Please help me put together a bunch of the disorganized concepts floating in my mind.
I have just learned the (not so hard) way how using a typical floating SMPS power adapter and pursuing
my best intentions in ESD safety rules led me to just the opposite effect of what was intended. You can refer to the thread...
https://www.eevblog.com/forum/reviews/ruining-mosfets-with-a-transistor-tester-and-tweezers/...where the consensus of the much more experienced than me forum members made a conclusion that the Leakage Current from the SMPS power adapter coupled with my grounding wrist strap resulted in permanent damage to a few (cloned Ali) MOSFETs. Not a big a deal but the most unpleasant moment in that accident was that I weren't even aware of the potentially dangerous practice I tried to adopt.
I'll try to formulate my questions so that they make sense which is not easy because of my lack of the fundamental knowledge in EE.
- The three closely related but not directly dependable concepts that I try to put together are: Personal electrical safety; Earth ground referencing of the workbench equipment and DUTs; ESD safety. As I mentioned above due to my lack of experience the first and the last of the three mentioned concepts played a joke on me and I rendered useless couple of the cloned FETs (luckily not the genuine ones).
- Let's skip over the most important of the three concepts, the 'Personal electrical safety' in the context of this discussion. The Earth grounding lines are there for safety and being aware of the hot side of PSUs should be enough to avoid the electrical shock in the first approximation (well, not really but...)
- Regarding the Earth ground referencing of the equipment and DUTs.
Suppose we've got the following setup on the workbench:
1. A breadboard with an STM32 devboard and some peripherals (LEDs, LCD, half a dozen of sensors). This setup is powered from the floating SMPS wall adapter (a 5V "USB" charger). The STM32 board and the breadboard commons are floating and unreferenced to Earth ground.
2. An arbitrary Waveform Generator (AWG), say an amateur favourite UNI-T UTG962E. This AWG is powered by an external floating SMPS power adapter as well. Not Earth ground referenced. The AWG provides signals to the STM32 devboard.
3. A Rigol DS11020-E oscilloscope which is powered by an internal SMPS and the oscilloscope is Earth ground referenced and Earth grounded at the same time.
4. Not permanent (on demand), the PC USB connection to the STM32 USB port when the PC interaction/MCU flashing is required.
5. An electronics enthusiast (the operator) wearing a grounding wrist strap connected to the mains Ground through a 1 MOhm resistor.
In the above configuration we've got 3 units of the equipment (setting the PC as a unit aside) with the 2 of the former having floating circuit commons (the devboard and the AWG) and the third (the oscilloscope) having the Earth ground common both internally and also externally (at the BNC connector).
Assume the operator is aware of the Leakage Current from the SMPS adapter powering the board and (s)he tries tries to be careful not to touch any of the STM32 board pins with the fingers when manipulating the Dupont wires but that's just taking risks, not the remedy.
By the way, there are two transient, potentially safe configurations in this setup when the board's common is effectively Earth ground referenced: 1. When the oscilloscope ground lead is connected to the board's common and/or 2.- when the USB (naturally having the Earth ground provided) is connected to the board's USB port. However one can't rely on these transient connections in the equipment safety context.
Questions: - If one doesn't replace the board's power supply in the form of SMPS AC/DC adapter with the battery power block (which is preferable as it was explained to me in the 'Blowing the MOSFETs' thread mentioned above) but continues to use the SMPS as the power supply for whatever reason, would it make the described setup much safer by connecting the Earth ground directly to the board's common, effectively providing the Earth ground to the secondary of the SMPS?
In other words, simply connecting the secondary of the SMPS to the Earth ground.
What's the Good and the Bad of this modification?
- Leaving the STM32 board's power supply for a moment (having provided two options, the battery or the Earth ground) let's move on to the Waveform Generator, the UNI-T UTG962E.
My first question would be: In the described setup with the UNI-T having floating power supply and unreferenced BNC output terminals are there any potential problems presented to the AWG itself (for instance from the operator's grounding wrist strap) and from the AWG to the rest of the equipment?
I mean - the lack of the Earth ground reference in the AWG stock setup.
There is a YT channel by Tony Albus and I think I've seen his posts on this Forum. In his #189 episode on YT he describes adding what he calls the "JUNTEK PSG 9080 External Reference and Ground Mod" to his AWG.
While in his video Tony shows his procedure of adding the External Earth ground reference to the device he doesn't talk much about the reason - Why he did that. So, why did Tony modified his Juntek (and all of his AWGs for that matter) with the external Earth ground reference?
Does it have to do something with:
- The device safety itself;
- Safety of the rest of the setup to which he connects the now Earth ground referenced AWG?
- Both or anything else? - I.e. not the circuit safety concerns but rather functional requirements, that is - signal equalization and rectification, etc.?
Why does Tony suggest in his YT video that in some situations the Earth ground reference he built into his AWGs is beneficial and in other cases not? He provided a switch at the back of the AWG for Earth ground reference On/Off.
I've already composed a rather lengthy post and I feel I need to stop here and not to go into further details at this point.
To summarise my post, my major question is whether it is necessary, having being maintained the operator ESD safety setup, to Earth ground reference all the equipment in the workbench configuration and what are the downsides of this.
For instance, in his famous 'EEVblog #279 - How NOT To Blow Up Your Oscilloscope!' YT video Dave suggests that having the "Arduino" powered by an SMPS AC/DC adapter is beneficial in the sense that the oscilloscope ground lead can be placed anywhere in the circuit without much worry but once the MCU board is Earth ground referenced one has to be aware of that.
Well - it's not a problem provided the operator watches closely and understands where (s)he connects the probe BUT - Dave said nothing about the inherent SMPS secondary Leakage current and he wasn't actually supposed to mention it in the context of that great video. It's just another question and another potential problem that needs to be addressed separately.
Please feel free to share your thoughts on the subject. I'll be away from my PC for the next five days (old school, sorry) but I'm looking forward to your replies that hopefully will help me demystify this really confusing bunch of the ideas and concepts.
To Earth ground reference or NOT to Earth ground reference? That's the question. And if YES - how far one wants to go?