1. What is CPU?

CPU, also known as central processing unit, is generally composed of arithmetic unit, controller, internal memory, input/output device, interface circuit and bus. However, with the advancement and update of technology, its functions and structure are constantly expanding – the original CPU peripheral circuit is also integrated into the device. When its hardware equipment is expanded to a certain scale, so that it can independently complete a more complex control function, this device is called a microprocessor. In the microprocessor family, in order to be suitable for a certain application field, the hardware structure is different from the general microprocessor (such as 80C51) – there is a certain uniqueness. For example, the microprocessor often used in the inverter referred to in this article has a six-way PWM wave output function and can realize specific control functions. It is also called a microcontroller, alias: single-chip microcomputer. Because people in the industry generally refer to the circuit board of the inverter single-chip microcomputer as the CPU motherboard, from the perspective of convention and concise definition, this book also calls the microcontroller (single-chip microcomputer) CPU.

The microcontroller used for inverter control, also known as “high-performance microcontroller”, should be a dedicated control chip developed specifically. It should have at least six PWM wave forming and output circuits and ports to output the six inverter pulses required by the drive circuit; it should have an A/D conversion circuit, and some should also have a D/A conversion circuit to adapt to the processing of analog input and output quantities; it should have high-speed pulse signal input and output ports and serial port sending and receiving pins to process various digital signals and communication instructions; it should contain program memory and data memory to store programs, original data and rewritable data; of course, it should also have port drivers, various buffers and other circuits, which will not be repeated here.

What goes in and out of each CPU pin is nothing more than some “signal flows”. Some only go in but not out, and go but not return, such as the switch signal input by the control terminal; some only go out but not in, and go but not return, such as the relay signal output by the CPU on the control terminal; some go in and out, such as the signal transmitted between the CPU and the memory, and the operation display panel. This type of signal is bidirectional. All signals can be divided into two categories based on their nature: digital and analog. Voltage detection signals and speed control command signals are often analog, while the start and stop signals of the inverter and keyboard input signals are digital (switch quantity). Some analog signals are processed by the CPU external circuit, such as A/D conversion, before being sent to the CPU. Some directly enter the CPU pins and are processed by the internal A/D circuit. The different CPU hardware function circuits used will inevitably lead to different external circuits.

The integration of microcontrollers is high enough. It is impossible to integrate all the circuit elements required for operation without limit. It needs the active cooperation of external circuits to work. There are three working conditions that are necessary for microcontrollers, the so-called three elements of CPU operation: +5V power supply, working clock, and reset signal. The generation of working clock is generated by the oscillation circuit composed of the internal circuit of the microcontroller and the crystal oscillator element connected to the pin; the reset signal is generated by the external reset circuit when the power is on. A low-level (or high-level) pulse is sent to the reset pin of the microcontroller, and the internal circuit controls the program reset and enters the standby state. The program memory and data memory (referred to as memory) inside the microcontroller have limited capacity and use. External memory-electrically erasable memory is often required to complete some data storage tasks (especially user program storage tasks), which should constitute the four elements of the normal operation of the microcontroller. In order to accept the user’s instructions or inform the user of the working status of the inverter, the microcontroller needs a human-machine interface-operation display panel, communicate with the user, and normal communication with the operation display panel becomes the five elements of the microcontroller’s operation. In addition, the quality of the external circuits of each pin of the microcontroller will affect the operation of the microcontroller. Since then, seven or eight or even dozens of elements of the microcontroller’s operation have appeared. In fact, in my opinion, from the perspective of the microcontroller (or microprocessor, CPU) itself, three elements are necessary for the work. Without them, the microcontroller cannot meet the most basic working conditions. If the microcontroller does not work due to other reasons, it is the external circuit, not the microcontroller itself that is not working, right?

The external circuit and external framework of the microcontroller have been built, but a “body” alone cannot work. It also needs a “soul” to command the operation of the body – the software control program. The program capacity of the inverter is large, generally up to thousands of lines. The control function of the microcontroller is concentrated on two points. One is the control of the output PWM wave. In this regard, there is a clear difference between high-quality and low-quality inverters. Some PWM waves are very optimized, while others are a bit bad – the output torque is small, the operating noise is large, and the carrier interference is also large; the other is the state detection and protection of the inverter module. This task is completed in conjunction with the external circuit, and it is also the focus of the inverter circuit.

Microcontroller – single-chip computer technology has become an important technical branch of automation control technology. I hope that friends can master more relevant knowledge themselves. I will not go into details here.

The failure rate of the CPU motherboard is relatively low, accounting for about 20% of the total failure rate. Failures mostly occur in the fault detection circuit and the control terminal circuit. The inspection and maintenance of the fault detection circuit has become an important inspection content of the CPU motherboard. When the fault detection circuit (subsequent circuits for voltage and current detection, temperature detection circuit) itself is damaged, it is a bit like “falsely reporting military intelligence” and deliberately making trouble. The main circuit is obviously good, but it reports an “output short circuit” fault or an output phase failure fault. The fan is obviously good, but it reports an overheating fault, etc., which prevents the inverter from being put into normal operation. The fault of the control terminal is mostly caused by the user mistakenly connecting to a high voltage, which burns the terminal power supply 24V, the terminal input circuit is open-circuited and damaged, and the input side circuit of the photocoupler is damaged.

The damage rate of the CPU chip itself is less than 2%. Due to the technical blockade involved, the internal program is not easy to crack. General maintenance personnel do not have the relevant conditions to repair the chip. They can only purchase original manufacturer parts or replace the CPU motherboard, the so-called “board-level repair”. However, for partial damage to the CPU chip, workarounds can be used to try to repair it.

The inverter CPU motherboard circuit, which is the basic circuit for CPU operation, is composed of power supply, crystal oscillator circuit, reset circuit, external memory circuit and operation panel display circuit. The reset circuit is composed of special three-terminal reset components IMP809M and R188. It provides a low-level pulse to the 48th pin of the CPU at the moment of power-on, just like shouting the slogan “Everyone take your place”, realizing system reset and starting the program to run. The 3rd, 4th, 6th and 8th pins are connected to the U2 (93C66) memory. The user control program has been stored inside when leaving the factory. During the debugging and use process, the user needs to modify certain parameters at any time to meet the control requirements. The modified parameter values are stored by U2. The four pins connecting the CPU and the memory are all connected to +5V by pull-up resistors.

