Diagnosing a transformer tap-changer’s performance, in order to determine whether the transformer is still fit for service or needs to be taken out of service for repair, can be done in a number of ways. This article describes a dual assessment method which evaluates an oil sample and measures the tap changer’s dynamic resistance.
To improve the reliability of diagnostics applied to online tap changers (OLTCs), the TDA, or “tap changer dual assessment”, programme combines two tests. The first is a laboratory evaluation of an oil sample from a tap changer (TASA, or “tap changer activity signature analysis” provided by TJH2b laboratories). The second test is a dynamic resistance measurement, an off-line electrical diagnostic test obtained using DV Power’s tap changer analysers.
The title of the second test method, Dynamic Resistance Measurement, is somewhat misleading – DRM is a method of recording electrical current during winding resistance testing. Test current is recorded at a high sampling rate during this exercise, while the tap changer positions are varied from position 1 to the end and back. This method has been used as a diagnostic tool since 1994  on resistive-type OLTCs in Europe. DRM has proven to have very good diagnostic abilities.
Knowledge about performance features of a OLTC provided by a DRM test include [2, 3]:
Having assessments from two different angles (materials laboratory and electrical performance test) adds greater reliability to an evaluation in the following ways:
We will not elaborate on problems OLTC can cause to transformer operation, as there are many papers presented on this topic, like the one showing a German experience  with 40% participation. Maintenance interval is in tens of thousands of operations for OLTC, but a study referenced by Allard  showed 12% required maintenance before the manufacturer’s suggested period. Furthermore, P Kang, et al , show that about 33% of OLTC failures are caused by incorrect maintenance and poor reassembly. All these presentations prove the need for viable diagnostics of an OLTC either before or after any maintenance is performed.
Tap changer activity signature analysis (TASA)
Historically, early applications of dynamic gas analysis (DGA) to evaluate a load tap changer’s condition were based on experience with transformers. Threshold limits were developed for the gasses produced by overheating, both individually and in combination. Many factors such as design, operations, ventilation, and online filtration affect gas levels. Consequently, this gas threshold approach offered limited success but proved the potential usefulness of fluid testing for OLTC condition assessment.
Since gas data alone cannot provide sufficient information to fully assess an OLTC’s condition, new approaches were required for OLTC evaluations. The search for this new approach led to the development of TASA, which provides a condition assessment of the load path components.
In addition to providing useful information for the maintenance of insulating fluid, fluid assessment tests are used in conjunction with OLTC gas data to provide diagnostic information about the condition of on-load tap changers.
Keeping the oil free of water, arc decomposition products and other contaminants is essential for proper operation of the on-load tap changer. Particle profiling provides important information about the deterioration of materials which result in particle production. This includes information about in-service processes such as fluid degradation, contact deterioration and mechanical wear of moving parts and rust formation. Two of the most important fluid degradation processes to be evaluated are charring of oil and coke formation.
Dynamic resistance measurement (DRM)
The principle of tap changer operation is to make contact with the new position before breaking from the previous. Resistive and reactive tap changers are constructed differently and produce distinctly different DRM results.
Resistive tap changers
When a resistive-type tap changer switches from one position to the next, resistors are introduced into the circuit to minimise arcing and lower circulating current during the brief period when the tapped portion of the winding is shorted (Fig. 1).
Reactive (reactance-type) tap changers
The reactance-type tap changer has two arms which alternatively go from one position to the other and in this way operate in bridging and non-bridging positions. An autotransformer restricts circulating current, so this model can continuously operate in a bridging tap position.
Changing tap position from bridging (odd positions) to non-bridging (even positions) and vice versa provides characteristic fingerprints.
Several interesting cases have been collected which show the correlation between the TDA and problems found with the OLTCs.
Case study 1: transformer in 67 kV substation
This case study is of a transformer tested in a 67 kV substation in California. TASA indicated an overheating condition and contact problems. DRM was utilised to obtain further confirmation and localisation of this problem. Defects were found on the reversing switch and other contact assemblies.
The first test was performed to obtain the resistance value of the winding at all tap positions. Fig. 2 is a static resistance graph which shows values for all three phases at different test currents.
Case study 2: series transformer
Another overheating problem was discovered and attributed to extended transition times of a particular tap changer with a series transformer. Extended transition times caused resistors to overheat and generate combustible gases in excess of acceptable norms.
