{"id":766,"date":"2020-09-03T21:05:29","date_gmt":"2020-09-04T02:05:29","guid":{"rendered":"http:\/\/polywaterv2.wpengine.com\/?p=766"},"modified":"2025-06-24T07:35:00","modified_gmt":"2025-06-24T12:35:00","slug":"how-temperature-and-water-influence-the-aging-of-transformers","status":"publish","type":"post","link":"https:\/\/www.polywater.com\/en\/knowledge-hub\/how-temperature-and-water-influence-the-aging-of-transformers\/","title":{"rendered":"How Temperature and Water Influence the Aging of Transformers"},"content":{"rendered":"
<\/div>\n

1. Introduction<\/h2>\n

A variety of factors influence the aging of transformers with oil\/celluosic systems and accelerate the aging process. In this article we will discuss:<\/p>\n

1.\u00a0 Specific influences of temperature and water on transformer aging.
\n2.\u00a0 Pathways by which moisture and water enter the transformer.
\n3.\u00a0 How temperature and water accelerate transformer aging.<\/p>\n

\"\"<\/p>\n

\"\"
\nWe will also describe how to:<\/p>\n

1.\u00a0 Reduce the influences of these accelerators.
\n2.\u00a0 Dry waterlogged transformers.
\n3.\u00a0 Minimize the influences of temperature.
\n4.\u00a0 Prevent accelerated aging altogether.<\/p>\n

In addition, we will outline factors that should be specified early in the transformer design phase to ensure a reliable asset over the long term. We’ll address acid and oxygen and their role as accelerators in a subsequent article.<\/p>\n

2. The Influence of Temperature<\/h2>\n

High temperatures have a significant impact on the transformer aging processes.<\/p>\n

2.1 Montsinger\u2019s Rule<\/h3>\n

In the 1930s, Montsinger helped to quantify the degree of accelerated transformer aging due to temperature. He determined that a temperature increase of 8\u00b0C will halve the service life of a transformer.<\/p>\n

Montsinger also showed that the aging process is negligible at temperatures below 50\u00b0C.\u00a0These principles continue to hold to this day, even with the development of new materials used in transformers in the intervening 70 years.<\/p>\n

Based on these principles, let\u2019s generalize that a 10\u00b0C increase will cut the service life in half and that monitoring a transformer\u2019s temperature makes sense only when it is over 50\u00b0C.\u00a0 Below this temperature, aging will be so slow as to rule out any failure at any time during the transformer\u2019s estimated service lifetime.<\/p>\n\n\n\n
Related Content: <\/strong>How Oxygen and Acids Influence the Aging of Transformers<\/a><\/td>\n<\/tr>\n<\/tbody>\n<\/table>\n

3. The Influence of Design<\/h2>\n

There is no such thing as a typical or identical trajectory of transformer aging, because there is no such thing as a \u201cstandard\u201d transformer.\u00a0 As stated above, temperature is an aging accelerator. More precisely, aging depends on the temperature in the hottest area of the transformer.\u00a0 It makes no sense to consider an average temperature.\u00a0 Aging is not average and will always go together with a specific temperature at a specific location inside the transformer.
\nTo assess transformer one needs to start with the hottest zone.\u00a0 If the cellulose has lost its mechanical strength due to thermal aging at a certain location in the transformer, it will not matter if the degree of polymerization (DP) value is still \u201cgood on average.\u201d In such a case, the transformer will not be \u201con average\u201d broken.\u00a0 It will completely fail unless it has already reached its end of life (EOL will need to be repaired at the very least. Therefore, it is important to determine during the specification stage of the transformer that the design of the internal cooling system be optimized for the conditions in which the transformer will operate.<\/p>\n

It will be impossible to correct a faulty internal cooling system, which we will address in detail in Section 5.1.<\/p>\n

The Influence of Water and Moisture<\/h2>\n

4. Water as an Aging Product<\/h2>\n

Many papers have been published on the subject of \u201cWater as an Aging Product.\u201d Some claim that water content can be used to assess a transformer\u2019s age. The technical basis of such claims seems plausible enough since it is a matter of chemistry. Every time long-chain cellulose molecules are broken up by thermally driven chemical processes, water is released.\u00a0 As a result, it could be said that the aging process is a source of water inside the transformer.\u00a0 What is less clear is the extent to which this phenomenon, called cellulosic pyrolysis, plays a significant role in waterlogging transformers.<\/p>\n

From my personal experience, all old transformers show increased water content, but pyrolysis is not a significant source.\u00a0 External sources of water\/moisture are more important in accelerated aging of transformers.<\/p>\n

4.1 Water and Moisture Gateways<\/b><\/h3>\n

Water content in different transformers of similar age and operating conditions is by no means similar enough to infer a universal aging mechanism. Because of the differences in water content in transformers, we must conclude that external causes generate the differences.<\/p>\n