The general model of the inverter, the operation display panel, has been used as an independent device to communicate with the CPU. It accepts user instructions and transmits relevant monitoring data. The operation display panel contains CPU, decoding driver, LED display and other circuits, and can perform two-way data transmission with the CPU. Between the operation display panel and the CPU, the RS442/RS485 transceiver realizes communication transfer. The user operation signal is input from the A and B differential input terminals and sent to the CPU from the R receiver output terminal; the data signal output by the CPU enters from the D transmitter input terminal and enters the operation display panel from the Y and Z driver output terminals.

In order to adapt to the new control requirements, the control terminal of the inverter is also equipped with an RS485 communication port. In the figure, U6 (15176B) is an RS485 transceiver. D is the driver input terminal, connected to the CPU’s TXD1 serial port sending pin; R is the receiver output terminal, connected to the CPU’s RXD1 serial port receiving pin; A and B are the receiver input and driver output terminals; DE and DR are the driver and receiver enable signal terminals. The working status of the driver and receiver is controlled by the level signals of these two pins.

CPU basic circuit inspection and maintenance:

The failure rate of the CPU (single-chip microcomputer) itself is extremely low. Except for abnormal situations such as damage caused by lightning strikes introduced by the inverter, electrical failures are rare. The CPU is damaged because it contains running programs. For the sake of technical confidentiality, the manufacturer has taken some confidentiality measures to the greatest extent possible. It is difficult to decrypt the program and re-copy the chip. General maintenance personnel do not have such technical means. Does this also involve the issue of intellectual property rights? Therefore, after damage, it is necessary to purchase a chip with the program copied from the manufacturer, or replace it from a circuit board of the same model, or simply replace it with a CPU motherboard.

The inspection of the CPU basic circuit mainly includes the inspection of its three working elements and other working conditions, and fault repair.

The failure of the CPU basic circuit (three-element circuit) is typically characterized by: after power-on and when the power supply is normal, the operation panel has no display, or displays a fixed character, the inverter has no initialization process, and all operations on the operation display panel fail, similar to the phenomenon that a computer cannot be turned on and “freezes”.

The nature of the fault: A. At least one of the three elements of CPU work is missing, the CPU cannot complete the initialization operation, and the program is “stuck”; B. The CPU detects the existence of a dangerous fault signal during the self-test process and is in a fault lock state. All operations are rejected. This is a “CPU motherboard pseudo-fault” phenomenon. After checking and eliminating the cause of the fault, the CPU “strike” phenomenon will disappear immediately; C. The CPU chip is damaged due to lightning strike or abnormal power supply.

Note: If the program is “stuck”, be sure to eliminate the “CPU mainboard pseudo-fault” first, and then inspect the three elements of the CPU and other circuits. Focus on inspecting the OC fault alarm circuit, see the relevant content of Chapters 4 and 5 for details.

To determine whether the CPU is working or the three-element circuits are normal, you can first make a rough judgment:

（1）.When the inverter is powered on, listen carefully to see if there is a “click” sound from the charging relay or contactor. If there is, it means that the three-element circuits are normal and the CPU is working normally. The inverter is in a fault lock state;

（2）Observe the operation display panel. Generally, there is a “power-on character” that flashes and finally stabilizes to a certain character. This process indicates that the CPU has also entered the working state.

（3）If you know the power-on self-test process of the inverter and the potential status of each pin, you can detect the voltage changes and level status of the relevant pins to determine whether the CPU is in operation. Use the key signal input of the operation display panel and the voltage changes of the key points of the detection circuit to determine whether the CPU is in working state. If you press the reset button on the panel, the inverter status signal output relay may make a “click” opening and breaking sound, and at the same time, the reset signal input pin of the drive circuit has a corresponding level change. This means that the CPU can accept the reset signal input and can output the fault reset signal to the drive circuit. This means that the CPU is working normally.

（4）If it is determined that the CPU is not operating normally, the basic working circuit of the CPU can be checked.

Fault inspection of the three-element circuit:

A、 Check the +5V power supply circuit. Check the CPU’s VDD, VSS, Vcc, GND and other power pins to confirm that the power supply is normal. The +5V power supply circuit is often connected to a filter capacitor with a large capacity of thousands of microfarads. When its capacity drops seriously, the CPU program will run disorderly and easily enter a program “dead loop”;

B、Check the reset circuit. The reset circuit provides a pulse voltage for the reset pin of the CPU during the power-on period. The duration of the pulse voltage is μs . Therefore, if a low pulse is required for reset, the static voltage of the CPU reset pin should be +5V. If a high level pulse is required for reset, the static voltage of the CPU reset pin should be 0V low level. Detection methods for the reset circuit:

a. According to the CPU reset pin’s requirement for high or low pulse voltage, measure whether its static potential is normal. If the static voltage is abnormal, check the CPU external reset circuit. You can disconnect the CPU pin to determine whether the abnormal reset pin voltage is a reset circuit failure or the CPU reset pin’s internal circuit is damaged.

b. If the static voltage is normal, you can use the artificial forced reset method to determine whether the CPU can work normally. The method is: if the static voltage of the CPU reset pin is +5V, use a metal wire to quickly short the reset pin and the power supply ground to artificially form a low-level signal input; if the static voltage of the reset pin is 0V, use a wire to quickly short the reset pin and the power supply +5V to artificially form a high-level signal input.

c. After forced reset, if the CPU can work normally, as shown by changes in the content of the operation display panel and the ability to modify parameters, it indicates that the external reset circuit is faulty and the damaged components must be replaced. For dedicated three-wire reset components, if there are no original components to replace, the RC component circuit can be connected for emergency repair;

d. If forced reset is invalid, the crystal oscillator circuit should be further checked.

