On the issue of common ground interference

bigbat · Published on 2022-11-25 16:37

On the issue of common ground interference [Copy link]

The following figure is: a high-voltage pulse discharge system. The green part and the yellow part (external control) are grounded. The function of the green part (discharge) is to generate a -400V voltage pulse with the same width as the input PWM pulse. The principle is that the charge pump current is not more than 30mA. The green module uses a 12V input, which is supplied by the yellow part. The blue part MCU main control and external control control the output PWM signal through the fast optocoupler 6N137. However, it is found that as long as the analog circuit of the external control module enters the discharge module, it will be disturbed. If it is not connected to the power generation module, it will work stably. Now it is necessary to rectify the circuit to eliminate the interference.

1. Add two diodes between the power supply terminals of the external control and discharge modules to prevent interference.

2. Connect a 12V voltage regulator tube in parallel to the power supply of the external control part.

3. Add a diode at the PWM output to control the current direction. Also add a 12V voltage regulator

Are these three measures effective? Do you have any better measures?

maychang · Published on 2022-11-25 18:09

[Works stably without connecting to power generation module]

Which is the “power generation module”? What color is it?

bigbat · Published on 2022-11-26 08:45

Sorry, it's the green one in the "discharge module". The voltage regulator is prepared as shown in the figure below.

maychang · Published on 2022-11-26 10:22

Interference problems are the most complicated and are difficult to analyze without seeing the electrical schematics and structure diagrams.

The three measures mentioned in the first post are probably useless. Moreover, this interference is probably not caused by "common ground".

maychang · Published on 2022-11-26 10:25

The first post says that "the function of the green part (discharge) is to generate a -400V voltage pulse with the same width as the input PWM pulse". 400V voltage is not small, and the waveform is a pulse with the same width as the input PWM pulse. It is more likely that this 400V pulse affects the yellow part through spatial coupling.

bigbat · Published on 2022-11-26 17:02

The current situation is: this negative 400V pulse affects the yellow part through spatial coupling. The yellow part is placed in an iron shell. My output is:

CLK is the output of PWM. The yellow discharge module is not made by me, so I cannot provide the circuit diagram, but I know that it uses a diode voltage doubler circuit to boost the voltage. The green and yellow are grounded.

maychang · Published on 2022-11-26 17:52

Can you explain in detail how the green part interferes with the yellow part? What happens?

bigbat · Published on 2022-11-27 11:14

There is a DA chip in the yellow part, which is used to generate a voltage of 0-4.8V. The chip is powered by 5V. In order to work stably, 78M05 is used to generate a maximum current of 500mA at 5V. At the same time, it also powers the 6N137 optocoupler. 6N137 has 3 chips, all of which are outputs. They are all PWM signals. The three channels have the same frequency but different phases (the frequency is 2HZ-100HZ). The two channels have a pulse width of 0.2-10mS milliseconds, which is the wider working pulse. One channel has a fixed pulse width of 0.2mS, which is the starting pulse. This channel is connected to the green discharge board to control the on and off of the high-voltage pulse, but does not directly provide the generation of high-voltage negative 400V. The high-voltage negative 400V pulse is passed into the gas discharge tube module. The gas discharge module works under a DC power supply of 800V. The module current is small and has been below 10 mA. After the gas discharge tube is discharged and broken down, the discharge tube is started. The two pulses are about 600V and continue to supply the discharge tube working pulse. The current is about 3A. That is, the power is 1800W. The gas discharge module is in an iron cabinet together with the discharge module, and both are connected via 12V and PWM control cables. No shielded wire is used.

The specific performance is as follows:

1. After the discharge module is connected, the voltage output by the DA chip will occasionally jump. For example, the DA output is 3.6V, and suddenly a full-scale output of 4.8V appears. The discharge tube will work suddenly.

2. When using an oscilloscope to measure, the voltage of the oscilloscope will jump irregularly during discharge, so it is impossible to measure the accurate waveform of DA. 4.8V is the highest voltage that will cause high voltage protection, so it is speculated that it is full-scale output.

3. When using an oscilloscope to measure, the circuit will occasionally have pulses when not working and sometimes have a negative voltage for a long time.

maychang · Published on 2022-11-27 12:34

Still not clear.