The most heavily waterlogged transformers are those that have undergone frequent load changes and thermal stress in the tank area. Transformers with critical components, such as the L.V. high-current bushings in electrolysis plants, are inherently leaky and show higher water content. The same applies to generator step-up (GSU) transformers.<\/p>\n

On the other hand, grid and transmission transformers normally have rather low water content. These transformers are both hermetically sealed and openly breathing transformers with similar water content after many years. This indicates that most of the moisture\/water in the transformer comes from external sources.<\/p>\n

4.2 Cooling Water<\/b><\/h3>\n

Another source of water is from leaking water coolers, especially those that are single walled.\u00a0 It makes no difference that the static pressure needs to be higher on the oil side than on the water side.\u00a0 Pipe elbows are especially prone to micro-cracks and high partial pressure which allow water to penetrate to the inside of the transformer.<\/p>\n\n\n\n
Related Content: <\/strong>Mapping Transformer Populations<\/a><\/td>\n<\/tr>\n<\/tbody>\n<\/table>\n

5. Preventing Accelerated Aging<\/h2>\n

To effectively reduce premature transformer aging, accelerators must be detected and managed. This needs to be addressed early in the design stage. It is critically important to insist on correct sealing system designs on the one hand and low thermal loads on the other. With an adequate cooling system, maximum temperatures can be kept lower, thereby significantly extending the service life of a transformer.<\/p>\n

\"\"<\/p>\n

5.1 Ensuring Correct Thermal Design<\/b><\/h3>\n

The manufacturer must prove, even during the project design phase, that the temperature profile of its transformer is appropriate. None of the admissible temperatures should ever be exceeded at any point, not even for short time periods. Even at the technical specifications stage, it is imperative to select the cooling system appropriate to the transformer\u2019s function. For example, for transformers expected to undergo frequent load changes, the internal cooling system must be designed to eliminate temperature spikes during rapid load increases, e.g., OD (Oil Directed) cooling. For transformers where the oil needs to be pumped, e.g., OFWF (Oil Forced\/Water Forced), or in the case of remotely installed oil-to-air radiators, the cooling in the windings must be based on ON (Oil Natural) design so that the oil is distributed as evenly as possible inside the transformer.<\/p>\n

To check a transformer’s thermal design after completion, fiber optics are used to generate an extensive profile of the temperatures inside the transformer. Ideally, there should be at least several sensors in each winding per leg. The optimal sensor distribution is on top, 2\/3 of the upper half, 2\/3 of the lower half, and on the bottom. Typically, the middle leg in a three-phase transformer should be equipped in every winding (HV, LV, tertiary, fine, and coarse tapping windings).<\/p>\n

5.2 Monitoring Gassing Behavior<\/b><\/h3>\n

During operation, a transformer’s gassing behavior will provide conclusive data about the effectiveness of its design. The technical specification of the transformer must stipulate that the presence of thermally induced gases during operation, e.g., C2H4, C2H6, C3H4, C3H6, C3H8, is prohibited. Their presence indicates temperatures above 150\u00b0C. The value of these gases in dissolved gas analysis (DGA) must be “0”. If not, the thermal conditions present in the transformer are unacceptable. I don\u2019t advise applying any limits on values.<\/span><\/p>\n

5.3 Reduction of Water Content<\/b><\/h3>\n

If testing reveals increased water content, e.g., during the oil quality test, the water content must be reduced. This value is by the sampling temperature and must always be checked using the Nielsen diagram.<\/p>\n

5.3.1 Measurement of Water Content<\/b><\/h3>\n

Since water is present in the transformer, its value needs to be collected directly or indirectly from the transformer. The water content of an oil sample alone tells us nothing.<\/p>\n

A useful method to measure water content is to use an oil moisture assessment in conjunction with a Nielsen diagram. To do this, the temperatures at the top and bottom of the transformer need to be monitored over an extended time period to ensure that the transformer load is balanced. Systems using a capacitive probe are particularly helpful for this purpose. The measurements must be recorded in the ppm range.<\/p>\n

A tried and tested direct dielectric measurement system for water content is the fire and gas detection (FDS) system. The data from this measurement can be regarded as accurate and can be supported by the method above.<\/p>\n

The following data should be used as reference values for the assessment of water content:<\/p>\n

\"\"<\/p>\n

Remedies and Sealants<\/b><\/h3>\n

If the water content exceeds the appropriate level, the cellulose requires drying. There are different drying processes available. The simplest and least problematic method is to use bypass systems. They extract water from the transformer during operation by draining it through the oil path. The use of oil treatment plants is not advisable because they only dry the oil and cannot remove more than 1% of the stored water from the transformer.<\/p>\n

Proper sealing of the transformer must be a high priority to prevent the ingress of humidity and water.\u202fThe use of technologies designed specifically for transformer sealing will ensure the long-term performance of the transformer.<\/p>\n

Key factors in choosing transformer sealants include the following:<\/p>\n