3. Check the crystal oscillator circuit. The crystal oscillator circuit has few external components, usually only two capacitors and a crystal oscillator. Common circuit faults include the following:

a. Because the crystal oscillator contains quartz crystal, it is easy to break and fail after severe vibration;

b. If the crystal oscillator or capacitor leaks electricity, the signal transmission loss will increase, causing the oscillation to stop;

c. The internal oscillation circuit of the CPU is damaged and the CPU needs to be replaced.

Measurement method: a. The oscillation pulse is a rectangular square wave, and its pin voltage is about the middle value of 0V and +5V. The voltage values of the two pins are slightly different, about 0.3V. Among them, the X2 pin is 2V, and the X1 pin is 2.3V. Please use the voltage range of the digital multimeter when measuring. If you use a pointer meter, due to the low internal resistance, it may cause oscillation to stop, making the measurement result inaccurate; b. If the crystal oscillator leaks slightly or the performance deteriorates, when the crystal oscillator pin is lightly ironed with an electric soldering iron, the CPU motherboard resumes normal operation. It may be that the crystal oscillator is inefficient and the crystal oscillator should be replaced; c. If the crystal oscillator is suspected to be bad, it is best to replace it with a good crystal oscillator for testing. When removing the crystal oscillator for inspection, you can shake the crystal oscillator and carefully observe whether there is a slight clattering sound inside it. If there is, it means that the crystal oscillator is damaged by vibration. Measure the resistance value of the two pins, which should be infinite. If there is a resistance value, it means leakage. If there is a capacitance meter to measure the two pins, a good crystal oscillator has a PF-level capacitance, and its capacitance value decreases with the increase of the nominal frequency. e. There is another rare situation of defective crystal oscillator. Due to structural deformation or mechanical aging, the circuit oscillation frequency is lower than the nominal frequency value, and the frequency of CPU clock pulse is reduced. First, the system operation slows down. Second, due to the change of time reference value, the CPU sampling of input current and voltage signals will have errors, and the display values of running current and output frequency will also have corresponding deviations. In severe cases, the CPU may stop by mistake. The occurrence of this fault is manifested as a difficult fault.

Check the fault of CPU external memory. The inverter can be operated and the parameters can be modified, but after power failure, the modified parameter value cannot be stored, indicating that the machine has an external memory fault. Check the power supply of CPU external memory and the status of the connection line with CPU. Because the “pulse stream signal” is transmitted between CPU and external memory, it is difficult to judge whether it works well or not by the voltage of its pins. You can remove a good memory from the same model circuit board and replace it for testing. Note: If a new blank memory chip is used, the machine will not work. The user control parameters are already stored in the memory when it leaves the factory. If conditions permit, the original storage content can be copied to the new chip. Or purchase the memory chip from the manufacturer and replace it.

Inspection and maintenance of the operation display panel. 1. The buttons and speed regulating potentiometers on the operation display panel are wearing parts. Due to dust and humidity at the work site, they may cause poor contact, unstable output frequency or failure to write parameters. They can be replaced and repaired. 2. The LED display strokes are incomplete. Vibration causes internal drive circuit pins to be poorly soldered, copper foil strips to break, etc., which can be repaired by welding. 3. The power supply is normal, but there is no display, or a fixed character is displayed. You can replace the operation panel with the same model for testing. If the operation display panel is faulty, you can purchase a complete replacement from the manufacturer. 4. If replacing the operation display panel is invalid, check the data communication module between the CPU and the operation display panel – RS442/RS485 transceiver and other circuits.

[ Failure Example 1 ] :

    A 7.5kW INVT inverter could not hear the sound of the charging relay closing when powered on, and all control operations failed. The voltage of the CPU reset control pin 48 was measured to be 2.3V, which should be 5V under normal circumstances. It was determined that the three-wire reset element IMP809M was defective. After replacement, the fault was eliminated.

[ Failure Example 2 ] :

A Fuji 5000G9S 11kW inverter had a fixed character displayed on the operation panel and could not be operated. The “program stuck” phenomenon occurred and was determined to be a CPU mainboard failure. The static voltage of the CPU reset control pin was measured after powering on and was normal. The manual forced reset method was ineffective. The fault disappeared when the crystal oscillator soldering pin was heated with a soldering iron. After replacing a high-quality crystal oscillator component and two ceramic capacitors, the fault was eliminated.

[ Failure Example 3 ] :

A Fuji 5000G9S 47kW inverter, the operation panel displays a fixed character, cannot be operated, and the “program stuck” phenomenon occurs, which is judged to be a CPU mainboard failure. After starting the machine for inspection, power on, and measuring the CPU power supply, it is normal, but the CPU chip is hot and has an abnormal temperature rise. It is judged that the CPU chip itself has a short circuit fault. A CPU chip of the same model is removed from an old circuit and replaced to eliminate the fault.

[ Failure Example 4 ] :

An INVT INVT-G9-004T4 low-power machine was found to have a damaged inverter module. The CPU mainboard and power driver board were powered on first, and the inverter module was purchased after the driver board failure was repaired. After powering on, the operation display panel showed H:00, and all the buttons on the panel failed to operate. It was determined to be a failure of the CPU basic circuit. The three working elements of the CPU were checked first, and no abnormalities were found; the other peripheral circuits of the CPU were checked, and no abnormalities were found. For a while, I was at a loss and did not know where to start, and the maintenance work came to a deadlock.

Later, when checking the current detection circuit, the voltage of the 8th and 14th pins of the current signal input amplifier U12D was 0V, which was normal; the 14th pin of U13D was negative 8V, which was an overcurrent signal output error. But logically, the CPU should report OL or OC, SC faults, and the program should not stop running, right? Try to cut off the fault signal so that it cannot be input into the CPU, power on, and the operation panel can actually be operated!

