International Journal of Computer and Electrical Engineering, Vol. 3, No.

6, December 2011

Maintaining Low Resistance in Conductive

Joints
Samarjit Bhattacharyya, Anandita Chowdhury, Hitesh R. Jariwala, M Sharaschandra Shetty, and
Rajulkumar Engineer
into account.
Further the role of changes in thermal and electrical
resistance that can occur in a clamped joint is very important
as both can affect the contact force and current flow across
the joints A purely metallic contact joint occurs only in
vacuum. In free air oxide layers form on the contact surfaces.
The hardness of the contact material also affects the
resistance to current flow across the joint [2]. A microscopic
view (Fig. 1) of the contact surfaces shows that they are
rough and irregular. Current flows across constricted areas
where these rough surfaces make contact. Our aim in this
paper is to optimize the above mentioned factors in order to
achieve excellent joint efficiency and zero hot spots in
switchgears.

AbstractIn the electric power industry, optimizing power

flow is a primary concern, in the generation, transmission and
distribution process. One key ingredient is providing and
maintaining low resistance conductive joints. Field experience
and laboratory studies have shown that this is especially true in
the case of bus bars and bolted high current connections.
Unplated joints are generally less reliable due to potential
oxidation of the surfaces. If unplated joints are used special care
must be taken to clean the two surfaces and petroleum based
greases should be applied to slow down the oxidation process.
The development of Belleville washers permitted some
significant advances in the ways busbars could be joined. This
paper describes how plated bus bars outperform unplated bus
bars by providing stable contact resistance and a low maximum
operating temperature that will increase the service life of the
bus joint.
Index TermsAluminum bus bar, belleville washer, copper
bus bar, contact resistance, , silver plating

I. INTRODUCTION
Earlier in the electric power industry the aluminum or
copper bus bars were installed uncoated and left that way.
While the performance of an uncoated bus joint may have
been sufficient years ago, todays increasing demands for
power, given the limited capacity and economies of the
marketplace, are forcing the producers to improve the
efficiency and performance of the entire system. Engineering
surfaces are never absolutely smooth and the surface
irregularities become apparent when observed under a
microscope. As a result, constriction resistance arises in
practical electrical interfaces because contact is made only at
few discrete spots as defined by the roughness of contacting
surfaces and applied contact pressure
The resistance of a joint [1] is affected mainly by two factors:

Fig. 1. Current flows across constricted areas [3]

II. PROBLEM FORMULATION

Following bus bar combinations are considered in this
paper for measuring contact resistance.
a) Bare Aluminum busbar and Bare Aluminum busbar
b) Bare Copper busbar and Bare Copper busbar
c) Buffed Aluminum busbar and Buffed Aluminum
busbar
d) Buffed Copper busbar and Buffed Copper busbar
e) Silver plated Copper busbar and Silver plated
Copper busbar
In each case milli-volt drop across the busbar joints at
different loads will be measured. Lower the voltage drop
lower will be the contact resistance for the same load.
Different factors [4] that will determine the efficiency of
the joint with possible remedies are as follows:
a) Streamline effect
b) Effect of oxides in contact resistance
c) Condition of the contact surfaces

a) Streamline effect or spreading resistance Rs, the

diversion of the current flow through a joint.
b) The contact resistance or interface resistance of the
joint Ri.\
The total joint resistance Rj=Rs+Ri
The above equation is valid only for a d.c current. Where
a.c. currents are flowing, the changes in resistance due to
proximity and skin effects in the joint zone will also be taken
Manuscript received July 6, 2011; revised August 12, 2011.
Samarjit Bhattacharyya, Anandita Chowdhury, and Hitesh R. Jariwala are
with Department of Electrical Engineering of S. V. National Institute of
Technology, Surat.india.
M Sharaschandra Shetty and Rajulkumar Engineer are with Reliance
Industries Limited, Hazira, Surat, Gujarat, India.

