Designing a matching circuit is fairly straightforward for a buffered input ADC because it offers almost constant impedance across all frequencies. As we have already seen, this is not the case with an unbuffered ADC. The aim of designing a matching circuit is to match the track mode impedance of the ADC so that the source sees only real (i.e., resistive) impedance at the signal frequency. It is worth noting that impedance matching is particularly important for high-frequency applications since an unmatched input would cause prominent reflections of the input signal and thus heavily distort it.
There can be many configurations of a matching circuit, and the exact implementation depends upon the end application. However, the general format remains the same as shown in Figure 1 . Here, the impedance matching block takes Zin into consideration and aims at making the effective impedance as seen by the source (Rin1 ) equal to Rs , which is the output impedance of the source.
For a buffered input ADC, the input impedance is very high and resistive. A simple resistive shunt can do the job of impedance matching in this case. Consider the example circuit shown in Figure 2 . Here, the input impedance of the buffer is Rin_buff . To match this impedance to the impedance of the source (Rsource ), a resistive shunt Rsh is added at the input terminal of ADC. The objective of this shunt is to match the effective input impedance (Rin ) of the ADC to the source impedance.
Equation 1 gives the solution for the design of a simple resistive shunt type matching circuit for buffered ADCs.
The design of an impedance matching circuit for an unbuffered ADC, however, is not as straightforward. It involves resistive as well as reactive elements of impedance. The key point to remember in this case is that the matching circuit is usually designed for a specific signal frequency. This is because the input impedance of the ADC itself is frequency dependent. Thus we have to first define a signal frequency, then design the impedance matching network considering the ADC’s input impedance at that signal frequency. There are various ways to achieve impedance matching once the signal frequency is known. Let us explore a few examples.
An impedance matching circuit for unbuffered ADCs can consist of a simple inductor. Since the ADC’s input impedance is capacitive, this inductor forms a resonant circuit with it (Figure 3 ). The inductor is designed such that the resonant frequency of the resultant circuit is equal to the signal frequency. Thus this circuit offers pure resistive impedance to the source at the signal frequency. Equation 2 can be used to find the value of inductance L for the matching circuit. In this equation, Cin is the input capacitance of the ADC, and fs is the signal frequency at which the matching network is to be designed:
This equation follows from the fact that L and Cin form a series resonant circuit. We can also design a matching circuit with parallel resonance. Once we have gotten rid of the reactive component of impedance, the resistive component can be matched with a resistive shunt similar to the one we have seen before.
For RF or IF applications, transformer-based matching circuits are more popular. The advantage of a transformer-based impedance matching circuit is that the transformer does not add any noise by itself. Besides this, a transformer-based circuit provides inherent single-ended to differential conversion, which, in turn, extends most of the advantages of differential input configuration to a single-ended signal. The actual matching circuit between the secondary winding of the transformer and the inputs of the ADC can be built in many ways per the specific needs of the application.
A resonant circuit similar to the one shown in Figure 3 can be used to nullify reactive the component of the ADC’s input impedance (Figure 4-A ). An additional consideration for a transformer-based circuits is that the impedance reflection seen by the secondary winding of the transformer towards its primary winding depends on the turn ratio of the transformer. Another commonly used transformer-based impedance matching circuit at baseband frequencies is shown in Figure 4-B . In this diagram, resistances R1 and R2 act to give sufficient isolation to the input signal from the ADC’s input capacitance.
The impedance matching block shown in Figure 1 is not always restricted strictly to the impedance matching circuit. This block may occasionally encompass an anti-aliasing low pass or band pass filter, depending on the system requirements. The effect of these filters on the impedance cannot be ignored while designing the matching circuit. Furthermore, when such a filter is present, it is recommended to place it at the transformer’s secondary winding. This helps to limit the noise that gets through to the ADC inputs.
Now that we have had a look at the commonly used impedance matching networks, we can complete the discussion of ADCs in general and their properties. We have seen how these properties affect the measurement systems either positively or negatively. We have also discussed how to address the negative effects of some of the properties of ADCs. In our next installment, we’ll discuss various types of ADCs, their architectures, and their peculiarities.
About the authors:
Sachin Gupta is working as a Senior Applications Engineer in the PSoC 1 Applications group with Cypress Semiconductor. He holds a Bachelor’s degree in electronics and communications from Guru Gobind Singh Indraprastha University, Delhi. He has several years of experience in mixed signal application development. He can be reached at firstname.lastname@example.org.
Akshay Phatak is an Applications Engineer with Cypress Semiconductor. He holds a Bachelor's degree in electronics and telecommunications from the College of Engineering, Pune, India. He likes to work on mixed-signal embedded systems. He can be reached at email@example.com.