1. "There is a DA chip in the yellow part." The green part does not need to use the DA output. The green part only gets the switch signal (three CLK signals) from the yellow part. So what is the purpose of putting the DA chip in the yellow part?

maychang · Published on 2022-11-27 12:39

Second, "high voltage negative 400V", "800V", "about 600V...current about 3A", these power supplies obviously do not get energy from the 12V supplied by the yellow part, 12V cannot supply such a high power. However, these power supplies obviously share the same ground with the yellow part, because the three CLK signals obviously control high-power MOS tubes, which must share the same ground. Are these hundreds of volt power supplies powered by AC mains? Is the AC mains input terminal well protected against interference?

maychang · Published on 2022-11-27 12:41

3. "After the gas discharge tube breaks down, the discharge tube starts." Are the "gas discharge tube" and "discharge tube" the same thing?

maychang · Published on 2022-11-27 12:47

4. "After the discharge module is connected, the voltage output by the DA chip will occasionally jump." The DA chip is probably controlled by the blue part (marked with MCU) in the first post, but the signal that the MCU controls the DA chip is definitely not PWM. It may be serial or parallel, but it is definitely not the CLKP in the sixth-floor picture. So, the change in the DA chip output may be due to the change in the control signal (which may be serial or parallel, but must be a digital signal) output by the MCU. In other words, it is the blue part that is disturbed, not the green part.

maychang · Published on 2022-11-27 12:49

5. "When using an oscilloscope to connect to the measurement, the voltage of the oscilloscope will jump irregularly during the discharge"

When using an oscilloscope, you must open the iron shell (the yellow part is placed in an iron shell). If the iron shell is opened, the anti-interference function of the iron shell is completely lost, and it is almost the same as having no iron shell.

maychang · Published on 2022-11-27 12:52

6. "When using an oscilloscope to connect to the measurement, the circuit will occasionally have pulses when not working, and sometimes negative voltage will appear for a long time."

Which part is "not working"? The blue part? The green part? The yellow part? The discharge tube? The oscilloscope? At which point in the circuit is the "negative voltage for a long time"? How long? How many volts is the negative voltage?

maychang · Published on 2022-11-27 15:49

The 8th post mentioned gas discharge tubes. The on-time of a gas discharge tube can be controlled, but once it is on, when it is off cannot be controlled. So, I was confused when I saw the word PWM. PWM means "pulse width modulation", which changes the pulse width. But the on-time of a gas discharge tube can be triggered by a pulse, but the off-time cannot, so the change of pulse width cannot control the gas discharge tube.

As for why this circuit uses pulse width modulation, it is hard to say without seeing the electrical schematic.

maychang · Published on 2022-11-27 15:53

There is a device called a thyristor, also known as a silicon controlled rectifier. These are abbreviations, and the full name is "crystalline thyristor" or "silicon controlled rectifier". Thyristor is a semi-controllable device made using the characteristics of gas discharge. External signals can control the moment when it starts to conduct, but cannot control the moment when it turns off. Crystal thyristor means nothing more than: it is a thyristor made of crystal, and its characteristics are similar to those of gas discharge thyristors.

maychang · Published on 2022-11-27 15:59

I'm afraid the several measures you proposed in your first post are not very useful. Effective anti-interference measures require understanding the path of interference intrusion into the device before adding them to the circuit.

For example, your "gas discharge tube" is obviously a strong interference source, but how does this interference "strong electromagnetic disturbance" affect your circuit? One possibility is that the gas discharge tube affects the AC mains, and the AC mains affects the MCU by supplying power to the MCU. Is this the transmission path? We can only know this through experiments.

bigbat · Published on 2022-11-28 09:17

The blue part and the yellow part are both used to control the module ground in the iron cabinet. The DA in the yellow part is also connected to the module in the iron cabinet. The circuit in the iron cabinet is made by other manufacturers. I cannot provide the accurate working principle here. The power supply in the iron cabinet is 220V and 380V. The specific power supply is determined by the downstream manufacturers. We only control the control part. What we know is what is said above. Two of my three pulses are signals to control the operation of the "discharge tube", specifically controlling the power supply of 600V to 800V for the "discharge tube" to work. Every time the discharge tube works, it needs a higher voltage to guide the voltage -400V, which is generated by the green part. The power supply of the green part is 12V, and this 12V is provided by the yellow part.

I have never touched the discharge tube, but I heard that it is a "gas discharge tube" and the details are unknown. It is installed by downstream manufacturers. The power supply part in the iron cabinet is provided by the upstream equipment manufacturer. I am only responsible for the control part.

bigbat · Published on 2022-11-28 09:24

I guess it might be some kind of high-power light source, a bit like the principle of fluorescent lamps. First there is a high voltage to ionize the gas, and then a lower voltage is given to work.

maychang · Published on 2022-11-28 10:32

But come again.

"Every time the discharge tube works, it needs a higher voltage guide voltage - 400V, which is generated by the green part. The power supply of the green part is 12V, and this 12V is provided by the yellow part."

You don't need to tell me about gas discharge tubes. I just want to know that your -400V voltage is powered by the 12V output of the yellow part. Where does this 12V voltage come from? It can't be a battery, right? It is probably generated by AC mains through rectification, filtering and voltage stabilization by a small power frequency transformer or a small power switching power supply. Then it is certain that there are two AC mains incoming lines in the yellow part. The interference signals generated by the gas discharge tubes may enter the yellow part through the AC mains, and of course, may also enter the blue part.

On the issue of common ground interference [Copy link]

Latest reply

Comments

Comments

Comments

Comments

Comments

Comments

Comments

Comments

Comments

Visited sections