Visual Inspection of Transformers

Visual inspection of the transformer exterior reveals important condition information. For example, valves positioned incorrectly, plugged radiators, stuck temperature indicator and level gauges, and noisy oil pumps or fans. Oil leaks can often be seen which indicate a potential for oil contamination, loss of insulation, or environmental problems. Physical inspection requires staff experienced in these techniques. There are several required physical inspections that will not be covered here because they were addressed previously. Be sure to inspect the following: winding temperature indicators, pressure relief devices, sudden pressure relay, conservator bladder, conservator breather, and bladder failure relay.

Oil Leaks

Check the entire transformer for oil leaks. Leaks develop due to gaskets wearing out, ultraviolet exposure, taking a “set,” or from expansion and contraction, especially after transformers have cooled, due to thermal shrinkage of gaskets and flanges. Many leaks can be repaired by applying an epoxy or other patch. Flange leaks may be stopped with these methods, using rubberized epoxy forced into the flange under pressure. Very small leaks in welds and tanks may be stopped by peening with a ball-peen hammer, cleaning with the proper solvent, and applying a “patch” of the correct epoxy. Experienced leak mitigation contractors whose work is guaranteed may also be employed. Some leaks may have to be welded. Welding may be performed with oil in the transformer if an experienced, qualified, and knowledgeable welder is available. If welding with oil in the tank is the method chosen, oil samples must be taken for DGA, both before and after welding. Welding may cause gases to appear in the DGA and it must be determined what gases are attributed to welding and which ones to transformer operation.

Oil Pumps

If the transformer has oil pumps, check the flow indicators and pump isolation valves to ensure that oil is circulating properly. Pump motor(s) may also have reversed rotation, and flow indicators may still show that oil is flowing. To ensure motors are turning in the proper direction, use an ammeter to check the motor current. Compare results with the full-load-current indicated on the motor nameplate. If the motor is reversed, the current will be much less than the nameplate full-load-current. Check oil pumps with a vibration analyzer if they develop unusual noises. Have the DGA lab check for dissolved metals in the oil and run a metal particle count for metals, if the bearings are suspect. This should be performed as soon as a bearing becomes suspect. Bad oil pump bearings can put enough metal particles into the oil to threaten transformer insulation and cause flashover inside the tank, resulting in an explosive catastrophic failure of the transformer tank.

Fans and Radiators

Inspect all isolation valves at the tops and bottoms of radiators to ensure they are open. Inspect cooling fans and radiators for cleanliness and fans for proper rotation. Check for dirty or damaged fan blades or partially blocked radiators. Fans are much more efficient if the blades are clean and rotating in cool air. Normally, fans blow cool air through the radiators, they should not be pulling air through. Check to see if fans are reversed electrically (i.e., pulling air first through the radiators and then through the fan blades). This means that the blades are rotating in warm air after it passes through the radiator, which is much less efficient. Place a hand on the radiator opposite the fans, air should be coming out of the radiator against your hand. Watch the blades as they rotate slowly when they are starting or stopping to determine which way they should be rotating and correct the rotation if necessary. See IEEE 62-1995™. Also inspect radiators and fans with an infrared camera.

Age

Transformer age is an important factor to consider when identifying candidates for replacement or rehabilitation. Age is one indicator of remaining life and upgrade potential to current state-of-the art materials. During transformer life, structural strength and insulating properties of materials used for support and electrical insulation (especially paper) deteriorate. Although actual service life varies widely depending on the manufacturer, design, quality of assembly, materials used, maintenance, and operating conditions, the expected life of a transformer is about 40 years.

Infrared Temperature Analysis

Infrared analysis should be conducted annually while equipment is energized and under full load, if possible. IR analysis should also be conducted after any maintenance or testing to see if connections that were broken were re-made properly. Also, if IR is performed during factory heat run, the results can be used as a baseline for later comparison.

IR for Transformer Tanks

Unusually high external temperatures or unusual thermal patterns of transformer tanks indicate problems inside the transformer, such as low oil level, circulating stray currents, blocked cooling, loose shields, tap changer problems, etc. Infrared scanning and analysis is required annually for trending purposes by National Fire Protection Association. See also IEEE 62-1995™. Abnormally high temperatures can damage or destroy transformer insulation and, thus, reduce life expectancy. Thermal patterns of transformer tanks and radiators should be cooler at the bottom and gradually warmer ascending to the top. See Picture 1 for a normal pattern; the red spot at the top is normal showing a “hot spot” top of B phase, about 110 degrees Fahrenheit (°F). Any departure from this pattern means a probable problem which must be investigated. An IR inspection can find over-heating conditions or incorrect thermal patterns. IR scanning and analysis requires trained staff, experienced in these techniques.

Picture 1: Normal Transformer IR Pattern

IR for Surge Arresters

Surge arresters should be included when infrared scanning energized transformers. Look for unusual thermal patterns on the surface of lightning arresters (see the arrester IR image in Picture 2). Note that the yellow in the top right of the image is a reflection not associated with the arrester. A temperature profile of the arrester is shown as black lines. Note the hot spot (yellow) about a third of the way down from the top. This indicates that immediate de-energization and replacement must be undertaken. Catastrophic failure is imminent which can destroy nearby equipment and be hazardous to workers. Also compare thermal patterns to sister units or earlier scans of the same arrester. Scan all high-voltage connections and compare them to nearby connections for unusual temperatures.

Picture 2: IR Image of Defective Arrester

IR for Bushings

IR scans of bushings can show low oil levels which would call for immediate de-energization and replacement. This generally means that the seal in the bushing bottom has failed, leaking oil into the transformer. The top seal has probably also failed, allowing air and up the bushing. Another reason a bushing can exhibit a high oil level is that the top seal is leaking, allowing water to enter. The water migrates to the bushing bottom, displacing the oil upward. Remember, over 90% of bushing failures are attributed to water entrance through the top seal. Bushings commonly fail catastrophically, many times destroying the host transformer or breaker and nearby equipment and causing hazards to workers. Picture 3 shows low-oil level in a high-voltage transformer bushing. Compare previous IR scans of the same bushing with the current scan. Doble hot-collar testing possibly may show this problem. However, Doble tests are run infrequently, and the transformer has to be out of service, under clearance, and both primary and secondary conductors removed, while an IR scan easily can be performed at any time.

Picture 3: IR Image of Defective Bushing

IR for Radiators and Cooling Systems

Examine radiators with an IR camera and compare them with each other. A cool radiator or segment indicates that a valve is closed or the radiator or segment is plugged. The IR image (Picture 4) shows that the cold left radiator section is valved off or plugged. If visual inspection shows the valves are open, the radiator or segment must be isolated, drained, removed, and the blockage cleared. Do not allow a transformer to operate with reduced cooling which drastically shortens transformer life. Remember, an increased operating temperature of only 8 to 10 °C will reduce transformer life by one-half. IR scan all cooling systems, including heat exchangers, fans, pumps, motors, etc. Check inside control panels for overloaded wiring, loose connections, and overheated relays. Look for unusual thermal patterns and compare similar equipment.

Picture 4: IR Image Showing Blocked Radiators

Corona Scope Scan

With the transformer energized, scan the bushings and surge arresters and all high-voltage connections for unusual corona patterns. Corona should be visible only at the top of bushings and arresters, and corona at connections should be similar to sister connections. As a bushing deteriorates due to physical defects, the corona pattern will grow progressively larger. When the corona pattern reaches a grounded surface (i.e., the tank or structure), a flashover will occur, destroying the bushing or arrester and, perhaps, the transformer. The corona scope will reveal this problem long before a flashover.

