Flux gate sensor
The flux gate principle is familiar from the flux-gate magnetometers. These were developed during the second world war and used by low flying aircraft to detect submarines. These days fluxgate sensors are used in gyro compasses and in lab equipment to measure remanent magnetism for example. The flux-gate magnetometer can measure the individual components of the earth’s magnetic field from 0.5 to 1 nT accuracy. The flux gate sensor is comprised of a core of minimal cross sectional area and made from material with high permeability and steep and small magnetic curves. Around this plate a coil is wound. The small cross sectional area is saturated by a small current in the coil. Figure 1 shows the hysteresis curve of the flux-gate.
Figure 1. Flux gate hysterisis loop
In figure 2 we see the location of the flux gate in the magnetic circuit of our future current sensor. Assume I1 is zero and therefore there is no flux in the core of figure 2. We now apply a block wave to the coil of the flux-gate so that it becomes saturated. The current is as shown in figure 3.
Figure 2. Magnetic circuit current sensor
The current will exponentially rise, but in a short time the core is saturated and the self induction Lf can be neglected so that only the resistance Rf of the coil limits the current. The average value of the current in Nf is zero. If there is a current I1 present and therefore a flux in the core of the current sensor, then the flux-gate in one direction will be more quickly saturated resulting in the current in the coil looking like that shown in figure 4. The average value of the current will be positive or negative corresponding to the direction of the current I1 .
Figure 3. Current and voltage in flux-gate coil when I1 = 0
Figure 4. Current and voltage in flux-gate coil with current I1
Closed loop sensor
Figure 5. Magnetic circuit with hall sensor
In fact the construction of a closed loop sensor with flux-gate is comparable with the magnetic circuit in figure 5 in which a Hall sensor is now replaced by a flux-gate sensor. This is shown in figure 6. The current to be measured, I1, produces a flux Φ1 and a current I2 is sent through the secondary coil (flux Φ2) in such a way that Φ2 = −Φ1. Here, N1· I1 = N2· I2.
It follows that: .
We are continually trying to bring the flux-gate out of saturation to the point of symmetry of the hysteresis loop so that the resulting flux ( Φ1 + Φ2 ) is zero. As was the case with the zero flux transducer (figure 5), the voltage drop of I2 over a resistor produces the output voltage of the current sensor. Since this is a floating output it is sometimes fed into a differential amplifier with opamps.
The signal generator which supplies the winding Nf is comprised of a comparator circuit with hysteresis (Schmitt-trigger). The current change in If produces noise in the primary of the current sensor as a result of the transformer operation. A filter is required to remove the noise.
Figure 6. Adjusting closed loop
The firm LEM supplies the market with the following flux-gate sensors CAS/CASR, CFSR and CTSR.The photo below shows a didactic model of a flux-gate current sensor.
Photo LEM: Didactic example of a flux-gate current sensor
Current measurement in a photovoltaic installation
As discussed in chapter 5 and 15 we want a PV-installation to operate at the MPP (maximum power point). This means for every PV topology we need to measure the DC-output (current and voltage) of the solar panel. In addition a current measurement in the input of the control loop is necessary for protection against short circuit and over current. In installations without transformer the maximum DC-current that the PV installation can send into the power grid is a maximum of 10mA to 1A.
The value depends on the standard used in the different countries. (IEC 61727,…VDE 0126-1). From all these requirements current sensors used need to have: an accuracy better than 1%, low offset, low drift in amplification, operate well with DC and low frequencies.
A sensor with closed loop flux-gate technology meets the specified requirements. The firm LEM has developed their CTSR-series for this type of application.
In addition to the mentioned requirements (accuracy, low drift,…) the CTSR-series has a number of additional functions such as: reference pin (for self testing), demagnetizing function (also via the reference pin). The CTSR can be used for single and three-phase voltages. The CTSR also has a version with four individual primary conductors. Three conductors are used for three-phase systems and the fourth conductor is used for testing the operation or as neutral conductor of the three-phase net.
In addition the magnetic core of the CTSR has two magnetic screens to protect the flux-gate from external magnetic fields. The reference pin provides access to the 2.5 V reference voltage. This Vref. can be used as a reference for an AD-converter.
Figure 7. (photo LEM): Example of a leakage current
Figure 7 shows an example of how a leakage current can flow in a PV-installation without transformer. The capacitance between solar panel and the roof can be the cause of the leakage current which can result in the solar panel rising to the net potential. It is therefore necessary that any leakage currents are measured contact free and in a galvanically isolated manner.
Figure 8. (Photo LEM): Example of a fault current caused by an earth fault
In the case of an insulation fault an earth current can flow. It is also necessary to control (monitor) this. The sensors used for this should be able to measure AC and DC currents since the fault current can be either AC or DC, depending on the fault location. Figure 8 shows an example of how to measure ( Δi ) in order to detect the fault current. It is clear that the current sensors required in a transformer free PV-installation do need to meet the specifications already mentioned (low offset, accuracy, DC-signals,…). The current measurement on the output measures the current difference as the result of a leakage current.