0

I'm measuring voltage on an audio amplifier mono output. Amplifier is powered using a stabilized power supply. I connect oscilloscope ground probe to one of the two output wire and a probe to another. Before I had no issues when measuring voltage.

Now, when I first connect ground clip, I get a spark, short circuit indication on power supply, so I disconnect it.

Why do I get this spark?

rem
  • 251
  • 3
  • 13
  • Oscilloscopes often--usually, even--have all their ground probes connected to mains neutral. If you want to avoid sparks, isolate your power supply from mains. (do not isolate the oscilloscope. This is a bad idea for several reasons.) – Hearth May 19 '17 at 15:56
  • 1
    It's common in these situations to refer to Dave @ https://m.youtube.com/watch?v=xaELqAo4kkQ – Paul Uszak May 19 '17 at 20:13
  • @Felthry : "Oscilloscopes often--usually, even--have all their ground probes connected to mains neutral." Don't you mean that they are connected to mains *earth*? For example, [this Tektronix document](http://info.tek.com/rs/tektronix/images/3AW_19134_2_MR_Letter.pdf) explains: 'Most traditional oscilloscopes have the "signal reference" terminal connected to the protective grounding system, commonly referred to as "earth" ground or just "ground".' – SamGibson May 19 '17 at 22:13
  • @SamGibson You're correct, in fact! I was thinking of mains earth and wrote neutral instead, for some reason. Too late to edit, unfortunately, but the point gets across I think. Mains neutral is usually pretty close to mains earth and would cause similar problems anyway. – Hearth May 19 '17 at 22:15

3 Answers3

2

The loudspeaker outputs of audio amplifiers are often in fact both live connections to the output of a H-bridge. Please see e.g.

http://m.eet.com/media/1063664/fig3.JPG

That's a class D mono speaker amplifier output stage. Do you see neither of the loudspeaker pins is connected to ground?

Janka
  • 13,636
  • 1
  • 19
  • 33
2

schematic

simulate this circuit – Schematic created using CircuitLab

  • ground on the right is the scope probe ground clip.
Tony Stewart EE75
  • 1
  • 3
  • 54
  • 182
-1

Thanks for your help. So far, nothing is dead!

Links from :

were very useful

To sum up : always check if your oscillo is isolated from mains. If it is not isolated and you need to perform floating measures refer this link

For my case, ampli out are not connected to pins. My oscillo GND is connected to Ground earth => spark and shortcircuit!

To make floating measurement I will use A minus B technique. For this technique, Oscilloscope ground must be connected to Device Under Test ground. In this case, DUT ground must be isolated from mains ground i.e. not powered from a non isolated power supply. If not, I will have a ground loop giving some imprecise measure and potentially causing damages. In this this link you will find in glossary some concise info about ground loops.

In my case, I'm using lab power supply (isolated from mains), so connecting DUT ground to oscillo GND (mains earth) is fine. I'm just making my DUT circuit not floating anymore!

rem

rem
  • 251
  • 3
  • 13
  • 4
    DO NOT FLOAT THE SCOPE! That is the absolute wrong thing to do. The document you linked to even tells you NOT TO FLOAT THE SCOPE. The reasoning is quite simple, and is given the linked document, and is explained at some length [here.](https://electronics.stackexchange.com/questions/73991/why-do-we-need-an-isolation-transformer-to-connect-an-oscilloscope) – JRE May 22 '17 at 18:02
  • I never talked about floating the scope, it's just about doing floating measures. Maybe I was not clear when saying " always check if your oscillo is isolated from mains", it does not mean "isolate your scope if you can" but just , "check that oscillo GND is Mains earth" – rem May 22 '17 at 20:11
  • A scope should not be isolated from the mains. – JRE May 22 '17 at 20:22