Sweep Frequency Response Analysis (SFRA) Tests

These tests show, in trace form, the winding transfer function of the transformer and are valuable to determine if any damage has occurred during shipping or during a through fault. Core grounds, core displacement, and other core and winding problems can be revealed by this test. These tests should be conducted before and after the transformer has been moved or after experiencing a through fault. Results should be compared to baseline tests performed at the factory or as soon as possible after receiving the transformer. If the SFRA tests cannot be performed at the factory, they should be conducted as an acceptance test before energizing a new or rebuilt transformer to establish a baseline. A baseline should be established for older inservice transformers during a normal Doble test cycle. If possible, one should use the same test equipment for baseline and following tests, or the results may not be comparable.

CAUTION: Do not attempt to run a sweep frequency response analysis test immediately after any DC test. Energizing with DC will leave a residual magnetism in the core and will offset the results of the SFRA test.

For a delta/wye transformer, a test voltage of variable frequency (normally 20 hertz [Hz] to 2 megahertz [MHz]) is placed across each phase of the high-voltage winding. With this set of tests, low-voltage windings are isolated with no connections on any of the bushings. An additional set of tests is performed by short circuiting all the low-voltage windings and again placing the test voltage on each phase of the high-voltage winding. A third set of tests is made by isolating the high-voltage winding and placing the test voltage across each low-voltage winding. See the M5100 SFRA Instrument Users Guide, Doble Engineering Company for connection details. Picture 1 shows test traces on a new three-phase transformer. The top three traces were taken on the low-voltage side: X1-X3, X2-X1, and X3-X2. The two outside windings (A and C phases) have the same general shape with a “W” at the lowest point of the trace, while the inside winding (B phase) has a “V” at the bottom. The high-voltage traces (lower three) have the same characteristics. Note that, in both high- and low-voltage tests, the traces fall almost perfectly on top of each other for the outside windings (A and C), while the inside winding (B phase), is slightly displaced to the left. These are characteristic traces of a three-phase transformer in good condition. These traces will be the baseline for future tests on this transformer.

Picture 1: SFRA Test Traces of a New Transformer


By comparing future traces with baseline traces, the following can be noted. In general, the traces will change shape and be distorted in the low frequency range (under 5,000 Hz) if there is a core problem. The traces will be distorted and change shape in higher frequencies (above 10,000 Hz) if there is a winding problem. Changes of less than 3 decibels (dB) compared to baseline traces are normal and within tolerances. From 5 Hz to 2 kilohertz (kHz), changes of + or – 3 dB (or more) can indicate shorted turns, open circuit, residual magnetism, or core movement. From 50 Hz to 20 kHz +/- 3 dB (or more), a change from baseline can indicate bulk movement of windings relative to each other. From 500 Hz to 2 MHz, changes of +/- 3 dB (or more) can indicate deformation within a winding. From 25 Hz to 10 MHz, changes of +/- 3 dB (or more) can indicate problems with winding leads and/or test leads placement. The above diagnostics come from the Euro-Doble Client Committee after much testing experience and analysis. Note that there is a great deal of overlap in frequencies, which can mean more than one diagnosis.

Picture 2: SFRA Test Traces of a Defective Transformer


Picture 2 shows traces of a transformer with a problem. This transformer test results are used for illustration only. The traces have the same general positions on the graph as the good transformer. The lower traces are high-voltage winding tests, while the upper traces are the low-voltage winding tests. Note in the higher frequencies of the low-voltage traces that “A” phase (X1-X0 green trace) is displaced from the other two phases more than 3 dB from about 4 kHz to about 50 kHz. With a healthy transformer, these would fall almost on top of each other as the other two phases do. Also notice that “A” phase (H1-H3 Lsh) is displaced in the test with the low-voltage winding shorted. There is an obvious problem with “A” phase on the low-voltage side. After opening the transformer, it was found that the “A” phase winding lead had burned off near the winding connection and re-welded itself on the winding at a different location, effectively shorting out a few turns. The transformer was still working, but hot metal gases (ethylene, ethane, methane) were actively generating and showing up in the DGA. Although other tests could have revealed this problem, SFRA showed the problem was with “A” phase and, therefore, where to concentrate the internal inspection.

Transformer Testing

Transformer testing falls into three broad categories: factory testing when the transformer is new or has been refurbished, acceptance testing upon delivery, and field testing for maintenance and diagnostic purposes. Some tests at the factory are common to most power transformers, but many of the factory tests are transformer-specific. Not all of the tests are performed at the factory, and not all of them are performed in the field. Each transformer and each situation is different, requiring its own unique approach and tests.

DC Winding Resistance Measurement

If generation of ethylene, ethane, and perhaps methane in the DGAs indicates a poor connection, winding resistances should be checked. Turns ratio, sweep frequency response analysis (SFRA), Doble tests, or relay operations may give indications that dc testing is warranted. Winding resistances are tested in the field to check for loose connections on bushings or tap changers, broken strands, and high contact resistance in tap changers. Results are compared to other phases in wye-connected transformers or between pairs of terminals on a delta-connected winding to determine if a resistance is too high. Resistances can also be compared to the original factory measurements or to sister transformers. Agreement within 5% for any of the above comparisons is considered satisfactory. If winding resistances are to be compared to factory values, resistance measurements will have to be converted to the reference temperature used at the factory (usually 75 °C). To do this, use the following formula:

Rs = Rm * (Ts + Tk)/(Tm + Tk)

Rs = Resistance at the factory reference temperature (found in the transformer manual);
Rm = Resistance you actually measured;
Ts = Factory reference temperature (usually 75 ºC);
Tm = Temperature at which you took the measurements;
Tk = Constant for the particular metal the winding is made from: 234.5 ºC for copper, 225 ºC for aluminum.


It is very difficult to determine actual winding temperature in the field, and, normally, this is not needed. The above temperature corrections are necessary only if resistance is to be compared to factory values. Normally, phase resistances are compared to each other or to sister transformers at the same temperature, and actual winding temperatures and corrections are not needed. Compare winding resistances to factory values; change in these values can reveal serious problems. A suggested method to obtain an accurate temperature is outlined below. If a transformer has just been de-energized for testing, the winding will be cooler on the bottom than the top, and the winding hot spot will be hotter than the top oil temperature. The average winding temperature is needed, and it is important to get the temperature as accurate as possible for comparisons.

CAUTION: Do not attempt to run an excitation current test immediately after any direct current (DC) test. Energizing with dc will leave a residual magnetism in the core and will ruin the results of the excitation current test.

The most accurate method is to allow the transformer to sit deenergized until temperatures are equalized. This test can reveal serious problems, so it’s worth the effort. Winding resistances are measured using a Wheatstone Bridge for values of 1 ohm or above and using a micro-ohmmeter or Kelvin Bridge for values under 1 ohm. A multi-amp (now AVO) makes a good instrument for these measurements, which is quick and easy to use. Take readings from the top of each bushing to neutral for wye-connected windings and across each pair of bushings for deltaconnected windings. If the neutral bushing is not available on wyeconnected windings, take each one to ground (if the neutral is grounded) or take readings between pairs of bushings as if it were a delta winding. Be consistent each time so that a proper comparison can be made. The tap changer can also be changed from contact to contact, and the contact resistance can be checked. Make sure to take the first test with the tap changer “as found.” Keep accurate records and connection diagrams so that later measurements can be compared.

Core Insulation Resistance and Inadvertent Core Ground Test (Megger®)

Core insulation resistance and core ground test is used if an unintentional core ground is suspected; this may be indicated by the DGA. Key gases to look for are ethane and/or ethylene and possibly methane. These gases may also be present if there is a poor connection at the bottom of a bushing or a bad tap changer contact.

CAUTION: Do not attempt to run excitation or SFRA tests on a transformer immediately after using dc test equipment. Residual magnetism will remain in the core and ruin the excitation current and SFRA test results.

