Two “strange” faults of AMB VSD,How to solve it?

A. Strange “fault characters”
The user sent a domestically produced frequency converter for repair, which is an Anbang Xin AMB-G9/P9 22kW frequency converter. According to convention, first remove the damaged module and power on to check if the drive circuit is normal; Power on, the operation panel displays OC fault codes; After the short-circuit fault signal returns to the optocoupler, the OC signal will no longer jump. When operating the RUN button on the control panel, the charging relay momentarily disconnects (with a click), the panel indicator light also goes out, and the display screen flashes, displaying a string of “fault characters” that cannot be found in the fault code table. Suspect that there are still other fault signals present. The output terminals of the three-phase output current detection signal are all 0V, normal. For other signals, without surveying the circuit, it is difficult to quickly identify their origins.

Occasionally, when turning off the power and then on again, it is discovered that the so-called “fault characters” mentioned above are actually startup characters! The essence of the malfunction is that there may be a short-circuit load on the load side of the switching power supply, especially in the driving circuit. When the startup signal is activated, the power supply voltage is pulled to an extremely low level, and even the switching power supply will stop vibrating due to this. Except for the charging relay being released due to insufficient suction voltage, the CPU determines that it is re powered on and displays the startup character! In fact, it is a process that is equivalent to turning on power again.
Check the driving circuit. There are two power amplification tubes connected in a push-pull form behind the driving IC, which are used to amplify the pulse output by the driving IC and then drive the inverter module. The upper and lower arm driving power amplifier circuits of the U-phase both have a transistor damaged due to module damage and impact. When there is no triggering pulse, a single transistor breakdown short circuit does not form a short circuit to the driving power supply. The arrival of pulse signals caused an instantaneous short circuit to the driving power supply due to the conduction of good and bad transistors, resulting in a momentary shutdown of the switch and power outage. The starting signal was also interrupted due to power outage, and the short circuit state of the power amplifier to the transistor after driving the IC was also relieved due to power outage. Then the CPU determines that the frequency converter is being powered on again, so the operation panel displays the power on character.
After dismantling the module, I was eager to power on and check the quality of the drive circuit, but I did not measure and judge the circuit in detail, so I wasted some time on this startup character. After dismantling the module, a thorough inspection should be conducted before powering on the driver board.

B. Fault characters that are also not present in the alarm code table
This Anbang Xin frequency converter is also quite unique. After installing the new module, the 530V DC voltage of the DC circuit is not connected first, and a 24V DC power supply is added for testing. After startup, the Br Tr FeiLuRe character jumps again, but can be reset by pressing the reset button; If the 24V power supply is disconnected, this fault still occurs but cannot be reset. Check the fault code for this item, consult the manufacturer, and answer that it is a brake circuit fault, feeling a bit out of place. The external circuit of the terminal is not connected to the brake resistor, and there is no short circuit in the brake components inside the measured terminal. It is possible that after the power supply of the inverter is shut down, the control power supply (the power supply of the switching power supply) may be connected in series, causing the detection circuit to return a fault signal.

After shutting down the 24V power supply, there is still a voltage of about 6V at the inverter power supply terminal. This voltage enters the fault detection circuit through certain links, reaching the alarm level of Br Tr FeiLuRe or acting as a Br Tr FeiLuRe fault signal. Isn’t it?
The negative pressure and pulse positive voltage of the six drive circuits are normal, especially with the guarantee of cut-off negative pressure. After the 530V DC voltage is sent to the DC circuit, there is at most an output phase failure, but it is impossible to damage the module. For safety reasons, replace the original 75A quick release fuse with a 2A one and try it directly on. Everything is normal.
It can be seen that if the 75A quick release fuse is broken or the IGBT tube of the brake control inside the module is short circuited (which may cause voltage drop in the DC circuit), an alarm signal of Br Tr FeiLuRe may be generated. The source of this signal may be reported to the CPU by the fault detection circuit after detecting an abnormal low voltage in the DC circuit. But is it not appropriate to define it as a fault in the brake circuit?
However, the occurrence of “faults” makes it impossible to implement the method of detecting inverter circuits in low-voltage power supply. It also caused some setbacks in the maintenance cost.