The protection sequence of INVT G9/P9 inverter is roughly as follows: when the power inverter output part is detected to be faulty at power-on, even if the start/stop signal is not received, the SC–output short circuit fault code will still be displayed, and all operations will be rejected; when the overcurrent signal from the current detection circuit is detected at power-on, H.00 will be displayed, and all operations will still be rejected at this time; when the power-on detection has a thermal alarm signal, most other operations can be performed, but the start operation is rejected, perhaps the CPU thinks that the output module is still in a high temperature rise state, and it will wait for it to return to normal temperature before allowing it to start running. For module short circuit faults and overcurrent faults, in order to ensure safe operation, all operations are simply rejected! However, this protective measure is often mistaken for the program entering an infinite loop, or a CPU peripheral circuit failure, such as abnormal reset circuits and crystal oscillator circuits.

After repairing the current detection circuit and checking that there was no abnormality in the drive circuit, the fault was eliminated after replacing the power module.

If you frequently repair a certain brand or several brands of inverters, you should have the terminal diagram of the connection cable between the CPU motherboard and the power/driver board. With it, the maintenance efficiency will be greatly improved. I spent a lot of effort to survey and map the terminal diagrams of the CPU motherboards of three inverters. I dare not keep it secret, so I share it here with you.

The above figure shows the function/destination (50 lines) of the TECO 7300PA inverter mainboard wiring terminal 7CN. The inverter of this model has excellent versatility in CPU mainboards, ranging from 22kW to 300kW, and can be interchanged. You only need to adjust the parameter “inverter capacity”. That is to say, this brand of inverter, due to different power levels, except for the power supply/driver board, actually uses the same CPU mainboard, so this terminal diagram can be used for “all sizes” during maintenance.

Using the terminal diagram:

1. Check the input and output control signals. Most of them are switch signals, and the high and low level values are very obvious. For example, the detection signal terminals 2 and 4 of the charging contactor and temperature sensor have a terminal voltage of 24V when normal and 0V when abnormal. The static and dynamic (start/stop) voltage values of the six-way pulse signal terminals are obvious; although the output current detection signal is an analog quantity, it should be 0V when static and unloaded. Generally, it is about 2V when fully loaded.

2. You can “play tricks” on the terminals. For example, if the module overheats, you can try to short-circuit terminal 4 and pin 28. If the fault disappears, it means that the external temperature relay (sensor) at terminal 2 is faulty. If the fault still exists, it means that the fault is in the subsequent temperature signal processing circuit.

If you connect a potentiometer between terminals 10 and 29 and connect its center arm to the output current detection terminal (terminal 7 or 9), you can simulate a load test to check whether the subsequent current detection circuit is normal.

If an analog signal only deviates from the normal value (and still works), and falsely reports an “overvoltage” fault under normal power supply, you can use parallel and series resistance methods at terminals 23 to make its voltage value fall back to the normal detection value for emergency repair.

[ Failure Example 1 ] :

    A 300kW7300PA inverter has an output frequency display on the operation display panel, but there is no three-phase voltage output on the U, V, and W terminals. After disconnecting the power supply of the inverter circuit, the six pulse signals of the 11, 13, 15, 17, 19, and 21 pins of the 7CN terminal were measured and all were normal. It was determined that the fault was in the pulse pre-stage signal circuit of the power supply/driver board. From these six terminals, it was quickly found that U12 (MC14069) was defective. After replacement, the fault was eliminated.

[ Failure Example 2 ] :

    A 300kW7300PA inverter reported a “DC circuit undervoltage” fault after power-on and could not start running. The DC circuit voltage detection signal of terminal 23 of the wiring terminal 7CN was normal. After power-on, the measurement of terminal 2 was always low level, indicating that the auxiliary contact of the charging contactor was not closed. The CPU detected that the charging contactor was not closed, so it reported an undervoltage fault.

Try to short-circuit terminal 2 and terminal 28, and it runs normally.

Check from terminal 2 to the power supply/driver board and find out that the normally open contact of relay KA1 is connected in series in the 2-terminal circuit. The contact of KA1 is in poor contact. Replace KA1 and the fault is eliminated.

[ Failure Example 3 ] :

A 7300PA37kW inverter reported an OC fault as soon as it was powered on. The power supply of the current detection circuit was detected from the terminals. The -15V of terminals 40, 41, and 42 was 0V. Due to the abnormal power supply, the output of the current detection circuit was offset and an OC fault was reported. The -15V rectifier diode of the switching power supply was checked and found to be open. The fault was eliminated after replacement.

In the above three faults, with the guidance of the 7CN terminal diagram, the faults were found quickly and accurately in a very short time and the faults were eliminated efficiently.

The above picture is the Alpha inverter CPU motherboard CNM wiring terminal diagram, a total of 24 lines. Although the CPU motherboard of this model cannot be replaced – the overload protection and other parameters are fixed and the relevant parameter values cannot be modified; the CPU motherboard circuits are also different. I have seen two motherboard circuits, but the CNM terminals are the same. Therefore, the CNM terminal function/destination diagram is still very valuable as a “guide” for maintenance.

[ Failure Example 1 ] :

An Alpha 2000 15kW inverter tripped an undervoltage fault as soon as it was started, and no charging contactor closure sound was heard when it was powered on. The CNM terminal 1 was tested and was always at a high level of 24V, indicating that the charging relay control circuit was faulty. From terminal 1 to the CPU mainboard, it was determined that the circuit inside the CPU control pin was damaged.

Emergency repair: Short-circuit terminal 1 and terminal 7. After the switching power supply starts to oscillate, force the charging relay to be energized and closed to eliminate the fault.

[ Failure Example 2 ] :

An Alpha 2000 18.5kW inverter, with a load rate of less than 50%, but the on-site power supply fluctuates greatly, sometimes as low as 320V. The inverter tripped an undervoltage fault, and the dealer and the user required measures to make the inverter run. Cut the cable of CNM terminal 8, connect a 4.7kΩ semi-variable resistor between terminals 25 and 7 , connect the center arm to terminal 8, and adjust it to a fixed voltage of 3V.

The inverter no longer trips the undervoltage fault and operates normally. (This emergency measure must be used with caution in specific applications!)

