re: sizing connectors for dc
8 oct 2003
bill kaszeta / photovoltaic resources wrote:
> my question has to do with the cabling. i'm using a two foot gauge
> 2 wire. it's rated 130 amp. i'll have a 100 amp fuse in parallel with
> a 30 amp fuse connected to the positive terminal.
this might work better with two 60 amp fuses. or two of those 2' wires in
parallel, or smaller 2' wires, with individual fuses. table 301.6 of the 2002
nec says gauge 2 wires with the right insulation can take 130 a at 90 c in
30 c air. the itt ref data for radio engineers book says fusing current is
kd^1.5, where d is the wire diameter in inches... 12 ga copper wire (0.0808")
fuses at 235 amps, so k = 10,232, and 2 ga should melt at 10,232(0.2576)^1.5
= 1,338 amps. i wouldn't try this at home... 16 ga copper melts at 117 amps...
16 ga aluminum melts at 87 amps.
>...i'd thought that two fuses from the same company would be of the same
>construction and therefore the resistance should be proportional to the
>size of the conductor.
sounds reasonable, with some tolerance... 2' of 2 awg wire has 0.0004 ohms...
>i can't get a good reading from any of my three multimeters. it looks
>like the 100 amp fuse is .2 ohms and the 30 amp is .3
both sound high. with similar construction, if the 100 a fuse melts at
100^2x0.2 = 2 kw, you'd think the 30 a fuse would be 2kw/30^2 = 2.2 ohms.
>my more sensitive meter actually shows the resistance drops when i try
>to measure the bigger fuse.
interesting, and backwards, compared to a light bulb, and counterproductive
in this application, if so. copper wires share current better, with a tempco
of + (plus) 0.0039/c. small warm copper wires in series with the fuses might
improve their current-sharing behavior. we might use a single small fuse and
a small warm copper wire shunt guaranteed to melt when the fuse blows. lots
of older french houses have lead "fuse wires" between binding posts in a fuse
box. you can buy a card with some 5 and 10 and 30 amp fuse wire in a hardware
store or a supermarket. france used to run on 120 v, until they doubled that
to improve capacity. quel surprise.
>i was taught in basic electicity classes (many years ago) that if a
>wire can carry 10 amps then two wires of the same guage in parallel
>will carry 20 amps. does that break down too, or is that still valid
>(assuming same guage and length)?
sounds good to me, although the nec derates multiple wires in a conduit,
where heat and temperature can build up.
>extracted from the 1993 version of the national electrical code (nec)...
>1. only conductors of 1/0 gauge or larger can be paralled and the lengths of
>the parallel runs must be the same.
>2. overcurrent protection devices such as fuses or circuit breakers are not
>to be paralled.
section 240.8 of the 2002 nec says:
fuses and circuit breakers shall be permitted to be connnected in
parallel where they are factory assembled in parallel and listed as
a unit. individual fuses, circuit breakers, or combinations thereof
shall not otherwise be connected in parallel.
the handbook commentary continues:
section 240.8 prohibits the use of fuses or circuit breakers in parallel
unless they are factory assembled in parallel and listed as a unit.
section 404.17 prohibits the use of fuses in parallel in fused switches.
it is not the intent of 240.8 to restore the use of standard fuses in
parallel in disconnect switches. however, 240.8 recognizes parallel low-
voltage circuit breakers of fuses and parallel high-voltage circuit
breakers or fuses if they are tested and factory assembled in parallel
and listed as a unit.
high-voltage fuses have long been recognized in parallel when they are
assembled in an identified common mounting.
section 404.17 says:
a fused switch shall not have fuses in parallel except as permitted in 240.8.
so parallel fuses are not exactly prohibited, just required to be assembled in
a "factory" ("factory" (unlike "garage") isn't defined in the nec) and required
to be listed.
who knows why the nec authors required this? if one fuse is blown, an
electrician may assume a circuit is dead when it isn't, but if so, why
are listed assemblies acceptable? maybe they have a warning label.
if the assembly is badly designed, and one fuse always blows first, at
a lower current than the combined rating, that seems to be more of an
inconvenience or inefficiency (these matters are not usually addressed
by the nec) than a matter of safety.
perhaps the nec "parallel-fuse" standards committee was composed of listed
parallel-fuse-assembly manufacturers with a desire to ensure a monopoly :-)
some standards have more to do with politics than physics.