A new technique for calibrating piezoresistive stress sensor chips

Microfabrication and calibration of a single-polarity piezoresistive three-dimensional stress sensing chip

“…If the effect of the end connector load is eliminated from (10) assuming that the first incremental load is the first dead weight, then…”

“…The current 4 PB fixture setup utilizing ZIF connectors has lower percentage error in the calculated uni-axial stress compared to using micro probes, which produces high stress errors if the applied dead weight forces are small. For example, the percent error in stress can reach around 50% if the ratio between probe force and dead weight is 1/5 [10,23]. The stress percentage error due to weight and length measurement, loading symmetry and beam rotation is estimated to be a maximum of 1% compared to 5% if probe forces are added [22].…”

“…The applied hydrostatic pressure induces a temperature increase in the hydraulic fluid, which increases the resistance of the piezoresistors by (αT) as given in (10). A resistance temperature detector (RTD) is installed in the pressure vessel to measure the slight temperature change with the applied pressure.…”

“…The calibration of B 1 and B 2 , and the crystallographic PR coefficients in other applications, has mostly been performed using a fourpoint-bending fixture to apply a known uni-axial stress to a silicon beam; however, the resistance measurement has been performed differently among researchers. Although the most common approach is to probe the silicon beams [7], other researchers used a printed circuit board (PCB) bonded to the silicon beam and connected to the piezoresistors through wire bonds [10,11] to eliminate the effect of probe forces on the beam. Lwo et al adopted a similar approach through connecting the wire bonds to enameled wires with silver-filled epoxy adhesive to avoid bonding a PCB to the silicon beam [12].…”