You are on page 1of 7

Technical White Paper

Advancing the Accuracy of Sag-Tension Calculations


for Enhanced Line Designs and Performance
By Jacques Hamian

1.0 Introduction
Conductor stress-strain coefficients are used to determine sag-tension data, which is in turn used to
predict the performance of transmission and distribution lines under varying conditions. As a result, the
cost to construct and maintain a line is influenced by stress-strain coefficients and sag-tension
calculations. Designs based on data that does not reflect actual conductor performance can result in
needless expenditures for the deployment of an over-designed line, higher repair costs for an under-
designed line, or exposure to hidden safety hazards.

The Graphic Method has been widely used to successfully calculate sag-tension based on stress-strain
values that are estimated. To enhance the output of this proven method, Itron has developed a process to
calculate stress-strain coefficients. As this paper illustrates, these calculated coefficients more closely
reflect actual conductor performance when the lines are subjected to higher stress-stain conditions. When
used with the Graphic Method, the improved accuracy of the sag-tension values help generate cost-
effective line designs by optimizing the stress-strain utilization of conductors.

In addition to improving the accuracy of sag-tension data, this paper illustrates how the user-friendly
interfaces of TL-Pro™ programs make it easier to more accurately specify conductor parameters and
conductor variables. The accuracy of this input data provides designers with enhanced, easy-to-
understand, information about creep, tension, inclined spans, and low point sags. Collectively, the
refinements to inputs and expanded output data enable designers to maximize conductor capabilities,
while minimizing the cost to deploy a line design. These advancements in transmission line and design
are included in both TL-Pro Designer and TL-Pro Design Studio, collectively referred to in this paper as
“TL-Pro.”

2.0 Input Refinements


To predict the performance of installed conductors, TL-Pro requires two types of inputs: conductor
parameters and conductor variables. Conductor parameters, such as stress-strain coefficients, are
properties affiliated with a particular conductor type. Conductor variables refer to external factors that
influence the performance of a conductor once it is installed.

2.1 Conductor Parameters


TL-Pro is designed to minimize data entry. Conductor parameters are automatically defined once the
conductor is selected from a window. This list includes commonly used conductors including those made
of two components (e.g. ACSR), homogeneous conductors (e.g. AAC), and fictitious two-component
conductors made from a material “equivalent” to a single component conductor.

Unlike traditional sag-tension programs that use a database to define conductor properties, TL-Pro uses
values calculated by an auxiliary program developed by Itron, which seamlessly interfaces with TL-Pro.

1
Technical White Paper

The calculated properties associated with the selected conductor are used in analysis calculations and
include:
• Stress distribution between the outer and inner components.
• Conductor tension at higher temperatures, which is carried entirely by the inner component
resulting in smaller sags. This feature can be turned on or off at the discretion of the user.

As new conductors are released, the auxiliary program requires minimal input: conductor type, stranding,
and size (e.g. ACSR, 26/7, and MCM). Using this data, the program calculates the required properties.
These calculated values can be reviewed, modified, and stored for use by the sag-tension program.
Coefficients for new conductors are also calculated using known coefficients originating from third
parties, such as The Aluminum Association, ALCOA, and standard or conventional stress-strain-creep (S-
S-C) charts. This feature makes it faster and simpler to analyze new conductors.

2.1.1. Stress-Strain Properties


Traditionally, stress-strain (S-S) properties used in the Graphic Method have been predicted on
polynomials using four or five coefficients for each component. While these polynomials have
been accurate and satisfactory, as transmission line designers strive to create more cost-effective
networks, the predicted conductor performance at higher conductor strains derived by these
coefficients varies from the performance demonstrated by the conductor under laboratory testing.

1-hr S-S data for the Steel Portion of ACSR


Typical Coeffs Itron Coeffs Actual curve
250

200
Stress (Ksi)

150

100

50

0
0 1 2 3 4
Strain (%)

Graph 1: Comparison stress-strain values.

For example, Graph 1 illustrates the tested performance of an ACSR conductor. Predicted
behavior for this type of cable, as determined by traditionally-used polynomials, is also charted.
Next, the cable’s sag-tension characteristics are plotted based on data obtained using Itron’s

2
Technical White Paper

enhanced sag-tension program. As the graph illustrates, this proprietary program exactly maps to
the laboratory-tested cable performance. These more accurate stress-strain values are calculated
by the Itron program using either coefficients from third parties, standard laboratory test data, or
published cable information. Itron’s stress-strain values are then used with the Graphic Method to
output sag-tension values used in evaluating conductor performance in the installation
environment.

2.2 Conductor Variables


2.2.1 Ruling Spans & Actual Spans
Sag-tension programs universally provide sags and tensions only for specified ruling spans,
without consideration for the actual spans within the ruling span. TL-Pro differentiates between a
ruling span and the actual span range that can be encountered within each ruling span. It also
enables the design engineer to easily define the difference in elevation between the anchor points
of any actual span.

Usually, instead of making the necessary calculations for the actual span, transmission line
designers have simply substituted the ruling span sag and tension data for the actual span sag and
tension data. This substitution introduces unnecessary errors, inefficiencies, and/or increases costs
in the line design. Additionally, the actual spans are rarely, if ever, level spans as implied in the
ruling span calculations.

TL-Pro allows designers to input the ruling span range and adjust output values based on input
values associated with actual and inclined spans. These calculations are fully integrated with the
flow of the main program and do not require reruns or other manual calculations.

2.2.2 Stringing Table


The standard program output shows the stringing sags for a number of actual spans ranging
between the minimum and maximum expected actual spans within each ruling span. The design
engineer has the option of specifying the required number of intermediate spans.

2.2.3 Tolerances
The actual performance of a transmission line is dependent, in part, upon the ability of the design
program to consider installation tolerances. In the past, these sag-tension tolerances were often
predefined. When installation conditions were expected to vary from these tolerances, designers
had to manually adjust output, or process the output through an alternative program, to obtain a
more accurate representation of field performance, or elect to accept the inaccuracy. Using
TL-Pro, design engineers can define tolerances that are used by TL-Pro to determine sags,
tensions, temperatures, and arc lengths. This has eliminated the manual or supplemental
calculations and further enhanced modeling accuracy.

2.2.4 Annealing factor


Historically, the impact conductor annealing has on line designs has been calculated based on a
limited number of data points, which are extrapolated to predict performance at elevated
temperatures. TL-Pro uses the available conductor data and a proprietary algorithm to more
accurately account for the annealing factor. The accuracy of this data facilitates the creation of
lines designed to operate at elevated temperatures for extended periods of time and determines the
impact temperatures (and the annealing factor) will have on ground clearances after occurrence of
operation under emergency conditions.

3
Technical White Paper

3.0 Refined Outputs


Through the use of more accurate input parameters and variables, designers obtain enhanced information
about factors such as creep, tension, and inclined spans. This output provides designers with the
information needed to more accurately assess the true cost and performance of line designs.

3.1 Creep
The accuracy of the stress-strain coefficients provides additional options for creep calculations. In
addition to providing sag-tension information for the standard ten year creep, TL-Pro can determine creep
after any number of years of operation (refer to Graph 2). The time period can be specified by the design
engineer by changing one input parameter. This is a unique feature as it does not require new creep tests
or any modification in the conductor database. (Software packages are available to translate coefficients
from third parties and creep information from standard laboratory tests or from published information.)

itron 10-Yr itron 1-hr alcoa 1-hr alcoa 10-yr

1 -hr and 10-Yr S/S for complete Conductor


39,800

34,800

29,800 1-hr (Itron & Alcoa)

24,800
Stress (psi)

19,800

14,800

9,800

4,800
10-Yr (Itron & Alcoa)
-200
-0.01 0.09 0.19 0.29 0.39 0.49 0.59 0.69 0.79
Strain (%)

Graph 2: Comparison data predicted using the Graphic Method and Itron calculations for
one hour, and ten years.

3.2 Tension
Traditional programs provide only the conductor tensions and leave the calculation of the structure loads
to the design engineer. TL-Pro provides the vertical, transverse, and longitudinal components of the
tensions at the upper and lower points of support for specified actual spans, and initial and final
conditions. In particular, the vertical load at the lower point of support indicates whether or not the
structure is under uplift conditions.

