Date: Thu, 25 Oct 2012 10:11:39 -0400
Reply-To: David Beierl <dbeierl@ATTGLOBAL.NET>
Sender: Vanagon Mailing List <vanagon@gerry.vanagon.com>
From: David Beierl <dbeierl@ATTGLOBAL.NET>
Subject: Re: Ampacity was Re: Headlight Upgrade Double Check (fuses)
In-Reply-To: <1CD89E84-7AEF-42FA-A295-88DA2507C9CE@gmail.com>
Content-Type: text/plain; charset="us-ascii"; format=flowed
At 11:04 PM 10/24/2012, turbowesty wrote:
>May I ask, where does the gauge # come from and how is it defined?
>The relationship between gauge and strands is quite linear
>(inversely? I'm not clear on what the context for 23/25/etc is).
The American Wire Gauge is similar to the Browne & Sharpe steel plate
gauge which I believe was defined in terms of thickness of a
dimensioned sheet of steel having particular weights. This is also
analogous to shotgun gauge, defined as diameter of a lead ball that
runs so many to the pound. Because of how they're defined gauge
numbers typically run contrary to intuition, i.e. big number -->
small dimension.
Modern AWG is defined at two gauges and interpolated/extrapolated
outside of them; see http://library.bldrdoc.gov/docs/nbshb100.pdf .
Stranding is heavily constrained by geometry because the strands have
to pack solidly together in circular form. Within those constraints
you get to choose the number of strands and sub-strands based on
requirements. Small-diameter wire rope, for example, is commonly
available in 1x19 (stiff) and 7x19 (flexible, because each of the 19
strands is itself composed of seven sub-strands twisted in opposite
handedness). In general stiff wire is cheaper to make than flexible,
so in electric wiring many gauges have several commercially available
stranding options. Test lead wire for example is finely stranded so
as to be as close to limp as possible. These stranding differences
lead to minor differences in conductor area in a particular gauge,
but such differences are really only of interest if you're using wire
as a resistor and need to calculate precise length of a wire to
obtain a desired resistance. Wire tables give resistance figures in
both resistance per foot/kilofoot/kilometer and per
pound/kilogram. See reference above for details of wire construction
and stranding in commercial practice and in general a lot more than
you want to know about how it all works.
That reference is probably authoritative in the US subject always to
physics and typography. There are many much more convenient tables
available on the web but beware of typographic or calculating errors.
In your van (RVC) you're interested in ampacity ratings to prevent
overheating; and resistance per unit length so you can determine
voltage drop in a specified supply circuit and keep it within
acceptable limits. The longer your circuit the more likely that the
second factor will govern. My Bosch book is as I've said mislain,
but I'll go out on a limb and say that for most 12v automotive
circuits a total wiring drop including return of a quarter to a half
volt is probably acceptable. Starter circuits expect a somewhat higher drop.
With regard to headlights, particularly halogen ones which are
universal now - halogen lamps by design burn very hot for two
reasons: efficiency, because high (hundreds of psi cold) internal
pressure helps keep tungsten on the filament where it belongs; and
because the halogen reaction that keeps vaporized tungsten from
depositing on the lamp envelope** needs high temps to run. But this
means they're intolerant of over-voltages.* Lamps for automotive use
are no doubt made and rated in anticipation of industry-practice
wiring drops, so if you get too carried away with big wires you may
find yourself with very bright lights that don't last long. This is
also an issue for people who are charging their batteries with a
smart regulator, or who simply have the voltage turned up high enough
for a standard regulator to fully charge (and eventually boil dry) a battery.
*In standard 750-hour household tungsten lamps operating at/near
rated voltage, the rule of thumb IIRC is that with increasing voltage
brightness goes up with the THIRD power of the voltage but lifetime
comes down with the FIFTH power. This give a very unforgiving
overvoltage vs lifetime curve (rated life of a 3200K photoflood bulb
is one hour, did you know?). The "run forever" tungsten lamps that
used to be hawked in mail-order catalogs and such are simply 750-hour
lamps with a rated voltage of 130v instead of 120v, so they give
crappy light for the power consumed but last a very long
time. Halogens are even less forgiving because their construction
lets them run normally at temperatures that would very quickly
blacken and burn out a conventional lamp.
** In ordinary lamps vaporized tungsten plates out on the cool lamp
envelope, sometimes to an astonishing degree. In instrument pilot
lights (#47 lamp for example) it's not uncommon to pull out a working
lamp and find that you are looking at a silver-black mirror and can't
see through when you hold it up to the window. One reason for the
pear shape of household lamps is to give lots of area so the film
will be still be pretty transparent when the filament burns
out. Another reason in the case of frosted lamps is to make the lamp
an area rather than pinpoint light source.
Yrs,
d