A. Streamline Effect:
The distortion of the lines of current flow at an overlapping
joint between two conductors affects the resistance of the
802

International Journal of Computer and Electrical Engineering, Vol. 3, No. 6, December 2011

jooint. In case off an overlappinng joint betweeen two flat coopper

baars the streamlline effect is dependent
d
onlyy on the ratio of
o the
leength of the ovverlap to the thickness
t
of the
t bars and not
n on
thhe width. Hencce the efficienncy of an overrlapping jointt does
noot increase as the length off the overlap increases
i
and from
thhe electrical pooint of view no
n advantage is to be gaineed by
em
mploying an unduly
u
long ovverlap.
The resistancce ratio e in fiig -2 is the ratio of the resisstance
off a joint due too streamline effect
e
Rs, to thhe resistance of an
eqqual length off single conducctor Rb:

wer contacts haave sufficientt electrical fiellds

fritting. Most pow
c
both A and B frittting. Therefo
ore, the contaact
to cause
resisstance is tempoorarily low, ass if only a thin
n film is preseent.
Th
his oxide form
mation can rappidly destroy unplated coppper
contaacts. Silver pllated contactss [7] are moree resistant to the
t
form
mation of this oxide film ssince silver does
d
not form
m a
stablle oxide. Coontact surfacees should bee flattened by
mach
hining and thooroughly cleanned before carrying out silvver
plating. However if sulphides arre present in the environmeent,
a thick sulphide fiilm will form oon silver conttacts.

w
where
a= breaddth of bar in mm.
m

C. Condition off the Contact Surfaces:

Th
he condition of
o the contact surfaces of a joint [8] has an
impo
ortant bearingg on its efficciency. The surfaces of the
t
copp
per should be flat and cleann. A microsco
opic view of the
t
contaact surfaces shows
s
that theey are rough and irregular as
show
wn in fig -3. Current
C
flows aacross constriccted areas wheere
thesee rough surfacces make conttact [9].

b= thickkness of bar in mm.

l= lengthh of overlap inn mm.
= resisttivity of the coonductor, mm.

.
Figg. 3. Magnified coontact surface [3]]

he surface area of the consttrictions depen

nds on:
Th
i) Hardneess of the mateerial
ii) Amouunt of contact force.

Fig. 2. Streamline efffect in overlappinng joints

D. Hardness off the Materiall:

Co
ontact hardneess is expresssed by the amount
a
of forrce
need
ded to cause peermanent defoormation.
Co
ontact hardneess is determinned by pressiing a round ball
b
into the surface of
o the contacct. The pressure causes the
t
surfaace to deflect a certain disttance while a load is applieed.
After the pressure is removed, a permanent in
ndentation is left
l
he surface. The
T differencee between th
he two lines of
in th
defleection d is thhe elastic defformation. Th
he depth of the
t
perm
manent deflecttion is the plasstic deformatio
on D. (fig-4)
Th
he contact harrdness is exprressed by the amount of forrce
need
ded to cause permanent
p
defo
formation. Forr a radius of ball
b
r.

From the graph it can bee seen that thhe effect falls very
raapidly for ratioos to two and then
t
very much slowly for values
v
upp to seven. This
T
means that in most caases the stream
mline
efffect has very little effect as the overlap iss of necessity much
m
grreater than sevven.
B. Effect of Oxides
O
in Conntact Resistancce:
In free air, oxide
o
layers foorm [5] on thhe contact surffaces.
Peetroleum greaases are used to slow downn the formatioon of
oxxide layers. Iff the petroleum
m layer is very thin the coontact
reesistance is neegligible or neearly equal to the resistancee in a
vaacuum.
Holm [6] deetermined thaat the contact resistance caan be
neeglected if thee thin film layyer is less thann 10-6 cm thickk. He
teermed this layyer quasimetaallic. The low
w resistance caan be
exxplained by tuunnel effect. Electrons
E
cann penetrate thee thin
fillm without ennergy loss, ass if they are passing throuugh a
tuunnel. If the laayer is thicker, electrons loose energy annd the
reesistance to cuurrent flow inccreases.
If there is a thick film on the contactt surfaces, theere is
addditional resisstance presentt. The extra reesistance is knnown
ass Fritting Ressistance Rf. This
T
occurs when
w
the petrooleum
cooating is greaater than 10-6 cm
c thick or an
a oxide or suulfide
fillm develops on the conttacts. This thhick layer caan be
peenetrated if thhe electrical fiield force is great
g
enough. Field
strrengths of 100-5 to 10-7 V/cm
m will overcom
me this thick layer.
l
Thhis breakdownn is known ass a fritting. Once the thick layer
beegins to dissollve, the area of
o spot contactt begins to inccrease
annd the resistannce is lowereed. The seconnd step is callled B