There is a trick to speed up the inspection, which is to make a diagram of the motherboard wiring terminals. Clearly mark the terminal serial number and the source and destination of the terminal signal. It would be even better if the dynamic and static voltage values of each terminal could be marked. Fortunately, there are many switching signals, which are easier to detect and judge. Analog signals also have obvious differences in dynamic and static, which are easier to detect.

The figure above is the function/destination diagram of the CN1 terminal. For terminals 4, 7, 8, and 9, the terminals are set by the manufacturer. Except for replacing the CPU motherboard, you can check the short-circuited state of the terminals. For maintenance, it is useless, so don’t worry about it. 24V, +15, -15V, +15V, and +5V power supplies are the 5 power supplies commonly used by the CPU motherboard. Some inverters have two 24V power supplies, one for the control relay coil and the other for the control sub-power supply. The power supply introduction in the terminal occupies most of the terminals, six inverter pulses, and some have one DC braking pulse, which occupies part of the terminals. The others are various input/output, switch quantity/analog quantity signal terminals. What we need to pay attention to and frequently test is this third type of terminal. When checking a certain circuit, you can also artificially change the voltage state of this type of terminal to see if the corresponding circuit reacts. By observing or measuring the normal or abnormal reaction of the corresponding circuit, you can clearly judge whether the circuit is good or bad.

[ Failure Example 1 ] :

An INVT G9/P9 5.5kW inverter has an unbalanced output three-phase voltage. After disconnecting the inverter power supply, the start signal is input and the six pulse signal terminals 19, 21, 23, 25, 27 and 29 of CN1 are detected. It is found that the voltage at pin 21 is fixed and there is no change in dynamic or static state. It is judged that the fault is in the pulse pre-stage circuit or the PWM pulse output pin of the CPU.

Check the inverter pulse pre-stage circuit U4 (LS07), the six-way inverter pulse at the input end is normal, and there is no signal voltage output at the output pin 8. Replace U4 and the fault is eliminated.

[ Failure Example 2 ] :

An INVT G9/P9 5.5kW inverter reported an undervoltage fault during operation. The CN1 terminal 39 pin was detected to be low level, indicating that the CPU has output a charging relay closing signal. The charging relay was checked and found to have been replaced. The coil solder joints were poorly soldered. After re-welding, the fault was eliminated.

(A) Power Supply Fault Inspection:

Fault Status: After powering on, the entire device does not respond, and the operation display panel shows no display. The measured 24V and 10V control power at the control terminals are both 0V.

Fault Essence: The VFD’s switching power supply is not working.

Troubleshooting Approach: (1) Switching voltage fault; (2) Pre-charging circuit fault.

Troubleshooting Method:
(1) First check the power supply source of the switching power supply and whether the DC loop has a normal 530V voltage. If the DC loop voltage is 0V, it indicates a fault in the pre-charging circuit, such as an open charging resistor, damaged half-wave rectifier circuit, or poor contact of the normally closed contact of the contactor or relay. Repair the pre-charging circuit first and then check the fault of the switching power supply. Often, after repairing the pre-charging circuit, the VFD is also repaired. It is not necessary to focus too much on the switching power supply circuit at first.

(2) If the 530V, 265V, or 300V DC power supply for the switching power supply is available, do not focus too much on its voltage stabilization and oscillation circuit. First, check if there are any faults such as short circuits in the secondary load circuit of the switching transformer, such as a damaged cooling fan, an IC short circuit in the fault detection circuit, or a breakdown of the rectifier diode. The fault rate on the load side of the switching power supply is higher, while the problems in the oscillation and voltage stabilization links are less common.

The troubleshooting approach and order determine the efficiency of the repair work. From the perspective of the entire circuit, checking the pre-charging circuit is a very important step and should be the first consideration when troubleshooting the fault of a non-working switching power supply.

(B) Fault Inspection of the Inverter Pulse Circuit:

The section from the six PWM output terminals of the CPU to the intermediate buffer circuit is called the pre-stage circuit of the inverter pulse, while the drive circuit is referred to as the post-stage circuit of the inverter pulse, collectively known as the inverter pulse circuit.

Fault Conditions:

(1) Normal start-up operation, with normal output frequency indication on the operation display panel, but no three-phase output voltage.
(2) Normal start-up operation, with normal output frequency indication on the operation display panel, but unbalanced three-phase output voltage.
(3) OC fault occurs immediately after pressing the start button.
(4) OC fault occurs during operation.
(5) Light-load operation is normal, but motor with load jumps or encounters OC fault.

Essence of the Faults and Inspection Approach (Corresponding to the Five Fault Conditions):

(1) Several factors may contribute: a. Loss of +5V* power supply on the input side of the drive circuit’s optical coupler; b. Damage to the buffer of the pre-stage pulse circuit; c. Uncertainty of the CPU’s relevant control signals or damage to related control pins; d. Misoperation of the fault protection circuit, resulting in the locking of the pulse pre-stage circuit by the fault signal.

Special attention should be paid to the pre-stage circuit of the inverter pulse signal, such as tri-state triggers and buffer circuits, which may be directly controlled by voltage and current detection and protection circuits. When the protection circuit misoperates, it may clamp and block the transmission of six pulse signals. The concept of the fault protection circuit participating independently in pulse transmission control should be kept in mind. Although faults caused by a and b are more common, those caused by c and d often constitute difficult faults, and a lack of inspection approach in this regard may lead to detours in repair.

(2) Three factors may contribute: a. Damage to the opto-coupler of the drive circuit, preventing the normal transmission of inverter pulse signals; b. Increased internal resistance of the inverter module, leading to poor conduction in three upper-arm IGBT modules. Therefore, the three drive circuits may not be equipped with IGBT voltage drop detection circuits, resulting in a failure to report OC faults; c. Malfunction of the pre-stage pulse circuit or the CPU inverter pulse output pin, causing the inverter pulse to be missing in one or two channels.

Don’t focus solely on the post-stage drive circuit, as the inverter pulse of the pre-stage may not be input to the drive circuit. Especially, consider whether the module is faulty or the internal resistance of the inverter module has increased. Failing to consider factor c may also lead to difficult faults.