Therefore, this test is only necessary if the winding resistance test shows all connections and tap changer contacts in good condition. The intentional core ground must be disconnected. This may be difficult, and some oil may have to be drained to accomplish this. On some transformers, core grounds are brought outside through insulated bushings and are easily accessed. A standard dc Megger® (1,000-volt Megger® is recommended) is then attached between the core ground lead (or the top of the core itself and the tank [ground]). The Megger® is used to place a dc voltage between these points, and the resistance is measured. A new transformer should read greater than 1,000 megohms. A service-aged transformer should read greater than 100 megohms. Ten to one-hundred megohms is indicative of deteriorating insulation between the core and ground. Less than 10 megohms is sufficient to cause destructive circulating currents and must be further investigated. A solid, unintentional core ground may read zero ohms; this, of course, causes destructive circulating currents and must be corrected before energization. Some limited success has been obtained in “burning off” unintentional core grounds using a dc or ac current source. This is a risky operation, and the current may cause additional damage. The current source is normally limited to a maximum of 40 to 50 amps and should be increased slowly to use as little current as possible to accomplish the task. This should only be used as a last resort and then only with consultation from the manufacturer, if possible, and with others experienced in this task.

CAUTION: This will generate gases which will be dissolved in the oil and will show up in the DGA! Take a sample for DGA with in 72 hours after burning off the unintentional core ground and compare this DGA with the prior one to determine what gases were created by this task.

Doble Tests on Insulation

Doble testing is important to determine the condition of a transformer, because it can detect winding and bushing insulation integrity and problems in the winding and core. Doble tests are conducted in the field on de-energized transformers using special test equipment. Generally, a Doble M-4000 test set is used along with accompanying software. The software automatically performs analysis of test results and responds with a four letter code: G = Good, I = Investigate, D = Deteriorated, and B = Bad. These codes refer to insulation quality. If a “D” or “B” code is encountered, the insulation should be re-tested, carefully investigated, and the problem explained before re-energizing. Other tests may have to be performed; and, perhaps, an internal inspection should be considered before the unit is re-energized. The Doble Company should be consulted, along with the transformer manufacturer, and other transformer experts. If the problem is severe, the unit may have to be taken out of service.

Insulation Power Factor Test

The purpose of this test is to determine the state of dryness of the windings and insulation system and to determine a power factor for the overall insulation, including bushings, oil, and windings. It is a measure of the ratio of the power (I2R) losses to the voltamperes applied during the test. The power factor obtained is a measure of watts lost in the total transformer insulation system, including the bushings. The power factor should not exceed 0.5% at 20 EC. Temperature correction of test results can be performed automatically on the Doble test set. The watts lost should not exceed one-half of one percent of the total power input (voltamperes) from the test. The values obtained at each test are compared to previous tests and baseline factory tests, and a trend can be established as the insulation system ages.

Capacitance Test

This test measures and records the capacitance (including bushings) between the high and low-voltage windings, between the high-voltage winding and the tank (ground), and between the low-voltage winding and the tank (ground). Changes in these values as the transformer ages and events occur, such as nearby lightning strikes or through faults, indicate winding deformation and structural problems, such as displaced wedging and winding support.

Excitation Current Test

The purpose of this test is to detect short-circuited turns, poor electrical connections, core de-laminations, core lamination shorts, tap changer problems, and other possible core and winding problems. On three-phase transformers, results are also compared between phases. This test measures current needed to magnetize the core and generate the magnetic field in the windings. Doble software only gives two indications on this test: “G” for good and “Q” for questionable. On a three-phase, wye/delta or delta/wye transformer test, the excitation current pattern will be two phases higher than the remaining phase. Compare the two higher currents only. If the excitation current is less than 50 milliamps (mA), the difference between the two higher currents should be less than 10%. If the excitation current is more than 50 mA, the difference should be less than 5%. In general, if there is an internal problem, these differences will be greater. When this happens, other tests should also show abnormalities, and an internal inspection should be considered. The results, as with all others, should be compared with factory and prior field tests.

CAUTION: Perform the excitation test before any DC tests. Excitation current tests should never be conducted after a DC test has been performed on the transformer. Results will be incorrect because of residual magnetism of the core left from the DC tests.

Bushing Tests

For bushings that have a potential tap, both the capacitance between the top of the bushing and the bottom tap (normally called C1), and the capacitance between the tap and ground (normally called C2) are measured. To determine bushing losses, power factor tests are also performed. C2 capacitance is much greater than C1. Bushings without a potential tap are normally tested from the bushing top conductor to ground and Ahot collar@ tests. These test results are compared with factory tests and/or prior tests to determine deterioration. About 90% of bushing failures may be attributed to moisture ingress, evidenced by an increasing power factor from Doble testing on a scheduled basis.

Percent Impedance/Leakage Reactance Test

This is normally an acceptance test to see that nameplate percent impedance agrees with the measured percent impedance when the transformer arrives onsite. Normally, a 3% difference is considered acceptable. However, after the initial benchmark test, the percent impedance should not vary more than 2% from benchmark. As the transformer ages or suffers events such as through faults, nearby lightning strikes, and other surges, this test is used in the field to detect winding deformation. Winding deformation can lead to immediate transformer failure after a severe through fault, or a small deformation can lead to a failure years later.
Percent impedance/leakage reactance testing is performed by short circuiting the low-voltage winding and applying a test voltage to the high-voltage winding. Reluctance is resistance to lines of magnetic flux. Reluctance to the magnetic flux is very high in spaces between the high- and low-voltage windings and spaces between the windings and core. Reluctance is very low through the magnetic core so that the vast majority of total reluctance is in the spaces. When winding movement (distortion) occurs, these spaces change. Therefore, the reluctance changes, resulting in a change in the measured leakage reactance. Changes in leakage reactance and in capacitance tests serve as an excellent indicator of winding movement and structural problems (displaced wedging, etc.). This test does not replace excitation current tests or capacitance tests but complement them, and they are used together. The excitation current test relies on reluctance of the core while the leakage reactance test relies on reluctance of the spaces. See Doble’s Leakage Reactance Instrument Users Guide, and IEEE®, Guide for Diagnostic Field Testing of Electric Power Apparatus-Part 1: Oil-Filled Power Transformers, Regulators, and Reactors (IEEE 62-1995™).

Gas Limits

Table shown on Picture 1 shows IEEE limits, compared with Doble in a study of 299 operating transformers. The table of gases from the Doble study seems more realistic, showing gas level average of 95% of transformers in the study. Note, with the last four gases, limits given by the IEEE (trial use guide) run over 70% higher than the Doble 95% norms. But with the first three gases, hydrogen, methane, and ethane, the IEEE limits are well below the amount of gas found in 95% norms in the Doble study. We obviously cannot have limits that are below the amount of gas found in normal operating transformers. Therefore, it is suggested that we use the Doble (95% norm) limits. The 95% norm limit means that 95% of the silicone oil transformers studied had gas levels below these limits. Obviously, 5% had gases higher than these limits. These are problem transformers that we should pay more attention to.

Picture 1: Comparison of Gas Limits


In table shown on Picture 2, the IEEE limits for L1 were chosen. For L2 limits, a statistical analysis was applied, and two standard deviations were added to L1 to obtain L2. For L3 limits, the L1 limits were doubled. Limits L1, L2, and L3 represent the concentration in individual gases in ppm. G1 and G2 represent generation rates of individual gases in ppm per month. To obtain G1 and G2 in ppm per day, divide the per month numbers by 30. Except for acetylene, G1 is 10% of L1, and G2 is 50% of L1. The generation rates (G1, G2) are points where our level of concern should increase, especially when considered with the L1, L2, and L3 limits. At G2 generation rate, we should be extremely concerned, reduce the DGA sampling interval accordingly, and perhaps plan an outage, etc.