Fig. 4. Determination of contact hardneess. [3]

E. Amount of Contact
C
Force:
Co
ontact resistannce is dependdent more on the
t total appliied
presssure than on the area of contact. If th
he total appliied
presssure remains constant
c
and tthe contact arrea is varied, the
t
total contact resisttance remains practically co
onstant. This can
c
803

International Journal of Computer and Electrical Engineering, Vol. 3, No. 6, December 2011

bee expressed byy the followinng equation:

S = stress, psi at
a the inside diiameter.
i
f = deflection, in.
t = thickness, inn.
E = modulus off elasticity of m
material, 30x 106 psi
= Poissons raatio for materiial ( 0.3 value for steel)
h = height, in.
meter, in.
D1 = inner diam
2 = outer diam
meter, in.
D2
Washer dimensionns are shownn schematicallly in fig-8. The
T
a M can be ttaken from fig
g-7.
consstants C1,C2 and

w
where
Ri = resistannce of the conttact.
P= total conttact pressure.
n=exponent between
b
0.4 and
a 1.
C= a constannt.
The greater the
t applied tottal pressure thhe lower will be
b the
jooint resistancee [10] and therefore for higgh efficiency joints
j
hiigh pressure is
i usually necessary. Thee Fig-5 show
ws the
efffect of pressuure on joint ressistance.

Figg. 5. Effect of preessure on contact resistance of a jooint between two copper

connductors
Fig. 7. Constannts for equation reelated to bellevillle washer [3]

The joint ressistance falls rapidly

r
with inncreasing presssure,
buut above a preessure of abouut 15N/mm2 thhere is little fuurther
im
mprovement [111]. These bussbar joint mayy heat up underr load
ass the contact pressure appplied with steel bolts tends to
inncrease becausse of the diffference in exppansion coeffi
ficient
beetween two dissimilar metals,
m
coppeer and steel. The
teemperature off the joint inccreases due too wattage losss and
thhen decreasess when the source is diisconnected. Such
reepeated therm
mal expansionss and contracctions of dissiimilar
m
metals
will varry the contact force and eveentually will cause
c
looose joints. To
T compensaate for thesee changes, spring
s
prressure shouldd be added to this
t joint. Belleville washerss [12]
arre frequently used
u
to compeensate for therrmal expansionn and
coontraction in clamped joinnts. It is im
mportant that these
w
washers
are nott over stressedd. Fig-6 showss a bolted jointt with
Belleville washher

Fig. 8. Dimensions
D
Of Belleville Washeers [3]

III. RESULTS BASED ON PRACTICAL EXPERIMENTS

Figg. 6. Belleville waasher in a bolted joint

j
[3]

The followinng method cann be used to calculate the sizze and

material properrties of Belleeville washerss (Refer Fig-007 &
m
Fiig-08). Load and stress are
a calculatedd according too the
foollowing equattions:
F((load)

h =

(1 -

M
MS(D

E f
) M(D

/ D

( h 2 )(h - f)(t

+ t 3 )

