Your questions answered: Meeting power quality needs

Several questions about power quality and harmonics are answered here

By Zia Salami and Tom Divine December 14, 2021
Courtesy: Consulting-Specifying Engineer

Devices using power electronics can produce distortion in electrical distribution systems, and it’s up to the electrical engineer to apply effective solutions to mitigate this added distortion. These solutions will ensure high-quality power is maintained within a nonresidential building, especially one with sensitive equipment or unique applications. 

This Critical power: Meeting power quality needs webcast on Dec. 7, 2021, covered the topic in depth, and several questions have been addressed in writing.  

Questions have been answered by: 

  • Tom Divine, PE, senior electrical engineer, Johnston, LLC, Houston 
  • Zia Salami, Ph.D., Management Specialist – SME: Electrical Power System, CDM Smith, Charlotte, North Carolina 

What’s the difference between harmonic mitigation versus power factor correction? 

Zia Salami: Harmonics (caused by non-linear loads) usually cause distortion power factor (hPF) for more heavily loaded systems. The addition of power factor correction capacitors alone won’t help this situation; harmonics mitigation equipment is needed. A low pf indicates that a power system is not being effectively utilized — unity is ideal — but if this occurs only when the system is lightly loaded and the reactive power component is quite low, there is no need for concern. 

Also, I need to point that the true power factor (TPF) is a combination of displacement power factor (dPF) and distortion power factor (hPF). TPF = dPF x hPF. TPF becomes very low if system has harmonic distortion. Displacement power factor (dPF) is the one that we typically know and is all about phase angle between fundamental voltage and current. The only time TPF is close to dPF that we remove the harmonic current from a system by different mitigations techniques. The addition of power factor correction capacitors will only help to improve dPF.  

What will be a bigger source of power quality issues, variable grid operations or power quality problems originating within a facility? 

Tom Divine: Conventional wisdom says that most of the power quality issues that a customer experiences will originate on his own site, rather than on the utility system. Late in the previous century, it was fairly common for a utility to maintain an in-house consulting services that would investigate power quality issues at customer facilities for a fee, with the kicker that the utility would forgo its fee if the issue were found to be on the utility system. Industry scuttlebutt said that the utilities won their bet at least 90% of the time. 

You spoke about modelling for passive filters. What about active filters? 

Zia Salami: I don’t recall any electrical software that can model active filter design within the harmonic study module. As mentioned in the presentation, some active filters can target not only a single harmonic, but can mitigate all harmonics (2nd to 50th). An active filter modeling work-around in ETAP (and perhaps other software packages) is to model several passive filters tuned to target each the worse harmonic offenders. 

Are there types of variable frequency drives (i.e., AC to AC, active front end, etc.) that do not cause harmonics back on to the power distribution system? 

Zia Salami: Simple answer: no. All VFDs produce harmonics, but the higher the pulse-rate (e.g., 18 or 24), the lower the harmonics. All design is a trade-off — higher pulse-rate = more complex design = higher cost. The best approach is to build a detailed power system model (design or existing system) and perform various what-if harmonic analyses to determine optimal design. 

Will a VFD connected to a pump/motor contain any harmonics from the pump/motor? 

Zia Salami: No. The VFD produces a voltage wave for the motor (optimized voltage magnitude and frequency). The current delivered to the motor is based on the power demand and impedance of the motor/pump. The harmonics produced by the VFD (and returned to the VFD power source) will be based on the VFD design pulse-rate (e.g., 6, 12, 18, 24), the VFD control mode (volts per hertz (V/Hz) and sensorless vector (SV) are the main modes) and the demand placed on the VFD by the motor. 

When we have power studies performed for Isc, coordination and arc flash, should we also have harmonic distortion study performed at the same time? Normally we have the power study performed when manufacturer’s equipment is being selected. 

Zia Salami: Simple answer: yes, if the design is expected to either generate harmonics or be expected to handle harmonics from the grid or standby sources. In all likelihood, yes — new commercial and/or industrial facilities almost always contain VFDs and computer systems — both generate harmonics. Do the study to see if harmonics mitigation equipment (including higher pulse-rate VFD design) is needed. 

Is it better to use transformer that are K-rated or transformer with zigzag secondary (0° or -30° displacement between the primary and the secondary)?  

Zia Salami: It is all depend on your system and type of issue that you are facing with. Existing harmonic distortion or equipment overloading and heating issue? In general, zigzag transformer is more effective to trap the lower harmonic and reducing the distortion. A K-factor transformer is an oversized transformer with more copper to protect itself against heat generated by harmonics currents. 

Should we specify transformers with K-factors to protect from harmonic loads? And how do we determine the K-factor value? 

Tom Divine: To calculate the K-factor for a transformer: 

For each harmonic  including the fundamental  take the square of the product of the harmonic number and the current at that harmonic. Sum those results for all harmonics and divide by the square of the transformer’s rated current. The result is the K-factor for the transformer under the specified loading conditions. 

