Figure 2-2 shows the input bandwidth and input drive level trade-offs across a frequency up to 10GHz comparing the five different front ends. The front-end bandwidth for each design gives an indication of the –3dB bandwidth and input drive level required to reach –6dBFS at 1.4GHz. For example, looking at the TRF1208 design, this only takes a –16dBm input signal to reach –6dBFS of the full-scale value of the ADC. Conversely, this takes approximately 1dBm to achieve the same level using a WB balun. Between the two, this is a difference of 17dBm of signal strength that is required from the previous stage. The balun and wideband interface network create loss and therefore drive up the noise figure number of the entire signal chain. Baluns create loss, which is also true of the LNA and FDA front-end designs that include a balun for the single-ended to differential signal conversion.

Figure 2-2 Frequency Response Comparison of the Five Front-End Designs

As discussed, baluns have loss, so the wideband balun interface requires the most signal drive strength, which requires a +1dBm signal level at the primary of the balun to achieve –6dBFS on the output of the ADC. Since all other comparisons use an active amplifier device (all of which have inherent gains), the input drive level required is much less: anywhere from –5 to –16dBm. Further analysis can be done on each front-end network work to even out the gains and input network losses. However, this information does provide some indication of what to expect before diving deeper into AC performance.

Frequency sweeps were conducted over the same bandwidth to capture and compare the AC performance, i.e. - SNR, SFDR and IMD3 performance. These three tests are typical standards used to make comparison trade-offs when designing with high-speed converters.

Figure 2-3 shows the trade-offs in SNR between the various configurations.

Figure 2-3 SNR Comparison of the Five Front-End Designs

Looking at the purple curve as the baseline performance, this can be seen that the wideband balun interface offers the best SNR performance over the entire bandwidth of the converter. The green curve representing the LNA approach is second, as these types of active devices typically have a very low noise figure, with about 1dB to 2dB of added noise. The FDA comes in third place, as the FDA has more wideband noise than the LNA but less than the TRF1208 design. There is a slight issue with common-mode noise cancellation when using the FDA in a single-ended input configuration, since the inherent design on the input is to expect a fully differential input signal. Using this type of configuration affects the SNR slightly.

The TRF1208 comes in last; however, the device has more output noise because this has a higher gain than the FDA in comparison. Remember that a higher active gain can gain self-generated noise. For example, with a 2GHz analog input signal, the TRF1208 has a gain equal to 16dB and a noise figure equal to 8dB at –166.7dBm/Hz, which yields -150.7dBm/Hz of output noise. The FDA has a gain equal to 10dB (S2D) and a noise figure equal to 11dB at –163.3dBm/Hz, yielding –153.3dBm/Hz of output noise.

All of the designs are configured for the widest bandwidth possible, as shown in Figure 2-2 . In any active design, reducing the bandwidth by using an anti-aliasing filter in-between the outputs of the amplifier and the inputs of the ADC helps reduce the wideband noise outside the bands of interest and the noise the converter sees, therefore pushing the SNR back toward the baseline performance, as shown in Figure 2-1 (WB Balun and 5200RF configuration).

Figure 2-4 shows the SFDR dynamic range from a linearity perspective over a 10GHz frequency sweep between the various front-end configurations. SFDR is a single-tone measurement that provides a good perspective on any limiting harmonics, for example, second harmonic, third harmonic, fourth harmonic, and so on within the frequency of interest.

Figure 2-4 SFDR Comparison of the Five Front-End Designs

Looking at the purple curve as the baseline performance, the wideband balun interface yields the best SFDR possible over the entire bandwidth of the converter. The green curve representing the LNA shows very degraded performance, particularly at the lower band up to 5GHz, as the even order distortion or HD2 always dominates given the single-ended nature of the LNA. The HD2 eventually falls out of the bandwidth of the ADC.

The FDA design has some third-order domination in the 0.5 to 3.5GHz area when using the differential front-end approach. More even-order degraded dominance is evident in the 0.5GHz to 5GHz range when using the single-ended approach.

The TRF1208 design stays on par with the passive baseline front end all the way, giving true testament that this amplifier is an option for wideband front ends that require an active device.

Another common converter test metric is two tone, which gives rise to IMD3 results or third-order inter-modulation distortion and more quickly emulates real-world system application signals. Simply put, two-tone measurements actively measure two signals injected into the front-end interface at the same time. These two signals are typically offset by 10MHz from each other and driven to the same level, or –7dBFS each. Figure 2-5 shows IMD3+ (2 × F1 + F2 and2 × F2 + F1) results. While captured, Figure 2-5 does not include IMD3– (2 × F1 - F2 and 2 × F2 - F1) for clarity in illustrating the performance differences.

Figure 2-5 IMD3 Comparison of the Five Front-End Designs

With the purple curve again illustrating the baseline performance, the wideband balun interface yields the best IMD3 performance possible over the entire bandwidth of the converter. The green curve, representing the LNA, shows degraded performance relative to the wideband balun interface. The blue and black curves representing the FDA interfaces are degraded in performance as well relative to the baseline up to 5GHz. The TRF1208 design stays on par with the passive baseline front end for the entire frequency sweep. This amplifier yields a clear advantage that this amplifier is an option for wideband front-end requirements.

Note in this front-end comparison is that the FDA evaluated has two power supplies, one negative, and consumes 1.8 W of power to keep the noise low. This is a method to drive the noise down, increase the headroom of the amplifier and throw more power at the design. The LNA dissipates the least amount of power; only 0.275 W with a single 5V supply. The TRF1208 runs off a single 5V supply and consumes 0.675W.