A Waukesha tap changer, type UZDRT made in 1996 and installed in an 84 MVA 138/26 kV transformer in a Wisconsin substation, was investigated in 2011. TASA results indicated gassing associated with overheated oil. The unit had been overhauled in September 2009 followed by TASA tests in August 2009, September 2010 and January 2011.
In April 2011 this unit was DRM tested and a newly patented technique was implemented to obtain the dynamic resistance graph on transformers equipped with a series transformer.
The series transformer is a separate three-phase transformer built in the tank of the large transformer itself; the objective being to minimise the current the tap changer has to carry by magnetically coupling it to the main winding. This magnetically coupled winding, where the tap changer is located, is not electrically connected to the transformer terminals and it is impossible to perform dynamic resistance measurement following a standard procedure. This investigation was the first ever test of this type.
To perform an OLTC condition assessment, a special procedure was devised to obtain the signature of the tap transitions.
Waukesha tap changer models UZD… are resistive types, i.e. resistors are inserted in the circuit of circulating current during the transition from one tap position to the other. These resistors are rated to withstand currents up to 600 A, but only for very short periods of time: typically 20 ms.
Similar constructions of resistive tap changers allow these resistors to heat up to 350°C for a short period of time. Although they are immersed in an oil that removes the heat quickly, if resistors are subjected to this high current for longer than normal, overheating occurs and methane, ethane and ethylene gasses are generated. Gas analysis results indicated that these key gasses were elevated.
The transformer showed a characteristic profile for both the heating of oil and the heating of solid insulation containing carbon to oxygen bonds.
The relationship of carbon oxide (CO) production is a characteristic of the heating temperature and the particular insulating material. Notice the insulating material that was burned by overheating adjacent to resistors in Fig. 6.
Time is of the essence
It is not surprising that resistive tap changers are called “fast tap changers”, because speed is important during transitions.
The graphs obtained in case study 2 (Fig. 7) indicate transition times (TT) from one position to another in the order of 60 ms (see Table 2).
These times are dependent on the adjustment of “switch fingers”, or contact fingers; a special device was made to achieve proper finger adjustment. Fig. 8 shows contact fingers with the rolling contacts in stationary position. The middle finger is touching the fixed contact. The right and left outer fingers, connected through the resistors R1 and R2, move over the fixed contacts only during the short transition time.
When resistors R1 and R2 are alone in the circuit, they carry load current, but when they are both inserted, during the middle transition period, circulating current heats them with additional energy (over and above load current).
Original design calculations of time and energy dissipation were 20/20/20 ms. The actual measured middle period (R1+R2, with circulating current) of 33 ms (Table 2) represents a 50% time increase over designed TT; i.e., a 50% increase in heating energy applied to the OLTC resistors.
Following all the data evaluation, opening the compartment revealed burn marks on the tap resistor epoxy support structure, as shown on the Fig. 6.
Checking the alignment of the contact fingers determined that some of them were more distant from centre than others; this increased distance increased the time these resistors were in the circuit.
Case study 3: substation LTC
TASA had been conducted on this substation transformer regularly over a couple of years. Results for this Reinhausen RMV-1 LTC showed acetylene increased to 1026 ppm and methane over
30 000 ppm.
These results showed an extremely abnormal dissipation of energy which indicated a terminal fault or wear activity. The dynamic resistance graph showed the characteristically problematic area associated with the X1 phase reversal switch and several other contact operations.
A visual inspection confirmed that there was an issue with the X1 reversing contacts. Heavy coking was observed on the slip-ring, upper stationary and moving reversing contact. Pitting on the stationary and slip-ring where the contact fingers normally sit on all three phases was also noted. This bank had undergone a substantial fault at some point in its life. No other issues were noted in the LTC compartment (see Figs. 9 to 14).
|R1 and R2||33 ms|
To remediate the problems with this tap changer, all the coking was cleaned from the slip-ring and the upper stationary and moving contacts on the X1 phase were replaced. Micro-ohm (µΩ) testing was performed on all three phases of the reversing contact assemblies. Testing verified that they were all less than 17 µΩ. The LTC compartment was cleaned, and new oil was installed.
