IEE Transactions on Power ApparatusandSystems,Vol. PAS-95, no.

1, JanuaryfFebruary 1976

OPTIMIZED GROUNDING GRID DESIGN USING VARIABLESPACING TECHNIQUE

J. G. Sverak, Member IEEE United Engineers & Constructors Inc.

Philadelphia,Pennsylvania

ABSTRACT

The envelope of e a r t hs u r f a c ep o t e n t i a lc u r v e s i s dis- and K i l i s an i r r e g u l a r i t yc o r r e c t i o nf a c t o rt oa l l o w

t i n c t l y convex f o r grounding mats with many meshes. fornon-uniformityofgroundcurrentflowfrom
Such a condition i s d i f f i c u l t t o a n a l y z e w i t h t h e e s t - d i f f e r e n t p a r t s ofthegrid,determinedempir-
ablishedmethods,whichareprimarilydevisedforcal- i c a l l y as:
culating.thecorner mesh voltage.Forthisreason, a
large,equallyspacedgrid may beoverdesignedtoward
t h e c e n t e r and underdesignedtoward the perimeter. Aim- K i l = 0.65 + 0.172 n ; (3)
i n gt o remedy such problems, thispaperanalyzes and
modifiestheexisting method of IEEE Guide No. 80, t o L i s t h et o t a l e n g t h of buriedconductors,in
allow a r e c u r s i v e p o i n t by point integration of surface meters;
gradientsthroughconsecutive meshes. A computerpro-
gram foroptimizedgriddesign i s described and t h e I i s the ma
x
m m t o t a l rms
iu current,flowingbe-
e f f e c t ofspacinggeometry i s documented . b y computer tween ground g r i d and e a r t h , as a d j u s t e d by t h e
plotted voltage profiles. f i b r i l l a t i o n decrement factor for a possible dc
currentoffset,in amperes;
INTRODUCTION
0 is t h ea v e r a g er e s i s t i v i t y of theground, in
Following i t s p u b l i c a t i o ni n 1961, IEEE Guide No. 801 meter-ohms.
hasprovedtobe a verypopularguideforsubstation
groundingapplications.
Thisequation(1)gives somewhat lower values for grids
The a b i l i t y t o p r o v i d e a simple, yet conceptually solid with more than four meshes, i n comparison
withthe
approach t o groundingsystemanalysis, and even more values obtained from e x p e r i m e n t a l r e s u l t s o r c a l c u l a t e d
importantly, a method forobtaining a reasonablesol- by o t h e r a n a l y t i c a l m e t h 0 d s ? , ~ * 7
u t i o n by p e n c i l and .paper,havebeenundoubtedlythe
most appreciated features. I t s proceduresfordeter- Also, a contradicting design requirement may result
miningthesaferangeofstep,touch and mesh v o l t a g e s from t h e u s e o f t h e e q u a t i o n f o r s t e p v o l t a g e , g i v e n as
are generally accepted by the industry.

However, i f compared with more rigorous methods, an un-

r e s t r a i n e d a p p l i c a t i o n of t h e p r e s e n t IEEE method occa-
s i o n a l l y may p r o d u c ei n c o n s i s t e n tr e s u l t sf o r a large where Ksl i s t h e s t e p p o t e n t i a l c o e f f i c i e n t , c a l c u l a t e d
groundinggrid. i n s i m p l i f i e d form, as

I np a r t i c u l a r ,t h ee q u a t i o nf o rt h e mesh voltageof a
rectangular,equallyspacedgrid,concerningthefirst
mesh from t h e p e r i m e t e r i s :
when a t t e m p t i n g t o s a t i s f y b o t h t h e s t e p and mesh v o l t -
age limits i n a r e c t a n g u l a r a r e a which i s not too wide,
hsh(1) = K
ml Kil ? I / L (1) and i tfh es o irl e s i s t i v i t y i s high. For instance,
when t h e number ofconductors i s i n c r e a s e di no r d e r to
where K
m l i s a surface potential coeffident taking in- lower mesh p o t e n t i a l s , t h e combined e f f e c t of h i g h e r K i
to account the effect of number n,spacing D , and Ks c a n o f f s e t t h e i n f l u e n c e o f t h e i n c r e a s e d l e n g t h
diameter d, and depth of b u r i ahl o, tfh e L i n ( 4 ) , so t h a t an increase, rather than the expected
grid conductors, which a r e m u t u a l l y p a r a l l e l , simultaneous decrease of calculated step voltageasults.
c a l c u l a t e d as:
For c l a r i t y , l e t i t be notedthat:

1. Touch v o l t a n e i s the Dotential difference

I
between
t h e ground e l e c t r o d e p o t e n t i a l rise and t h e earth
s u r f a c ep o t e n t i a l a t t h ep o i n t where a man i s
standing on his feet, while simultaneously having
h i s hands i nc o n t a c tw i t h a grounded s t r u c t u r e ,
during a f a u l t ;

2. Mesh v o l t a g e i s the worst possible value of touch

v o l t a g e t o be found w i t h i n a mesh of a grounding
gridu , s u a l l yi fs t a n d i n g a t , ornearthecen-
t e r of the mesh;

Paper F 75 526-4, recommended and approved by the IEEE Substations Corn- 3. Stepvoltage i s t h e d i f f e r e n c e i n s u r f a c e p o t e n -
mittee of the IEEE Power Engineering Society for presentation at the IEEE PES t i a l s of two p o i n t s a t one meterdistance,ex-
Summer Meeting, San Francisco, Calif., July 20-25, 1975. Manwript submitted perienced by a man b r i d g i n gt h i sd i s t a n c ew i t h
February 5,1975; made available for printingMay 5,1975. hisfeet,withoutcontactinganythingeise.

362
-
SCOPE equidistantpositionofsuch a p l a n ew i t hr e s p e c tt o
any o t h e r real c o n d u c t o r so rt h e i rimages. Sincehhis
The i n t e n t o f t h i s p a p e r i s t o shopr that the p r i n c i p a l c o n s t i t u t e s a geometricindependenceforeachconductoq
proceduresof IEE No. 80 can beeasilyextended and t h e CJt term can be set equal t o zero, and t h er i g h t
improved t o compare f a v o r a b l yw i t ht h e more accurate, s i d eo f (6) i s p e r f e c t l yv a l i d . Fig. la.
b u t a l s o more e l a b o r a t e methods, based on a non-uniform
current density principle.

Therefore,the well-known fundamentalequationsofthe A' B'

uniform current density method and t h e r o l e o f t h e ir-
r e g u l a r i tcyo r r e c t i ofna c t o r are b r i e f ldyi s c u s s e d Dab,h.d j h FIG. l ab below
left,
first.

Subsequently, a fornula i s developed t o allow a recur-

sive, p o i n t by p o i n ti n t e g r a t i o no fs u r f a c eg r a d i e n t s
through a l l parallel grid contbctors, for'boththe
equal and unequalspacings.
The r e l a t i o n between the envelope of surface touch pot-
e n t i a lc u r v e s and spacinggeometry i s demonstrated by
computer p l o t t e d p o t e n t i a l p r o f i l e s o f g r o u n d i n g elec-
trodes,usingvariousdegreesofprogressivelyreduced
spacing toward perimeter.