C2
/D 1 ) 2 (1 2 )
f
+
+
t
2
C1
EfC 1

w
where
F = load, lb
804

In
n the figures and
a the charts given below we are trying to
comp
pare results of
o the follow
wing busbar combinations
c
at
diffeerent loads vaarying from 1100 Ampere to
t 603 Ampeere.
Heree we are meaasuring voltagge drop across the contaccts.
Low
wer the voltagee drop lower w
will be the con
ntact resistance.
a) Bare Aluuminum busbaar and Bare Aluminum busbbar
b) Bare Coopper busbar aand Bare Copp
per busbar
c) Buffed Aluminum
A
buusbar and Bu
uffed Aluminuum
busbar
C
busbarr and Buffed Copper
C
busbaar
d) Buffed Copper
e) Silver plated
p
Coppeer busbar and
d Silver platted
copper busbars.
b
Siize of the bus bar in all the ccases:
Leength: 100 mm
m, Width: 30 m
mm, Thicknesss: 5 mm
Ov
verlap betweeen two conneccting busbars in
i all the cases:
Leength: 30 mm
m, Width: 30 m
mm
Bo
olts used for jointing busbaars:
Siize : M10 , MO
OC: MS with zink plated
To
orque applied while conneccting busbars in
i all the casess:
12 nM.
hus in all thesee cases we havve kept same overlap
o
betweeen
Th
the bus
b bars, appllied optimum pressure and
d used Belleviille
wash
hers. A thin film
f
of petrolleum coating was applied on
each
h contact surfaaces to avoid fformation of oxide
o
layer.

International Journal of Computer and Electrical Engineering, Vol. 3, No. 6, December 2011

Chart-1
Al-Al (Without Buffing)

Al-Al (With Buffing)

Current (A)

mV

Current (A)

mV

100

14.1

118

9.7

200

19.6

204

10.7

300

25.4

307

12

406

32.4

405

13.1

503

39.1

526

17.8

597

46.9

611

19.1
Fig. 11.

Chart-2
Cu-Cu (Without Buffing)

Cu-Cu (With Buffing)

Current (A)

mV

Current (A)

mV

103

13.5

105

9.3

205

17.3

208

10.6

310

21.7

300

11.8

405

26.2

404

14

506

31.1

510

15

601

35.7

600

15.9

Chart-3
Cu-Cu (Silver Coated)
Current (A)

Fig. 12.

mV

107

9.1

199

10.3

302

11.6

408

13.1

508

13.9

606

14.2

From the results shown in chart-1 we can see that joints

between two buffed aluminum busbars are better then joints
between two aluminum busbars which are not buffed.
In chart-2 we can see that joints between two buffed
copper busbars are better then joints between two copper
busbars which are not buffed. In chart-3 we can see that a
buffed copper joint can be further improved by silver plating
of the mating surfaces. This is shown in graphical form in
fig-9.In Fig-10 we have compared the results between buffed
aluminum joints and buffed copper joints.Fig-11 shows
supremacy of silver plated copper joints over buffed
aluminum joints and buffed copper joints.Fig-12 shows
overall comparison of the all five cases considered in the
experiment.

In figs 9,10,11,12 X axis of the curve is showing current

flowing through the bus bar joints and Y axis is showing
milli-volt drop taking place in the joints. Lesser the milli-volt
drop across a joint better will be the joint, as there will be
lesser contact resistance, lower temperature rise and lesser
power loss.

IV. CONCLUSION
The performance of contact joint is dependent on
maintaining low resistance. It is evident from the
experimental results that by just smoothening or buffing the
surface of both copper and aluminum busbars we can get very
low resistance across a joint. This can be further improved by
silver plating of the mating surfaces in copper busbars.
Minimum resistance in conductive joints can be achieved
by using silver plated Copper busbars as compared to
unplated bare bus bars while keeping other constraints like
overlap between the bus bars, applied optimum pressure, size
and type of washer, application of thin film of petroleum
coating on contact surfaces unchanged. Buffing also helps to
minimize contact resistance of the joints.
Stable and minimum contact resistance of joints will reduce
the need for frequent maintenance, decrease overall
downtime of equipment and maintenance costs and greatly
reduce the risk of catastrophic failures.

