Tag: POWTRAN

Repairing a PI-18 type 11kW universal frequency converter, many “secrets” were discovered during the processing. The summary is as follows: 1. The damage to the module is not only a short circuit or open circuit in the main current terminals R, S, T and U, V, W, but also possible short circuits between the triggering terminals and the main terminals, short circuits between the triggering terminals, and open circuits inside the triggering terminals. Measuring that there is no short circuit in the main terminal does not confirm that the module is not damaged; 3. Assuming that there are no issues with the measurement of the main terminal and trigger terminal, it cannot be completely confirmed that the module is not damaged. There are still hidden damages such as leakage and poor performance in the module. The key is to take measures to verify its quality and ensure that the final installed block is a good one. And during the power on debugging process, it will not cause new faults, thereby expanding the scope of faults and causing human trouble. Special attention should be paid to the selection of disassembled modules. There is no abnormality measured with a multimeter, but there may be potential damage. Choose a good module, and the DC voltage of the main circuit should not be easily delivered. It is necessary to first use a lower DC power supply and verify that there are no abnormalities before connecting the DC bus circuit and conducting a startup test.

A. Abnormalities encountered during repair:
Measure the terminals of the main circuit for any abnormal phenomena such as short circuits, and especially test the other pins of the inverter module once to confirm that they can be powered on for testing; Power on, display normal, press the start button when unloaded, trip OC protection and stop.
To verify the source of this overcurrent signal, check the input processing circuit of the current transformer, which is LM347 (with the same structure as LM324). If the input and output states are manually changed, the frequency converter does not respond. It appears that the OC signal is not output by the current detection circuit.
Considering that it is still fed back by the inverter module, a thorough inspection of the driving circuit, including the filtering capacitors of the four power sources in the circuit, and a capacity check have also been conducted. No abnormalities.
The driving circuit of this machine adopts three integrated circuits, namely A4504, MC33153, and P521. A4504 is the isolation optocoupler between the CPU input trigger pulse and the main circuit, MC33153 is the module driver, and P521 is used to feedback abnormal conditions of the inverter module to the CPU, in order to achieve rapid shutdown protection. During power on, whether in standby or startup mode, short circuit any output terminal of the six P521 circuits, and the frequency converter will trip OC protection to stop the machine. The circuit is very sensitive. The start under no-load conditions triggers the OC, which is mostly fed back to the CPU by these six optocouplers.
It was determined that the inverter output module SKM75GD124D was damaged. After the power outage, a multimeter was used to carefully check and no abnormalities were found. The measurement results of the six main terminals and other trigger terminals are the same when compared with a good block. Give it a try by changing it. Only the replacement module has been tested.
Purchase a dismantling product of the same model, disconnect the DC power supply of the module after welding, activate the operation panel button, and the panel displays normal frequency output. Measure the DC voltage of the six drive outputs -0V when the module trigger end is not started and running, and around 7.6V after starting, all of which are normal. Using an oscilloscope to measure the amplitude and variation of the six trigger pulses is also normal, and it is determined that the driving circuit and connecting wires are accurate and error free.
I still didn’t dare to rashly connect to the DC bus. I first connected the DC24V switching power supply and tried to start it. The frequency display of the frequency converter was normal, and the output voltage of U, V, and W was measured. At 50 Hz, the voltage was only 13V, and there was a periodic contraction in the output amplitude, but it could still be triggered and operated normally. Due to unfamiliarity with the output waveform, this phenomenon was not noticed, and a DC power supply of around 200V was connected. When powered on, the OC still jumped! I feel like there is still a problem with the module! Replace with a confirmed good module on the ground, connect to a 24V switching power supply, and then measure the output voltage of U, V, and W. At 50Hz, the voltage value has risen to 17.8V, and the output amplitude is constant without any contraction phenomenon. This is the good block!
The internal IGBT tube of the purchased disassembled module was not directly broken down, but it was damaged and had a large leakage current. When connected to a 24V power supply, although the output dropped to 13V, it did not cause a fault protection action. However, when connected to a DC power supply above 200V, the leakage current had reached a certain value, leading to fault action.
I purchased another disassembled product and after repeating the above steps, everything was confirmed to be normal before proceeding with installation. Finally fixed.
Disassembling the machine module is not unusable, but it must be used with caution. Due to the extremely difficult process of soldering the module onto the circuit board and removing it, it is best to connect a lower DC power supply before installation and check its quality. If the block is damaged, clean the solder on the pins and it can be returned or replaced. After confirming the good block, solder it into the circuit board.

B. Reiterate the repair steps:

1. All six phase output trigger pulses are normal, and the inverter module can be soldered;
2. First, use a 24V switching power supply to conduct a power on test, and if there are no abnormalities, then send the DC bus voltage (such as abnormal driving circuits and leads, adding a 24V switching power supply will not damage the module. Pay attention to measuring whether the three-phase AC output is balanced and whether there is DC component in the output. If there is an abnormality, there is often a lack of trigger pulse in one arm or abnormal trigger pulse. The detection in this stage is crucial, and the hidden dangers of faults are often exposed);
3. Connect the DC power supply circuit in series with the light bulb, then connect the inverter module, conduct no-load power transmission test, measure the three-phase balance of the output, and finally connect the original DC power supply. During this process, if there are any potential faults in the module, it will be further exposed.