On 04/08/2016 09:37, GB wrote:
> On 04-Aug-16 7:42 AM, AndyW wrote:
>>
>> The bottom line is this.... if you go into a sub station with a plank of
>> wood then Ohm's law means nothing, you are nothing more than a big
>> reduction in dielectric strength carrying another reduction in
>> dielectric strength. Arc flash can occur from about 440V upwards and
>> some substations carry 11Kv up to about half a mega-volt. My 11KV
>> systems have a prohibited arc-flash range of a metre and that is inside
>> in a warm, dry and controlled environment. An outside substation
>> carrying over 11KV in damp conditions with rain, accumulated dirt,
>> clouds of insects (there is an interesting case of a substation near a
>> river arc-flash frying a cloud of mayfies) and wet ground this can
>> extend to several metres.
>
> I understand exactly the mechanism you are describing, but I am really
> surprised at the numbers you are giving.
>
>
https://en.wikipedia.org/wiki/Dielectric_strength
>
> This gives the dielectric strength of air as 3MV/m. This means (if I
> have understood correctly) that you wouldn't expect 11Kv to arc over
> until the gap got down to around 3.67mm.
>
> That's of course in ideal conditions - dry air, no dust, etc. I can
> certainly understand why you would create a safety zone of 1m, but if my
> 3.67mm calculation is correct 1m seems like quite a big safety margin.
You are almost certainly correct in your calculation but it is a perfect
static calculation in perfect circumstances and ignores surges,
atmospheric changes, effects of poor ships husbandry* etc that can cause
a flash.
* I was involved in the aftermath of one of these after an HVAC filter
was removed from a system and allowed crud and dust to build up causing
a flash in an air circuit breaker.
For a fixed conductor in good conditions (eg a cable or terminal) then
you *should* be safe at an inch or so.
For a moving conductor (eg a contactor or circuit breaker) when it moves
you can (but should not due to suppression) get an arc produced that
will usually arc between the terminals until it reaches a critical gap
and then it extinguishes.
If there is a good ground nearby or even as much as a long human hair
floating past that could provide a better path then air it can jump to
that. There are cases of flashes bursting out of cabinets to hit people
nearby.
In addition your static calculation does not take account of the surges
experienced when client equipment shuts down and the generators are
still dumping energy into the system. These surges are usually dumped to
massive resistor banks but there is a delay especially if the offload is
an emergency or failure and not planned and carefully controlled.
Also an arc flash will produce a blast of heat reaching into the tens
of thousands of degrees Celsius that produces an expanding plasma cloud
and lumps of burning metal even if the suppressions systems cut in. PPE
is required just to enter the compartment due to the risk of injury or
death due to secondary effects of the flash. You could actually survive
the electricity and be killed by the gas and metal.
Once the arc is created then it can expand to the arc flash limit which
could be tens of metres away.
> Once the arc occurs, the ionisation would reduce the resistance, and
> hence my initial reaction was surprise at your comment that the arc
> often dies down after a few cycles. I suppose the ionised air heats up
> and rises upwards, creating a longer path?
A lot of HV kit have arc detection and suppression systems that will
prevent it building but I did say "Basically you get hit by lightning,
usually it acts for a few cycles <where the suppression system cuts in>
but you can get a persistent arc like that created by an arc welder;
once the arc has initiated it can self-sustain by incrementally reducing
the arc-path impedance though its own plasma. <if it does not kick in> "
There are multiple cases in the InMarEST proceedings and MCA Guidance
Notes where suppression systems have failed resulting in arcflashes and
resulting death or sever injury.
My main thrust was that going into a substation with a piece of 'dry'
wood to knock away an ex-human charcoal statue is like looking for a gas
leak with a lighter.... you may get just away with it but if it goes
wrong at all it goes wrong big. It is an all or nothing thing really.
James was talking about actually hitting a live conductor which is just
asking for trouble 'dry' wood or not.
James's physics is fine for static ohm's law but what kills people in
these circumstances does not follow Ohm's Law.
I seem to have gone on a bit. Sorry about that.
Andy