Doing a test with relatively long cables for strain measurements (about 10 meters+ or so) using Simcenter SCADAS and VB8 strain cards.
Any important things to keep in mind when using long cables to measure strain?
There are probably two big concerns:
Noise or Interference on Measurements
Strain gauges produce very low level voltage signals which are transmitted through the cables. Because the levels are low, the signals are susceptible to even small amounts of electromagnetic and electrostatic interference.
Electromagnetic interference comes from from nearby electric power sources, including electric motors and electric transformers as shown in Figure 1.
Current flowing through an electrical device generates a magnetic field. Any wire carrying a measurement signal passing through this magnetic field picks up electromagnetic interference.
Electrical charges can accumulate on a signal wire, when electrical charges are discharged from a source (Figure 2). Ever experienced a spark when shaking hands with someone during the winter? This is an example of electrostatic discharge. In fact, the spark can happen without actual contact between the two hands. The hands just have to be close enough that the spark jumps from one to the other.
In measurements, electrical sources like fluorescent lights can create electrostatic interference on an unshielded signal wire.
It is important to have good shielding (Figure 3) around the signal wire to prevent electrostatic interference. Charges accumulate on the shielding, rather than the signal wire.
The shield takes these accumulated charges away from the signal wires via a low resistance path to ground. Shielding is a thin metal foil wrapped around the signal and supply wires. A braided shield gives the cable flexibility so it can bend easily when routing the cables.
What can be done?
Some things that can be done to abate electromagnetic and electrostatic interference:
Calibration Factor Sensitivity
A calibrated strain gauge is provided with a sensitivity or calibration factor (for example 1000 uE/mV). The calibration factor, used to convert the voltage signal produced by the gauge to strain, depends on the voltage supply level used when calibrating.
For example, suppose the gauge was calibrated with the supply at five volts, and the resulting calibration factor is 1000 uE/mV. If the supply was changed to one volt, the calibration value would be less. For a one volt supply, it would be 200 uE/mV. The calibration factor changes with the voltage supply.
Typically, strain gauges are calibrated in a lab, and do not require long cables when calibrating. With a short cable, the voltage supplied and the voltage that reaches the gauge, are the same.
However, if a long cable is used while measuring, the voltage supplied and the voltage that reaches the gauge may be different (Figure 4). The longer the cable, the higher the electrical resistance. This resistance reduces the voltage across the cable.
For example, the voltage could be reduced across the supply wire by 10% (for example, supply of 5 volts reduced to 4.5 volts at the gauge). This would result in a 10% amplitude error on the measurement.
There are different methods to account for the drop in the supply voltage when using a long cable:
These three options can be handled with the Simcenter SCADAS VB8 card as described in the next section.
Simcenter Testlab and Simcenter SCADAS
This section describes how the previously mentioned methods ensure an accurate strain measurement using Simcenter Testlab and the Simcenter SCADAS VB8 card.
Calibrate with Long Cable
If using Simcenter Testlab to calibrate the strain gauges with long cables, the calibration resistor should be placed at the gauge. Details about using Simcenter Testlab to calibrate a strain gauge are given in the knowledge base article ‘Using Simcenter Testlab to measure strain gauges’.
In the knowledge base article, internal shunt resistors of the SCADAS hardware are used. This would not give the correct result when using long cables. Instead, a shunt resistor placed at the gauge itself should be used. The “ExternalSense” field should be set then to “External Shunt, No Sense Lines” as shown in Figure 5.
Lead Wire Resistance Estimation
The number of ohms of resistance of the cable can be entered in a field (for one wire only) in the Channel Setup workbook. Using the ohms per meter or ohms per foot and multiplying by the corresponding signal wire length, and the total ohms of resistance can be calculated.
This value can be entered into the software:
Sense lines are dedicated wires that measure the voltage being supplied at the gauge. These sense wires are in addition to the signal wires and voltage supply wires (Figure 7). These wires have no current so they can measured the voltage at the gauge correctly.
Sense wires get connected to Pins 2 and 5 on the SCADAS VB8 card as shown in Figure 8.
The voltage supply will be automatically adjusted by the SCADAS data acquisition hardware to maintain a certain supply level at the gauge.
To utilize sense lines:
Sense lines increase the amount of wiring that is required. For example, a full gauge increases from 4 wires to 6 wires!
Questions? Email firstname.lastname@example.org or post a reply to this article.
SCADAS and other Acquisition Hardware links: