We are going to dip into the mailbag for this week’s Tech Corner article. A correspondent asked a question that we have also fielded recently at a number of seminars. “When is it necessary to sample and test the oil in my transformers more often than once a year?” There are actually two parts to the answer. The first deals with abnormal test results leading to a recommendation for a shorter than normal interval for retesting. The second deals with an owner determining that an annual interval is too long for proper routine monitoring for a specific piece of equipment in the customer’s system.
Most of the time, the results of testing routine insulating oil samples fall within ranges of values that we consider to be normal and acceptable. In such cases, we recommend that the customer take no special action beyond resampling and retesting the equipment at the next normal, routine sampling interval. We recommend a routine insulating liquid sampling and testing interval of one year for most equipment.
When the test results appear to indicate abnormal conditions, the most appropriate response may be to obtain another sample, at a shorter than normal interval, and repeat some or all of the analyses that were performed. It also may be appropriate to include other analyses that were not run on the sample that appears to be abnormal. In general, we try to confirm abnormal results and develop an idea of how the results are trending and changing over time. Changes in results may be caused by development of real problems with contamination, oil or insulation aging, developing fault conditions, or changes in the operation of the equipment. Apparent problems may, however, also be caused by temporary disruptions in the normal operation or by obtaining unrepresentative samples.
Retesting allows us to develop sufficient information to determine whether we are dealing with a real problem. We can develop additional useful information for management of the system by performing a retest at a shorter than normal interval. We also minimize the risks that the problem, if real, will lead to permanent damage if we define the shorter than normal interval properly.
For a transformer where we are testing for the first time, abnormal results may have been caused by past problems that were corrected. Retesting at a shorter interval will indicate that the abnormal condition is no longer active. Establishing a trend for the abnormal is a much more reliable management tool for managing electrical equipment oil testing and maintenance decision than just interpreting one snapshot of the data.
This general concept is applicable to virtually all electrical equipment insulating liquid tests. Two areas, in particular, where this is of increased importance are in moisture testing and dissolved gas analysis.
A newly “discovered” wet condition in electrical equipment may be caused by incursion of moisture into the equipment. The apparent moisture incursion is not certain, however, and could have been caused by an unrepresentative sample. The shorter interval retest allows us to determine which scenario is most likely so that we can direct appropriate action concerning the abnormal results. Determination of the length of the shorter interval will depend on how potentially serious the problem may prove to be.
Similarly, sudden changes in DGA may be caused by development of an incipient fault condition. Evaluating and interpreting the DGA results will provide an indication as to whether operation is normal or whether there is an apparent fault condition. To determine the severity of the fault condition, it is frequently necessary to establish the generation rate for the combustible gases that are being produced.
That cannot be done reliably from one set of abnormal data, but instead requires that the unit be retested. Selection of appropriate retest interval is critical to maintaining the integrity of the system while getting enough information to manage the operation of the equipment being tested.
Six months, three months, and one month intervals are relatively common recommendations. In areas where the potential problems are more immediately of concern, intervals of one or two weeks or even daily are used. Immediate retests are typically done to confirm very severe conditions prior to investigating causes of the problem and performing maintenance to correct those conditions.
A “resample” recommendation is very important, even though it is not encountered all that frequently. What “resample” typically means is that the results are such that we cannot interpret them reliably. We need additional information concerning the condition of the oil, equipment, and insulation in order to direct the continuing management of the equipment.
For some types of equipment and insulating liquid analyses, the normal interval should be shortened. Some typical cases where this is done include furnace transformers and the large rectifiers used in metals processing and the chemical industry. Service duty for this type of equipment is relatively severe, so they tend to develop problems more frequently and more rapidly than other electrical equipment. A typical recommendation is to run all of the routine (and maybe some of the non-routine) analyses on an annual basis while also performing a DGA every quarter.
Other types of equipment where this strategy is frequently pursued may include GSUs, large transmission units, generation auxiliary or station service transformers, main substations for industrial facilities, load tap changers, units still under warranty where problems may be expected or where the manufacturer has recommended a testing strategy, and equipment that may be designated as being critical to the operation of the system. In some cases, in addition to DGA, other tests such as moisture, particles and filming compounds analysis, liquid power factor, and/or D1816 dielectric breakdown voltage may be repeated at routine intervals shorter than one year.
We will be discussing another topic next week.