By recognizing that the conductor tensions are different at various locations along a catenary, TL-Pro
applies the specified tension limits to the maximum conductor tension (T-tension) in each ruling span,
while evaluating the conductors’ S-S-C properties for the average tension (P-tension) in each ruling span.

TL-Pro provides two options for calculating P-tension:


• Conventional: Typically, tension output is calculated for the low point of the catenary using the
horizontal tension (H-tension) and the conductor tension (T-tension). This alternative is inaccurate
in up-lift conditions, where the low point in the catenary is imaginary (outside the catenary).

4
Technical White Paper

• Accurate: This tension output views the catenary as many sections, or nodes. For each node, TL-
Pro determines the conventional tension, and then averages the conventional tensions for all nodes
along the catenary to derive the accurate P-tension.

3.3 Inclined Spans & Low Point Sags


Traditional sag-tension programs assume level ruling spans, where low point sags, sighted sags, and mid-
span sags are one and the same. In the real world, spans are characterized by differences in elevation
between the two points of support. TL-Pro recognizes this and provides the low point sags and the sighted
sags for each span (the mid-span sags are almost identical to the sighted sags, even for inclined spans).
For inclined spans, the program provides a warning if there are uplift conditions and identifies the sags at
the low point of the catenary, as well as the sighted sags. Under uplift conditions, the low point sags
become imaginary and only the sighted sags have significance.

4.0 Calculations
This section illustrates the non-proprietary calculations used to derive some of the values discussed in this
paper.

Define State1 (Initial condition)


1. Define the load case of temperature (t1), ice thickness (i1), wind pressure (w1), and NESC k-
factor (k1), corresponding to State1.
2. Select a conductor strain x1 corresponding to 1.
3. Calculate the conductor tension P1 corresponding to steps 1 and 2 above:
4 4
P1 = InitialF(x1,t1) = ∑ Aoz(x1) * x1z + ∑ Aiz(xi(x1,t1)) * xiz(t1)
z =0 z =0
4. Calculate the horizontal component, H1, of the tensions in the ruling span corresponding to P1:
S* W1r 0.25*S* W1r 0.5*S* W1r
P1 = 2 * H1* cosh + 2 * H1* cosh + H1* cosh where
2 * H1 2 * H1 2 * H1
W1r is the resultant unit weight of the conductor in state1, and
S is the ruling span
2 * H1 W1r *S
5. Calculate the conductor arc length, L1, corresponding to H1: L1 = *sinh
W1r 2 * H1
Calculate State2 (Initial condition)
6. Define the load case of temperature (t2), ice thickness (i2), wind pressure (w2), and NESC k-
factor (k2), corresponding to State2.
L2 − L1
7. State2 is calculated: = x2 − x1 + (t2 − t1) * α (x2, x1)
L1
where
L2 and x2 are the conductor arc length and strain respectively in state2
(L2 calculated through steps 2 through 5 above)
Calculate Final condition corresponding to the initial condition
8. Calculate initial condition at the everyday temperature (no load), StateEDTinitial, using steps 2
through 7 above: Defined through a conductor strain xEDTinitial and length of LEDTinitial.
9. Select a final conductor tension at the everyday temperature, PEDTfinal then calculate its
corresponding initial strain, x’EDTinitial, at the everyday temperature (no load) through:
PEDTfinal = InitialF(x'EDTinitial,EDT)

5
Technical White Paper

4
10. Calculate the creep corresponding to PEDTfinal: Creep(PEDTfinal) = ∑ Cz * PEDTfinalz
z =0
11. Calculate LEDTfinal using steps 4 and 5 above.
12. Verify the accuracy of the selection obtained in step 9 above through:
LEDTfinal − LEDTinitial
= Creep(PEDTfinal) + x'EDTinitial − xEDTintial
LEDTinitial
repeat steps 9 through 12 above to find the final unloaded EDT condition after creep.
13. Calculate the final modulus of elasticity of the conductor:
MOEcnd = ho*Eo+hi*Ei
where
ho is the area of the outer component as a percent of the total conductor area,
Eo is the modulus of the outer component
(similarly hi and Ei are the same parameters for the inner component)
14. Calculate the strain, xmax, that satisfies the following equation:
InitialF(xmax,EDT) = MOEcnd *[xmax − (Creep(PEDTfinal) + x'EDTinitial)]