(3) Several factors may contribute: a. Defects in the post-stage drive circuit itself; b. Insufficient load capacity of the power supply of the drive circuit, such as loss of capacitance in filter capacitors and low efficiency of rectifying diodes (increased forward resistance and decreased reverse resistance); c. Defects in the inverter module.

Dynamic and static testing (voltage testing) of the drive circuit may appear normal, but it is necessary to test the current output capability of the drive circuit. Pay attention to factors b and c.

(4) Several factors may contribute: a. Load capacity of the drive circuit and internal resistance detection of the inverter module; b. Three-phase output current detection circuit; c. Reference voltage circuit in the fault detection circuit; d. User load-related reasons.

Pay attention to the influence of factors b, c, and d. Defects in the three-phase detection circuit itself or shifts in the operating point may cause false OC fault reports. Deviations in the reference voltage of the fault detection circuit may lead to inaccurate current detection and false OC fault reports. If all checks are fine, look for the cause on the production site, not excluding issues related to the load. Factors b and c may again fall into the category of difficult faults.

(5) Three factors may contribute: a. Insufficient current (power) output capability of the drive circuit; b. Defects in the inverter module, resulting in increased internal resistance; c. Issues with the load circuit, such as a faulty motor, not necessarily a fault of the VFD.

Abnormal operation of the VFD does not necessarily indicate a problem with the VFD itself. It is recommended that users try replacing the motor. Consider factors b and c, and sometimes factors outside the VFD.

The main circuit of the fault detection circuit is still composed of an operational amplifier. Usually, the operational amplifier is connected to the following types of circuits, completing three tasks of signal analog amplification, comparative output, and precision rectification.
(A)、 Inverting amplifier circuit:

An operational amplifier has excellent characteristics such as high input impedance (without using signal source current), low impedance (with good load characteristics), amplification of differential mode signals (the difference between two input signals), suppression of common mode signals (with the same polarity and size of the two input terminals), and the ability to provide linear amplification for both AC and DC signals.
The figures (1), (2), and (3) in the circuit form are inverting amplifiers, where the output signal is in opposite phase to the input signal, also known as inverting amplifiers. The circuit has a dual amplification effect of voltage and current on the input voltage signal, but in small signal circuits, only the amplification and processing of the voltage signal are emphasized. The voltage amplification factor of a circuit depends on the ratio of R2 (feedback resistance) to R1 (input resistance). R3 is the bias resistor, and its value is the parallel value of R1 and R2. Due to the different values (ratios) of R2 and R1, three types of signal transmission functions can be completed, namely, forming three signal processing circuits: inverter amplifier, inverter, and attenuator. (1) The circuit is an inverting amplifier circuit with an amplification factor of 5; (2) The circuit is a inverter, which plays a reverse phase output role on the input signal and has no amplification factor, so it cannot be called an amplifier. Or input a 0-5V signal, then output a 0-5V inverted phase signal; (3) The circuit is an attenuator circuit. If a 0-10V signal is input, the output is 0-3. A 3V inverted signal is a proportional attenuator.
Figures (1), (2), and (3) have two characteristics of the circuit: 1. The input and output signals are reversed; 2. Whether it is an amplification, attenuation, or inverting circuit, the output signal maintains a proportional output relationship with the input signal, which can be broadly referred to as an inverting amplifier because the amplification factor of the inverter is 1, and the attenuator precisely utilizes the amplification effect of the circuit.
It is interesting that these three types of inverting amplifiers have applications in current and voltage detection circuits. Taking the current detection circuit as an example: This is because the current transformer connected in series to the three-phase output terminal has an amplifier built-in, and the output signal has reached a voltage amplitude of volts, while the input signal amplitude of the CPU must be within a voltage amplitude of 5V or less. Therefore, some reverse current signal processing circuits use reverse amplifiers with a certain amplification factor; Some use inverter circuits, which only perform inverter processing on the signal based on the polarity requirements of the CPU input voltage signal, without the need for further amplification; Some circuits are adapted to the signal amplitude range of the later stage circuit, and even use attenuator circuits to attenuate the voltage signal from the current transformer before sending it to the later stage circuit.
The power supply of the analog signal circuit in the detection circuit is generally powered by both positive and negative 15V dual power sources according to the requirements of amplifying AC signals. Based on the circuit form of the inverter amplifier and the circuit characteristics of the operational amplifier, we can find the corresponding detection method:

1. According to the characteristics of the inverting amplifier, when a positive signal voltage is input, the input voltage must be negative below 0V, and vice versa, the output voltage must be positive above 0V, with 0V (ground) as the reference potential. To determine whether it is in a normal state based on the static circuit values of the circuit and input/output pins;
2. Identify whether the circuit in this stage is an amplifier, inverter, or attenuator. Based on the ratio of input resistance to feedback resistance, the output voltage value can be roughly calculated to determine whether the circuit is in a normal state;
3. According to the characteristic that the circuit has an amplification (or attenuation) effect on differential mode signals and a zero amplification effect on common mode signals, when short circuiting two input terminals, the output voltage should be close to the zero potential value; Alternatively, if there is a positive voltage (or negative voltage output) at the output terminal, but two input terminals are short circuited, the output voltage immediately drops (or rises) to around 0V. The circuit is good and can transmit signals normally.
4. If the input voltage value can be artificially changed, the output voltage will inevitably change accordingly, which can be used to determine whether the amplifier is in a normal state.
   [Fault Example 1]
   After a certain frequency converter is powered on, an OC fault is reported, and the fault reset is invalid. The current detection circuit, as shown in the diagram (1), has an output voltage of+12V. The CPU reports an OC signal after power on due to a serious overcurrent signal input. Short circuit the 2 and 3 pins of the operational amplifier with metal tweezers, measure that there is no change in the output circuit of pin 1, and it is still+12V. It is judged that the operational amplifier is damaged. After replacement, the fault is eliminated.
   [Fault Example 2]
   A certain frequency converter outputs an undervoltage signal after being powered on, and detects the electrogram (2) circuit. The input voltage is -3V, but the output voltage is 0. 7V, indicating a malfunction of the amplifier in this stage. A 10k resistor was connected in series with an external DC 12V power supply, and the output voltage did not change when input to the reverse input terminal. It was determined that the amplifier in this stage was damaged, and the fault was resolved after replacement.
   (B)、 In phase amplifier and voltage follower circuit:

