Input RMS voltage (V)

20

30

40

50

60

70

80

90

100

100P

1m

10m

100m

D001

50 dB Gain

33 dB Gain

Frequency (Hz)

-50

-40

-30

-20

-10

0

10

20

30

40

50

60

100

1k

10k

100k

1M

10M

D005

Gain setting = 33 dB

Gain setting = 50 dB

11

OPA607

www.ti.com

SBOS981 – OCTOBER 2019

Product Folder Links: OPA607

Submit Documentation Feedback

Copyright © 2019, Texas Instruments Incorporated

The signal chain discussed here, is with a signal frequency at 200 kHz. The signal chain can be tweaked to

support lower frequencies with an appropriate tuning of the band-pass filter. Because the reflected signal

amplitude varies based on the distance of the reflecting object, one of the prime challenges in a design is to have

a wide dynamic range.

To achieve a SNR of greater than 40 dB for signals from 300 uVrms to 30 mV, a variable-gain front-end stage

based on the OPA607 was chosen. The front-end stage has two gain settings: 6 V/V and 31 V/V. The SW

(switch, relay, or analog mux) can be dynamically toggled to ensure maximum sensitively to the receiving signal.

The OPA607 proves to be an attractive solution for this front-end signal chain because of the high input

impedance of the OPA607. Besides the high input impedance, the OPA607 also has very low quiescent current,

making the device very suitable for a dense, high-channel-count system. The ultrasonic receive sensors (piezo

crystal) have source impedance in the range of a few tens of kilohms. The OPA607 has an input bias current of

20 pA (maximum). This small bias current results in reduced distortion when compared with a bipolar amplifier

with input bias currents in the range of a few hundreds of nano-amperes. The OPA607 large-gain front-end is

followed by a narrowband band-pass filter that is tuned to a 200-kHz center frequency. The narrowband filter is

stage of the OPA836 is a low-impedance output of the OPA607, hence the higher input bias current of the

OPA836 is not a cause of concern. The OPA836-based band-pass filter was designed using the techniques

mentioned in the Filter Design in Thirty Seconds application report. A 1-µF capacitor is placed in the feedback

network of the OPA607 (as in Figure 4) to reject the DC bias in the received signal since the AC component is of

interest in the received signal. The 1-µF capacitor gives the OPA607 a high-pass response with a low cutoff

frequency that ensures any DC signal picked up by the Rx node is filtered and only the AC signal gets gained.

Figure 6 shows the frequency response of Figure 4. As shown in Figure 6, the frequency response is a high-Q

factor band-pass filter centered around 200 kHz. Designing such a high-Q band-pass filter helps eliminate white

band noise along with other interferences present in the circuitry, resulting in a high SNR signal chain. The

OPA607 front-end and the OPA836-based band-pass filter together help achieve a total gain of 33 dB (44 V/V)

or 50 dB (316 V/V) based on the SW (switch) position. Assuming the maximum full-scale input (FSR) of a dual

supply (±2.5 V) powered ADC to be approximately 2.2 V, the 50-dB gain mode can be used to achieve greater

than a 40-dB SNR for signals from 300 µVrms to 5 mVrms. Because the received signal is a sine wave at 5

of the following ADC, as per the FSR assumptions. Beyond 5 mV, the 33-dB gain mode is used and helps

extend the maximum allowable input voltage from 5 mV to 50 mV. Figure 5 shows the achievable SNR as a

function of the input voltage. Figure 5 shows that operating on the 50-dB gain mode is always favorable until the

input Vrms voltage nears 5 mV to achieve higher SNR.

The SNR value is for the op-amp-based signal chain only. The SNR value of the ADC or fully differential amplifier

(FDA) further effects the reported value in Figure 5 and must be vectorially added to arrive at the total signal-

chain SNR.

8.2.1.3 Application Curves

Figure 5. Signal-Chain SNR vs Input

Figure 6. Gain vs Frequency