Building a more accurate electronic scale with the AD620 Operational Amplifier


Class: BE209
Group: R3
Members: BANSAL, AMAR D; CHUAN, PEIYING; GOEL, SANGITA; HU, PATRICK C; LAU, JASMIN; YUM, JASON W
Date: December, 2002

Full Text

INTRODUCTION:

In this experiment, an electronic scale was constructed using an AD620 operational amplifier and a strain gage.  The primary goal was to test the accuracy of the electronic scale by comparing weight readings from the electronic scale to those measured by a regular weighing scale.  A strain gage is a resistor that changes resistance linearly with elongation, and the AD620 is an op-amp with a modifiable gain, which can be changed by varying the resistance attached between two pins of the op-amp.  The AD620 allows us to amplify millivolt signals to signals in the volt range.  This may increase accuracy by allowing us to better pinpoint the voltage readings.  Thus, we tested the differences in accuracy between narrow and broad weight ranges in our calibration, and the accuracy of higher gain versus lower gain.

We hypothesize that broad-range calibrations will produce a more accurate calibration than a narrow-range calibration.  A calibration range is defined as “a range between the highest and lowest standards used for calibrations, from which the properties of unknown samples can be determined.” 1  Basically, this means that a calibration curve should be used to measure properties (in our case, weight) within the calibration range.  Therefore, it is hypothesized that a narrow range of 0 to 60 pounds will not be as accurate as a broad range curve for measuring body weights over 100 pounds.  The inaccuracy of the narrow range may be caused by greater extrapolation of the calibration line.  Also, any variability in deformation of the bar and strain gage may not be taken into account with the narrow range calibration.  We are also hypothesizing that altering the gain on the AD620 should not have any effect on the accuracy of the scale.  Assuming that the output signal from the chip is large enough to be picked up by the DMM, amplifying that signal even further would also amplify any noises in it.  Thus, a higher gain should not have any clear advantage.