Picture 2: Suggested Levels of Concern (Limits)


Except for acetylene, generation rate levels G1 and G2 were taken from IEC 60599 which is used with mineral oil transformers. Any amount of ongoing acetylene generation means active arcing inside the transformer. In this case, the transformer should be removed from service. These criteria were chosen because of an absence of any other criteria. As DGA criteria for silicone oils becomes better known and the quantified table shown on Picture 2 will change to reflect new information.
As with mineral oil-filled transformers, gas generation rates are much more important than the amount of gas present. Total accumulated gas depends greatly on age (an older transformer has more gas). If the rate of generation of any combustible gas shows a sudden increase in the DGA, immediately take another oil sample to confirm the gas generation rate increase. If the second DGA confirms a generation rate increase, get some outside advice. Be careful, gas generation rates increase somewhat with temperature variations caused by increased loading and summer ambient temperatures. However, higher operating temperatures are also the most likely conditions for a fault to occur. The real question is, has the increased gas generation rate been caused by a fault or increased temperature from greater loading or higher ambient temperature?
If gas generation rates are fairly constant (no big increases and less than G1 limits above), what actions need to be taken if a transformer exceeds the L1 limits? Begin to pay more attention to that transformer, just as with a mineral oil transformer. It may be necessary to shorten the DGA sampling interval, reduce loading, check transformer cooling, get some outside advice, etc. As with mineral oil transformers, age exerts a big influence in accumulated gas. Be much more concerned if a 3-year old transformer has exceeded the L1 limits than if a 30-year old transformer exceeds the limits. However, if G1 generation rates are exceeded in either an old or new transformer, the level of concern should be stepped up.
If accumulated gas exceeds the L2 limit, it may be wise to plan to have the transformer de-gassed. Examine the physical tests in the DGAs and compare them to the Doble/IEEE table (table shown on Picture 3). The oil should be treated in whatever manner is appropriate, if these limits are exceeded. If both L1 limits and G1 limits are exceeded, be more concerned. Reduce sampling intervals, get outside advice, reduce loading, check transformer cooling and oil levels, etc. If G2 generation limits are exceeded, be extremely concerned. It will not be long before L3 limits are exceeded, and consider removing the transformer from service for testing, repair, or replacement.
If acetylene is being generated, the transformer should be taken out of service. However, as with mineral oil transformers, a one-time nearby lightning strike or through fault can cause a “one-time” generation of acetylene. If you notice acetylene in the DGA, immediately take another sample. If the amount of acetylene is increasing, an active electrical arc is present within the transformer. It should be taken out of service. With critical silicone (or mineral oil-filled transformer), such as a single station service transformer or excitation transformer, a spare should be available at another facility. If there are no other possible spares, consider beginning the budget process for purchasing a spare transformer.

Physical Test Limits

Table shown on Picture 3 lists test limits for service-aged silicone filled transformer oil. If any of these limits are exceeded, treat the oil in whatever manner is appropriate to return the oil to serviceable condition.

Picture 3: Doble and IEEE Physical Test Limits for Service-Aged Silicone Fluid


If only one number appears, both Doble and IEEE have the same limit. If the above limits are exceeded in the DGA, the silicone oil should be filtered, dried, or treated to correct the specific problem.

Silicone Oil-Filled Transformers

Silicone oils became more common when polychlorinated biphenols were discontinued. They are mainly used in transformers inside buildings and are smaller than generator step-up transformers. Silicone oils have a higher fire point than mineral oils and, therefore, are used where fire concerns are more critical. As of this writing, there are no definitive published standards. IEEE has a guide, and Doble has some service limits, but there are no standards. Information below is taken from the IEEE publication, from Doble, from articles, from IEC 60599 concepts, and from Delta X Research’s/Transformer Oil Analyst rules. Silicone oil dissolved gas analysis is in the beginning stage, and the suggested methods and limits below are subject to change as we gain more experience. However, in the absence of any other methods and limits, use the ones below as a beginning.
Silicone oils used in transformers are polydimethylsiloxane fluids, which are different than mineral oils. Many of the gases generated by thermal and electrical faults are the same. The gases are generated in different proportions than with transformer mineral oils. Also, some fault gases have different solubilities in silicone oils than in mineral oils. Therefore, the same faults would produce different concentrations and different generation rates in silicone oils than mineral oils.
As with mineral oil-filled transformers, three principal causes of gas generation are aging, thermal faults, and/or electrical faults resulting in deterioration of solid insulation and deterioration of silicone fluid. Overheating of silicone oils causes degradation of fluid and generation of gases. Generated gases depend on the amount of dissolved oxygen in the fluid, temperature, and how close bare copper conductors are to the heating. When a transformer is new, silicone oil will typically contain a lot of oxygen. Silicone transformers are typically sealed and pressurized with nitrogen. New silicone oil is not de-gassed; and, as a rule, oxygen concentration will be equivalent to oxygen solubility (maximum) in silicone. The silicone has been exposed to atmosphere for some time during manufacture of the transformer and manufacturer and storage of silicone oil itself. Therefore, carbon monoxide and carbon dioxide are easily formed and dissolved in the silicone due to the abundance of oxygen in the oil, resulting from this atmospheric exposure. In normal new silicone transformers (no faults), both carbon monoxide and carbon dioxide will be generated in the initial years of operation. As the transformer ages and oxygen is depleted, generation of these gases slows, and concentrations level off. See Picture 1 for the relationship of decreasing oxygen and increasing carbon monoxide and carbon dioxide as a transformer ages. This curve is for general information only and should not be taken to represent any particular transformer. A real transformer with changes in loading, ambient temperatures, and various duty cycles would make these curves look totally different.

Picture 1: Relationship of Oxygen to Carbon Dioxide and Carbon Monoxide as Transformer Ages


After the transformer is older (assuming no faults have occurred), oxygen concentration will reach equilibrium (Picture 1). Reaching equilibrium may take a few years, depending on the size of the transformer, loading, ambient temperatures, etc. After this time, oxygen, carbon monoxide, and carbon dioxide level off, and the rate of production of these gases from normal aging should be relatively constant. If generation rates of these gases change greatly (seen from the DGA), a fault has occurred, either thermal or electrical. Rate of generation of these gases and amounts can be used to roughly determine what the fault is. Once you notice a significant increase in rate of generation of any gas, it is a good idea to subtract the amount of gas that was already in the transformer before this increase. This ensures that gases used in the diagnosis are only gases that were generated after the fault began.

Comparison of Silicone Oil and Mineral Oil Transformers

Some general conclusions can be drawn by comparing silicone oil and mineral oil transformers:

1. All silicone oil-filled transformers will have a great deal more CO than normal mineral oil-filled transformers. CO can come from two sources—the oil itself and from degradation of paper insulation. If the DGA shows little other fault in gas generation besides CO, the only way to tell for certain if CO is coming from paper degradation (a fault) is to run a furan analysis with the DGA. If other fault gases are also being generated in significant amounts, in addition to CO, there obviously is a fault, and CO is coming from paper degradation.
2. There will generally be more hydrogen present than in a mineral oil-filled transformer.
3. Due to “fault masking,” mentioned above, it is almost impossible to diagnose what is going on inside a siliconefilled transformer based solely on DGA. One exception is that if acetylene is being generated, there is an active arc. Look also at gas generation rates and operating history. Look at loading history, through faults, and other incidents. It is imperative that detailed records of silicone oil-filled transformers be carefully kept up to date. These are invaluable when a problem is encountered.
4. If acetylene is being generated, there is a definitely an active electrical arc. The transformer should be removed from service.
5. In general, oxygen in a silicone-filled transformer comes from atmospheric leaks or was present in the transformer oil when it was new. This oxygen is consumed as CO and CO2 are formed from the normal heating from operation of the transformer.
6. Once the transformer has matured and the oxygen has leveled off and remained relatively constant for two or more DGA samples, if you see a sudden increase in oxygen, and perhaps carbon dioxide and nitrogen, the transformer has developed a leak.

Oil Treatment Specifications

After the oil is treated, the results should be as follows (Picture 1).