The circuit shown in Figure (1) is a typical circuit form of a in-phase amplifier, which is also one type of amplifier circuit. The input signal enters the same phase end of the amplifier, and the output signal is in the same phase as the input signal. The voltage amplification factor of the circuit is 1+R2/R1. It is also used for amplifying analog signals in fault signal detection circuits. When R2 is short circuited or R3 is open circuited, the output signal has the same phase and equal magnitude as the input signal, so (1) the circuit can further evolve into (2) and (3) circuits.
The above figures (2) and (3) show the voltage follower circuit. The output voltage completely tracks the amplitude and phase of the input circuit, so the voltage amplification factor is 1. Although there is no voltage amplification effect, it has a certain current output ability. The circuit plays a role in impedance transformation, improving the load capacity of the circuit and reducing the mutual influence of high impedance in the signal input circuit and low impedance in the output circuit. When used as a circuit follower, sometimes a single power supply is also used.
(1) The circuits (2) and (3) are also used in fault detection circuits for amplifying analog signals and processing reference voltage signals.
Based on the characteristics and functions of the circuit, the detection method can be obtained as follows:

1. (1) The circuit is a in-phase amplifier circuit, and the output voltage amplitude and polarity are proportional to the input voltage. The voltage amplification factor of this stage is about 6 times. When the input voltage value is 1V, the output voltage is approximately 6V. It is possible to determine whether the circuit is in a normal state based on the calculation of input and output voltage values;
2. (2) (3) The circuits are all voltage follower circuits, and the output voltage is completely tracked by the input voltage. The output voltage should be equal to the input voltage, which can be used to determine whether the circuit is in a normal state.
3. By short circuiting two input terminals or manually changing the input voltage, the corresponding changes in the output voltage can be measured to determine whether the circuit is in a normal state.
   [Fault Example 1]
   A certain frequency converter experienced an OH fault when powered on. The reference voltage circuit of the temperature detection circuit, as shown in Figure (2), has an output voltage of 1V. This machine is a voltage comparator circuit, and its input voltage is measured to be 5V. Under normal conditions, the output voltage should also be 5V. After cutting off the output load circuit, the output voltage remains at 1. 2V, it was determined that the amplifier in this stage was damaged, and the fault was resolved after replacement.
   (C) Precision positive and negative half wave rectifier and full wave rectifier circuits:

The AC voltage signal from the current transformer needs to be rectified into DC voltage through subsequent half wave or full wave rectification circuits, and then sent to the CPU for current display and control. Precision half wave or full wave rectification circuits are also used for processing and amplifying analog signals. Ordinary rectification circuits use diodes as rectification devices, but diode rectification has drawbacks such as nonlinear distortion and dead zone voltage. Especially when used for small signal rectification, it will cause output signal distortion and output errors. By utilizing the amplification effect and deep negative feedback effect of the operational amplifier, a diode is added to the amplification circuit. By utilizing the unidirectional conductivity characteristics of the diode, different depths of negative feedback are introduced to the input positive and negative half wave signals. The input μ V level signal can be rectified in a dense manner, and the circuit itself also has a voltage following or amplification effect.
The above figure (2) shows a precision negative half wave rectification circuit. The circuit will perform precision rectification on the input negative half wave signal and output it in reverse phase. For the positive half wave input signal, the connection of D1 introduces deep negative feedback to the amplifier; At the beginning of the negative half wave input signal, due to the signal input amplitude being smaller than D1 and D2, both are in the cut-off state, and the circuit is in an open-loop amplification state. A small signal input will cause the output pin voltage to be greater than -0. 7V, D1 conduction, D2 reverse bias cutoff; The series connection of D2 and R125 introduces moderate negative feedback (the resistance of R125 can determine whether the current circuit is a rectifier or a rectifier amplifier, and the current circuit is a precision rectifier with no amplification effect), which is equivalent to an inverter amplifier, and the output is inversely correlated with the input signal.
The difference between the circuit in Figure (1) and Figure (2) is that the polarity of the two diodes in the circuit is opposite, making it a precise rectification circuit for the input positive half wave signal. The principle of rectification is the same.
By adding a half wave rectifier circuit and an inverse summation circuit, as shown in Figure (3), the positive and negative half waves are input and output in reverse to obtain a full wave output voltage waveform, forming a high-precision full wave rectifier circuit.
In fault detection circuits, rectifier circuits are often used to sample the three-phase output current signal, which is rectified and amplified as an analog voltage signal (current detection signal) input into the subsequent fault signal processing circuit and CPU circuit, used for overload alarm and sampling processing of operating current.
The input of the circuit is an AC voltage signal, while the output is a DC voltage signal. Most circuits are rectifiers, and some circuits are rectifier amplifiers.
Detection method:

1. Rectifier circuit: The input side is AC voltage, and the output side is DC voltage. The two measured values are relatively close.
2. Rectifier amplifier, with AC voltage on the input side and DC voltage on the output side. The output DC voltage value is higher than the input AC voltage value.
3. By short circuiting two input terminals or manually changing the input voltage, the corresponding changes in the output voltage can be measured to determine whether the circuit is in a normal state.
   [Fault Example 1]
   A certain frequency converter experienced an OC fault when powered on. The output voltage of the current detection circuit as shown in Figure (2) was 13V. After unplugging the lead terminal of the current transformer, the amplifier in that stage still had 13V. It was determined that the rectifier circuit was damaged and the fault was resolved after replacement.
   (C) Circuit for voltage comparator, step voltage comparator, and window voltage comparator:
   The above-mentioned circuits are all used for amplifying and rectifying analog signals, and their output signals still have analog signals, which can be called analog signal (amplification) processing circuits. However, in the following voltage comparison and other circuits, the output is a switch signal. The circuit has left the scope of analog amplification and seems to have entered the field of “digital circuits”, using analog circuits as digital circuits for application.

The function of a voltage comparator is to compare the magnitude of two input voltage signals. In Figure (1) of the circuit, the voltage at the same phase input end of the amplifier is the voltage divider value 2 of R2 and R3 resistors to+5V. 5V, known as the reference voltage value, compares the input signal with this reference value. When it is higher than this value, it outputs a 0V low-level signal. When it is lower than the low value, it outputs a+15V high-level signal. Circuit, also known as a single value comparator, the output state of the circuit depends on a value (a point) of the input signal voltage -2. 5V.
If the two-stage voltage comparator is connected to the circuit shown in Figure (2), it becomes a stepped voltage comparator. The circuit has one input signal and two output signals. The N1 and N2 voltage comparators input the same signal voltage, but the reference voltage values at the same phase input terminals of the two-stage circuit are different. The N1 reference voltage is 6. The reference voltage value for 6V and N2 is 3. 3V. When the input signal gradually increases from 0V to 3. When the voltage is above 3V, the output state of N2 first changes to low level; N2 has an input signal value greater than 6. At 6V, there is only a low-level signal output. When the circuit in Figure (2) is used for voltage detection in the DC circuit, when the regenerative energy generated by the load motor is fed back to the DC circuit, causing the DC circuit voltage to rise to a certain value, N2 first outputs a braking action signal, connects the braking resistor to the DC circuit, and consumes the voltage increment; If the voltage continues to rise, N1 will output an OU overvoltage signal, and the frequency converter will shut down for protection.
If the two-stage voltage comparator is connected to the circuit shown in Figure (3), it constitutes a window voltage comparator circuit. Compared to single-stage voltage comparator circuits, window voltage comparators can be referred to as dual value comparators. The circuit has two benchmark comparison values and outputs one signal. When the input signal is ≥ reference voltage 1 ≤ reference voltage 2, the circuit output state transitions. Within a range of the intermediate value of the input signal, the output state remains unchanged. The circuit in Figure (3) is a ground fault signal processing circuit. The in-phase end of the N1 amplifier is the voltage divider of R46 and R50 to+15V, while the reverse end of the N2 amplifier is the voltage divider of R81 and R69 to -15V. The input three-phase current sampling signal enters the reverse input terminal of N1 and the in-phase input terminal of N2, and is compared with the positive and negative partial voltage values, respectively. Whether it is the positive half wave or negative half wave of the input signal, as long as it is greater than the two reference values, a ground short circuit signal will be reported.
The voltage comparator uses digital circuits, which can flexibly set the reference voltage based on the signal amplitude, making it more convenient than using digital circuits. In addition, the circuit in Figure (3) adopts an operational amplifier circuit with an open collector output, which can achieve parallel output at the output end, making the circuit more concise. If a regular amplifier is used, the output signal also needs to be isolated by two diodes and connected together in parallel.
Three types of voltage comparator circuits are commonly used to convert analog signals of detected current or voltage into switch signal – fault signal output, for implementing control actions and shutdown protection.
Detection method:

1. The amplifier output has only two level states, low level, close to the ground level of the power supply or negative power supply value; High level, close to positive power supply value;
2. If the voltage value of the inverting input terminal is lower than that of the in-phase input terminal, the output is low level; otherwise, the output is high level.
3. By short circuiting two input terminals or manually changing the input voltage, the corresponding changes in the high and low levels of the output terminals can be measured to determine whether the circuit is in a normal state.
   (E) Hysteresis comparator circuit:
   It is also one of the voltage comparator circuits. The circuit in Figure (3) of the voltage comparator, also known as the hysteresis voltage comparator circuit. The voltage comparator circuit can be upgraded to a hysteresis comparator circuit by introducing an additional positive feedback circuit. Hysteresis comparator circuits are referred to as voltage comparator circuits with hysteresis characteristics. If ordinary voltage comparison is regarded as “voltage point comparison”, hysteresis comparator can be regarded as a comparator circuit for “voltage range comparison”. Usually, we hope that the output state of the circuit is stable enough, and comparing the voltage at a “point” can cause instability in the output state due to frequent output. Improving the “point” comparison of the input circuit to “segment” comparison can effectively solve this problem – within a “segment value” of input voltage variation, the output state remains unchanged. Figure (2) shows a positive feedback branch composed of R4 and D1, which converts the “point” comparison characteristic of the circuit into a “segment” comparison characteristic.

The control principle is briefly described as follows:
Assuming that the circuit in Figure (2) is used for processing the braking action signal, the input signal is the voltage sampling signal of the DC circuit. When the voltage of the DC circuit rises abnormally due to the energy feedback from the load motor, reaching 680V, the input voltage value of Vin reaches 9. When the voltage is above 5V and higher than the reference voltage at the reverse end of the amplifier, the amplifier outputs a low-level signal, and the subsequent braking circuit acts, connecting the braking resistor to the DC circuit to consume the voltage increment; Due to the consumption effect of the braking resistor, the input voltage value of Vin quickly drops to 9. Below 5V, but the braking signal is still being output, and it does not mean that the DC circuit voltage slightly drops, causing the braking signal to disappear. This indicates the role of the hysteresis comparator. The braking circuit continues to operate until the DC circuit voltage returns to below 620V, and the sampled input voltage is below 7. The braking circuit only stops working at 5V.
When the circuit is static, the voltage at the same phase end of the amplifier (7.5V) is higher than the voltage at the opposite phase end, and the output voltage is a high-level voltage of nearly 15V. R4 and D1 are introduced into the same phase end circuit, artificially raising the same phase end voltage to 9. 5V. When the input voltage is above 9. At 5V, the circuit output state reverses and the output end becomes low level. D1 reverse bias cutoff, feedback loop interruption, and reference voltage at the same phase end restored to 7. 5 V partial voltage value. In this way, when the input sampling voltage is below 7. At 5 V, the brake signal stops outputting.
Hysteresis comparator circuits are commonly used for voltage detection in DC circuits, outputting braking signals and overvoltage/undervoltage fault signals.
The detection method is the same as the voltage comparator, omitted.