
re: pv to house = 500' ... is it doable?
28 nov 2002
bill kaszeta / photovoltaic resources wrote:
>there are 3 basic designs for which to estimate the costs...
>1. inverter with remote array (48vdc recommended nominal voltage),
>2. 120 vac system with remote batteries and inverter,
>3. 120/240 vac system (depends on size of loads)with remote batteries
> and inverter (onehalf the current if loads can be balanced).
one might add:
4. remote gridtied 240 v inverter with no batteries.
>minimum wire size is based on code requirements of ampacity
>and allowable voltage drop (5% per usa nec 31015).
section 31015 of the 2002 nec has 3 pages of text, but none of them
seem to mention 5%. section 31015(a)(1), fpn no. 1 says
ampacities provided by this section do not take voltage drop into
consideration. see 210.19(a), fpn no. 4, for branch circuits and
215.2(d), fpn no. 2 for feeders.
210.19(a)(1), fpn no. 4 says
conductors for branch circuits as defined in article 100, sized to
prevent a voltage drop exceeding 3 percent at the farthest outlet...
where the maximium total voltage drop on both feeders and branch
circuits to the farthest outlet does not exceed 5 percent, provide
reasonable efficiency of operation.
there is no section 215.2(d)... 215.2(a)(4), fpn no. 2 says
conductors for feeders circuits as defined in article 100, sized to
prevent a voltage drop exceeding 3 percent at the farthest outlet...
where the maximium total voltage drop on both feeders and branch
circuits to the farthest outlet does not exceed 5 percent, provide
reasonable efficiency of operation.
it looks like these are just recommendations vs requirements.
ampacity is a requirement, but section 310.15(c) provides an alternative
formula for use under engineering supervision:
i = sqrt((tc(ta+dtd)/(rdc(1+yc)rca), where
tc = conductor temp (c)
ta = ambient temp (c)
dtd = dielectric loss temp rise
rdc = dc resistance of conductor at temp tc (in ohm/cm?)
yc = ac resistance resulting from skin effect and proximity effect
rca = effective thermal resistance from conductor to ambient (ccm/w)
ignoring the ac bits, it looks like this is just ohm's law for heatflow:
dt = prca, where dt is the conductorambient temp diff and p is the heat
power and rca is the conductorambient thermal resistance. p = i^2rdc,
so dt = i^2rdcrca, and i = sqrt((tcta)/(rdcrca))...
i wonder how this formula works. pa soil is about 12 c, and table 310.16(a)
says uf12/2 can carry 1.08x25 = 27 a max at 25 c. assuming tc = 60 c and
1.93 ohms/1000', ie rdc = 0.000063. with ta = 12 and dtd=yc=0 and rca = 60
(from nec appendix b), i = sqrt((6012)/(0.000063x60)) = 112 a, half the
itt book's barewire fusing current of 235 a :) but as bill says, this is
probably uneconomical. well, it might be, with a solar rebate of $5 per
panel peak watt and no wire rebate and solar panels that only cost $3.75.
nick