Picture 1: Oil Treatment Specifications

Taking Oil Samples for DGA

Sampling procedures and lab handling are usually areas that cause the most problems in getting an accurate DGA. There are times when atmospheric gases, moisture, or hydrogen take a sudden leap from one DGA to the next. As has been mentioned, at these times, one should immediately take another sample to confirm DGA values. It is, of course, possible that the transformer has developed an atmospheric leak or that a fault has suddenly occurred inside. More often, the sample has not been taken properly, or it has been contaminated with atmospheric gases or mishandled in other ways. The sample must be protected from all contamination, including atmospheric exposure. Do not take samples from the small sample ports located on the side of the large sample (drain) valves. These ports are too small to adequately flush the large valve and pipe nipple connected to the tank. In addition, air can be drawn past the threads and contaminate the sample. Fluid in the valve and pipe nipple remain dormant during operation and can be contaminated with moisture, microscopic stem packing particles, and other particles. The volume of oil in this location can also be contaminated with gases, especially hydrogen. Hydrogen is one of the easiest gases to form. With hot sun on the side of the transformer tank where the sample valve is located, high ambient temperature, high oil temperature, and captured oil in the sample valve and extension, hydrogen formed will stay in this area until a sample is drawn.
The large sample (drain) valve can also be contaminated with hydrogen by galvanic action of dissimilar metals. Sample valves are usually brass. A brass pipe plug should be installed when the valve is not being used. If a galvanized or black iron pipe plug is installed in a brass valve, the dissimilar metals produce a thermocouple effect, and circulating currents are produced. As a result, hydrogen is generated in the void between the plug and valve gate. If the valve is not thoroughly flushed, the DGA will show a high hydrogen level. Oil should not be sampled for DGA purposes when the transformer is at or below a freezing temperature. Test values which are affected by water (such as dielectric strength, power factor, and dissolved moisture content) will be inaccurate.

CAUTION: Transformers must not be sampled if there is a negative pressure (vacuum) at the sample valve.

This is typically not a problem with conservator transformers. If the transformer is nitrogen blanketed, look at the pressure/vacuum gauge. If the pressure is positive, go ahead and take the sample. If the pressure is negative, a vacuum exists at the top of the transformer. If there is a vacuum at the bottom, air will be pulled in when the sample valve is opened. Wait until the pressure gauge reads positive before sampling. Pulling in a volume of air could be disastrous if the transformer is energized.
If negative pressure (vacuum) is not too high, the weight of oil (head) will make positive pressure at the sample valve, and it will be safe to take a sample. Oil head is about 2.9 feet (2 feet 10.8 inches) of oil per psi. If it is important to take the sample even with a vacuum showing at the top, proceed as described below. Use the sample tubing and adaptors described below to adapt the large sample valve to ⅛-inch tygon tubing. Fill a length (2 to 3 feet) of tygon tubing with new transformer oil (no air bubbles) and attach one end to the pipe plug and the other end to the small valve. Open the large sample (drain) valve a small amount and very slowly crack open the small valve. If oil in the tygon tubing moves toward the transformer, shut off the valves immediately. Do not allow air to be pulled into the transformer. If oil moves toward the transformer, there is a vacuum at the sample valve. Wait until the pressure is positive before taking the DGA sample. If oil is pushed out of the tygon tubing into the waste container, there is a positive pressure, and it is safe to proceed with DGA sampling. Shut off the valves and configure the tubing and valves to take the sample per the instructions below.

DGA Oil Sample Container

Glass sample syringes are recommended. There are different containers, such as stainless steel vacuum bottles and others. Using only glass syringes is recommended. If there is a small leak in the sampling tubing or connections, vacuum bottles will draw air into the sample, which cannot be seen inside the bottle. The sample will show high atmospheric gases and high moisture if the air is humid. Other contaminates such as suspended solids or free water cannot be seen inside the vacuum bottle. Glass syringes are the simplest to use because air bubbles are easily seen and expelled. Other contaminates are easily seen, and another sample can be immediately taken if the sample is contaminated. The downside is that glass syringes must be handled carefully and must be protected from direct sunlight. They should be returned to their shipping container immediately after taking a sample. If they are exposed to sunlight for any time, hydrogen will be generated, and the DGA will show false hydrogen readings. For these reasons, glass syringes are recommended, and the instructions below include only this sampling method. Obtain a brass pipe plug (normally 2 inches) that will thread into the sample valve at the bottom of the transformer. Drill and tap the pipe plug for ⅛-inch national pipe thread, insert a ⅛-inch pipe nipple (brass if possible) and attach a small ⅛-inch valve for controlling the sample flow. Attach a ⅛-inch tygon tubing adaptor to the small valve outlet. Sizes of the piping and threads above do not matter. Any arrangement with a small sample valve and adaptor to ⅛-inch tygon tubing will suffice. See Picture 2.

Picture 2: Oil Sampling Piping

Taking the Sample

♦ Remove the existing pipe plug and inspect the valve opening for rust and debris.
♦ Crack open the valve and allow just enough oil to flow into the waste container to flush the valve and threads. Close the valve and wipe the threads and outlet with a clean dry cloth.
♦ Re-open the valve slightly and flush approximately 1 quart of oil into the waste container.
♦ Install the brass pipe plug (described above) and associated ⅛-inch pipe and small valve, and a short piece of new ⅛-inch tygon tubing to the outlet of the ⅛-inch valve.
♦ Never use the same sample tubing on different transformers. This is one way a sample can be contaminated and give false readings.
♦ Open both the large valve and small sample valve and allow another quart to flush through the sampling apparatus. Close both valves. Do this before attaching the glass sample syringe. Make sure the short piece of tygon tubing that will attach to the sample syringe is installed on the ⅛-inch valve before you do this.
♦ Install the glass sample syringe on the short piece of ⅛-inch tubing. Turn the stopcock handle on the syringe so that the handle points toward the syringe. Note: The handle always points toward the closed port. The other two ports are open to each other. See Picture 3.

Picture 3: Sampling Syringe (Flushing)

♦ Open the large sample valve a small amount and adjust the ⅛-inch valve so that a gentle flow goes through the flushing port of the glass syringe into the waste bucket.
♦ Slowly turn the syringe stopcock handle so that the handle points to the flushing port (Picture 4). This closes the flushing and allows oil to flow into the sample syringe. Do not pull the syringe handle. This will create a vacuum and allow bubbles to form. The syringe handle (piston) should back out very slowly. If it moves too fast, adjust the small ⅛-inch valve until the syringe slows and hold your hand on the back of the piston so you can control the travel.

Picture 4: Sampling Syringe (Filling)

♦ Allow a small amount, about 10 cubic centimeters (cc), to flow into the syringe and turn the stopcock handle again so that it points to the syringe. This will again allow oil to come out of the flushing port into the waste bucket.
♦ Pull the syringe off the tubing, but do not shut off the oil flow. Allow the oil flow to continue into the waste bucket.
♦ Hold the syringe vertical and turn the stopcock up so that the handle points away from the syringe. Press the syringe piston to eject any air bubbles, but leave 1 or 2 cc of oil in the syringe. See Picture 5.

Picture 5: Sample Syringe Bubble Removal

♦ Turn the stopcock handle toward the syringe. The small amount of oil in the syringe should be free of bubbles and ready to receive the sample. If there are still bubbles at the top, repeat the process until you have a small amount of oil in the syringe with no bubbles.
♦ Reattach the tygon tubing. This will again allow oil to flow out of the flushing port. Slowly turn the stopcock handle toward the flushing port which again will allow oil to fill the syringe. The syringe piston will again back slowly out of the syringe. Allow the syringe to fill about 80% full. Hold the piston so you can stop its movement at about 80% full.
♦ Close the stopcock by turning the handle toward the syringe. Oil again will flow into the waste container. Shut off both valves, remove the sampling apparatus, and reinstall the original pipe plug.
♦ Return the syringe to its original container immediately. Do not allow sunlight to impact the container for any length of time. Hydrogen will form and give false readings in the DGA.
♦ Carefully package the syringe in the same manner that it was shipped to the facility and send it to the lab for processing.
♦ Dispose of waste oil in the plant waste oil container.