F i n a l l y ,t h e computerprogram i s described and several

p a r t s of t h e program are l i s t e d t o show i q l e m e n t i n g o f I
t h e method i n F o r t r a n , However, thisindependence-canneither be declared fol
Cx when analyzing ( 7 ) , norfortheequation (11, i n
which t h ef a c t o r Km i s a directoutgrowthofusing (6)
APPLICABILITY OF EXISTING EQUATIONS and ( 7 ) together.Sincetheirregularityfactor K i is
f u r t h e r added a s a s i m p l e m u l t i p l i e r a f t e r i n t e g r a t i o n ,
R e f e r r i n gt ot h e Appendix I o f t h e Guide', t h e a n a l y t - i . e . a f t e r computing P ( x ) from thedistance X1 = 0, t o
i c a l l y most s i g n i f i c a n t e q u a t i o n s o f t h e ZEE g r a d i e n t t h ec e n t e ro ft h ef i r s t mesh inward, t o X2 = -D/2,the
a n a l y s i s method, are: o r i g i n a l components of (1) can no longer be r e t r e i v e d
f o r c o r r e c t i o n s on t h e i n d i v i d u a l b a s i s o f r e l a t i v e dis-
tance.Considering (l), thenextquestionthen, is :
Can themutualinfluenceoftheindividualgrid
elements be sufficiently covered by t h e f a c t o r K i ,
and as simply as a p p l i e di n( l ) ,e l i m i n a t i n gt h e need
f o r non-zero term Cx e n t i r e l y ?

No is t h e answer. For i n s t a n c e ,i ft h eo p p o s i t e were

true,then,followingtheprincipleofsuperposition,
it s h o u l db ep o s s i b l e :1 )t oc a l c u l a t et h ev o l t a g ei n
where, i n a d d i t i o n t o t h e symbols already mentioned, t h ec e n t e ro ft h ef i r s t mesh by ( 1 ) ; 2) t oc o n t i n u e
from t h i sv a l u e by P n t e g r a t i n gp o i n t by p o i n t toward
Ey i s t h ev o l t a g ed i f f e r e n c ei nv e r t i c a lp l a n e be- t h ec e n t e ro ft h eg r i d ; 3) t os t o p a t t h ep o i n t ex-
tween a horizontallyburiedconductor and the a c t l y a b o v e , t h e most c e n t r a lc o n d u c t o r ; 4 ) t o estimate
p o i n t on the earth surface inmediately above it; t h ev o l t a g e a t t h i sp o i n td i r e c t l y as E y by (6); and
5) t og e tf a i r agreementofresults,since,insuch a
Ex i s t h e relative v o l t a g e d i f f e r e n c e i n h o r i z o n t a l symnetrical or
almost synmetrical configuration, all
p l a n e between any two p o i n t s on t h ee a r t hs u r - t h e superimposed Ex terms canceleachother.(Fig.lb.)
f a c e ,d e n o t i n gt h e i rd i s t a n c e s XI, X2 from t h e
perimeter with a p l u s s i g n i n an outward d i r e c t - A substantial disagreement results in practice.
i o n from t h e g r i d , and w i t h a minus s i g n f o r t h e
points inside the grid area; Interestingly,Sctwartz3hasbeenvery much awareof
somi d i f f i c u l t i e s as e a r l y as 1957,mentioning i t a l s o
i i s t hceu r r e nf lt o w i nign t o ground p eurn i t in h i s d i s c u s s i o n on t h e AIEE ConmitteeReport?Paper
length of conductor. 58-98, which had preceeded the publication of Guide No.
80.
For a r e c t a n g u l a r g r i d , c o n s i s t i n g o f n x m conductors,
at least a set of n p a r a l l ecl o n d u c t o r h s a tsob e Withoutelaboratingfurther,theequation (1)had best
considered. notbeinterpreted as givingthecorner mesh v o l t a g e
estimate.Instead,forreasonswhich w i l l follow,the
Therefore, i t i s u s e f u l t o a n t i c i p a t e t h a t t h e i n t e g r a l equation may serve as a f a i r estimate f o r t h e p e r i p h e r -
solutionsof (6) and (7) canformallybeexpected as a l meshes i n c e n t r a l axis, providedthegriddoesnot
Ey = F(y) + Cy and Ex = P(x) + Cx, where Cy, Q are un- c o n s i s t o f many meshes, o r t o p r o v i d e a reasonable avep
determinedintegral terms, which may f u r t h e r e f l e c t
some f a c t o r ofn-conductor gemtry.
ageestimateforanygrid -
ifanoptimized,-ally
spaced p a t t e r n of conductors i s a n t i c i p a t e d .
Examined, Cy doesnotrepresent any problem. The l e f t
side of equation (6) e x p l i c i t l y r e f l e c t s t h e a s s u m p t i o n DISCUSSION
t h a tw i t h i nt h ei n t e g r a t i o np a t h from a c o n d u c t o rt o
i t s image, p r o c e e d i n gi n vertical p l a n e ,t h ee f f e c to f Simple rectangular g r i d s i n u n i f o r m s o i l were rigorous-
otherconductorscanbeneglectedduetothenearly l y examined by E. T. B. Gross e t al., duringtheyears

3 63
1953-56.Using Maxwell’s method of subareas, the ground TABLE I11 (1904m2 g r i d area, 0.03m cable dia.; 10 kA)
e l e c t r o d e was d i v i d e d i n t o a g r e a t number of elementary
sub-conductors,toobtain a s i m u l t a n e o u ss o l u t i o nf o r MESH SPACING C
Om MESH VOLTAGE STEP
theindividualelementswithlinearchargeden~ity.~,s SUM 6 (v) (v) (v)
4 21.82 7865.52* 2553.84 9171.19
Recently,Dawalibiand Mukhedkar reported’lY8s 9 b e t t e r 16 10.91 3513.37* 5305.18 2245.79
than 90% agreementwithexperimentalresultsfortheir 36 7.27 1886.61* 2240.44 3497.77
i n t e g r a t i o n mekhod, whichcombines t h e p r i n c i p l e o f i n - 64 5.45 1121.85* 2465.83 2339.70
finitesimal subdividing with a family of sinple correc- 100 4.36 698.0W 1774.32 2489.36
tive coefficients to simulate the effect of non-uniform 144 3.64 438.34* 1265.06 2668.19
currentdensity,increasing from t h e g r i d c e n t e r toward 196 3.12 268.18* 866.23 2866.05
perimeter. I n comparisonwithothermethods,theycon- 256 2.73 151.17* 540.29 3077.38
cluded that: 324 2.42 67.77* 265.52 3298.86
400 2.18 6.66* 28.40 3528.32
1. I fo n l yu n i f o r mc u r r e n td e n s i t y i s assumed, an 484 1.98 -39.07* -179.97 3764.31
i n t e g r a t i o n method g i v e s s a t i s f a c t o r y r e s u l t s f o r
p e r i p h e r a l meshes, but an o v e r l y s t e e p p o t e n t i a l *)indicates values for the irregularity factor K i = 1.
pyramid r e s u l t s i n t h e area of inner meshes. See a l s o Appendix f o r aFortranformulationofthe
applied equaticms.
2. I f t h e c o m e r mesh voltage i s c a l c u l a t e d by t h e
IEEE method, e i t h e r w i t h o r w i t h o u t some correc- Obviously, not the ultimate number ofmeshes,butspac-
t i o nf o rc u r r e n ti r r e g u l a r i t y ,t h eo b t a i n e dv a l - ing and depth of buriallcoductor ratio are the factors
u e sa r ei n c r e a s i n g l yo p t i m i s t i cf o rg r i d sw i t h undermining a p p l i c a b i l i t y o f t h e method.
more than four meshes, and the calculated voltage
may erroneously f a l l t oz e r of o rt h eg r i d sw i t h Fora m a j o r i t y of p r a c t i c a l c a s e s , t y p i c a l l y w i t h i n t h e
more than 40 meshes. 0.25m
exceeding
- lm depthrangeandconductorsizesusuallynot
500 M a , i.e. with nominal
a diameter of
Both these points have merit, d e s p i t e t h e f a c t that t h e 0 . 0 2 5 ~ 1o rl e s s ,t h ec r i t i c a ld e c l i n e of t h e mesh wlt-
second conclusion i s r a t h e r m i s l e a d i n g and require6 the a g e so c c u r sf o rt h er a t h e ri m p r a c t i c a ls p a c i n gv a l u e s
following qualification: -
between 1 2.5 meters.
The Dawalibi and Mukhedkar c a l c u l a t i o n s a p p e a r t o b e F u r t h e ra n a l y s i s of t h e s e and similarcomputerseries,
based on thedimensionsof a scaled-down l a b o r a t o r y as w e l la s c o m p a r i s o nw i t ht h ea l t e r n a t er e s u l t s from
d e l , c h a r a c t e r i z e d by a 1 6m x 16m g r i d frame, bur- j t h e rs o u r c e s ,s u p p o r tt h ef o l l o v i n g resume:
i e d 0.6m below earthsurface,with100amperes as-
sumed t o flow i n t o 200 ohm-meter e a r t h . And, i f t h e a. The simplifiedequations(11,(2),(31,(41, (5)
model data a r e b e l i e v e d , c o n d u c t o r s o f a relatively of IEE method p r o v i d e s a t i s f a c t o r y means o f e s t -
monstrous s i z e were used, with O.lm (i.e. 3.937 inclj irmting the grounding performance of smaller sys-
diameter. tems, t y p i c a l l y up t o 138 kV c l a s s , i f t h e ground
gridcanbeapproximated by arectangularsystem
of
several
parallel conductors, equally spaced
A check with the simplified equations of IEEE method
a p p l i e di n camputer series, Table I, seems t o con- more than4 -
5m (15 f t . ) a p a r t .
f i r m t h i s d a t a , though. b. For p h y s i c a l l y l a r g e g r i d s w i t h many meshes,typ-
i c a l of many o u t d o o r i n s t a l l a t i o n s i n t h e 230 t o
Were t h e model c o n s i d e r e d a s a 1 1 3 r e p r e s e n t a t i o n o f 765 kV c l a s s e s ,t h ee q u a t i o n s( l ) ,( 2 ) , (3) may
a p r a c t i c a l i n s t a l l a t i o n , a 0.01111a l u e of conductor
diameter i s more appropriate. Comparative results
give10 -
40% l o i r e rv a l u e sf o rt h ec o r n e r mesh
voltage.
f o r such a r e p r e s e n t a t i o n and f o r a 48m x 48m, f u l l
s i z e g r i d c a r r y i n g 1 0 kA, a r e shown i n Tables I1 and C. The stepvoltageequations(4) and (5) t e n dt o
111. producepessimisticvalues.Sincethestepvolt-
age i s c a l c u l a t e d by e x t r a p o l a t i n gt h e maxiaum
TBBLE I (256m2 g r i d ’ a r e a , 0.lm c a b l e d i a . ; 0.1 ld) gradient over one meter d i s t a n c ei n, s t e aodf
finding the voltage difference between two p o i n t s
by i n t e g r a t i o no ft h es u r f a c eg r a d i e n t sv i t h i n
t h ed i s t a n c e ,i n some c a s e st h ee r r o rc a nb ea s
h i g h a s 1009..
8.00 108.07* 126.01 78.25
4.00 31.95* 48.25 79.39 d. A l l t h es i m p l i f i e de q u a t i o n sa r e ill s u i t e df o r
2.67 5.83* 10.82 88.68 treatmentofdense umts, p a r t i c u l a r l y when t h e
2.00 -4.88* -10.74 100.90 numerical value of spacing between the conductors
approachesthe order of the conductor diameter and
of the burial depth values.