Tabular data presentation
Table 3 provides an example of the way DRM results are represented numerically. Powerful DV Win software extracts important features from DRM graphs and presents them in tabular form for easier comparison and analysis.
|Date and time||Connection||Current [A]||R1(25°C) [mΩ]||R1(75°C) [mΩ]||V1 [mV]||Ripple %||Tap Position||Transition time [ms]|
|5/8/2010 13:58||1U – 1N||10,74||291,2437||347,36||3126,641||0||1||0|
|5/8/2010 13:59||1U – 1N||10,93||285,0616||339,9868||3116,955||8,5||2||46,8|
|5/8/2010 14:00||1U – 1N||11,15||279,0591||332,8277||3112,185||8,5||3||48,1|
|5/8/2010 14:00||1U – 1N||11,37||272,5099||325,0166||3097,266||8,4||4||46,8|
|5/8/2010 14:01||1U – 1N||11,57||266,7406||318,1357||3085,091||8,4||5||45,5|
|5/8/2010 14:02||1U – 1N||11,78||260,5021||310,6952||3067,431||8,8||6||45,7|
|5/8/2010 14:02||1U – 1N||11,99||254,1737||303,1474||3046,654||9,5||7||44,4|
|5/8/2010 14:03||1U – 1N||12,25||246,5858||294,0975||3019,804||10,4||8b||50,6|
|5/8/2010 14:04||1U – 1N||12,25||245,8175||293,1812||3010,838||11,2||9||48,5|
|5/8/2010 14:04||1U – 1N||12,52||239,5066||285,6543||2997,603||11||10||50,8|
|5/8/2010 14:05||1U – 1N||12,74||233,2372||278,1769||2972,571||10,9||11||48,8|
|5/8/2010 14:05||1U – 1N||13||227,2802||271,0721||2954,72||10,9||12||47|
|5/8/2010 14:06||1U – 1N||13,24||221,025||263,6117||2926,333||10,6||13||45|
|5/8/2010 14:07||1U – 1N||13,52||214,9339||256,347||2906,878||11,2||14||45,6|
|5/8/2010 14:07||1U – 1N||13,77||208,6241||248,8214||2873,16||12,2||15||45,7|
Each new test methodology must address not only the selection of proper parameters but even simple things like defining common terminology, outlining features that are of interest, avoiding inappropriate connections, methods of addressing substation interference, etc. Also, various test instruments in use around the world operate at different current values, with different sampling rates, with single-phase or three-phase recording. This makes a simple comparison more difficult requiring greater effort in the standardisation of procedures.
In October 2010, the AMforum association formed a working group of experts and practitioners with extensive experience in this field to exchange knowledge on OLTC testing methods and the interpretation of results. A workshop was organised in Madrid to review present practices . As a result of working together and sharing different methods of applying the test, the working group is in a position to make procedure recommendations, including the optimum technique for testing.
 HFA Verhaart: “Tussenrapport over faaloorzaken van distributie- en koppeltransformatoren op basis van de onvoorziene nietbeschikbaarheid.”, KEMA report 43613-T&D 94-102278, 1994.
 HFA Verhaart: “A diagnostic to determine the condition of the contacts of the tap changer in a power transformer.”, CIRED, Brussels, 1995.
 “LM Perea, AR Gómez, Unión Fenosa Distribución; and JLL Gómez, Norcontrol Soluziona: “Dynamic resistance measurements in LTC”, Proceedings of the EuroDoble Colloquium 2000.
 “Transformer failure causes in Germany”, Institute of electric power systems, Division of high voltage engineering, IEH Liebnitz University, Hanover.
 L Allard: Presentation on OLTC diagnostics, Euro TechCon, Chester, 2010.
 P Kang, D Birthwistle, et al: “Non-invasive on-line condition monitoring of OLTC”, IEEE Proceedings, January 2000.
 IBEKO Power: “Manual of Winding Resistance Ohmmeter RMO60TD”, February 2011.
 MR OLTC instruction manual.
 Report of the First Workshop on DRM, Madrid May 16, 2011.
 R Samsudin, A Berhanuddin, Y Zaidey, M Haneef, “Experience on Dynamic Contact Resistance Test on Eroded and Worn-Out Tap-Changer Contacts”, TNB Research, Malaysia.
 JJ Erbrink, R Leich, et al: “Reproducibility of Dynamic Resistance Measurement Results of On-Load Tap Changers – Effect of Test Parameters”, Delft University of Technology, 2010.
 Jur Erbrink: “OLTC Diagnosis on High Voltage Power Transformers using Dynamic Resistance Measurements”, Doctoral thesis, Delft University of Technology, 2011.
Contact Jacques Franken, Action Instruments, Tel 011 403-2247, firstname.lastname@example.org
Source: EE plublishers