In practice, the magnitude of the current harmonics isn’t known at design time, for a couple of reasons: manufacturers rarely publish harmonic loading information for their equipment and owners often delay final selections of electronic equipment until construction is nearly complete, in order to take advantage of technological advances and price reductions that are likely to occur between design and occupancy. The process of determining whether a K-rated transformer is required for a particular installation and, if so, what that K-rating should be, is often as intuitive as it is analytical. 

Transformer manufacturers often provide rules of thumb about applying K-rated transformers, though they often vary from one manufacturer to another. I’m reluctant to provide one here. You may want to consult your preferred transformer manufacturer for guidance in selecting K-rated transformers. 

For power inverter, does a “pure sine” inverter will eliminate the harmonics or it’s just a marketing point of view and you can’t reach true sine wave? 

Zia Salami: No known VFD will produce a “pure sine-wave.” Pure sine-wave is an ideal; even synchronous generators don’t produce the ideal. 

Is it possible to cause electrical system resonance with a passive or active harmonic filter? 

Zia Salami: There is possibility to have parallel resonance by having passive filter within a system especially if there is a capacitor bank. Much lower chance (to none) with active filter. 

Should new switchboards or switchgear be specified with power quality meters to monitor power quality and capture poor power quality events for analysis? 

Zia Salami: Of course, this is subjective answer. Cost prohibitive for SWBDs (there could be many). The probable best approach is to procure a high-quality PQ meter for continuous duty and with high internal memory data capture capability (several days continuous data gathering). For SWGR, specify test connection jacks (like ABB FT-1) at the PT compartment or incoming (from grid, but at bus-side) breaker compartment for easy power quality meter input connections. 

Comment on 50 Hertz system? 

Zia Salami: Most of what we discussed will apply on 50 Hz system as well with the difference that the fundamental frequency instead of 60 is 50 Hz.  

What are best practices to eliminate excessive THD? 

Zia Salami: There are several techniques as pointed out, passive and active filters, higher pulse converters, de-rated standard transformer or using K-rated transformer, adding 3% to 5% reactor in series in front of non-line loads (e.g., VFD), balancing loads specially if you have many single-phase loads, sizing neutral conductor properly and more. Selection depends on if it is a new design or existing system, the cost, medium- or low-voltage system, amount of harmonic distortion in the system, etc.  

Is it true that the neutral current can be 1.73 times the line current in a perfectly balanced 3-phase, 4-wire load comprised of multiple single-phase circuits rich in harmonics? For example, a panelboard feeder? 

Tom Divine: That’s the theoretical maximum neutral current under the conditions described. What’s necessary is that each phase deliver current for only a third of each half-cycle. The phase currents won’t overlap and they’ll add independently in the neutral. The RMS neutral current will then be, in spreadsheet-like notation, 

IN = sqrt ( 3 * IP^2) = sqrt(3) * IP, where IP is phase current and IN is neutral current. 

Sqrt(3) is very nearly 1.73. So, yes, under these conditions, the neutral current can reach that level. 

If a facility has a great number of harmonic loads throughout the facility, what can be done to mitigate the potential for a great amount of harmonic distortion throughout the system? Would power factor correction equipment help? 

Zia Salami: I have discussed several techniques in similar questions. Also as discussed before, power factor correction equipment (e.g., capacitors) alone won’t help this situation; harmonics mitigation equipment is needed.  

What is rough magnitude of cost difference between passive and active filter technologies?

Zia Salami: Rough magnitude, probably 5 to 8 times more for active filter compared to passive filter. Active filter is very expensive, but with higher efficiency.  

The higher pulse converters reduce the more troublesome harmonics, but do they have the potential to create any other problems? 

Zia Salami: Higher pulse converters (i.e., VFDs/VSDs) disadvantages are higher cost and consequent larger footprint. I am not aware of any known disadvantages. 

Are there any measurement devices that also indicate the best method of mitigation or is this always an engineering decision? 

Zia Salami: No known measuring devices that indicate the best method of mitigation. Best method for such indication is harmonics analysis with what-if case analysis. 

Converters with higher pulse counts reduce harmonics for troublesome harmonics, but do they have the potential to create other problems? 

Zia Salami: Simple answer is probably no. Of course, all design is a trade-off — higher pulse-rate is more efficient to reduce harmonic distortion lower, however it is more complex with higher cost. For the impact study, I suggest building a detailed power system model (design or existing system) and perform various what-if harmonic analyses to determine optimal design and possible system impact. 

How do you balance harmonics reduction and power factor correction? 

Zia Salami: As I discussed in similar questions, improving power quality such as reducing harmonic distortion within a system will improve total system power factor (i.e., true power factor) 


Author Bio: Tom Divine, PE, senior electrical engineer, Johnston, LLC, Houston. Zia Salami, Ph.D., Management Specialist – SME: Electrical Power System, CDM Smith, Charlotte, North Carolina