Existential Angst wrote:

"David Nebenzahl" wrote in message
.com...
On 11/6/2009 11:08 PM Existential Angst spake thus:

If you have 20 2-wire bx cables stopping, say, 10 feet short of a 20
circuit panel, then, it seems to me, you can de facto *make* 10 edison
circuits up to the splice points, and therefore need only 10 neutrals
going between the panel and those 20 cables.

Thus, the edison circuit business is dependent on the actual *geometry of
the wiring*, ie, "partial length" edison circuits are possible.

Thus, your seemingly restrictive "only circumstance" for two hot wires
sharing one neutral is in fact a fairly non-limiting requirement in this
splicing circumstance, as ultimately, if a 20 circuit 240 V panel is
filled, you will *always* be able to make 10 edison circuits, for at
least part of the run.

So you do understand the concept of "Edison" circuits, right? That you
can't just grab any two hots, run a neutral with them and call it a day?
Please tell us you understand this, or don't try it.

My first reply to dpk would indicate that I do understand edison circuits.
Your previous post, and this one, indicate that you have a problem
understanding that I understand.


Do you know *why* you can use just one neutral to serve two hots with this
kind of circuit?

If I didn't, would you tell me?

Here's the deal:

If the loads on each service leg are "balanced" (recall that vague
reference?), ie, electrically identical ito impedance, you wouldn't need a
neutral at all. The total voltage applied to the two loads (effectively in
series here) is 240 V, and the voltage drop across each load on a leg is
then 240/2 = 120 V, which is the appliance rating.

But what if one load is a 100 W bulb is on one leg, and a 1,000 W toaster is
on another, and the neutral is lifted.
The voltage drop across the bulb is 10/11 * 240 and that across the toaster
is 1/11 * 240. Thus, the bulb will burn out, but not the toaster.

Thus, the neutral acts as a kind of centertap for unbalanced loads on each
leg, proly best analyzed with Thevenin's theorem -- in wiki, if anyone is
innerested, altho the sample problems there are not very illuminating.
Thevenin's theorem is essentially Ohm's Law on effing steroids.... wow.....

This Edison business explains why, in the presence of a a marginal or
inadequate neutral, adding or switching on another load will actually make
lights *brighten*, because now current from one side of the svc is not being
driven through a high-resistance neutral, and instead flows better through
the now-balanced load on the other svc leg.

This neutral business clearly explains why 240 V appliances are inherently
superior from an electrical loading pov -- both legs are balanced by
definition, and no neutral is required at all.
Plus, at 240 V, there is much less IR drop, and much less I^2R transmission
loss. It's better all the way around.

Europe operates on 220-240 V, but according to one poster here, that is
between one leg and neutral, so they have the same "balancing problem" we
do, just at a higher voltage. The key is to have 240 V *between two hot
legs* -- and, of course, corresponding 240 V appliances.

Having said all this, I'll proly not go Edison-crazy, and just splice all
neutrals directly to the new panel. This way, there is no possibility of
future incompatibilities resulting from unwitting changes.

--
EA



--
Who needs a junta or a dictatorship when you have a Congress
blowing Wall Street, using the media as a condom?

- harvested from Usenet

Perhaps just put a suitable pull box where the lines will reach, and
install a ground buss bar kit and an isolated neutral bus bar kit, and
tie them to the ground and neutral in the panel with appropriately sized
conductors. Since it's a short run, the same gauge as the panel neutral
and ground connections would certainly be sufficient.

When I have done this type of extension for a panel replacement, I have
normally done something like this for the grounds, but not the neutral.