Post by drhenley on Aug 3, 2018 15:51:34 GMT -5
Here are the reasons people have problems with electronic scales (strain gauge type)
Temperature coefficient of gauge factor (TCGF) is the change of sensitivity of the device to strain with change in temperature. This is generally compensated for by the introduction of a fixed resistance in the input leg, whereby the effective supplied voltage will decrease with a temperature increase, compensating for the increase in sensitivity with the temperature increase. This is known as modulus compensation in transducer circuits. As the temperature rises the load cell element becomes more elastic and therefore under a constant load will deform more and lead to an increase in output; but the load is still the same. The clever bit in all this is that the resistor in the bridge supply must be a temperature sensitive resistor that is matched to both the material to which the gauge is bonded and also to the gauge element material. The value of that resistor is dependent on both of those values and can be calculated. In simple terms if the output increases then the resistor value also increase thereby reducing the net voltage to the transducer. Get the resistor value right and you will see no change.
Zero shift with temperature - If the TCGF of each gauge is not the same, there will be a zero shift with temperature. This is also caused by anomalies in the force collector. This is usually compensated for with one or more resistors strategically placed in the compensation network.
Linearity is an error whereby the sensitivity changes across the pressure range. This is commonly a function of the force collection thickness selection for the intended pressure and the quality of the bonding.
Hysteresis is an error of return to zero after pressure excursion.
Repeatability - This error is sometimes tied-in with hysteresis but is across the pressure range.
EMI induced errors - As strain gauges output voltage is in the mV range, even μV if the Wheatstone bridge voltage drive is kept low to avoid self heating of the element, special care must be taken in output signal amplification to avoid amplifying also the superimposed noise. A solution which is frequently adopted is to use "carrier frequency" amplifiers which convert the voltage variation into a frequency variation (as in VCOs) and have a narrow bandwidth thus reducing out of band EMI.
Overloading – If a strain gauge is loaded beyond its design limit (measured in microstrain) its performance degrades and can not be recovered. Normally good engineering practice suggests not to stress strain gauges beyond ±3000 microstrain.
Humidity – If the wires connecting the strain gauge to the signal conditioner are not protected against humidity, such as bare wire, corrosion can occur, leading to parasitic resistance. This can allow currents to flow between the wires and the substrate to which the strain gauge is glued, or between the two wires directly, introducing an error which competes with the current flowing through the strain gauge. For this reason, high-current, low-resistance strain gauges (120 ohm) are less prone to this type of error. To avoid this error it is sufficient to protect the strain gauges wires with insulating enamel (e.g., epoxy or polyurethane type). Strain gauges with unprotected wires may be used only in a dry laboratory environment but not in an industrial one.
Temperature coefficient of gauge factor (TCGF) is the change of sensitivity of the device to strain with change in temperature. This is generally compensated for by the introduction of a fixed resistance in the input leg, whereby the effective supplied voltage will decrease with a temperature increase, compensating for the increase in sensitivity with the temperature increase. This is known as modulus compensation in transducer circuits. As the temperature rises the load cell element becomes more elastic and therefore under a constant load will deform more and lead to an increase in output; but the load is still the same. The clever bit in all this is that the resistor in the bridge supply must be a temperature sensitive resistor that is matched to both the material to which the gauge is bonded and also to the gauge element material. The value of that resistor is dependent on both of those values and can be calculated. In simple terms if the output increases then the resistor value also increase thereby reducing the net voltage to the transducer. Get the resistor value right and you will see no change.
Zero shift with temperature - If the TCGF of each gauge is not the same, there will be a zero shift with temperature. This is also caused by anomalies in the force collector. This is usually compensated for with one or more resistors strategically placed in the compensation network.
Linearity is an error whereby the sensitivity changes across the pressure range. This is commonly a function of the force collection thickness selection for the intended pressure and the quality of the bonding.
Hysteresis is an error of return to zero after pressure excursion.
Repeatability - This error is sometimes tied-in with hysteresis but is across the pressure range.
EMI induced errors - As strain gauges output voltage is in the mV range, even μV if the Wheatstone bridge voltage drive is kept low to avoid self heating of the element, special care must be taken in output signal amplification to avoid amplifying also the superimposed noise. A solution which is frequently adopted is to use "carrier frequency" amplifiers which convert the voltage variation into a frequency variation (as in VCOs) and have a narrow bandwidth thus reducing out of band EMI.
Overloading – If a strain gauge is loaded beyond its design limit (measured in microstrain) its performance degrades and can not be recovered. Normally good engineering practice suggests not to stress strain gauges beyond ±3000 microstrain.
Humidity – If the wires connecting the strain gauge to the signal conditioner are not protected against humidity, such as bare wire, corrosion can occur, leading to parasitic resistance. This can allow currents to flow between the wires and the substrate to which the strain gauge is glued, or between the two wires directly, introducing an error which competes with the current flowing through the strain gauge. For this reason, high-current, low-resistance strain gauges (120 ohm) are less prone to this type of error. To avoid this error it is sufficient to protect the strain gauges wires with insulating enamel (e.g., epoxy or polyurethane type). Strain gauges with unprotected wires may be used only in a dry laboratory environment but not in an industrial one.