EDN Access -- 12.22.94 Design approach simplifies signal conditionin
Design Ideas:December 22, 1994
Design approach simplifies signal conditioning
Robert S Villanucci,
Wentworth Institute of Technology, Boston, MA
The low cost and wide availability of 8-bit microcontrollers, such as Motorola's MC68HC11, allow you to easily incorporate intelligence in pressure-measurement systems. Your main challenge is to signal-condition the sensor's small, differential bridge signal into a single-ended output voltage that the µC's A/D converter, running off 5V, can accept. You can easily identify circuit functions and design hardware by graphing the required system response and using basic math. You can apply this general design technique to all linear sensors.Consider a design that must convert pressure ranging from 0 to 5 psi to a 0.5 to 4.5V signal. Fig 1a illustrates the design requirements of the signal-conditioning circuit. The equation for the analog interface of this linear system design is
VADC = (800 mV/psi)PIN + 0.5V.
In Eq 2, the sensitivity of the pressure sensor equals 5 mV/psi when you drive the sensor from a 5V reference
V1 - V2 = (5 mV/psi)PIN.
The signal-conditioning circuit's design equation describes the electronics needed to interface the sensor to the µC. You can write this equation by solving Eq 2 for PIN and substituting
VADC = 160 (V1 - V2) + 0.5V.
The design should amplify the pressure sensor's differential output signal, V1 - V2, by a gain of 160. You must convert the result to a single-ended voltage, VADC, and offset this voltage by 0.5V. The gain requirements are moderately high. Consequently, you should amplify the signal in two stages: an instrumentation amplifier with a gain of 10 followed by a gain/offset stage. Eq 4 characterizes the instrumentation amplifier voltage (VIA)
VIA = 10 (V1 - V2).
You must solve Eq 4 for V1 - V2 and substitute the resulting expression into Eq 3 to complete the interface design. The result of this equation yields Eq 5, which describes the final gain/offset stage:
VADC = 16 (VIA) + 0.5V.
Fig 1b shows the implementation of this design. The SCX15DNC sensor measures pressure and outputs a differential signal (see Eq 2). The differential signal drives the single-supply instrumentation amplifier, IC1. The circuit wires IC1 for a gain of 10. IC2's noninverting amplifier and the passive adder, comprising R1, R2, and R3, complete the design by satisfying Eq 5. A passive adder can only attenuate signals. Therefore, you must first divide Eq 5 by 20--or any gain greater than 16--to yield a linear equation of the form Y=mX+b as follows:
VA = 0.8 (VIA) + 0.025V.
The passive adder combines the m and b terms. IC2's gain-of-20, rail-to-rail op amp reestablishes system gain requirements.
Apply 0 psi and adjust the zero-trim potentiometer until VADC equals 0V to calibrate the circuit. Next, apply 5 psi and adjust the gain-trim potentiometer for a 4.5V output. The zero and gain trims interact. Therefore, you should repeat these adjustments until both points stay fixed. Finally, adding C1 to the passive adder before the final gain stage creates a lowpass filter with a pole at approximately 5 Hz. This filter prevents power-frequency noise from corrupting the sensor's output. (DI #1643)EDN Magazine. EDN is a registered trademark of Reed Properties Inc, used under license.