The Distribution Lightning Program (DLP), created for FPL, was developed to study the effects of both direct and induced lightning strokes on overhead distribution line designs for the purpose of predicting the performance of shield wires, arresters and grounding using a digital software program. The advantage of this program, over most existing programs involving traveling wave analysis, is its easy-to-use format that is menu driven with only a few commands. Engineers with little or no background in the area have produced expert analysis with a minimum of outside help. Field studies by FPL have confirmed the prediction of these digital simulations. The DLP program allows the user to create, analyze and store characteristics of sections of feeders, which include the feeder geometry, pole BIL, pole arrester location and rating, pole grounds, shield wires and neutral wires (Fig. 1). Sections of these basic feeders can then be assembled to form a complete feeder system for the purpose of making a whole system analysis. The flexibility of the program allows the characteristics of each pole to be individualized in terms of such important items as grounding, arrester spacing and hardware.

Grounding Grounding may or may not be important, depending on the type of line protection used. For example, Fig. 2 shows the effect of grounding when a shield wire is used. As can be seen, even for this wood crossarm example, where the BIL is high, low values of resistance are mandatory, illustrating that utilities with poor soil conditions that result in poor grounds, would not be helped much by a shield-wire approach. Armless construction, with even lower BIL, would be considerably worse. On the other hand, ground resistance has little effect on line protection using arresters on all three phases, as can be seen in Fig. 3, which shows that there is little, if any, difference between having grounds of 25 ohms or 250 ohms.

Arrester Spacing Because of the high rates-of-rise of lightning current, the voltage crest on the overhead line builds up very rapidly. Although the conduction of the arrester creates a negative component that reduces this buildup, if the arrester is too far from the stroke location, the arrester may not come into play. The higher the rate-of-rise for current and voltage, the closer the spacings of the arresters must be. For example, if the BIL of the line is about 200 kV, arresters must be spaced at every pole and on every phase.

Hardware Since equipment and support hardware can severely reduce BIL, consideration must be given to guy wires that can be a major factor in reducing a structure's BIL. For mechanical advantage, guys are generally attached as high on the pole as possible and are close to the phase wires. These guys provide a path to ground and when they are high on the pole, the BIL will be reduced. In addition, fuse cutouts represent a prime example of unprotected equipment that can lower the pole BIL, where a fuse cutout has a 95-kV BIL for a 15-kV class system. The BIL can be lowered further when the neutral wire on wood pole lines is close to the phase wire as can the use of concrete and steel structures. Indeed, metal crossarms and metal hardware used on wood structures may have the same effect as an all-metal structure if the hardware is grounded.

Field Tests FP&L's overhead line design to prevent flashovers includes the establishment of a minimum of 300 kV BIL to protect against induced strokes, the standardization of insulation materials by using wood and 35-kV porcelain tie-top insulators rated for 210 kV BIL each and the use of vertical framing to maximize shielding, economics and maintainability in R/W roadside construction. In south Florida, FPL built two model lines (Figs. 4a & b), one mi (1.6 km) long using these framing requirements. One had arresters every 600 ft (1829 m) with 25 ohm grounding, and the other had an overhead ground wire, which was grounded at every pole to 10 ohms, maximum. Both lines had high exposure in the same geographic area about 20 mi (32 km) apart. The ground flash density (GFD) was 25 flashes per sq mi for both lines. The results of these field tests showed that the arrester-protected line displayed 0.228 interruptions per mile per year, while the OHGW line had none. The DLP analysis projected interruptions per mi per year of 0.34 and 0.10, respectively.