IEC 60076-10 measures transformer noise with microphones placed on a contour around the unit, at 0.3 m with the fans off or at 2 m with the fans running. The mean sound pressure and the area of that surface give the sound power level, which is the value that is guaranteed and compared.
What the standard measures
Transformer noise is expressed as an A-weighted sound power level, in dB(A). It is not measured at a single point. Sound pressure or sound intensity is measured at several points along a line around the unit, averaged, and converted to power using the area of that line and its height. On the sound power level of a transformer: transformer sound power and far field.
IEC 60076-10 describes how to do this so that the result can be repeated in another laboratory. This guide walks through the measurement decisions in the order they are taken: method, instruments, radiating surface, contour, microphones, environment and validity.
To understand where the hum comes from, see why a transformer hums.
Two valid methods
Clause 5 admits two methods. They are agreed between manufacturer and purchaser when the order is placed.
| Method | Reference | What is measured |
|---|---|---|
| Sound pressure | ISO 3746 | Sound pressure level at each microphone |
| Sound intensity | ISO 9614-1 | Sound intensity level normal to the surface, plus sound pressure |
The standard states that both methods give a reproducibility standard deviation of 3 dB or less between different laboratories.
Instruments and calibration
Clause 4 calls for a type 1 sound level meter complying with IEC 60651 for sound pressure, and a class 1 intensity instrument complying with IEC 61043 for sound intensity. The frequency range must cover the transformer spectrum.
The equipment is calibrated immediately before and immediately after the measurement sequence. If the calibration changes by more than 0.3 dB, the measurement is declared invalid and repeated.
Load conditions
Load conditions are agreed when the order is placed (clause 6.1). The normal test is at no load, with rated voltage of sinusoidal or nearly sinusoidal waveform and rated frequency (clause 6.2). The tap changer is on the principal tap unless agreed otherwise, and this must be stated in the report. For more on where a transformer’s hum comes from: why a transformer hums.
Load-current noise is measured separately and added to no-load noise. That addition is explained in the guide on transformer sound power.
The principal radiating surface
Clause 7.2 defines this surface with an imaginary string. The string surrounds the unit and is projected vertically from the tank cover down to the tank base.
- Included: the tank, tank stiffeners, cable boxes, tap changers and cooling equipment less than 3 m from the tank.
- Excluded: bushings, turrets, oil pipes, the conservator, valves, control cabinets and cooling equipment 3 m or more from the tank.
If the cooling is on a separate structure 3 m or more away, clause 7.3 gives a surface of its own for that structure. Dry-type transformers without an enclosure have their own definition in clause 7.4.
The prescribed contour
The contour is the horizontal line on which the microphones are placed. Its distance from the radiating surface depends on the cooling state (clause 8).
| Condition | Contour distance |
|---|---|
| Fans off | 0.3 m |
| Fans off, dry type without enclosure, for safety | 1 m |
| Fans running | 2 m |
The height of the contour depends on the tank height.
- Tank lower than 2.5 m: one contour at half the height.
- Tank 2.5 m or taller: two contours, at one third and two thirds of the height. The height is lowered only for safety reasons.
If only the cooling equipment is energized, the standard applies the same rule to the cooler structure, with a 4 m height limit.
The measurement position can be modified for safety. With horizontal high-voltage bushings, for example, the contour is limited to the safe area.
- Principal radiating surface. String projected from the cover to the tank base, including cooling equipment less than 3 m away
- Contour distance. 0.3 m with fans off or 2 m with fans running
- Contour height. Half the height if the tank is lower than 2.5 m; one third and two thirds if it is 2.5 m or taller
- Microphones. At most 1 m apart and no fewer than six
The microphones
Clause 9 calls for roughly equidistant positions around the contour, no more than 1 m apart and at least six in total. A storage device with averaging is permitted: the microphone travels along the contour at roughly constant speed and at least as many samples are taken as there are positions. Only the energy average is recorded in the report.
The measurement surface area
The sound power level is obtained by adding to the mean pressure level the logarithm of the measurement surface area S (clause 13). That area depends on the distance (clause 10).