r
TABLE I1 (256m2 g r i d a r e a , 0.01m cable dia.; 0.1 U)
STEP
However, in c o n s i d e r i n gt h el a r g eg r i d s ,a na d d i t i o n a l
SUM (v) (9 (v) point should b e emphasized:
4 184.41* 215.02 78.25
16 77.76* 117.41 79.39 e. A s t h ed i f f e r e n c ei nc u r r e n td e n s i t yv a l u e s be-
36 38.55* 71.47 88.68 tween t h e i n n e r and t h e p e r i p h e r a l meshes i n c r e a s s
64 2.00 20.56* 45.19 100.90 w i t h an i n c r e a s e i n t h e number ofmeshes,anysin-
100 1.60 10.83* 27.52 114.62 gle calculation of the comer mesh v o l t a g e becomes
144 1.33 5.01* 14.45 129.30 i r r e l e v a n tt o a p r a c t i c a ld e s i g n . Economizing by
196 1.14 1.29 4.18 144.66 a separate subdivision of few p e r i p h e r a l meshes i s
256 -1.19 -4.25 160.55 arule.Furtherevaluationofinner mesh v o l t a g e s
becomes then necessary.

3 64
MODIFIED METHOD in"compensationforcurrent irregularities. Using an
e q u a ls p a c i n gf i r s t , i t i s p o s s i b l et o :
R e f e r r i n gt ot h e experimental evidence' s 8 s 9 i t i s as-
sumed that the envelope
of
ground
surface
potential 1. c a l c u l a t e t h e e n t i r e v o l t a g e p r o f i l e , u s i n g a n ar-
curves i s rather convex f o re q u a l l ys p a c e dg r i d sw i t h b i t r a r y v o l t a g e level as t h e i n t e g r a t i o n c o n s t a n t ,
many meshes.
2. f i n d t h e p r o f i l e maxlmun)which usually corresponds
For t h i s reason, t h ea b i l i t yt oe v a l u a t et h ee n t i r e tothepointdirectly above theconductorlocated
mesh v o l t a g e p r o f i l e o f a l a r g e g r i d , from p e r i m e t e r t o close to the geometric center of the grid,
perimeter, becanes a design necessity.
3. assume t h e true v o l t a g eo ft h i sp a r t i c u l a rp o i n t
And, f o r t h e ~ a m ereason,anunequallyspacedgridde- equal to the grid potential rise minus t h e v o l t a g e
s i g n p r d s e s t o b e t h e s u p e r i o r alternative t o t h a t o f d r o p i n vertical p l a n e , c a l c u l a t e d b y (6),
equal s p a c i n g ,t oa v o i dt h e c m problemofoverde-
signingtowardthecenter, when t h e number of g r i d sub- 4. r e c u r s i v e l y r e c a l c u l a t e t h e e n t i r e v o l t a g e p r o f i l e
d i v i i i o n s i s high. t o matchtheassumption 3, o b t a i n i n g a s o l u t i o n
which c o r r e l a t e s well w i t ho t h e r , more s o p h i s t i -
Similar to the equation (7), t h e g r a d i e n t G(x) ofnx m c a t e d methods, f o rt h ep e r i p h e r a l mesh rows.
conductor grid with unequal spacings in the longitudid
d i r e c t i o n ,D ( l ) ,
pressed as
D(2),D(3) ,......,
D(n-l),canbe ex- Fig. 2 i l l u s t r a t e s t h i s c o n c e p t , andthefollowing
of computer program l i s t i n g , w r i t t e n i n t h e Kronos ver-
part

sionof F ORTRAN, Fig. 3, providesdetailedinformation

x2+h2
L)+X D(
(D(1)+x)2+h2
(D(1)+D(2)+x)*+h2
+ D(l)+D(2)+x
+. ] (8) on theprocedure,
PROFIL (JN), where s
as implemented i nt h e sub-program

JN i s a number of consecutive meshes through

Assigning formally D(0) = 0, t h e e q u a t i o n f o r t h e v o l t - which t h e v o l t a g e p r o f i l e i s taken,
age difference between two p o i n t s on g r o u n d s u r f a c e , i n
h o r i z o n t a ld i r e c t i o n ,i n s i d et h eg r i d , is P(1) i s a s o i lr e s i s t i v i t y ,i n kohm-meters,

P(3) i s a ground ' f a u l t c u r r e n t , i n kA,

P(8) i s a d e p t ho fb u r i a l ,i n meters,

P(17) i s a t o t a l g r i d p o t e n t i a l rise, i n kV,

S(1) i s the p e r i m e t e r - t o - p e r i m e t edri s t a n ci en

the given profile plane, in meters,

S(8) i s a t o t a l e n g t ho fb u r i e dc o n d u c t o r s i, n
meters,
wheren i s t h e d e r ofparallelconductorswhich are Q(15) i s a d i a m e t eortfhge r i cdo n d u c t o ri ,n
p e r p e n d i c u l a rt ot h ep l a n eo ft h ec a l c u l a t e dv o l t a g e meters.
profile.

Thus, on t h e b a s i s o f equal c u r r e n td e n s i t y ,t h ee q u a - Both t h et r a p e z o i d a al n dS i q s o n ' si n t e g r a t i o nr u l e s

tions f o rs t e p and t o u c hp o t e n t i a l s of theunequally were t e s t e dg, i v i n gp r a c t i c a l l yi d e n t i c a rl e s u l t s( a
spaced grid can easily be defined and computerized. 0.3% d i f f e r e n col er s s ) . However, Simpson's rule
has b e e n r e t a i n e d f o r t h e method as b e i n g t h e o r e t i c a l l y
When t h e e f f e c t o f c u r r b n t i r r e g u l a r i t y e n t e r s t h e p i c - more a d a p t a b l e t o t h e character of calculated curves. A
ture, a problem arises. I m p l i c i t l ya, n yi r r e g u l a r i t y 0 . b (10cm) i n t e g r a t i o n s t e p i s used.
f a c t o r s i m i l a r t o K i o f ( I ) , i s a two-coordinate affair,
nothwithstanding that t h pe r e s e n t IEEE Guide No. 80
recognizesonlyn-conductors from ,the n x rn g r i d con-
ductor set, n e g l e c t i n g t h e e f f e c t of cross-connections
to
the
remaining m-conductors. %ere i s l i t t l e ex-
perimental evidence 1, a v a i l a b l feoar n y attempt t o
develop a s u i t a b l e f o n u u l a f o r t h e i r r e g u l a r i t y e f f e c t s kV

I
i n i n n e r meshes.

Nevertheless, a s u b s t i t u t e s o l u t i o n is possible. W17)

total
As i t h a s b e e n a l r e a d y i n d i c a t e d , t h e i n t e g r a t i o n meth- potential
ods t e n d t o p i l e e x c e s s i v e l y t h e v o l t a g e d i f f e r e n t i a l s rise
i n t h e area of inner meshes, i f a uniform current den-
s i t y i s assumed.
I .

-1 0 dist.- meter
The Author has found that, i nt h ep a r t i c u l a r case of TLegend:
IEE method, t h i s phenomenon c a n b e u t i l i z e d t o a g r e a t ADJV = a n a d j u s b e n t v a l u e
advantage i ne l i m i n a t i n gt h e need f o r a n ys p e c i a l ir-
regularity factor. W R T = a voltage drop i n v e r t i c a l
p l a n e , p e r theEquation(6)
Withoutsacrificingtheconceptual sinplicity o tf h e
existing I E E method, the following sequence of comput-
e r i z e dc a l c u l a t i o n s will r e s u l ti n a g r a d u a l ," b u i l t - Fig. 2

365
I ~ ( I ~ P . G T . I ) G O T Oe
IFCIKP.EP.0)LOOP-I
WTO 34

IF(IKP.Ee.2)KOX: I
IP(IKP.EP.3)KOX:2

FIG. 3
Depending on t h ei n p u t data sequence, i.e. i n whator-
d e rt h eg r i dl e n g t h andwidthparameters are entered,
t h ep o t e n t i a lp r o f i l ec a n b et a k e nt h r o u g ht h ef i r s t
ray of meshes, e i t h e r in t h e l o n g e r , o r i n t h e s h o r t e r
cuds.

Generally,theuse of unequalspacingtechniquesdoes
notguarantee a r e d u c t i o ni nt h e number of g r i d con-
d u c t o r s ,b u tr a t h e ra s s u r e st h e i rb e t t e ru t i l i z a t i o n .
The b e s t p r a c t i c a l s o l u t i o n i s usually achieved by pro-
granming t h eo p t i m i z i n gc r i t e r i o nt oe x i tt h eu n e q u a l
spacingbuild-uploop, as soon as thenearlyoptimal
A spacingfunctioncan-bedeterndnedfor obtaining a balance i s achieved for a l l b u t t h e c o r n e r meshes.
patternofgridconductors, which vi11 a s s u r e a f a i r l y
uniform level of maxinnnn mesh p o t e n t i a l s t h r o u g h o u t t h e In thosecaseswherethecorner mesh voltageexceeds
g r i d , ncwhereexceedingtheallowablestepandtouch s i g n i f i c a n t l y t h e p e r m i t t e d limits, i t i s moreeconom-
voltages. i c a lt os u b d i v i d es u c h meshes by additionalconductors
on an i n d i v i d u a l b a s i s , r a t h e r t h a n t o make a program-
Forinstance,the Fig. 4b i l l u s t r a t e s the e f f e c t o f op- ming allayance f o r more conductors.
timized unequal spacing, in comparison with t h e e q u a l l y
s p a c e da l t e r n a t i v e of t h e same griddesign. Hg. 4a.
366
 MESH VOLTAGE -
COMPARISON CHART CASE I
The comparison charts at right,CASES I, 11, 111, show
the difference between the rrvurimum surface potentials
calculated for individual meshes along the longitudinal
8 Uniform soil 30 ohm-meters, equally spaced grid of profile, as obtained for the equally and the unequally
8 8 x 3 conductors, 161.4m x 32.5111 grid area, indiv-spaced grids, in three typical examples.
mesh size 23.05m x 16.275111, depth of burial
i7 idual
' .5m, 25 k4 grid current, 0.3 sec. fault
time, size4/0 (211.6 MU4) copper conductors.
duration
Fig. 5, further documents the effect of a gradually
intensified spacing function. As shown, the function
base is one for equal spacing, and 'greater than one for
IEEE
Method I ModifiedMethod unequal spacings.
Mesh Equal Spacing Unequal Spacing;
DESCRIPTION OFCOMPUTER PROGRAM
No
SPACING MESH MESH MESH SPACING I
METER Kv Kv Kv METER Given the grid areadhnsions, magnitude and duration
of ground fault current, plus other design parameters
1Y7 23.06 1.601 15.91 1.902 like the depth
2.099 of burial, average length of ground rods,
2,6 23.06 1.626 1.398 21.23 soil and surface layer resistivity data, etc., the pro-
395 23.06 1.377 1.371 27.04 gram first evaluates size of the main grid conductors.
4 23.06 32.96 1.427 1.298
Since a ground grid should withstand any thermo-mechan-
Safely tolerable touch voltage limit.......1.657 kV ical abuse resulting from the prolonged flows of fault
Station grid potential rise calculated.....4.522 kV current, possibly caused by malfunction of the primary
Station grid totalresistance.............0.181 ohm relays, a GuideNo. 80 recorrmended valueof 4'secondd
is automatically applied in the calculation. However,
the program has a provision to override such a nominal
MESH VOLTAGE COMPARISONCHART CASE I1 - time setting, if the effect of back-up relaying, or a
deliberate change of the conductor size, are required.
w uniform soil 165 ohm-meters, equally spaced grid of
2VJ 9 x 7 conductors, 252.h x 22Cm grid area, indiv- Then, the lknits for tolerable step and touch voltages
idual mesh size 31.495mx 36.667~1, depth of burial during a ground fault are calculated, following the
.5m, 30 kA grid current, 0-15 sec. fault duration
3 time, 500 MCMcopper conductors.
well-known fonuuli of IEEE Guide, transcribed below as

IEEE Method I Modified

Method Etouch= (1+1.5xP(2))xO.l65/W) (kV;kohm-m,A,sec) (11)
Mesh Equal Spacing Unequal Spacing where P(2) is the surface layer resistivity parameter,
No

1,8
SPACING
METER
MECH-
Kv
MECH
KV
3.186
MESH
Kv
2.848
! SPACING
METER
20.33
31.49
,

2.298
P(4)
and

.
is the anticipated time for clearing the
fault

2,7 31.49 - 2.480 27.34 2.002Within a certain range of conductor total length, a
3Y6 31.49 - 2.072 35.17 1.938 series of consecutive grid designs is generated, re-
4Y5 31.49 - 1.880 1.989 43.14 flecting the constraints of approximate grid width and
L
length, individual mesh side ratio, minimum permissible
Safely tolerable touch voltage limit.......2.343 kV distance between grounding rods, etc.
Station grid potential rise calculated....10.835kV
Station grid totalresistance.............O.361 ohm The first, equally spaced grid pattern is determined
semi-empirically, combining a computer selected total
length of horizontally buried conductors with theto-
MESH VOLTAGE CWARISON CHART - CASE I11 tal length of grounding rods, as required by the input
data, Consequently, the step and touch potentials are
Uniform soil 165 ohm-meters, equally spaced of gridcalculated for the corner mesh, using fhe equations (1)
15 x8 conductors, 290.5~1 x 139m grid area, indiv- to (5), and the results are compared with the permis-
idual mesh size 20.747~1 x 19.857111, depth of burial iible limits. Thenceforth, corrections in the total
Irl .5m, 30 kA grid current, 0.15 sec. fault duration length estimates are made until the solution is found
3 time, 500 MCM copper conductors. for a match between the calculated and limiting voltage
values.

Mesh
IEEE Method I ModifiedMethod
Unequal Spacing
Although these simplified calculations are not accurate
Equal Spacing enough, it is expedient to determine the basic design

,
~ No SPACING
METER
MESH
Kv
MESH
KV
MESH
KV 1 SPACING
METER
pattern, resistance and potential rise of an equally
spaced grid inthat stage, and then to proceed with the

1
~
1,14
2,13 20.75 -
3.720 2.069
2.393
1.700
10.73
20.75
13.70
2.930
more accurate, second part of calculations,

technique,
PROFIL(JN),
to improve the grid design by means of unequal spacing
' 3,12 20.75 - 2.366 17.06 1.521
4,11 20.75 - 1.952 1.453 20.66 Such an optimized design is characterized by a gradually
~

5,lO 20.75 - 1.660 24.31 1.457decreasing spacing toward the perimeter, under the con-
6,9 20.75 - 1.472 27.80 1.503 straint of nraintaining thesame number of conductors as
1 798 20.75 - 1.562 30.97
1.379 determined by the estimating routine.

Safely tolerable touch voltage limit.......2.343 kV An example of complete computer run is shownin Fig. 5.
Station grid potential rise calculated...,11.008kV The printout, corresponding to the already tabulated
Station grid totalresistance..........,..O,367 ohm and plotted results of CASE I, is self-explanatory to a
large degree.

367
 0 .***+**+*
0 ;i:irM
1.657 kV limit

. Y t +

.. +++* * *
.... I
+*+***+*****
+**I***
I
1.80
twQ-m-(IDI*m
I
1.143 Id.t
SWUI m m r rtv
.................. ..................x .........
.la1 .U
1

0
C .. *+** STATIM .)OIO mmnu menu MI
C ~ 1 m O uR O A D mlm71& menu MI .+.*
0
C I#.-
.*
.+ * -
.*
v
.+
C .* C
0 .*
.*
.*
***
0
e .*
.+
.*
N.* . ** s
0 .*
.+ 0
0 .*
.+
.*
i
0
0
0
e 0
0 .*
0
.*
.+
.+
0
.+ 0

p
.+
0 .*
41.-
.. ** + + * e

'I"
0 .+ + 0
0
.,+
.+
.. ** + +
0
.*+ ++ +
0
$1.1)
. +*+ + ++ 0
8 1 .++ ++
(?
..+
* * + ++ 0
.+ +
0
.*
.+ + 0

m.Yc.++ ... *
.* .*
* * ++ +* **++*,*u*
+*****.* 0

. * .+
+ * ..+,*****
+* 0
.* +
.+ : 1
.*
.*
.*
.*
j t j
:e:
+*
++
0

0
0 11.-
.*+ 0
.* *+ **
.+
.*
.?
+ ++ 0

- SI.1)
.*+p*+d
.. +** ' + *+ ++ + + * +
.+
+ * t +

. * + ++ ++ ++ +* * +
.. +*+ +*+ *++* +* **
0
,-

0
+
0
+
* i:t:::*

'i"
* * * * L.....
+
+
0 *
+ +++
0 *
f t *+
***
0
+
+ +++
+ +*
0 +* + + +
CASE I + * ++
+ + * ++ + + +
::k:* CASE I
0

..*
cj.t ii
.+**+++ +
0 .*+**++ + + FIG. 4b
IlI.UI- + + ++ I I#.& + +.+ * * +
. * + I * * +
+ ++++**
0 + +++
0
+ **
+*+* +**
+++
ii$*
+ + 4 * ++
C + + + *+
+ +
+
+ +
0 + +
**
0 _ C * * * * + + * * * Y
+ **
+ **
++ +*
+**
0
*++*
+*** .'* ++ +
0 +*++ +
+++*
+*++ *
0 *+** +
**++ . r r l +
+*** +
C .*+* + +
**+*
.
+*** +
*+*+
****
*...
** .**. **
+**.
0
IU.HC
*:
++
ir
.++*+++
u
*. +
+
+
+
***
e e 0
+
368
" ....

e
*#
e
e
0

0
H)lltlB? I 01.15 1.962 a83
a.63 1.517 .2l6
0 24.12 I .3w .ne
25.e I .S4E -254
24. I2 I .S88 .252
0 22.63
21.15
1.511
1.982
a16
.283
0
#

0
1.2 BASE. unmuu S~ACIIS
0
119sw SPACIIS W.lOucI( MX.SIU
m. In1 IIVI IKVl
w 1. M E S H CORDUCTOR ADOITIOI: IUlIBRl 1
0

0
m
0 0
TOUCR-YOLTASE CRlTfRIOI DLSIQI
0
ITR LWGTH
APPLIED LEI6TH USED WLIAGECALCUL.
10. I ~ ~ R S I umms1 IKVI
0
2.23
2.92
I.I3 0
I .q2
1.60
0

c
0
G?ID
CFVDED ARU IflX III..................
(RID amEnT SIDE RATIO USED
W I D SPACIIICS LLEnElT SIZE lfl X RI........
TOTAL ORIDC O ~ D U C T O R L E ~ G T W1111
IOIU LEIQTH or ROO ELECTRODCS In)
...............
...................
................
161.4
I
23.159
.
X 52.5
.ins
X 16.215
144.~4~
se.nc+a
0

0
TOTAL LERGTH OF BURIED COIDUCTORS I R I . . . . . . . . . . . . . 894.645
0 luREER OF R O D S V S E D . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 24.000 0 #

,181 0
4.522
I902
I .61 1 0
.668

S A m Y mLDABLE STP-WKTAGE IKVI .......................... 5.124

0
S A m Y TOLRABLF TOUCW-MLTAQE IKVI.........................
ADJ. C U R U ? , BY DECRD(U1OF 1.10, IKAI ................... 1.651
21.10 0
0

3 69
The programincludes an o p t i o n t o bypass t h e i t e r a t i v e 4. E.T.B. Gross, B.V. Chitnis, L.J. S t r a t t o n , ttGround-
subroutine foi: determiningthetotalconductorlength, ingGridsforHigh-VoltageStations,"Ibid.,Vol.72,
andsuch a l e n g t hc a nb ee n t e r e dd i r e c t l y . %is pro- pp. 799-810,August1953.
vision, together with other input data, p e r m i t s a check
5. E.T.B. Gross, R.B. Wise, "GroundingGridsforHigh-
o f. t h ef i n a le n g i n e e r i n gd e s i g n ,w h i c ho f t e nd i f f e r s
frau t h eo r i g i n a l l yc a l c u l a t e d
anceofthegroundgrid
one. The t o t a lr e s i s t -
i s c a l c u l a t e d by Schwartz's
VoltageStations
u l aP
-
r l a t e s , "I b i d . ,
I1 ResistanceofLargeRectang-
Vol. 74, pp. 801-809 October
method.3 A l s o ,t h ee f f e c to fa d d i t i o n a ls u b d i v i s i o n s , 1955.
a si n d i v i d u a l l ya p p l i e df o rt h ec o r n e r meshes,canbe
6. E.T.B. Gross, R.S. H o l l i t c h , "Grounding
Grids
for
estimated.
Vol. 75, pp.
-
High-Voltage S t a t i o n s I11 ResistanceofRectang-
ularGrids,"Ibid., 926-935,
October
The program f i e l dl e n g t h i s 0.043 KPR u n i t s . For com-
parison,thefieldlengthoftheprogram GRD 2, as 1956.
l i s t e d i n t h e Appendix, i s 0.004 KPR u n i t s .
7. F. Dawalibi, D. Mukhedkar, "Optimum Designof Sub-
U t i l i z i n gt h e computertimesharingsystem KRONOS of s t a t i o n Grounding i n a -0 LayerEarthStructure,
the Control Data Corporation, we use the memorg assign- P a r t 1 -.-AnalyticalStudy," IEE Trans., Vol. PAS-94
ment of MA = 60,000 o c t a lu n i t s . The averagerunning (to be printed).
time i s approximately15-20systemseconds.
7a J. G. Sverak,"Discussionoftheabovepaper, IEEE
Trans., V O l . PAS 94, pp. p(btreoi n t e d ) .
CONCLUSIONS
8. F. Dawalibi, D. Mukhedkar, "Optimum Designof Sub-
s t a t i o n Grounding i n a Two LayerEarthStructure,
The described method f o rs u b s t a t i o ng r o u n d i n ga n a l y s i s
combines a number o f p r a c t i c a l a d v a n t a g e s : -
P a r t I1 Comparisonbetween
imentalresults,"Ibid.,bl.
t h e o r e t i c a l and exper-
PAS-94, ( t o be printed).
1. It extends the presently well established concept
of IEE methodof calculatingthedangerousstep 9. F. Dawalibi, D. Mukhedkar, "Optimum Designof Sub-
s t a t i o n Grounding i n a Two LayerEarthStructure,
andtouchvoltages,using a computerizedevalua-
t i o n of t h e s u b s t a t i o n p o t e n t i a l p r o f i l e t h r o u g h - -
P a r t 111 Studyofgroundinggridsperformance
new electrodesconfiguration,"Ibid.,
and
Vol. PAS-94,
out the entire grounding grid.
(to be printed).
2. It r e c t i f i e s t h o s e s h o r t c o m i n g s o f t h e s i m p l i f i e d
approach,which a r e no l o n g e r t o l e r a b l e f o r l a r g e
grid calculations.
L i s t e d below i s t h e program GRD 2,which hasbeenused
3. An advancedconceptofoptimizedgroundinggrid f o re v a l u a t i o n of IEEE s i m p l i f i e d method.However, the
d e s i g n ,u t i l i z i n gt h et e c h n i q u e of p r o g r e s s i v e l y Fortranformulationofequations(l), (2), (3),(4)and
unequalspacings, i s implemented. (5), i n c l u d e di nt h es u b r o u t i n e s MESHC and STEPC, i s
g e n e r a l l yu s a b l ef o r any computerizedapplicationof
4. The modest memory requirements make thecomputer t h e e x i s t i n g IEEE method.
a p p l i c a t i o n s u i t a b l e f o r time s h a r i n g f a c i l i t i e s . m O G R A M CJ(D2CINPUT OUTPUT)
PRINT,*AREA,SOIL,D~PTH,DIAMETER,CURRENT+,
READ,~R_EA_,S_O_fL,PP_TH,CDIA,C_UURR
SIDE-SQRT(AREA)
1
-
A(XNOWLEDGEMWT PRINT,*LOOP*, S READ,NN
PRINT 18
The Authorwould l i k e t o thank Mr.
E.F. Jones,Chairman PRINT 19
W 1 1 I=l,AN
in: [m2,0hm-m,m~m,A]
of t h e IEE Task-Force P468.17 of WorkingGroup 70.1, NSZ 2* I NS is n&nberspacings
of
Groundingof Gas Insulated Substations," and Manager of NC: (NS+1)*2
the Substation Engineering Department, United Engineers NM: NS**2
CTL=ROAT(NC)*SIDE
6 Constructors, Philadelphia,
for
encouragement
and SPACZSIDE/FLOAT(IS)
s u p p o r t o f t h i s work. CI: .SS+. I?L*FLOAT(HS+O
CALL MESHC(CM, NS,SPAC, DPTH, CDIA)
WESH~CM*CI*SOIL*CUR~/CTL
Permission i s grantedtoreproduceorquote from t h e CALL STEPCCCS HS SPAC DPTH,CDIA)
l i s t e d p a r t s of the computer programs RENA 2 and GRD 2. ESTEP:CS*CI*S~IL;CURR~CTL
However, credit should be given to United Engineers, as Er(FS@=EmFSH/CI
PRINT 20,N?l,SPAC,~ESO,EW~H,ESTEP
thesource. 1 I CONTINUE
21 FCRWAT(/23X 3H*) *INDICATES VALUE*/26X *FOR KIRREG.:¶
REFERENCES 2 E FORMAT(I4,4~,F5.2:2X.FR.2,2H* ,2(F8.2,2h) etc.-
19 FORMAT(* NO.*,3X,*[W1*,3(7X,*tVI*)/)
18 FORMAT(//* MEsH*,2X,*SPACING*,2X,*CORNER MESH VOLTAGE*
1. LeEE No. 80fMarch 1961/71, "Guide f o r S a f e t y i n A l - PRINT 21 etc.- A-
t e r n a t i n g C u r r e n t S u b s t a t i o n Grounding,"byAmerican PRINT,////
STOP S END
I n s t i t u t e o fE l e c t r i c a lE n g i n e e r s , New York,1961,
SUBROUTINE STEPCCCS, NS,SPAC,DPTH,CDIA)
reaf finned 1971. R:(1./(2*DPTH))+(l./(SPAC+DPTH))
Do 2 IZ2,NS, 1
2. AIEE Camittee Report,
"Voltage
Gradientsthrough R=fS+(I./ (FLOAT(I)*SPAC))
2 CONTINUE
t h e Ground Under Fault Conditions,"AIEETransactions
on Power ApparatusandSystems,
692,October1958.
Vol. 79, pp. 669 - CS:FS/3.142
Fs=0.
RETURN f END
SUBROUTINE MESHC(CM,NS,SPAC,DPTH,CDIA)
3. S. J. Schwarz, "AnalyticalExpressionsforthe Re-
Fm:. 15
NF:2*NS- I
s i s t a n c e ofGroundingSystems,"Ibid., Vol. 73, pp. DO 1 115 NF 2
F~:~*~~AT~I)/FLoAT(I+I)
1011-1016,August 1954. 1 COHTInUE
CMZ(ALOG(SPAC**~./(I~.*DPTH*CDIA)))/~.~~~+ALOG(F~~)/~.I~~
F
m
: 0.
RETURN S END

3 70
 Discussion 1:

F. D a w a i (The Shawinigan Engineering Co., Montreal, Quebec) and

D. Mukhedkar (Ecole Polytechnique Montreal, Quebec): Mr. Sverak is
to be congratulated for a very timely paper. His efforts to improve and
update the IEEE Guide No 80 are welcome.It is unfortunate that inour
days where power systems are studied in increasingly greater detail by
use of sophisticated computer programs,grounding is still looked at by 13
semi-empirical methods.
We agree with most of the conclusions of the paper. However a
number of points in thepaper are misleading and need to be clarified:
- In his abstract and in the text the author states that "The en-
velope of earth surface potentialcurves is distinctly convex for ground-
ing mats with manymeshes. This is true when uniform current distribu-
tion is assumed in the conductors of the mat. The use of an exact cur-
rentdistributioninthemat will flatten the envelopeand make it
practically horizontal. In order to illustrate this point we have studied
with our comuuter urogram case I11 of Mr. Sverak's uauer. The mid is 10
shown in Figuk I of t h e discussion. We have calculated b e earth &face
potentials for the three directions shown and for both conditions of
uniform current distribution and exact current distribution. However
only the results for direction 3 are shown in Figure 2 and 3. These last
figures c o n f m that the convexity of the potential envelope is caused
by a uniform current assumption in the grid, such a hypothesis is only
justified forgrids with small number of meshes,
Mr. Sverak refers to our publication (7) in order to justify his
above statement. The curves of our experimental measurements which
were reported in the mentioned paper show clearly flat envelopes for
the experimental curves. Perhaps our labeling of the curves plotted was
not clear enough or misleading.

Direc, Direc. Dlret.

3 2 1
4 4 4
I I I

+ x axis
Fig. 1. Grid configuration

Manuscript received August 29,1975.

3 71
 - Mr. Sverak is right about the model data used in reference 7of The concentration and varying depth of ground rods at “critical”
his paper. The reducedmodelused could not have extremely mall locations in the substationare not described at any length in the paper.
conductor radius for obvious minimum mechanical rigidity needed for I would appreciate the author’s comments regarding the use of long
handling and installation. Our opinion is that the results and conclusionsground rods in soil layers with varying degrees of resistivity, also, with
given in (7) are notinfluenced by the radius size selection. However we groundmg material installed in deep wells or in adjacent lakes and also
emphasize again that the conclusions given in our paper were peculiar how this could affect the step, touch, andmeshvoltages in his cal-
for the case studied. Much more work is still needed in order to deter- culations.
mine the influenceof all the parameters. Finally, I would appreciate the author’s comments concerninghis
- In page 3 of the discussed paper, paragraph d we find the state- feeling about not connecting the substation ground to the substation
ment in contradiction with the text content. Can the author elaborate fence ground in order to reduce faultvoltage on the fence to the general
more on this? public.
- Attempts to develop methods for the calculations of current In reply to Mr. A. B. Purdy; we are in agreement with most of his
distribution in grounding electrodes have been carried out recently. A concise comments. With regard to the profde printout, the suggestion
number of publications are available (Reference (1) and (2) of the dis- “to adjust all theconductors to one voltagelevel”wouldmean to
cussion). neglect theirregularity problem entirely and to align all theprofde
- The method of compensation for current irregularities seems “peaks” to the P(1O)UVERT voltage level. Ifthis is what Mr.Purdy had
very interesting. Is it possible with this method to take into account the in mind, the resultwould be as follows:
presence, in close proximity of the main grid, of return grounds and/or With the exception of peripheral meshes, such a rectified profie
buried metallic structures? Such close structures will obviously distort should come close to a true representationof the centralvoltage profie
the voltage profie. in Direction 4, Fig. 7. Thus, from the practical point ofview, one
would be able to obtain two limiting profdes fromeach computer run.
REFERENCES The “best” one, correspondingto the central row line, and the “worst”
one, asumed tobe taken through the peripheral mesh rows,as described
T. N. Ciao, M. P. Sarma “Effect of two layer earth on the electric before.
fieldsnear HVDC electrodes” IEEE Transactions, PAS,Vol. 91 The remark that the final computer printout should show the sur-
November 1972 pp 2356-65. face potentials outside the grid, is valid. In fact, we neglected to show
F. Dawalibi,D. Mukhedkar. “MultiStep Analysis ofInterconnected this part of the curve only for the rather petty technical reason that the
electrodes” IEEE Summer Power Meeting, San Francisco, 1975 potential profie reduces very sharply outside the perimeter,and so the
paper No F75 522-3. scale of the computer plot would have to be drastically compressed.
But, we agree that it is essential to show the outside surface potential
data, and we intend to accomplish that with an additional, separate
A. B. Purdy (AppalachianPower Company, Roanoke, Virginia): Mr. printout.
Sverakand his company are to becommended for theireffortsin
presenting a paper that is both educational and useful. The author has
developed a needed computer program for solving the longexisting
problem of calculating surface potentialsand gradients in electric power J. G . S v d : First, I would like to thank all discussers for theirvaluable
stations when the system is faulted. comments and for their interest in this paper.
The basic principlesin IEEE Guide No.80 for Safety in Alternating The described program represents nothing more than an adequate
Current Substation Grounding are modified and further developed and engineering tool for specific design purposes. There are,and have been,
some of the simplifying assumptions are analyzed. Many of the short- several other methods which may offer solutions analytically superior
comings of the Guide result from the simplifications. to thatof RENA 2. In Mis respect, our goals weremore utilitarian.
Equation (2) of this paper is Equation (66)in IEEE Guide No.80 For instance, we looked into the earlier work done by Professor
simplified for X=D/2 and D>>h. As the numberof meshes is increased Gross andhis students, and tried to benefit fromsome of their advanced
neither of the assumptions is valid and calculations may show mesh calculations. Yet, it soon became obvious that for large grids, the re-
voltages lessthan zero. quirements on the computer memory tend to skyrocket.
Equation (3) is of dubious value except forsquare grids with square The final worth of any grounding calculation depends largely on
mesh. Guide No. 80 states that Ki is the “Non-Uniformity Factor” to the quality of soil sampling in the field. In practice, this fact acts as a
take into account the higher gradients at corners. The empirical equa- deterrent against the use of more-precise, but elaborate methods.Very
tion is based only on ground conductor distribution. often the mean value of soil resistivity readings has such a wide margin
The above facts justify statementsa, b and d in theDiscussion sec- of confidence limitsthat its statistical value is only a bit better than an
tion of this paper.The latter partof statement cis not so evident. arbitrary guess. So, we felt that if there is any positive way to extend
Equation ( 8 ) is valid only for conductors of infrnite length.When the relativelysimpleIEEE method, that this would be of a general
the distance from the conductor to the point under investigation a p advantage.
proaches the length of the conductor the value obtained is very pessi-
mistic.
The final computerprintoutshould be made to adjust all the
conductors to one voltage level and the surface potential outside the
grid shown for at least one half the spacing of the two outside con-
ductors.
This paper shows the great importance of IEEE Guide No. 80 and
at the same time makes parts of the Guide obsolete.
Manuscript received August 14,1975.

George L. Finley (Duke Power Company, Charlotte, N.C.): Mr. Sverak

is t o be congratulated for his fine contribution
to the state of the art of
substation grounding.
This paper shows that by the use of modem tools, such as the
computer, one can accomplish safer andbetter grounding grid results in
large substations than could possibly have been achieved in the earlier
years of our industry,and with less grounding material.
It would be interesting to makeacomparison of thecost of
grounding two identical substations by using the former uniform grid
method and the method outlined in this paper, as well as the safety
comparison described in the paper. Also, I believe it would be valuable
to make field voltage readings of step, touch, and mesh voltages with
typical fault current values introduced into these grids at realistic fault
entrylocations, such as at breakers,transformers, and grounding Fg 6.
switches.

Manuscript received September5,1975. Manuscript received September26,1975.

