31-05-2012, 02:09 PM
Thermal Overload Protection of Power Transformers Operating Theory and Practical Experience
Thermal Overload Protection of Power Transformers.pdf (Size: 577.85 KB / Downloads: 173)
Introduction
TRADITIONAL methods of protecting power transformers use functions based on measured
current and voltage. These functions are useful in detecting short circuits and other transient
electrical fault events in the transformer. However, for liquid-immersed power transformers, the
temperature of the winding hot-spot is the important factor in the long-term life of the
transformer. The insulating oil temperature is dependent on the winding temperature, and is used
to indicate the operating conditions of the transformer. Many numerical transformer protection
relays available today include protection functions that operate on insulating oil temperatures,
calculated loss-of-life due to high oil temperature, and predicted oil temperatures due to load.
These types of functions are not routinely applied, often since protection engineers may lack an
understanding of the operating principles of these functions, and transformer operating
conditions, to properly determine a settings methodology. A factor to consider when looking at
these temperature-based functions is the risk of accelerated aging, and transformer failure, is
increasing. Modern utility operating practices try to maximize the utilization of power
transformers, which may increase the occurrence of over-temperature conditions, and
transformer aging. Over-temperature conditions and accelerated aging are adverse system events
that must be identified and protected against.
The most common function provided for thermal protection of power transformers is the
thermal overload (ANSI 49) function. To properly set this function, the protection engineer must
understand the basics of the thermal performance of power transformers, and the basic design of
the specific implementation of the 49 function.
Northeast Utilities has implemented thermal protection of substation power transformers. The
temperature protection is combined with distribution automation to manage transformer load.
Thermal overload levels of the transformers force an automatic load transfer through feeder
circuit re-configuration. Predictive overload alarms warn the Distribution System Operators of
the pending automatic forced load transfer, to allow manual intervention. The settings criteria,
control logic, and operations criteria for the thermal overload protection are discussed, as well as
an overview of the operating experience.
Thermal Overload Protection of Power Transformers – Operating Theory and Practical Experience
2005 Georgia Tech Protective Relaying Conference
2
Pump
(optional)
MAIN TANK RADIATOR
Insulating Oil
Windings
Paper Insulation
Basics of Transformers Thermal Performance
THIS paper considers only the performance and protection of liquid-immersed power
transformers. Power transformers are transformers used between the generating stations and
distribution network, and are larger than 500 kVA in size. Power transformers use a laminated
steel core with copper or aluminum windings. The windings have a solid insulation of refined
paper, and highly refined mineral oil is the insulating and cooling medium for the entire
transformer. The core, windings, and insulation all have specific thermal capabilities. Losses in
the winding and core cause temperature rises in the transformer, which are transferred to the
insulating oil. Failure to limit these temperature rises to the thermal capability of the insulation
and core materials can cause premature failure of the transformer.
Figure 1: Simple transformer representation
A transformer is rated at the power output the transformer can continuously deliver at rated
voltage and frequency, without exceeding the specified temperature rise. This temperature rise is
based, in part, on the thermal limitations of the core, winding, and insulation. Therefore, the
MVA rating of the transformer is based on the maximum allowable temperature of the
insulation. Design standards express temperature limits for transformers in rise above ambient
temperature. The use of ambient temperature as a base ensures a transformer has adequate
thermal capacity, independent of daily environmental conditions.
Thermal Overload Protection of Power Transformers – Operating Theory and Practical Experience
2005 Georgia Tech Protective Relaying Conference
3
Transformer Heating
No-load losses and load losses are the two significant sources of heating considered in thermal
modeling of power transformers. No-load losses are made up of hysteresis and eddy loss in the
transformer core, and these losses are present whenever the transformer is energized. Hysteresis
loss is due to the elementary magnets in the material aligning with the alternating magnetic field.
Eddy currents are induced in the core by the alternating magnetic field. The amount of hysteresis
and eddy loss is dependent on the exciting voltage of the transformer.
