Abstract : Based on the introduction of the working principle of the signal generator for insulation positioning, this paper elaborates on the hardware and software design of the signal generator. Products based on this design have passed testing and can be applied to IT systems, providing a safe and reliable power supply solution for application environments.
Keywords : IT system signal generator fault location
0 Introduction
This paper designs a signal generator for insulation fault location (hereinafter referred to as the signal generator), which is installed in the IT system to work with the insulation fault location device to achieve the insulation fault location function. When an insulation fault occurs in the IT system, the signal generator is activated and generates a location signal, which is injected between the IT system and ground. The insulation fault location device uses sensors to inspect each branch. When a location signal is detected flowing through a certain branch, the branch where the insulation fault is located can be identified. At this time, the operator can perform targeted power cut-off or other protection operations on the faulty branch without having to cut off power to each branch individually, thereby improving work efficiency and ensuring the continuity of system power supply.
1. Signal Generator Principle
Signal generator working principle: When a single-point grounding fault occurs in the IT system, a location signal is injected alternately between a certain line in the system and the ground so that the insulation fault locator can detect the location signal on the faulty branch. The principle of the signal generator is shown in Figure 1.
Figure 1 Schematic diagram of the signal generator
In IT systems, the effective value of the injected test signal must be small enough to avoid causing too much interference to the IT system or even harming the system load; but it must also have a large enough peak value to generate a large enough current on the faulty branch so that the current transformer of the fault locator can monitor it normally.
Considering both scenarios above, this paper adopts a pulse signal as the test signal. A sufficiently large pulse signal amplitude and a sufficiently narrow pulse width can achieve the desired goals of a sufficiently small effective value and a sufficiently large peak value. From a design simplification perspective, it is unnecessary to directly generate a high-voltage pulse signal on the signal generator; this can be achieved by extracting the peak value of the AC signal in the IT system.
For a single-phase AC IT system, the voltage between lines L1 and L2 is AC220V, with a peak value of 220V, which meets the requirement of a sufficiently large pulse peak. To meet the requirement of a sufficiently small effective value, this paper sets the voltage threshold to 50V according to the standard IEC61557-9, which states that "the effective value of the positioning signal voltage shall not exceed 50V". Based on this, the pulse width can be calculated (since the pulse width is very small, for ease of calculation, this peak pulse can be considered as a rectangular pulse with an amplitude of 220).
When the AC voltage period is 50Hz, the pulse width is:
When the AC voltage is 60Hz, the pulse width is:
By utilizing the timer function of the microcontroller in conjunction with an optocoupler, a peak pulse of 0.4ms can be accurately captured. Since 0.4ms < 0.4304ms < 0.5165ms , and in the actual captured pulse signal, except for the peak, the amplitude of all other points is less than V, its effective value will definitely be less than the set threshold (50V), thus satisfying the requirement that the effective value of the pulse is sufficiently small.
2 Hardware Design
The hardware functional modules of the signal generator mainly include a power supply module, a central control module, a monitoring module, a signal generation module, a communication module, and an indicator light module. The hardware design principle block diagram is shown in Figure 2.
Figure 2 Hardware Design Block Diagram
After the signal generator is powered on, the CPU monitors the voltage of the IT system in real time through the monitoring module, measuring the AC frequency of the IT system. When a ground insulation fault occurs in the system, the signal generator determines the pulse width and pulse frequency of the test signal based on the measured frequency, extracts the system peak, generates a test signal, and applies it alternately between L1-PE and L2-PE. Due to the insulation fault, the faulty branch can be equivalent to a small resistor, connecting the faulty line of the IT system to the ground, forming a current loop. Therefore, the test signal can generate a test current on the faulty branch. When the insulation fault locator inspects each branch, it can determine that the branch is the faulty branch if it detects this test current on a certain branch. In this design, the central control module uses ST's 32-bit ARM-M3 core microcontroller STM32F103. This chip boasts high processing speed, with a maximum operating speed of 72MHz. It features abundant on-chip and peripheral resources, including 20KB of on-chip RAM and up to 64KB of FLASH memory. It also includes a multi-channel 12-bit A/D conversion module and multiple communication interfaces such as SPI, C, and CAN, greatly simplifying the design of peripheral circuits.
3 Software Design
The signal generator's control program is written in C language and employs a structured programming approach, facilitating code maintenance, portability, and upgrades. After power-on, the system first initializes and performs self-tests on each module to ensure system reliability. Then, once all hardware circuits are confirmed to be functioning correctly, it automatically enters normal operating mode. The main program flowchart is shown in Figure 3.
Figure 3 Flowchart of the System Main Program
To ensure the signal generator operates accurately and reliably and to prevent malfunctions, specific program algorithms are used in the software for processing.
(1) Digital filtering algorithm. The signal generator uses a digital filtering algorithm to filter out harmonics, noise and other interference in the signal, and only allows the useful signal to participate in the result calculation, so as to make the calculation result more accurate and reliable.
