of the ADC. This DC offset results in a tone present at DC (or 0 Hz). The frequency folding tool is assuming an ideal ADC which would have no DC offset. The DC offset tone at 0 Hz is shifted down in frequency to -98 MHz. Due to the complex mixing and decimation this DC offset tone folds back around into the first Nyquist zone in the real frequency domain. Recall that tones alias back into the first Nyquist zone as shown in the frequency folding tool. When looking at a complex input signal where a tone shifts into the second Nyquist zone in the negative frequency domain it will wrap back around into the first Nyquist zone in the real frequency domain.
Since we have decimation enabled with a decimation ratio of two, our decimated Nyquist zone 92.16 MHz wide (recall: fs = 368.64 MHz and the decimated sample rate is 184.32 MHz which has a Nyquist zone of 92.16 MHz). The DC offset tone is shifted to -98 MHz which is 5.84 MHz delta from the decimated Nyquist zone boundary at 92.16 MHz. When this tone folds back around into the first Nyquist zone it ends up at the same offset from the Nyquist zone boundary in the real frequency domain which is 92.16 MHz – 5.84 MHz = 86.32 MHz. This is exactly where we see the tone in the FFT plot above!
It is great to have a tool like the Frequency Folding Tool that can be used to identify where tones are. It also helps to understand a bit about the ADC behavior when performing complex or real mixing and/or decimation such that when we see ‘unexpected’ tones we can identify their origin. Using the frequency folding tool helps to see how tones alias with an ADC so we can identify and predict where tones will alias. I hope this example proves useful for those out there working with ADCs. I also hope this helps you to understand how tones alias not only in the real frequency domain but also in the complex (negative) frequency domain.