Transformer derating for Neutral Harmonic Currents - Harmonic Current Scaling? 2

[budhiman]

Hello

I have a harmonic current spectrum from a system test performed which could be only loaded 25% of its rated load (the remaining loads are still under construction). All the loads are L-N well distributed. There is no other type of load in this system.

To obtain the full load (100%) harmonic spectrum can the harmonic orders be scaled by four (measured spectrum multiplied by 4)? Is there an empirical formula that can be used here OR since all loads will be well-distributed scaling should be fine?

Will be using this data for transformer derating calculation based on IEEE C57.110-2018.

 

[waross]

What is the nature of the loads?
Is distortion power factor a factor?
--------------------
Ohm's law
Not just a good idea;
It's the LAW!

 

[budhiman]

What is the nature of the loads? - Non-linear loads. Only loads on the system are switching power supply units L-N (277V)
Is distortion power factor a factor? Yes

 

[Maximillianrichards]

The harmonic spectrum of a system is proportional to the load current, so it is possible to scale the harmonic orders by multiplying the measured spectrum by a factor that represents the desired load level. In this case, if the system is currently loaded to 25% of its rated load and you want to determine the harmonic spectrum for a 100% load, you could multiply the measured spectrum by 4.

However, it is important to consider the type of loads and their distribution when scaling the harmonic spectrum, as this can affect the accuracy of the results. In this case, since all the loads are L-N well-distributed, scaling the harmonic orders by 4 should be a reasonable approximation.

It's also worth noting that the accuracy of the scaling method may depend on the measurement method used to obtain the harmonic spectrum, as well as the quality of the measurement data. It is always a good idea to validate the results of any scaling calculation with actual measurements taken at the desired load level, if possible.

For transformer derating calculation based on IEEE C57.110-2018, it is also important to consider the load types and the contribution of each harmonic order to the overall harmonic distortion level. The standard provides guidelines for calculating the harmonic derating factor for transformers based on the harmonic content of the load current, as well as other factors such as the type of transformer and the ambient temperature.

 

[waross]

Triplen harmonics sum on the neutral.
The fundamental currents cancel on the neutral.
Neutral current will be the unbalanced in phase current plus three times the triplen harmonic currents.
Seat of the pants engineering:
Measure your line currents. If they are not balanced, drop loads until the line currents are closely balanced.
Measure the neutral current.
The in-phase components will have cancelled, and you will be reading the harmonic currents.
Then, you may multiply this current by the reciprocal of the PU line current to estimate the neutral current at full load.

well-distributed scaling should be fine?

YES.

200% rated neutral bus is common for panels intended for harmonic loads.
By code, there shall be no reduction in the ampacity of neutral conductors. (You may decide to increase the neutral conductor ampacities above 100%. With significant harmonics, derating may be applied to circuits with three phases and a loaded neutral in a conduit or cable.
(More than 3 current carrying conductors.)

--------------------
Ohm's law
Not just a good idea;
It's the LAW!