Fig. 9.

Fig. 10.

805

International Journal of Computer and Electrical Engineering, Vol. 3, No. 6, December 2011
Hitesh R. Jariwala received his B.E.(Electrical) degree
from S.V.Regional college of Engg. And Technology,
Surat, India in 1989 and M.Tech degree from Indian
Institute of Technology, Bombay, India in 2005 with
specialization in Power Electronics and Power System.
He is working as Associate Professor in Electrical
Engineering Department, S.V. National Institute of
Technology, Surat, India. His area of interest is Power
system Dynamics, HVDC and FACTS.

REFERENCES
S. S. J. Kindersberger and H. Lbl Joint Resistance of Busbar-Joints
with Randomly Rough Surfaces , Proceedings of the 21th Conference
on Electrical Contacts 2002, Zurich,
[2] A.K. Sawhney, A Course In Electrical Machine Design ,2006. ch. 2
[3] F. W. Kussy and J. L. Warren, Design Fundamentals For Low-Voltage
Distribution and Contro, Marcel Dekker Inc, 1987, pp.133-157..
[4] S. Bhattacharyya, A. Choudhury, and H.R. Jariwala, Department of
Electrical Engineering, S.V. National Institute of Technology,
Surat-395007, India. High quality joints of copper bus bars
International Journal of Engineering Science and Technology Vol.
2(8), 3808-3815. 2010.
[5] The Oxide Handbook, Ed. G. V. Samsonov, IFI/Plenum,N.Y.
[6] R Holm, Electrical Contacts, Springer-Verlag, New York (English
version) 1967.
[7] H. B. Chudnovsky, Degradation of Power Contacts in Industrial
Atmosphere: Silver Corrosion and Whiskers, Proceedings of the 48th
IEEE Holm Conference on Electrical Contacts,2002.
[8] R. L. Jackson, Significance of surface preparation for bolted
aluminium joints, IEE Proc. C, Gen, Trans. &Distrib., 128,(2), pp.
45-54, 1981.
[9] W. O. Freitag, Electric Contacts, Illinois Institute of
Technology,Chicago 1975, p. 17.
[10] J. L. Johnson and L. E. Moberley, Electrical Contacts, Illinois Institute
of Technology, Chicago, 1975, p. 53.
[11] S. M. Garte, Electric Contacts,Illinois Institute of Technology,1976, p.
65.
[12] Almen and Laszlo, Belleville Washers, Trans. ASME, Vol. 58, 1936
[1]

M Sharaschandra Shetty, received his BE in Electrical

Engineering from MIT Manipal, University of Mysore,
during the year 1984, Presently working as Head of
Electrical (Engineering and Maintenance) at Reliance
Industries Ltd., Hazira Manufacturing Division .

Rajulkumar Engineer received his BE in Electrical

Engineering from SVNIT- Surat, Gujarat. Presently
working as a Relay Testing and Protection Engineer at
Reliance Industries Ltd., Hazira Manufacturing Division

Anandita Chowdhury received her B.E. and M.E.

degree from University of Calcutta, and Ph.D. degree
from Indian Institute of Technology, Kharagpur.
Presently she is working as an Associate Professor in the
Department of Electrical Engineering of S. V. National
Institute of Technology, Surat, India. She is having
more than nineteen years of teaching experience. Her
area of research interest includes Electrical Machines,
Drives and Power system Stability.

Samarjit Bhattacharyya received his B.E. (Electrical)

degree from Jorhat Govt Engineering College Assam.
Presently he is working in Reliance Industries Limited,
HMD, Surat and also pursuing M-Tech in the
Department of Electrical Engineering of S. V. National
Institute of Technology, Surat. He is having more than
15 years of industrial experience.

806