No such thing as Safe Isolation. You are only proving dead at the time of testing, a few seconds later the installation or part of it could become live.

Best thing is do not touch anything conductive even if it is green/yellow covered.

JP

PEN faults, solar panels, batteries poor drawings - hands up if you have ever pulled the wrong fuse on a big site? There may be another incomer for something un-marked.

It is far from impossible for things to be live or worse for them to become live mid-job after all.

And then beyond that, the cases where it is meant to be. Live working may be OK, so long as you can justify it and have the tools and know how to do it well. Of course if not, then  you first slip-up maybe your last.

Mike.

Yes. There is more than one (mains) supply. I recall such an event being mentioned during my training, but not in a domestic installation.

But ...

There could be.

I am in the process of installing a new DB at home. The new one sits adjacent to the old one, but the supplies are different so there is scope for carelessness.

OK. As Mike says, a PEN fault. So next door is still connected and we share the same gas (or water) supply so neutral and earth go up to somewhere near line voltage. Hang on though - I could be touching an ECP without working on the installation.

But then again, both supplies are in plastic and we have no PV on the roof and no EV plugged in on the drive.

Hm?

Chris Pearson said:

It should not be possible for the installation to become live if locked off and the key is in one's pocket

Induced or capacitive coupling from another source (perhaps lighting ... but not always) has been known to take human life !

It's "energization" ...

Chris Pearson said:

There is more than one (mains) supply. I recall such an event being mentioned during my training, but not in a domestic installation.

But ...

There could be.

Island mode ... prosumers' electrical installations ... and other means of energization!

John Peckham said:

Can you think of way and installation could become live with the main switch off and the supply fuses withdrawn?

I can. Even without solar PV and battery storage ...

YES ... but ... that's only one option. And it's not just "PEN" problems. There are TWO other issues to consider here:

1. What if I said that "diverted earth currents" are not limited to PME, but possible in ANY system with main protective bonding?

2. The harmful current n may not come from a protective conductor, but a live conductor, or even a telecomms/auxiliary conductor?

Worth remembering that we don't need Amperes to kill, just mA !

Are we talking electromagnetic induction and conductive coupling when it comes to telecoms ?

AMK said:

Are we talking electromagnetic induction and conductive coupling when it comes to telecoms ?

Yes, but not limited to telecomms, inductive and capacitive coupling can also occur with power circuits ... also in three-phase networks (or installations) because isolators don't always disconnect the Neutral (esp. TN systems), if we isolate, then there's broken neutral in the supply to the board we're working on, could mean a nasty voltage on the Neutral !!!

There are LOADS of ways we can be side-swiped by "accidental, unintended, or unforeseen" voltages and currents.

Chris Pearson said:

It should not be possible for the installation to become live if locked off and the key is in one's pocket (electricians' padlocks are sold with only one key), or perhaps the main fuse(s) is.

What do you mean by "should" ... and I'm afraid, sadly, it's an incorrect assumption that a given installation can't become unexpectedly energized, if "properly isolated" and locked off. It should,. however, be designed so that the risk of that happening is kept to a minimum of (hopefully) very unlikely events.

Chris Pearson said:

It should not be possible for the installation to become live if locked off and the key is in one's pocket (electricians' padlocks are sold with only one key), or perhaps the main fuse(s) is.

What do you mean by "should" ... and I'm afraid, sadly, it's an incorrect assumption that a given installation can't become unexpectedly energized, if "properly isolated" and locked off. It should,. however, be designed so that the risk of that happening is kept to a minimum of (hopefully) very unlikely events.