CAUTION: Do not eject all the oil, or air will re-enter. Do not pull the piston. This will cause bubbles to form. Do not eject any bubbles that form after the sample is collected. These are gases that should be included in the lab sample.

Transformer Oil Tests

Transformer Oil Tests That Should Be Completed Annually with the Dissolved Gas Analysis

Dielectric Strength

This test measures the voltage where the oil electrically breaks down. The test gives a good indication of the amount of contaminants (water and oxidation particles) in the oil. DGA laboratories typically use ASTM D-1816. Using the D-1816 test, the minimum oil breakdown voltage is 20 kV for transformers rated less than 288 kV and 25 kV for transformers 287.5 kV and above. If the dielectric strength test falls below these numbers, the oil should be reclaimed. Do not base any decision on one test result or on one type of test. Instead, look at all the information over several DGAs and establish trends before making any decision. The dielectric strength test is not extremely valuable, moisture in combination with oxygen and heat will destroy cellulose insulation long before the dielectric strength of the oil has given a clue that anything is going wrong. The dielectric strength test also reveals nothing about acids and sludge. The tests explained below are much more important.

Interfacial Tension

This test (ASTM D-791-91), is used by DGA laboratories to determine the interfacial tension between the oil sample and distilled water. The oil sample is put into a beaker of distilled water at a temperature of 25 ºC. The oil should float because its specific gravity is less than that of water (specific gravity of water is one). There should be a distinct line between the two liquids. The IFT number is the amount of force (dynes) required to pull a small wire ring upward a distance of 1 centimeter through the water/oil interface. (A dyne is a very small unit of force equal to 0.000002247 pound.) Good, clean oil will make a very distinct line on top of the water and give an IFT number of 40 to 50 dynes per centimeter of travel of the wire. As the oil ages, it is contaminated by tiny particles (oxidation products) of the oil and paper insulation. Particles on top of the water extend across the water/oil interface line that weaken the surface tension between the two liquids. Particles in oil weaken interfacial tension and lower the IFT number. The IFT and acid numbers, together, are an excellent indication of when the oil needs to be reclaimed. It is recommended that the oil be reclaimed when the IFT number falls to 25 dynes per centimeter. At this level, the oil is very contaminated and must be reclaimed to prevent sludging, which begins around 22 dynes per centimeter. See FIST 3-5. If oil is not reclaimed, sludge will settle on windings, insulation, cooling surfaces, etc., and cause loading and cooling problems, as discussed earlier. This will greatly shorten transformer life. A definite relationship exists between the acid number, the IFT, and the number of years in service. The accompanying curve (Picture 1) shows the relationship and is found in many publications. (It was originally published in the AIEE transactions in 1955.) Notice that the curve shows the normal service limits for both the IFT and the acid number.

Picture 1: Interfacial Tension, Acid Number, Years in Service

Acid Number

Acid number (acidity) is the amount of potassium hydroxide (KOH) in milligrams (mg) that it takes to neutralize the acid in 1 gram (gm) of transformer oil. The higher the acid number, the more acid is in the oil. New transformer oils contain practically no acid. Oxidation of the insulation and oils forms acids as the transformer ages. The oxidation products form sludge and precipitate out inside the transformer. The acids attack metals inside the tank and form soaps (more sludge). Acid also attacks cellulose and accelerates insulation degradation. Sludging begins when the acid number reaches 0.40. It is obvious that the oil should be reclaimed before it reaches 0.40. It is recommended that the oil be reclaimed when it reaches 0.20 mg potassium hydroxide per gram (KOH/gm). As with all others, this decision must not be based on one DGA test, but watch for a rising trend in the acid number each year. Plan ahead and begin budget planning before the acid number reaches 0.20.

Test for Oxygen Inhibitor

Previously, the need to keep the transformer dry and O2 free was emphasized. Moisture is destructive to cellulose, and even more so in the presence of oxygen. Some publications state that each time you double the moisture (ppm), you halve the life of the transformer. As was discussed, acids are formed that attack the insulation and metals that form more acids, causing a viscous cycle. Oxygen inhibitor is a key to extending the life of transformers. The inhibitor currently used is ditertiary butyl paracresol (DBPC). This works similar to a sacrificial anode in grounding circuits. The oxygen attacks the inhibitor instead of the cellulose insulation. As this occurs and the transformer ages, the inhibitor is used up and needs to be replaced. Replacement of the inhibitor also generally requires treatment of the oil. The ideal amount of DBPC is 0.3% by total weight of the oil (shown on the transformer nameplate).
Test the inhibitor content with the DGA every 3 to 5 years. If the inhibitor is 0.08%, the transformer is considered uninhibited, and the oxygen freely attacks the cellulose. If the inhibitor falls to 0.1%, the transformer should be re-inhibited. For example, if your transformer tested 0.1%, you need to go to 0.3% by adding 0.2% of the total weight of the transformer oil. The nameplate gives the weight of oil. For example, 5,000 pounds—so 5,000 pounds x 0.002 = 10 pounds of DBPC needs to be added. Adding a little too much DBPC does not hurt the oil. Dissolve 10 pounds of DBPC in the transformer oil that you have heated to the same temperature as the oil inside the transformer. It may take experimenting some to get the right amount of oil to dissolve the DBPC. Mix the oil and inhibitor in a clean container until all the DBPC is dissolved. Add this mixture to the transformer, using the method given in the transformer’s instruction manual for adding oil. In either case, do not neglect this important maintenance function. It is critical to transformer insulation to have the proper amount of oxygen inhibitor.

Power Factor

Power factor indicates the dielectric loss (leakage current) of the oil. This test may be completed by the DGA laboratories. It may also be completed by Doble testing. A high power factor indicates deterioration and/or contamination by products, such as water, carbon, or other conducting particles, metal soaps caused by acids (formed as mentioned above), attacking transformer metals, and products of oxidation. The DGA labs normally test the power factor at 25 ºC and 100 ºC. Doble information indicates that the inservice limit for power factor is less than 0.5% at 25 ºC. If the power factor is greater than 0.5% and less than 1.0%, further investigation is required. The oil may require replacement or fullers earth filtering. If the power factor is greater than 1.0% at 25 ºC, the oil may cause failure of the transformer, so replacement or reclaiming is required. Above 2%, oil should be removed from service and replaced because equipment failure is imminent. The oil cannot be reclaimed.

Oxygen

Oxygen must be watched closely in DGA tests. Many experts and organizations, including EPRI, believe that oxygen levels in the oil above 2,000 ppm greatly accelerate paper deterioration. This becomes even more critical with moisture above safe levels. Under the same temperature conditions, cellulose insulation in low oxygen oil will last 10 times longer than insulation in high oxygen oil. It is recommended that if oxygen reaches 10,000 ppm in the DGA, the oil should be de-gassed and new oxygen inhibitor installed. High atmospheric gases (O2 and N2) normally mean that a leak has developed in a bladder or diaphragm in the conservator. If there is no conservator and pressurized nitrogen is on top of the oil, expect to see high nitrogen but not high oxygen. Oxygen comes only from leaks and from deteriorating insulation.

Furans

When cellulose insulation decomposes due to overheating, organic compounds, in addition to CO2 and CO, are released and dissolved in the oil. These chemical compounds are known as furanic compounds or furans. The most important one, for our purposes, is 2-furfuraldehyde. When DGAs are required, always request that furans testing be completed by the laboratory to check for paper deterioration. In healthy transformers, there are no detectable furans in the oil, or they are less than 100 parts per billion (ppb). In cases where significant damage to paper insulation from heat has occurred, furan levels may be at least 100 ppb and up to 70,000 ppb. Furanic content in the oil is especially helpful in estimating remaining life in the paper insulation, particularly if several prior tests can be compared and trends established.
Use the furan numbers in table shown on Picture 2 for assessment. Do not base any evaluation on only one test, use several DGAs over a period of time to develop trending. The first column in table on Picture 2 is used for transformers with non-thermally upgraded paper, and the second column is for transformers with thermally upgraded paper.

