Connecting ungrounded and grounded devices!
Background (copied).
Having bought new headphones with a microphone, and coming home, I was sad to discover that the microphone was creating extraneous noise. I returned to the store, where we checked on the laptop, there was no extraneous noise from the microphone. When I got home I started looking for the reason. I connected the old ones, the headphones do not make noise. I connected new headphones again, they are noisy. After some time, I accidentally touched the system unit with my foot, and lo and behold, the noise decreased significantly.
So, I came to the conclusion that there is some kind of interference on the system body. I immediately thought about grounding, and I started measuring the voltage of the case relative to the Earth. First, I took the neutral wire and was surprised to find that the potential difference was about 100V. I decided to measure the voltage relative to the heating battery, still ~100V.
I will not go into detail about where the voltage on the housings of refrigerators/washing machines comes from. Let me just say that the reason in 99% of cases is the same as in the system unit case.
You can find a more detailed description and explanation on Google.
In short, the same reason:
There is a filter in the computer power supply that dampens high-frequency interference and throws it to the ground.
Thus, we have 110V going to the ground (if the socket is 220V), but the current is only interference current, which means our current strength will be insignificant.
We get 110 Volts on the body only when it does not ground anything (including a person). With a “standard” grounding electrode of 30 ohms, the voltage on the case will be 220/(677k + 30)*30 = 10 mV (ten millivolts).
If the case is not grounded, then when a person touches it, the voltage will not be much higher: the resistance of the human body is 1 kOhm, and the voltage is 220/678 * 1 = 324 mV. This is very little voltage.
Now about the surge protector. The article says: “If you have N system units included in it, then the current filtered by each filter in the power supply will add up, and the location on the case of each system unit will be smaller.” But the current can only flow, it cannot “be in the housing”. The voltage, which is regulated in the housing, cannot change when connected in parallel.
In fact, we are talking about the fact that if you or someone else accidentally “grounds” by touching the body of the system unit, then it will simultaneously ground all other devices plugged into adjacent sockets. It is then that the currents flowing through their grounding connections add up, which leads to the addition of the voltages applied to the person.
And now to the essence of the issue.
My digital input burned out. Sometimes it is useful to have RCA terminals that are metal rather than plastic, or thermally insulated. Then you will be grounded when you take them with your fingers. Otherwise, potential can be transferred from an ungrounded device to a grounded one.
Many devices use high-frequency noise protection at the input. Those same small coils and nanofarad capacitors. So they reset half of the power potential to zero. 220/2. Yes, there is minimal current, but there is voltage. A spark appears when you connect it to another device.
And what happens when you plug the plug of an ungrounded device into a grounded one? Right. The ungrounded runs to be grounded by others.
If the terminals bodies are metal, then you temporarily become ground for the ungrounded. And you save the device. And if the terminals are plastic, then the second device receives an impulse.
There's an even funnier story there. Even if everything is grounded, it must be plugged into one socket.
Since the grounding resistance will be the same (the length of the ground route is the same).
And if you turn it on from different sockets, then the difference in the earth path will be the difference of length of this earth. Conventionally, one socket is 10 meters from the shield, and the second is 15 meters (socket on the opposite wall for example).
Now you're getting to the potential difference. Since there will be a potential difference, it will be the resistance of the length of your earth is 5 meters.
You think it's bullsh*t. 1 Meter +- 0.015 Ohm. 5 meters 0.075 Ohm.
If the device consumes 1A@220W, then at this current the voltage difference between them is 0.27V. (+0.13V on case).
And you just plugged it into another socket...
Perhaps I’m try to manipulating the concepts, and someone will say you’re a lousy electrician.
Well, to hell with it, you have potential on your case and it will go to the case of the other, and not into the signal circuit.
There won't be any sh*t.
But there is a nuance... The resistance of the analog input, for example (between the signal and zero is in kOhms. 10-20-40-100 kOhms. Nothing will happen to it. But the digital input is a 75 Ohm standard That is, a thousand times less.
Let me remind you that the operating voltage of the digital input is 0.3-0.6V only
This is a short story about how you can burn out a digital input just by connecting an ungrounded device to a grounded one.
Background (copied).
Having bought new headphones with a microphone, and coming home, I was sad to discover that the microphone was creating extraneous noise. I returned to the store, where we checked on the laptop, there was no extraneous noise from the microphone. When I got home I started looking for the reason. I connected the old ones, the headphones do not make noise. I connected new headphones again, they are noisy. After some time, I accidentally touched the system unit with my foot, and lo and behold, the noise decreased significantly.
