Maxim Integrated MAXQ7667 - Manual (Page 304)

17.5.2.2 Echo Receive Register Configuration
The input of echo reception stage can be configured to accept external (echo) signal on the ECHOP and ECHON pins or internally
generated diagnostic signals can be added to the echo signal (see Figure 17-2). Selection of the signal is done by setting the LNA
input mux select bits, LNAISEL[1:0] (RCVC.7:6). Table 17-6 gives a description for the different settings of the LNAISEL bits.
Table 17-6. Echo Reception Input Mux Selection
When selecting a new channel on the input mux, both the BPF and the LPF must be allowed to settle before the LPF output provides
a valid representation of the input signal. To reduce electrical noise, the echo signal does not pass through the mux and is always con-
nected to the LNA.
The output of the LNA can be monitored by setting the LNA output mux select bit, LNAOSEL (RCVC.8) to 1. The differential output of
the LNA gets connected to AIN0 and AIN1 of the SAR ADC (see Section 14). This action does not disconnect the LNA from the sigma-
delta ADC. When LNAOSEL is set to 1, the output of the LNA can be monitored by connecting an oscilloscope to pins AIN0 and AIN1.
Avoid loading AIN0 and AIN1 or injecting signals on them when LNAOSEL = 1, as this will affect the signal going to the sigma-delta
ADC. For lowest noise operation, set LNAOSEL to 0, keeping the differential output of the LNA disconnected from AIN0 and AIN1.
The receive gain bits, RCVGN4:0 (RCVC.[4:0]), set the gain of the echo receive path. The amount of amplification is adjustable over a
23.5dB range with an average gain step of 0.8dB. Changes in gain are achieved through a combination of analog and digital tech-
niques. Gain changes settle within one ADC conversion, and any switching glitches are removed by the LPF. This rapid settling time
allows a virtual time variable gain amplifier to be created by the software. In a typical application the software sets the gain to a low
value when the burst is first sent and then increases the gain (dynamically) with passing time. This allows strong echoes from nearby
objects to be processed without clipping, while small signals from distant objects are processed with the maximum gain. All the gain
takes place before the BPF. Therefore, ambient acoustic noise and electrical interference are amplified along with the echo signal. To
prevent clipping, the entire input signal must be considered when selecting the gain. The available gains and their corresponding
RCVGN settings are shown in Table 17-7.
Table 17-7. Effective Gain for the Echo Reception Stage
_________________________________________________________________________________________________________ 17-22
MAXQ7667 User’s Guide
LNAISEL[1:0] DESCRIPTION
00 Mux open. Only echo signal is connected to the LNA.
01
0V channel selected on the mux. This places a short of approximately 50 between ECHOP and ECHON, which reduces the input
signal to near 0V.
10
2mV P-P channel is selected on the mux. This connects the 2mV P-P square wave to ECHOP and ECHON. The 2mV square wave
matches the frequency and duty cycle of the burst signal. This feature allows an echo to be simulated that is processed by the entire
signal chain. The 2mV square wave is derived from the burst signal but is not affected by BGT or BTRI.
11 Reserved (do not use).
ECHO RECEIVE PATH GAIN FROM THE ECHO INPUTS TO THE LOWPASS FILTER OUTPUT
(GAIN IS CONTROLLED BY THE VALUE IN RCVGN BIT)
RCVGN
FULL SCALE
(mV P-P )
LPFD
RESOLUTION
(µV P-P /LSB)
RCVGN
FULL SCALE
(mV P-P )
LPFD RESOLUTION
(µV P-P /LSB)
0b00000 101.58 1.55 0b10000 25.56 0.39
0b00001 90.44 1.38 0b10001 22.94 0.35
0b00010 81.26 1.24 0b10010 20.32 0.31
0b00011 74.05 1.13 0b10011 18.35 0.28
0b00100 68.16 1.04 0b10100 17.04 0.26
0b00101 62.91 0.96 0b10101 15.73 0.24
0b00110 58.33 0.89 0b10110 14.42 0.22
0b00111 54.39 0.83 0b10111 13.76 0.21
0b01000 51.12 0.78 0b11000 13.11 0.2
0b01001 45.22 0.69 0b11001 11.14 0.17
0b01010 40.63 0.62 0b11010 10.49 0.16