3 72
 Of course, certain problems result from the uniform current con- Profile directions
*\
‘-.
cept. If, starting at the perimeter, an integration process is applied ‘4,3
throughout the grid, the natural result is that the voltage differentials
are piled excessively towards the center of the grid, i.ei in comparison
to what would be the cumulative value of voltage differentials propor-
tional to the true, non-uniform current distribution.
In terms of the voltage profile, any additional attempt to correct
v t t t
E D # *

the resulting curve by adjusting all the grid conductors to one voltage
level would not help much. In order to compensate for the effects of
uneven currentdistribution, either the inner or the peripheral mesh
voltages would have to be obtained with the help of some corrective
factor-in fact, much in the same way as in the original IEEE method.
However, the Author’s method retains the shape of voltage profile
curve exactly as it results from the equal current distribution. Instead of
trying to adjust the profile voltage at the perimeter, the whole curve is
floated toward a predetermined voltage level, which applies only to the
innermost, single “peak”. The point to be made is that by adjusting the
profile to the value of minimum touch voltage at this least dangerous
spot, a different effect, that of averaged unequal distribution, is achieved
Figure 6 shows a comparison of.the typical charge density curves
of a single quare loop6, and of the center and outermost profiles of a
square plate , versus an estimated segment of averaged density curves
resulting from the analysis of computer runs for square grids with from
9 to 64 meshes.
Since these last curveshavesomewhatless than ideal character-
istics for perfect simulation, the final outcome is a small mesh voltage
bias inthe central area of the grid,reducinggradually into a fair
approximation of mesh voltages, in the critical parts close to the peri-
meter.
With most errors on the conservative side, the approximationleans
toward pessimistic values, as pointed out by Mr. Purdy, but the errom
become significant only for very large grids with many meshes. In this
respect, the use of theunequal spacing technique acts as a partial
countermeasure.
This result is rather logical, considering that the total sum of all
conductor currents has to be the same for any distribution. Thus, for a
grid consisting of n-elements, the total ground current I=I]+Iz+ ... +In
and the grid potential rise Vo are known, once the grid design and the
total grid resistance are determined. All the individual conductor cur-
rents 11, I2 ...,In must satisfy the equations
v o = r 1 1 1 1 + r 1 2 1 2 + . . . +r l n I n

where rij represents the mutual resistance between elements i and j for
i # j, rd = rji; or the self-resistance for i = j.
Smilarly, once a certain potential level is predetermined correctly Fg 7.
for some point at the earth surface, it follows from the super-position
principle that an alternative set of equations can be defined for this or
any other potential on the earth surface, and for the related current
flows. Therefore, the individual currents, their mutual ratios and the
relative current distribution can be determined by simultaneous solu- IEEE method makes the terms like “central profile” or “peripheral
tions of such equationsets, step by step. profile” irrelevant to the calculation.
Dr. Mukhedkar and Mr. Dawalibi have indicated that the introduc- To the other points:
torystatements concerning the convexity of thepotential curve en- - Paragraph d relates more to the equations (1) to ( 5 ) and less to
velope can be misleading since the convexity results from the particular the equations ( 6 ) to (9). Yet, as this is a matter of degree, there is
method used - that is, from a ufiform current assumption. It is further hardly any contradiction.
suggested that the use of an exact current distribution in the mat would - The addition of recent references is welcomed.
flatten the envelope and make it practically horizontal. If the Discussers - The problem of separate conductive objects buried near the grid
mean only the envelope of the profile “peaks”, they are right. However, is the intricate one, as they disturb the ground field and alter the path
if the envelope of “valleys” is concerned, as originally shown in Fig. 4a, of ground currents emanating from the grid. Without going into detail,
b, the problem is more complex. the m e b d cannot take into accountsuch effects, and no modification
In large grids, thegeometry of the gradient field produces a rela- short of writing a new program would do.
tively sharp increaseof current density within the last two or three In reply to Mr. A. B. Purdy; we are in agreement with most of his
peripheral mesh rows, while the large area of inner meshes remains at a concise comments. With regard to the profile printout, the suggestion
much lower density level. “to adjust all the conductors to one voltage level” would mean to neg-
Using a 1 to 5 intensity scale, Figure 7 illustrates the usual spatial lect the irregularity problem entirely and to align all the profile “peaks”
pattern of grid segments with high current densities. The grid configura- to the P(I0)-UVERTvoltagelevel.If this is whatMr.Purdyhad in
tion refers to Case I11 in particular. mind, the result would be as follows:
It is apparent that at least six peripheral meshes in Direction 1, and With the exception of peripheral meshes, such a rectified profile
two meshes in Direction 3, arebound to have their “valleys” more should come close to a true representation of the central voltage profile
pronounced than the othermeshes, even with the exact current distribu- in Direction 4, Fig. 7. Thus, fromthe practical point of view, one
tion. From this point of view, it is somewhat unfortunate that the Dis- would be able to obtain two limiting profiles from each computer run.
cussers opted to show the profile curvein Direction 3, instead of 1 . The “best” one, corresponding to the central row line, and the “worst”
Nevertheless,Figure 3 of this discussionstillseems to offer enough one, assumed to be taken through the peripheral meshrows, as de-
evidence of the existing convexity. scribed before.
For the samereason, the voitageprofilesproducedbyRENA2 The remark thatthe final computerprintout should show the
should alwaysbeviewed as taken through the peripheral rows. This surface potentials outside the grid, isvalid. In fact, weneglected to
decision results more from the characterof averaged distribution curves, show this part of the curve only for the rather petty technical reason
as shown in Fig. 6, than from the method. The basic concept of the that the potential profile reduces very sharply outside the perimeter,

3 73
and so the scale of the computer plotwould have to be drastically com- representations used in PUGY, and the experience gained with its appli-
pressed. But, we agree that it is essential to show the outside surface po- cation, on some other occasion.
tential data, and we intend to accomplish that with an additional, sepa- Finally, to the question about connecting or not connecting the
rate printout. substation ground to the substation fence, the choice depends mostly
Responding to Mr. George L. Finley’s questions concerning the on local circumstances. When the fence is not connected to the ground-
use of ground rods in the computerized design: The program can handle ing grid, the arrangement will be safe for all ground faults within the
only ground rods of uniformly prescribed length. The other parameter station except when there is a direct feeding of the fault current into
which can be prescribed is the minimum spacing between the individual the fence. In such a case, there might be extremely steep voltage gradi-
rods. Depending on the final grid design selected by the computer, the ents around the fence which may constitute a danger for bothstep and
rods are placed either at each crossing point, or if the grid spacing is touch. For other faults, the fence remains at the same potential as the
is smaller than the required rod distance, then the rods are placed in a surrounding soil.
sort of checker board pattern. On the other hand,when the fence is connected to the ground grid,
Concerning the use of deep wells, or adjacent ground electrodes then during every fault within the substation, the ground grid potential
tied over some distance to the grounding grid, such an application is rise will be transferred to the fence and therefore it is rather mandatory
generally useful in conditions where the soil resistivity is high, but one to provide some kind of potential ramp in the vicinity of the fence, by
may encounter the problem of transferred potentials. However, this burying one or two loopsof ground conductor outside the perimeter.
particular program is not equipped for this type of consideration. As a So, in summary, the separately grounded fence mght be advan-
matter of fact, we have developed another program, named PUGY, for tageous in cases where the danger of a fallen overhead wire is low. The
that purpose. The program provides modelingof a distributed parameter tied-in fence seems to be the preferable arrangement for substations
lattice network system to evaluate the transfer of ground potentials with many incoming overhead lines, or where the outer loops outside
alongburiedlong metallic electrodes, like cable shielding sheaths, the fence may favorably contribute in lowering the total resistance of
counterpoise, water pipes, etc. We hope to describe the specific model the system.

CORRECTION TO “ASIMULATIONTECHNIQUEFORSTUDYING REAL ANDREAC-

TIVE POWERFLOW PA’ITERNSI” AND “AN APPROACH FOR PREVENTING SYSTEM
INSECURITIES ARISING FROM LINE AND TRANSFORMER OUTAGES~”

M. S. Sachdev and S. A. Ibrahim

Power Systems ResearchGroup
University of Saskatchewan
Saskatoon. Canada

In the above papers the first pages and corresponding discussions were interchanged. The
correct order for “A Simulation Technique for Studying Real andReactivePowerFlow
Patterns1’’ is page 2092,2102-2109,2100. The correct order for “An Approach for Preventing
System Insecurities Arising from Line and Transformer Outages2” is page 2101, 2093-2099,
21 I@-2114.

lM.S. Sachdev and S. A. Ibrahim, IEEE T m c t i o n s on Power Appmafur and Systems, vol. PAS-94, Novem-
ber/December 1975, p. 2092.
2M. S. Sachdev and S. A. IbrahimJEEE Tmnracrions onPowerAppamtusand Systems, vol. PAS-94, Novem-
ber/ 1975,p.2101.

3 74