Load losses are the more significant source of transformer heating, consisting of copper loss
due to the winding resistance and stray load loss due to eddy currents in other structural parts of
the transformer. The copper loss consists of both DC resistance loss, and winding eddy current
loss. The amount of loss is dependent on transformer load current, as well as oil temperature. DC
resistance loss increases with increasing temperature, while other load losses decrease with
increasing oil temperature. All of these factors are considered in calculations of thermal
transformer performance.
The basic method for cooling transformers is transferring heat from the core and windings to
the insulating oil. Natural circulation of the oil transfers the heat to external radiators. The
radiators increase the cooling surface area of the transformer tank. Pumps may be used to
increase the flow of oil, increasing the efficiency of the radiators. In non-directed flow
transformers, the pumped oil flows freely through the tank. In directed flow transformers, the
pumped oil is forced to flow through the windings. Forced air cooling is commonly applied on
large power transformers, using fans to blow air over the surface of the radiators, which can
double the efficiency of the radiators. For some large power transformers, water cooling may
replace large radiators. Large power transformers may also have additional ratings for multiple
stages of forced cooling. Normally, only two stages are applied, providing transformer ratings
equivalent to 133% and 167% of the self-cooled rating.
Both the IEEE and the IEC established standard designations for the various cooling modes of
transformers. The IEEE has adopted the IEC designations. The designation completely describes
the cooling method for the transformer, and the cooling method impacts the response of the
transformer insulating oil to overload conditions. Table 1 lists the common transformer cooling
designations.
Table 1: Transformer cooling designations
Old IEEE Cooling Designations IEC Equivalent
Self-cooled OA ONAN
Forced air cooled FA ONAF
Directed-flow forced liquid cooled FOA ODAF
Water cooled OW OFWF
Forced liquid and water cooled FOW OFWF
Impact of Oil Temperature on Power Transformers
INCREASING transformer load increases the temperature of the insulating oil, so loading above the
nameplate rating involves some risk. Transformers are rated at a maximum oil temperature rise
over ambient, with modern transformers rated at 65º C rise above ambient. These risks include
reduced dielectric integrity due to gassing, reduced mechanical strength and permanent
deformation of structural components such as the core and windings, or possible damage to
Thermal Overload Protection of Power Transformers – Operating Theory and Practical Experience
2005 Georgia Tech Protective Relaying Conference
4
auxiliary equipment such as tap changers, bushings, or current transformers. Oil temperature,
therefore, makes a good choice to use as the basis of a protection function, providing sensitivity
to a number of possible transformer issues. Standard temperature limits are defined in the IEEE
Guide for Loading Mineral-Oil Immersed Power Transformers, (described in the rest of this
paper as the Guide for Loading) are listed in Table 2.
Table 2: Standard temperature limits, 65º C rise transformer, 30º ambient temperature
Standard temperature limits
Average winding temperature rise 65º C Above ambient
Hot-spot temperature rise 80º C Above ambient
Top liquid temperature rise 65º C Above ambient
Maximum temperature limit 110º C Absolute
One factor in transformer over-temperature conditions is the loss of insulation life. Aging of
the refined paper insulation is based on temperature, moisture content, and oxygen content over
time. Modern oil preservation systems minimize the impact of moisture and oxygen on insulation
life. Therefore, aging studies of transformers use the hottest-spot oil temperature to determine
transformer life. [3]
The term “transformer life” is assumed to mean the insulation life of the transformer, not the
total operational life. “Loss-of-life” is assumed to mean loss of the total insulation life of the
transformer. For 65º C rise transformer operate at the maximum temperature, the Guide for
Loading uses 65,000 hours (7.4 years) as normal life expectancy, based on 50% retained
mechanical strength of the insulation. The Guide for Loading also states that 180,000 hours (20.6
years) is also a reasonable value for a normal life expectancy. This means, practically, that the
transformer can be operated at full load for 65,000 hours over the total operational life of the
transformer before the mechanical strength of the insulation is reduced by half, increasing the
likelihood of failure during short circuits. The relationship between oil temperature and
transformer life expectancy is given by the accelerating aging factor, FAA. FAA for 65º C rise
transformers is defined as: