Fire is a very complex phenomeon. It is complex
because ignition requires
* electrical power dissipation,
* component fault temperature exceeding the
fuel material ignition temperarture,
* electrical and thermal energy sufficient
to raise the fuel to ignition temperature,
* adequate thermal energy transfer rate from
the thermal source to the fuel material to
raise the fuel to ignition temperature,
* production of thermal energy for a duration
sufficient to bring the fuel material to
ignition temperature.
Despite the standards, there is no "magic" number
of watts that will cause (or NOT cause) ignition
of a material. However, the watts specified in
standards is quite reasonable and provides a
reasonable first-pass at protection against
ignition. Not perfect, but reasonable.
Potential for fire even when the product complies
with the standard? Be careful here. The original
incident, while generating a localized fire, did
not spread from the equipment, or even within the
equipment. It did not cause a "fire" as contemplated
by the standards. It did cause a local "fire" that
self-extinguished. It seems to me that the standard
indeed provided the constructional requirements to
prevent spread of fire.
The IEC 950 requirement is that all components --
even those in a Limited Power circuit -- be mounted
on V-1 or better material. This was the case in
this incident. The capacitor flamed, charred the
board, but did not spread because the board was
flame-retardant material.
This MIGHT not have been the case if the capacitor
was in an unlimited power circuit!
Very small fires such as a chip capacitor fire have
a very small amount of fuel. The world around them,
including the PWB, provides massive heat-sinking.
There's not much opportunity to transfer thermal
energy to another part if the fuel is highly limited
as in a chip capacitor (and the board is flame-
retardant).
As I have said before, fire safeguards are comprised
just as electric shock safeguards. First and fore-
most, prevent ignition by controlling fault-condition
power dissipation. Second, provide a mitigating
safeguard in the event flame actually does occur.
For this incident, the mitigating safeguard was the
flame-retardant PWB.
So, low-power circuits largely do not require a fire
enclosure (e.g., control of top, side, and bottom
openings) because the fire, if any, should be very
small in the first place, and is mitigated by mounting
components on flame-retardant materials.
I concur that we should never rely on the standards
to tell us that our construction is safe. Consider
this:
How would you make a product safe if there were
no safety standards?
Safety professionals MUST know the nature of hazards
and how they manifest themselves to cause an injury.
Best regards,
Rich
ps: Despite all the hypotheses posted here, we still
don't know the cause of the fire in this incident.
Good scientific methodology requires that the
incident be reproduced in the lab before any
conclusion can be made about this incident.
pps: Participate in safety standards committees? Yes,
but we need engineers who can offer tested and
proven safeguards, and have the backing of their
respective organizations to perform safety research,
not just BOGSAT.
---------
This message is coming from the emc-pstc discussion list.
To cancel your subscription, send mail to [email protected]
with the single line: "unsubscribe emc-pstc" (without the
quotes). For help, send mail to [email protected],
[email protected], [email protected], or
[email protected] (the list administrators).