(2) Adaptive AC Frequency Method for IT Systems. Due to the diversity of working environments, the working voltage is not necessarily 50Hz. The actual voltage frequency may be higher or lower. Therefore, the AC frequency of the IT system needs to be monitored in real time through a monitoring module. The monitoring module compares the voltage between L1 and L2 lines. The time when UL1 > UL2 and UL1 < UL2 are recorded as t1 and t2, respectively. Because there is a certain threshold voltage during voltage comparison, there will be a phenomenon where t1 > t2 or t2 > t1. If t1 + t2 = 20ms (i.e., the system AC frequency is 50Hz), and a system-to-ground insulation fault occurs, a pulse with a width of 0.4ms can be intercepted between L1 and L2 .
As shown in Figure 4, during each cycle of the system voltage, the signal generator captures two pulses: one at the peak of the positive half-wave of L1-L2 (second row in Figure 4), and the other at the peak of the negative half-wave of L1-L2 (third row in Figure 4). If the fault point is on line L1, the pulse waveform captured at the peak of the negative half-wave of L1-L2 will appear positive on the faulty branch and can be detected by the insulation fault locator. If the fault point is on line L2, the pulse waveform captured at the peak of the positive half-wave of L1-L2 will appear positive on the faulty branch and can be detected by the insulation fault locator.
Figure 4 shows the voltage between L1 and L2 and the extracted pulse voltage.
If t1+t2=10ms, considering the requirement that the effective value of the pulse is less than 50V, then instead of taking two pulses per cycle (L1-L2 positive half-wave, L1-L2 negative half-wave), we can choose to take two pulses every two cycles (L1-L2 positive half-wave, L1-L2 negative half-wave). The same logic applies to other frequencies.
The actual positioning signal generator is shown in Figure 5. It is powered by DC24V and has LED indicators on the panel that show the working status as "Run", "Communication" and "Test".
Figure 5. Actual picture of the signal generator
When the IT power distribution system is running without faults, the signal generator automatically monitors the system frequency. When a single-point grounding fault occurs in the IT power distribution system, the signal generator generates a test pulse signal, which, in conjunction with the insulation monitor and insulation fault locator, locates the faulty branch.
The signal generator has passed type testing, and all indicators meet national standards. It has been successfully applied in the intensive care unit of a hospital, as shown in Figure 6. The insulation monitor, insulation fault locator, and signal generator form a local area network via a communication line. After power-on, the signal generator automatically enters monitoring mode to monitor the frequency of the IT system. When the insulation monitor detects a ground insulation fault in the IT system, it activates the signal generator and insulation fault locator via the communication line, entering signal generation mode and fault location mode respectively.
Figure 6. Application diagram of the IT system in the intensive care unit of a hospital.
In practical engineering applications, the pulse waveform generated by the signal generator is shown in Figure 7.
Figure 7 shows the waveform generated by the signal generator.
As shown in Figure 7, the waveform contains a large amount of noise interference, and the peak value is smaller than the theoretical value (the sine waveform in the figure represents the system voltage for comparison), but it still meets the requirements for insulation fault location. The waveform monitored at the insulation fault locator, after filtering and other preprocessing operations, is shown in Figure 8.
Figure 8. Waveforms monitored by the insulation fault locator
As shown in Figure 8, the monitored pulse waveform is higher than the interference waveform, forming a significant drop. By setting an appropriate threshold and considering factors such as pulse width, it is possible to accurately determine whether a test signal passes through this branch, i.e., whether this branch has an insulation fault.
Upon detecting a faulty branch, the insulation fault locator displays the number of faulty branches and simultaneously transmits the faulty branch information back to the insulation monitor via the communication line. Upon receiving the information, the insulation monitor immediately alarms, displays the number of faulty branches on its interface, and simultaneously commands the signal generator and insulation fault locator to stop emitting signals and locating faults. The signal generator then re-enters monitoring mode.
The system was debugged on-site, simulating 100 insulation faults, and the insulation fault location rate was 100%, which fully proves the feasibility of the signal generator.
4. Conclusion
The signal generator designed in this paper features adaptive frequency control for IT systems, injection of high peak and low RMS pulse waveforms, and multi-system networking capabilities. Its current operating status can be displayed via panel indicator lights. This signal generator complies with relevant national standards and, when used in conjunction with insulation monitors and insulation fault locators, can provide a safe and reliable power supply solution for IT systems.
Source : Electrical Engineering Technology, Issue 4, 2014
References
[1] GB-50054-2011 Code for Design of Low-Voltage Power Distribution Systems [S]
[2] JGJ16-2008 Code for Electrical Design of Civil Buildings [S] .
[ 3 ] IEC61557-9Electricalsafetyinlowvoltagedistributionsystemsupto1000Va.c.and1500Vd.c. — Equipmentfortesting , measuringormonitoringofprotectivemeasures—
Part9 : EquipmentforinsulationfaultlocationinITsystems