- At 0.3 m: S = 1.25 · h · lm, where h is the tank height and lm the contour length. The factor 1.25 is empirical and accounts for the sound radiated by the top.
- At 2 m: S = (h + 2) · lm, where h includes the fans if the cooling is separate.
- At 1 m, dry type without enclosure: S = (h + 1) · lm.
The test environment
The aim is a nearly free field over a reflecting plane, with no sound-reflecting objects nearby (clause 11.1). The standard prohibits measuring inside cells or enclosures.
Indoors, the reflecting plane is the floor, preferably with an absorption coefficient below 0.1. The environmental correction K depends on the ratio between the absorption area of the room and the measurement area. A room is satisfactory if that ratio is 1 or more, which gives a K of 7 dB at most. The maximum permitted value of K is 7 dB.
Outdoors, K is approximately zero if there are no reflections. Do not measure with temperature or wind gradients, with precipitation or at high humidity.
Background noise and validity
Background noise is measured immediately before the test and repeated with the transformer de-energized, at the same positions and heights (clause 11.2). It must be roughly constant. If the background is more than 10 dB below the combined level, it is enough to measure it at a single position and no correction is made.
The standard requires waiting for the transformer to stabilize. Residual direct current, recognized by odd harmonics in the spectrum, can change the level for minutes and, in extreme cases, for hours.
Table 2 of clause 11.3 decides whether the test is valid.
| Measured level minus background | Initial minus final background | Decision |
|---|---|---|
| 8 dB or more | any | Accept |
| Less than 8 dB | Less than 3 dB | Accept |
| Less than 8 dB | More than 3 dB | Repeat |
| Less than 3 dB | any | Repeat |
There is one exception: if the uncorrected measured level is lower than the guaranteed one, the transformer is considered compliant and the test need not be repeated. This condition is noted in the report. The standard recommends aiming for a difference of at least 6 dB.
With the intensity method, the difference between the mean pressure and the mean intensity must not exceed 8 dB(A) (clause 12.3).
What the report must contain
Clause 16 lists the content: manufacturer, date, transformer data, guaranteed level, method, instruments with their calibration, a dimensioned sketch of the measuring points, voltage, current, frequency and tap, contour length, height and effective area, and the people present.
When you need it in your project
Noise is guaranteed in the contract and verified in the factory test. The standard leaves the method (clause 5) and the load conditions (clause 6.1) to agreement between manufacturer and purchaser. Define them in the specification, before placing the order.
At Tiza Group we test power transformers at the factory and on site, in Colombia or elsewhere on request, and we support acceptance tests. The guide on transformer factory tests shows where noise measurement fits in the test protocol.
Related guides
- Why a transformer hums and when to worry
- Transformer factory tests
- Power transformer
- Transformer Sound Power: No-Load, Load and Far Field
- Why a Transformer Hums and When to Worry
References
- IEC 60076-10: power transformers, determination of sound levels (clauses 4 to 13 and 16).
- ISO 3746: determination of sound power levels from sound pressure.
- ISO 9614-1: determination of sound power levels from sound intensity.
Frequently asked questions
How far from a transformer are the microphones placed?
At 0.3 m from the principal radiating surface when the fans are off, and at 2 m when they are running. Clause 8 of IEC 60076-10 allows 1 m instead of 0.3 m only for safety reasons on dry-type transformers without an enclosure. The microphones go at half the tank height if the tank is lower than 2.5 m, and at one third and two thirds of the height if it is 2.5 m or taller.
How many measuring points does the standard require?
At least six microphone positions, spread roughly uniformly around the contour and no more than 1 m apart (clause 9). A large transformer needs more than six, because the contour is long and the maximum spacing governs.
What makes a noise measurement have to be repeated?
Background noise. If the background level is less than 3 dB below the level measured with the transformer energized, the test is repeated. It is also repeated if the background changes by more than 3 dB between the start and the end and is less than 8 dB below the measured level. If the measured level is lower than the guaranteed one, the unit is considered compliant (clause 11.3, table 2). A calibrator that drifts by more than 0.3 dB between before and after also invalidates the measurement.