Picture 2: Furans, DP, Percent of Life Used of Paper Insulation


Testing is completed for five different furans which are caused by different problems. The five furans and their most common causes are listed below:

♦ 5H2F (5-hydroxymethyl-2-furaldehyde) caused by oxidation (aging and heating) of the paper;
♦ 2FOL (2-furfurol) caused by high moisture in the paper;
♦ 2FAL (2-furaldehyde) caused by overheating;
♦ 2ACF (2-acetylfuran) caused by lightning (rarely found in DGA);
♦ 5M2F (5-methyl-2-furaldehyde) caused by local severe overheating (hotspot);

Doble inservice limits are reproduced below to support the above recommended guidelines (table shown on Picture 3). Additional guidelines given in table shown on Picture 4 are useful.

Picture 3: Doble Limits for Inservice Oils


1
D 877 test is not as sensitive to dissolved water as the D 1816 test and should not be used with oils for extra high-voltage (EHV) equipment. Dielectric breakdown tests do not replace specific tests for water content.
2
The use of absolute values of water-in-oil (ppm) do not always guarantee safe conditions in electrical apparatus. The percent by dry weight should be determined from the curves provided.
3
ND = None detectable.
These recommended limits for inservice oils are not intended to be used as absolute requirements for removing oil from service but to provide guidelines to aid in determining when remedial action is most beneficial. Remedial action will vary depending upon the test results. Reconditioning of oil, that is, particulate removal (filtration) and drying, may be required if the dielectric breakdown voltage or water content do not meet these limits. Reclamation (clay filtration) or replacement of the oil may be required if test values for power factor, interfacial tension, neutralization number, or soluble sludge do not meet recommended limits.

Picture 4: Additional Guidelines for Inservice Oils


1 
When an oil is allowed to sludge in service, special treatment may be required to clean the core, coil, and tank.

Moisture in Transformer Insulation

Picture 1 shows how moisture is distributed throughout transformer insulation. Notice that the moisture is distributed according to temperature, with most moisture at the bottom and less moisture as temperature increases toward the top. This example shows almost twice the moisture near the bottom as there is at the top. Most service-aged transformers fail in the lower one-third of the windings, which is the area of most moisture. The area of most moisture is also the area of most electrical stress. Moisture and oxygen are two of the transformer’s worst enemies. It is very important to keep the insulation and oil as dry as possible and as free of oxygen as possible. Failures due to moisture are the most common cause of transformer failures. Without an accurate oil temperature, it is impossible for laboratories to provide accurate information about the M/DW or percent saturation. It will also be impossible for you to calculate this information accurately.

Picture 1: Water Distribution in Transformer Insulation


Experts disagree on how to tell how much moisture is in the insulation based on how much moisture is in the oil (ppm). At best, methods to determine moisture in the insulation based solely on DGA are inaccurate. The methods discussed below to determine moisture levels in the insulation are estimates, and no decision should be made based on one DGA. However, keep in mind that the life of the transformer is the life of the insulation. The insulation is quickly degraded by excess moisture and the presence of oxygen. Base any decisions on several DGAs over a period of time and establish a trend of increasing moisture.If the lab does not provide the percent M/DW, IEEE 62-1995 gives a method. From the curve (Picture 2), find the temperature of the bottom oil sample and add 5 ºC. Do not use the top oil temperature. This estimates the temperature of the bottom third (coolest part) of the winding, where most of the water is located.

Picture 2: Myers Multiplier Versus Temperature


From this temperature, move up vertically to the curve. From this point on the curve, move horizontally to the left and find the Myers Multiplier number. Take this number and multiply the ppm of water shown on the DGA. The result is percent M/DW in the upper part of the insulation. This method gives less amount of water than the General Electric nomogram (Picture 3).

Picture 3: Water Content of Paper and Oil Nomogram


This nomogram, published by General Electric in 1974, gives the percent saturation of oil and percent M/DW of insulation. Use the nomogram to check yourself after you have completed the method illustrated in Picture 2. The nomograph in Picture 3 will show more moisture than the IEEE method.
The curves in Picture 3 are useful to help understand relationships between temperature, percent saturation of the oil, and percent M/DW of the insulation. For example, pick a point on the ppm water line (10 ppm). Place a straight edge on that point and pick a point on the temperature line (45 ºC). Read the percent saturation and percent M/DW on the center lines. In this example, percent saturation is about 6.5%, and the % M/DW is about 1.5%. Now, hold the 10 ppm point and move the sample temperature upward (cooler) and notice how quickly the moisture numbers increase. For example, use 20 ºC and read the percent saturation of oil at about 18.5% and the % M/DW at about 3.75%. The cooler the oil, the higher the moisture percentage for the same ppm of water in the oil.
Do not make a decision on dry out based on only one DGA and one calculation, it should be based on trends over a period of time. Take additional samples and send them for analysis. Take extra care to make sure the oil temperature is correct. You can see by the nomogram that moisture content varies dramatically with temperature. Take extra care that the sample is not exposed to air. After using the more conservative IEEE method, if, again, subsequent samples show M/DW is 2.5% or more and the oil is 30% saturated or more, the transformer should be dried as soon as possible. Check the nomogram and curves above to determine the percent saturation of the oil. The insulation is degrading much faster than normal due to the high moisture content. Drying can be an expensive process, it is prudent to consult with others before making a final decision to implement dry out. However, it is much less expensive to perform a dry out than to allow a transformer to degrade faster than normal, substantially shortening transformer life.

Moisture Problems

Moisture, especially in the presence of oxygen, is extremely hazardous to transformer insulation. Recent EPRI studies show that an oxygen level above 2,000 ppm dissolved in transformer oil is extremely destructive. Each DGA and Doble test result should be examined carefully to see if water content is increasing and to determine the moisture by dry weight (M/DW) or percent saturation in the paper insulation. When 2% M/DW is reached, make plans for a dry out. Never allow the M/DW to go above 2.5% in the paper or 30% oil saturation without drying out the transformer. Each time the moisture is doubled in a transformer, the life of the insulation is cut by one-half. Keep in mind that the life of the transformer is the life of the paper, and the purpose of the paper is to keep out moisture and oxygen. For service-aged transformers rated less than 69 kV, results of up to 35 ppm at 60 °C are considered acceptable. For 69 kV through 230 kV, a DGA test result of 20 ppm at 60 °C is considered acceptable. For greater than 230 kV, moisture should never exceed 12 ppm at 60 °C. However, the use of absolute values for water does not always guarantee safe conditions, and the percent by dry weight should be determined. See table 19, “Doble Limits for In-Service Oils,” in section 7.6. If values are higher, the oil should be processed. If the transformer is kept as dry and free of oxygen as possible, transformer life will be extended.
The manufacturers dry new transformers specifies to no more than 0.5% M/DW during commissioning. That means that a transformer with 10,000 pounds of paper insulation, has 10,000 x 0.005 = 50 pounds of water (about 6 gallons) in the paper. This is not enough moisture to be detrimental to electrical integrity. When the transformer is new, this water is distributed equally through the transformer. It is extremely important to remove as much water as possible.
When the transformer is energized, water begins to migrate to the coolest part of the transformer and the site of the greatest electrical stress. This location is normally the insulation in the lower one-third of the winding. Paper insulation has a much greater affinity for water than does the oil. The water will distribute itself unequally, with much more water being in the paper than in the oil. The paper will partially dry the oil by absorbing water out of the oil. Temperature is also a big factor in how the water distributes itself between the oil and paper. See table on Picture 1 below for comparison.