So, I came to the conclusion that there is some kind of interference on the system body. I immediately thought about grounding, and I started measuring the voltage of the case relative to the Earth. First, I took the neutral wire and was surprised to find that the potential difference was about 100V. I decided to measure the voltage relative to the heating battery, still ~100V.
I will not go into detail about where the voltage on the housings of refrigerators/washing machines comes from. Let me just say that the reason in 99% of cases is the same as in the system unit case.
You can find a more detailed description and explanation on Google.
In short, the same reason:
There is a filter in the computer power supply that dampens high-frequency interference and throws it to the ground.
Thus, we have 110V going to the ground (if the socket is 220V), but the current is only interference current, which means our current strength will be insignificant.
We get 110 Volts on the body only when it does not ground anything (including a person). With a “standard” grounding electrode of 30 ohms, the voltage on the case will be 220/(677k + 30)*30 = 10 mV (ten millivolts).
If the case is not grounded, then when a person touches it, the voltage will not be much higher: the resistance of the human body is 1 kOhm, and the voltage is 220/678 * 1 = 324 mV. This is very little voltage.
Now about the surge protector. The article says: “If you have N system units included in it, then the current filtered by each filter in the power supply will add up, and the location on the case of each system unit will be smaller.” But the current can only flow, it cannot “be in the housing”. The voltage, which is regulated in the housing, cannot change when connected in parallel.
In fact, we are talking about the fact that if you or someone else accidentally “grounds” by touching the body of the system unit, then it will simultaneously ground all other devices plugged into adjacent sockets. It is then that the currents flowing through their grounding connections add up, which leads to the addition of the voltages applied to the person.
And now to the essence of the issue.
My digital input burned out. Sometimes it is useful to have RCA terminals that are metal rather than plastic, or thermally insulated. Then you will be grounded when you take them with your fingers. Otherwise, potential can be transferred from an ungrounded device to a grounded one.
Many devices use high-frequency noise protection at the input. Those same small coils and nanofarad capacitors. So they reset half of the power potential to zero. 220/2. Yes, there is minimal current, but there is voltage. A spark appears when you connect it to another device.
And what happens when you plug the plug of an ungrounded device into a grounded one? Right. The ungrounded runs to be grounded by others.
If the terminals bodies are metal, then you temporarily become ground for the ungrounded. And you save the device. And if the terminals are plastic, then the second device receives an impulse.
There's an even funnier story there. Even if everything is grounded, it must be plugged into one socket.
Since the grounding resistance will be the same (the length of the ground route is the same).
And if you turn it on from different sockets, then the difference in the earth path will be the difference of length of this earth. Conventionally, one socket is 10 meters from the shield, and the second is 15 meters (socket on the opposite wall for example).
Now you're getting to the potential difference. Since there will be a potential difference, it will be the resistance of the length of your earth is 5 meters.
You think it's bullsh*t. 1 Meter +- 0.015 Ohm. 5 meters 0.075 Ohm.
If the device consumes 1A@220W, then at this current the voltage difference between them is 0.27V. (+0.13V on case).
And you just plugged it into another socket...
Perhaps I’m try to manipulating the concepts, and someone will say you’re a lousy electrician.
Well, to hell with it, you have potential on your case and it will go to the case of the other, and not into the signal circuit.
There won't be any sh*t.
But there is a nuance... The resistance of the analog input, for example (between the signal and zero is in kOhms. 10-20-40-100 kOhms. Nothing will happen to it. But the digital input is a 75 Ohm standard That is, a thousand times less.
Let me remind you that the operating voltage of the digital input is 0.3-0.6V only
This is a short story about how you can burn out a digital input just by connecting an ungrounded device to a grounded one.
Last edited:
I have this problem with my pinecil soldering iron, the charger is ungrounded, so depending on which way I plug it in, I could be disharging 80V from the Y capacitors into my circuit. There is a grounding screw on the iron but not a single thought was given to allowing the iron to be grounded through the USB-C cable it already has.
a] be careful not to mingle Safety Ground/Protective Earth with Planet Earth. While there is continuity between them they are not the same thing.
b] when measuring AC voltages with a modern high impedance DMM (Digital Multi-Meter). You may get a high voltage reading that electricians call 'phantom voltage'. The high voltage disappears when measured with a low impedance meter or if a small load is attached.
b] when measuring AC voltages with a modern high impedance DMM (Digital Multi-Meter). You may get a high voltage reading that electricians call 'phantom voltage'. The high voltage disappears when measured with a low impedance meter or if a small load is attached.