15. Calculate the final stress-strain properties of the conductor:


1 1
P1final = FinalF(x1,t1) = Compressiono(x1) * ∑ Koz * x1z + Compressioni(xi(t1)) * ∑ Kiz * xiz(t1)
z =0 z =0
where
x1 is the strain of the outer component, final condition,
xi is the strain of the inner component, which is a function of t1.
Calculate State2 (Final condition)
16. Define the load case of temperature (t2), ice thickness (i2), wind pressure (w2), and NESC k-
factor (k2), corresponding to State2.
17. State2 is calculated through:

L2 − L1
= x2 − x1 + (t2 − t1) * α (x2, x1) where L2 and x2 are the conductor arc length and strain respectively in
L1
state2 (L2 calculated through steps 2. to 5. above), α is the coefficient of linear expansion of the conductor,
which is a function of of x2 and x1.

5.0 Conclusion
Since the introduction of the Graphic Method, the only significant advancement has been to computerize
the method. While the method is a crucial for transmission line design, Itron’s work has enhanced
the method, providing designers with a tool for today’s design environment.

These enhancements have been achieved by refining program inputs related to the conductor’s
parameters and variables: stress-strain, ruling spans, actual spans, stringing table, regularly and
rarely occurring loads, tolerances, and the annealing factor. As a result of refinements in these
areas, the TL-Pro software programs more accurately model the characteristics of the conductor
over its installed life. This output is essential for optimizing the design and operation of transmission
lines.

6
Technical White Paper

In addition to improving conductor behavior estimates, TL-Pro is an unprecedented software


tool. Whether using TL-Pro Design Studio or TL-Pro Designer, the software leverages the
advantages of computer technology to streamline the iterative design process, drastically
reducing the time it takes to evaluate various design options. The program’s ability to interface
with data sources, Graphic Method programs, and libraries helps eliminate errors and create on-
demand solutions. User-friendly interfaces create a tool suited for use by a variety of skill levels
and provides a single application program that can be used throughout the concept-to-completion
design process.

References
1. Theodore Varney, “ACSR Graphic Method for SAG-TENSION Computations,” Aluminum
Company of America, Pittsburgh, Pennsylvania, 1927.
2. Aluminum Company of America (ALCOA), “Graphic Methods for Sag-Tension
Calculations,” 1926, 1963.
3. Alcan, “Graphic Methods for Sag-Tension Calculations,” 1950.
4. Southwire Company, “Overhead Conductor Manual,” 1994.
5. Paul F. Winkelman, “Sag-Tensions Computations and Field Measurements of Bonneville
Power Administration,” American Institute of Electrical Engineers, paper 59-900, 1959.

Author Biography
Jacques Hamian is a principal engineer in the Transmission and Distribution division of Itron.
He has extensive experience with all aspects of electrical conductors, including conductor
design, conductor manufacturing, line design, and line construction. His work with the
analytical aspects of conductors includes detailed investigation of sag-tension behavior, Aeolian
vibration, galloping, structural loads, voltage gradients, corona, skin effect lightning effects, and
other phenomona. During his career he has worked with ALCAN, BC Hydro, Manitoba Hydro,
and as an independent consultant.

Itron Inc.
Itron is a leading technology provider and critical source of knowledge to the global energy and water
industries. Nearly 3,000 utilities worldwide rely on Itron technology to deliver the knowledge they require
to optimize the delivery and use of energy and water. Itron delivers value to its clients by providing
industry-leading solutions for meter data collection, energy information management, demand side
management and response, load forecasting, analysis and consulting services, transmission and
distribution system design and optimization, web-based workforce automation, C&I customer care,
as well as enterprise and residential energy management.

To know more, start here: www.itron.com

Itron Inc.
Corporate Headquarters
2818 North Sullivan Road
Spokane, Washington, U.S.A. 99216
Phone: 1.800.635.5461
Fax: 1.509.891.3355

You might also like