1

I was reading an article in a magazine in which the following sentence is written about a circuit that generates a certain waveform on a load R (in which one of its terminals is connected to the earth - not the circuital GND - ):

"Never connect a normal oscilloscope probe to the ends of the load. In fact, the oscilloscope's GND terminal is connected directly to the earth conductor and therefore you could make a direct short circuit between the circuit phase and the earth, with relative risks of explosion, electrocution or fire. You need to use a differential probe."

Can you explain me this on detail? I do not follow this reasoning. Why should the phase "touch" the earth voltage?

To clarify the problem, here there is a picture of the schematic (the load is connected to CH1 or CH2, while the input AC 230V enters x1):

enter image description here

Kinka-Byo
  • 3,340
  • 2
  • 19
  • 58
  • Dave from the EEVBlog made an excellent video explaining about ground and scopes. You should watch it and that might answer your questions. Here's the video: https://www.youtube.com/watch?v=xaELqAo4kkQ&t – Bimpelrekkie Sep 13 '19 at 06:06
  • 1
    For context, can you please add reference to the article itself? I assume the big unlabeled box with "A5/SCL" an Arduino or an Atmel ATMEGA328 microcontroller. Based on the triacs and optocouplers, I guess the idea is a PWM lighting dimmer commanded via I2C? But if jumper J2 is installed, then the "digital ground" will be directly connected to power grid Line -- a safety hazard, if a human can ever come into contact with any part of this circuit or its controls while it is connected to mains. Even jumper J1 could be hazardous, especially if line and neutral were reversed. – MarkU Sep 13 '19 at 06:24
  • Unfortunately I do not have any link, because it is an article on paper. Yes, it is an ATMEGA 328 microcontroller, and the whole circuit is a dimmer. Precisely, when you say that the connection between the digital ground and the power grid line is a hazard, which are the dangerous points which may result short circuited? – Kinka-Byo Sep 13 '19 at 19:51
  • As explained in my answer, the circuit must be treated as live. If, for example you have it connected to a programmer and your laptop and switch on power to this circuit you will destroy something. – Transistor Sep 13 '19 at 20:07

1 Answers1

2

The article is cautioning you from connecting the scope probes - and especially the ground clip onto the mains voltage wiring at the bottom left of the schematic. The neutral can have a few volts on it due to voltage drop caused by current in other circuits in the building and, if the 'scope's ground is connected to mains earth, then clipping the scope ground onto neutral may cause mains return current to flow through the scope. There is also the danger that you clip the probe ground lead onto the live which would be much worse.

A differential probe solves the problem because neither of the two probe wires are grounded.

Notice that circuit, despite having an isolation transformer, PT1, has J1 and J2 to directly connect either to the mains. The control logic must therefore be treated as live and the same rules apply to connecting the scope probes to that also.

Related:

Transistor
  • 168,990
  • 12
  • 186
  • 385
  • Thank you for the answer. So the problem is not the connection between the phase and the earth, but between the neutral and the earth, right? – Kinka-Byo Sep 13 '19 at 19:52
  • 1
    Either will be a problem. A connection from live to earth through your 'scope could be fatal. A connection from neutral to earth through your 'scope is less likely to harm you but could destroy the 'scope by burning up the ground traces inside. An insulated instrument such as a Fluke Scopemeter would be OK as it has no mains earth connection, the case is fully isolated and the sockets are finger-proof. – Transistor Sep 13 '19 at 20:05