Picture 1: Comparison of Water Distribution in Oil and Paper


Table on Picture 1 shows the tremendous attraction that paper insulation has for water and how the water changes in the paper with temperature. The ppm of water in oil shown in the DGA is only a small part of the water in the transformer. When an oil sample is taken, it is important to record the oil temperature from the top oil temperature gauge. Some laboratories give percent M/DW of the insulation in the DGA, others give percent oil saturation, and some give only the ppm of water in the oil. If you have an accurate temperature of the oil and the ppm of water, the Nomogram will give percent M/DW of the insulation and the percent oil saturation.
Where does the water come from? Moisture can be in the insulation when it is delivered from the factory. If the transformer is opened for inspection, the insulation can absorb moisture from the atmosphere. If there is a leak, moisture can enter in the form of water or humidity in the air. Moisture is also formed by the degradation of insulation as the transformer ages. Most water penetration is flow of wet air or rain water through poor gasket seals due to a pressure difference caused by transformer cooling. If a transformer is removed from service during rain or snow, some transformer designs cool rapidly and the pressure inside drops. The most common moisture ingress points are gaskets between bushing bottoms and the transformer top and the pressure relief device gasket. Small oil leaks, especially in the oil cooling piping, will also allow moisture ingress. With rapid cooling and the resultant pressure drop, relatively large amounts of water and water vapor can be pumped into the transformer in a short time. It is important to repair small oil leaks; the small amount of visible oil is not important in itself, but it indicates a point where moisture will enter.
It is critical for life extension to keep transformers as dry and as free of oxygen as possible. Moisture and oxygen cause the paper insulation to decay much faster than normal and form acids, sludge, and more moisture. Sludge settles on windings and inside the structure, causing less efficient transformer cooling, which allows temperature to slowly rise over a period of time. Acids cause an increase in the rate of decay, which forms more acid, sludge, and moisture at a faster rate. This is a vicious cycle of increasing speed forming more acid and causing more decay. The answer is to keep the transformer as dry as possible and as free of oxygen as possible. In addition, oxygen inhibitor should be watched in the DGA testing. The transformer oil should be dried when moisture reaches the specific values. An inhibitor (ditertiary butyl paracresol [DBPC]) should be added (0.3% by weight ASTM D-3787) when the oil is processed.

Water can exist in a transformer in five forms:

1. Free water, at the bottom of the tank.
2. Ice at the tank bottom (if the oil specific gravity is greater than 0.9, ice can float).
3. Water can be in the form of a water/oil emulsion.
4. Water can be dissolved in the oil and is given in ppm in the DGA.
5. Water can be in the form of humidity if transformers have an inert gas blanket.

Free water causes few problems with dielectric strength of oil, however, it should be drained as soon as possible. Having a water-oil interface allows oil to dissolve water and transport it to the insulation. Problems with moisture in insulation were discussed above. If the transformer is out of service in winter, water can freeze. If oil specific gravity is greater than 0.9 (ice specific gravity), ice will float. This can cause transformer failure if the transformer is energized with floating ice inside. This is one reason that DGA laboratories test the specific gravity of transformer oil.
The amount of moisture that can be dissolved in oil increases with temperature (see Picture 2). This is why hot oil is used to dry out a transformer. A water/oil emulsion can be formed by purifying oil at too high temperature. When the oil cools, dissolved moisture forms an emulsion. A water/oil emulsion causes drastic reduction in dielectric strength.
How much moisture in insulation is too much? When the insulation reaches 2.5% M/DW or 30% oil saturation (given on some DGAs), the transformer should have a dry out with vacuum, if the tank is rated for vacuum. If the transformer is old, pulling a vacuum can do more harm than good. In this case, it is better to do round-the-clock re-circulation with a Bowser drying the oil as much as possible, which will pull water out of the paper. At 2.5% M/DW, the paper insulation is degrading much faster than normal. As the paper is degraded, more water is produced from the decay products, and the transformer becomes even wetter and decays even faster. When a transformer gets above 4% M/DW, it is in danger of flashover if the temperature rises to 90 ºC.

Picture 2: Maximum Amount of Water Dissolved in Mineral Oil Versus Temperature

Dissolved Moisture in Transformer Oil

Moisture is measured in the dissolved gas analysis in ppm. Some laboratories also give percent saturation, which is the percent saturation of water in the oil. This is a percentage of how much water is in the oil compared with the maximum amount of water the oil can hold. Picture 2 shows it can be seen that the amount of water the oil can dissolve is greatly dependent on temperature. The curves below (Picture 3) are percent saturation curves. On the left line, find the ppm of water from your DGA. From this point, draw a horizontal with a straight edge. From the oil temperature, draw a vertical line. At the point where the lines intersect, read the percent saturation curve. If the point falls between two saturation curves, estimate the percent saturation based on where the point is located. For example, if the water is 30 ppm and the temperature is 40 ºC, you can see on the curves that this point of intersection falls about halfway between the 20% curve and the 30% curve. This means that the oil is approximately 25% saturated. Curves shown on Picture 3 are from IEEE 62-1995.

Picture 3: Transformer Oil Percent Saturation Curves


CAUTION: Below 30 °C, the curves are not very accurate.

Carbon Dioxide / Carbon Monoxide Ratio

This ratio is not included in the Rogers Ratio Method of analysis. However, it is useful to determine if a fault is affecting the cellulose insulation. This ratio is included in transformer oil analyzing software programs, such as Delta X Research Transformer Oil Analyst. This analysis is available from the TSC at D-8440 and D-8450 in Denver.

Formation of CO2 and CO from the degradation of oil impregnated paper increases rapidly with temperature. Calculate a normal operating CO2/CO ratio at each DGA, based on the total accumulated amount of both gases. Look at several DGAs concentrating on CO2 and CO. Experience has shown that, with normal loading and temperatures, the CO2 generation rates runs 7 to 20 times higher than CO. With a CO2/CO ratio above 7, there is little concern. With some transformers, ratios down to 5 times more CO2 than CO might be considered normal. However, be careful with a ratio below 7. If H2, CH4, and C2H6 are increasing significantly as well as CO and the ratio is 5 or less, there is probably a problem. Take time to know the particular transformer by carefully checking all prior DGAs and establishing a normal operating CO2 to CO ratio.

CAUTION: After a suspected problem (a substantial increase in the amount of CO), the ratio should be based on the gas generation of both CO2 and CO between successive DGAs and not on accumulated total CO2 and CO gas levels.

If a problem is suspected, immediately take another DGA sample to confirm the problem. Take the amount of CO2 generated between the DGAs and divide it by the amount of CO generated at that same time to establish the ratio. An excellent indication of abnormally high temperatures and rapidly deteriorating cellulose insulation is a CO2/CO under 5. If the ratio is 3 or under, severe and rapid deterioration of cellulose is certainly occurring. In addition to DGAs, perform the Furans test. Extreme overheating from loss of cooling or plugged oil passages will produce a CO2/CO ratio around 2 or 3 along with increasing Furans. If this is found, de-energization and internal inspection is recommended, the transformer is in imminent danger of failure.

Table shown on Picture 1 is adapted from IEC 60599. Some of the wording has been changed to reflect American language usage rather than European.

Picture 1: Typical Faults in Power Transformers


Notes:

1. X wax formation comes from Paraffinic oils (paraffin based). These are not used in transformers at present in the United States but are predominate in Europe.
2. The last overheating problem in the table says “over 700 °C.” Recent laboratory discoveries have found that acetylene can be produced in trace amounts at 500 °C, which is not reflected in this table. We have several transformers that show trace amounts of acetylene that probably are not active arcing but are the result of high-temperature thermal faults, as in the example. It may also be the result of one arc, due to a nearby lightning strike or voltage surge.
3. A bad connection at the bottom of a bushing can be confirmed by comparing infrared scans of the top of the bushing with a sister bushing. When loaded, heat from a poor connection at the bottom will migrate to the top of the bushing, which will display a markedly higher temperature. If the top connection is checked and found tight, the problem is probably a bad connection at the bottom of the bushing.