Bare busbar

I'd not be very surprised you cannot get a good contact to a bus bar covered with a layer of oxides and general grot built  up over the years -  it is however quite impressive if it reads totally dead. In terms of the live/dead test it is perhaps an argument for checking with a non-contact 'volt-stick' type indicator, although that  is not without potential to mis-read also.

The microscopic version of the same problem allows folk to get voltage drop and Zs/PSSC readings that do not reconcile, this is just a rather spectacular example.

The saving grace is that the adventitious insulation of the bars in this way may actually make it safer, if a spanner is dropped in the works it may act as some kind of primitive current limit. Mind you I'd prefer a design with some sort of cover, it is funny how your standard of what is reasonable to work in / on changes over the years.

Mike
PS a thought

The comments about flat caps may be a bit tongue in cheek, but there is a serious point here that re-inserting a fuse against a fault or your tools or wrist watch falling into the works may not end well in this sort of design, and a singed shirt sleeve or glove is much preferable to a burn that may need a skin graft.

As many newer folk will not have encountered a traditional skeleton board with lots of bare live metal, it is worth remembering that the penalty for a lapse in concentration is that bit higher than with a modern one, so dress accordingly, casual beachwear is not really appropriate. There is a very real arc flash hazard here, as the not-quite-zero resistance fault that you get as the oxides are blasted away with a blinding flash like striking a welder, is the kind  that creates the most volume of mobile hot metal.

PPS edit. In a resistive case, the absolute worst case arc is where the fault resistance is equal to the loop impedance of the supply - so an energy of half the PSSC of the fault loop, times half the pre-fault supply voltage, times the upstream ADS disconnection time is available to melt and throw metal about, and the volume of that metal can be estimated adiabatically . With older boards like this the risk is heightened as the upstream ADS is often also quite pedestrian by modern standards.

I'd not be very surprised you cannot get a good contact to a bus bar covered with a layer of oxides and general grot built  up over the years -  it is however quite impressive if it reads totally dead. In terms of the live/dead test it is perhaps an argument for checking with a non-contact 'volt-stick' type indicator, although that  is not without potential to mis-read also.

The microscopic version of the same problem allows folk to get voltage drop and Zs/PSSC readings that do not reconcile, this is just a rather spectacular example.

The saving grace is that the adventitious insulation of the bars in this way may actually make it safer, if a spanner is dropped in the works it may act as some kind of primitive current limit. Mind you I'd prefer a design with some sort of cover, it is funny how your standard of what is reasonable to work in / on changes over the years.

Mike
PS a thought

The comments about flat caps may be a bit tongue in cheek, but there is a serious point here that re-inserting a fuse against a fault or your tools or wrist watch falling into the works may not end well in this sort of design, and a singed shirt sleeve or glove is much preferable to a burn that may need a skin graft.

As many newer folk will not have encountered a traditional skeleton board with lots of bare live metal, it is worth remembering that the penalty for a lapse in concentration is that bit higher than with a modern one, so dress accordingly, casual beachwear is not really appropriate. There is a very real arc flash hazard here, as the not-quite-zero resistance fault that you get as the oxides are blasted away with a blinding flash like striking a welder, is the kind  that creates the most volume of mobile hot metal.

PPS edit. In a resistive case, the absolute worst case arc is where the fault resistance is equal to the loop impedance of the supply - so an energy of half the PSSC of the fault loop, times half the pre-fault supply voltage, times the upstream ADS disconnection time is available to melt and throw metal about, and the volume of that metal can be estimated adiabatically . With older boards like this the risk is heightened as the upstream ADS is often also quite pedestrian by modern standards.

it was a bit tongue in cheek because it sounds utterly absurd now but the guy showing me, on 100a BS3036 boards was deadly serious.

We used a cotton rag, insert it in the bottom half, then push smartly up while looking away

This was only 20 years ago. It is still advice I take to this day, with MCCBs always but I do have a set of insulated gloves that are too cheap not to buy.