PLC project debugging signal interference case analysis

[Introduction] During the debugging of a PLC project, you may encounter all kinds of strange problems, signal interference is one of them, and it is difficult to find the cause of the problem. Today I will share two cases, hoping to help you.

Example A

1. Description of the phenomenon:

The AO point in the Siemens PLC sends out a 4-20mA current control signal, which is output to the Siemens inverter, but cannot control the inverter to start.

(II) Fault finding:

(1) Suspected AO board problem, using a multimeter to measure the 4-20mA output signal, the signal is normal. (2) Then suspected that there is a problem with the inverter control signal input terminal, replaced with a inverter of the same model, the problem still exists. (3) Use a handheld signal transmitter as a 4-20mA output signal source, output standard current signal to the inverter, the inverter starts, so we rule out the analog output board and inverter faults. (4) It is speculated that the inverter interference signal is transmitted to the analog channel. (5) Therefore, a signal isolation module is added to the PLC analog 4-20mA output channel, the isolator input terminals 5 and 6 are connected to the analog output module, the output terminals 1 and 2 are connected to the inverter, and the terminals 3 and 4 are connected to the external 24VDC power supply, the inverter starts normally. (6) Based on this, it is concluded that the root cause of the problem is the inverter interfering with the analog channel.

(III) Notes:

(1) The PLC power supply and the power system power supply (inverter power supply) are configured separately, and the PLC power supply should use an isolation transformer;

(2) The power line should be separated from the signal line as much as possible, and the signal line should be shielded;

(3) Regardless of analog signal input or analog signal output, analog channels must use signal isolation modules;

(4) Design software filters in PLC programs;

(5) Design the signal ground and power ground separately.

Example B

Some time ago, I saw an analysis and solution to the problem of analog interference. In our actual application, we will encounter many similar problems. Share it with everyone: "There are 10 250KW motors in the workshop, and the load is a high-pressure pump. The inverter uses Schneider ATV71 and is connected to the PLC through DP. The PLC uses Siemens 300, the pressure transmitter is Siemens, and the transmitter to the PLC is 4-20mA analog, and the shielded line is used in the middle.

After debugging, everything was normal after running for a week. After the manufacturer left, pump No. 8 suddenly appeared when it was turned on. The pressure was set to 40 kg, and the actual value was 70 kg. The actual value was set to 80 kg, but it was 110 kg. At first, I suspected that the sensor was faulty, but everything was normal after replacing it with other pumps. After that, the inverter was fully turned on, and pumps No. 3, 4, 5, 6, 7, 9, and 10 also had similar problems. It was speculated that the pressure sensor was interfered by the inverter. The manufacturer suggested adding metal pipe shielding. However, considering the difficulty of on-site construction (the control room is more than 30 meters away from the motor, and all cables are underground).

And I think the harmonic interference of the inverter should be the pressure value fluctuating up and down, and it is rare to see interference causing a linear increase. At first, I suspected that there was something wrong with the manufacturer's program, because the pressure value displayed on the display was always 40 kg, but the inverter output a frequency of 70 kg. The manufacturer disagreed with this view and said that they definitely used the Siemens standard PID block.

I couldn't figure it out. I accidentally discovered that the negative and shielding layers of the manufacturer's sensor were connected to the M of the PLC analog input terminal at the same time. After removing the shielding wire and connecting it to the ground of the equipment, the fault was eliminated. Inference: For a 2-wire sensor, the positive pole has a 24V voltage provided by the PLC, and the negative pole is where the sensor outputs a 4-20mA current. After the shielding wire and the negative pole are connected together, the induced electromotive force on the shielding wire generates a current that enters the PLC input terminal together, causing a superimposed current, thereby forming a linear increase in the pressure value. As a result, the same situation occurred within two days of operation, and what was more serious was that one of the pressure sensors was removed and there was still 40 kg of pressure. Finally, it was found that the negative pole stripping on the PLC input side was too long, short-circuited with each other, causing the signals of other channels to string out. Then I remembered that when I first started debugging, the manufacturer asked me if the equipment ground and the cabinet ground were not on the same ground. After the two ends of the pressure sensor shielding wire were grounded, the interference was particularly strong. It couldn't be displayed. I didn't think much about it and said it was single-ended grounding, and it turned out to be normal. Now I think that the current output side of each sensor was connected together through the shielding wire, causing a short circuit. After that, the ground wire on the sensor side was removed. Since the shielding wires were not connected together, the signal was normal.