If you’d rather listen than read, feel free to play the audio file below for the rest of this article.
Harmonics and power quality basics
Before any busbar physics, the vocabulary has to be right. Most confusion about busbar harmonics and power quality comes from treating THD as one number rather than as a summary of a spectrum.
What “harmonics” actually mean in AC systems
A harmonic is a current or voltage component at an integer multiple of the supply frequency. On a 50 Hz system the 5th harmonic is 250 Hz; on 60 Hz it is 300 Hz.
Linear loads draw current proportional to the applied voltage. Nonlinear loads do not — they draw current in pulses, and a pulsed waveform is mathematically identical to a fundamental plus a series of harmonics.
Nothing is injected in a mystical sense. The load simply draws a non-sinusoidal current, and the network sees that shape as a spectrum of frequencies sharing one conductor.
THDi vs THDv (and why both matter)
THDi is current distortion. THDv is voltage distortion. They are related but not interchangeable, and confusing them wastes a lot of commissioning time.
Loads create harmonic currents. Those currents flow through system impedance and produce harmonic voltage drops, which is what turns THDi into THDv. A stiff supply with low impedance can absorb substantial harmonic current with little voltage distortion; a weak supply cannot.
The practical consequence: THDi tells you what your loads are doing. THDv tells you what the rest of the switchboard is being subjected to. Log both, at the same point, at the same time.
Readers who want to understand the reasoning behind this will find this detailed article very useful.
Why RMS heating gets worse under distortion
Heating in a conductor follows I²R, and the I in that expression is total RMS current, not fundamental current.
Add harmonic components and total RMS rises even when the fundamental stays constant. A 30% THDi load draws measurably more RMS current than a linear load of the same kW, and the busbar heats accordingly.
Then the second effect compounds it. R is not constant with frequency, which is where the busbar physics in the next sections begins.
Because the situation can change over time, we suggest checking the latest published information before you make a decision.
Distortion behaves differently depending on the conductor system carrying it. A short refresher on electrical busbars makes the physics in the following sections easier to place.
Common harmonic sources in switchboards
Identifying the source matters more than the total, because the mitigation choice follows the spectrum rather than the headline figure.
VFDs and rectifier front ends
The six-pulse diode bridge is the workhorse front end on drives and rectifiers, and it produces a characteristic signature: dominant 5th and 7th, then 11th and 13th.
Those are negative- and positive-sequence orders. They do not accumulate in the neutral, which is why a VFD-heavy plant can show high THDi with a perfectly ordinary neutral current.
What they do instead is load the phase busbars and interact with any capacitance in the system. That interaction is the resonance problem covered later.
UPS systems and data center loads
Modern UPS rectifiers are far cleaner than older designs, but the load behind them is not. IT equipment runs on switch-mode power supplies, and those are single-phase nonlinear loads.
That changes the spectrum entirely. Single-phase SMPS loads generate strong 3rd harmonic, and the 3rd is a zero-sequence component.
A data hall and a pump room can show similar THDi and present completely different risks to the switchboard. The spectrum, not the total, is what tells them apart.
Welders, LED drivers, EV chargers, SMPS loads
The nonlinear load population has grown quietly. LED drivers replaced fluorescent ballasts, EV chargers arrived in car parks, and every desk carries several SMPS.
Peer-reviewed modelling of a 400/230 V three-phase four-wire network shows how far this can go. Under highly distorted conditions the neutral current reached 180% of nominal phase current. Joule losses on the neutral ran more than three times those on a phase conductor.
Welders add a different problem — intermittent, unbalanced draw that produces flicker as well as distortion.
Which nonlinear loads end up on the same bar is decided long before commissioning. The way distribution boards are grouped often explains why one board sees a clean spectrum and another does not.
How harmonics impact busbars and switchboards
Here is where harmonic effects on busbars stop being an abstraction. Three separate loss mechanisms rise with frequency, and they act on the same conductor at the same time.
Extra RMS current and temperature rise
Ampacity tables assume sinusoidal current at the fundamental frequency. That assumption is doing a lot of quiet work.
Under distortion, total RMS current exceeds the fundamental component, so a busbar sized on a 50 Hz calculation is already running closer to its thermal limit than the drawing suggests. Temperature rise limits, not conductor melting, are what the design has to satisfy.
Copper works against you here. Resistance rises roughly 0.4% per degree Celsius, so a hot bar is a more resistive bar, which then runs hotter still.
Skin effect, proximity effect, and eddy-current losses
Skin depth falls as frequency rises. At 50 Hz, skin depth in copper is around 9.3 mm; at the 11th harmonic it is roughly a third of that.
The consequence for a wide, thin busbar is direct. Current crowds toward the surface and the edges, effective cross-section falls, and AC resistance climbs. Skin effect in busbars is a fundamental-frequency problem already — harmonics make it materially worse.
Proximity effect in busbars adds to it. Adjacent phase conductors distort each other’s current distribution, and that distortion also worsens with frequency. Eddy current losses in busbars appear in nearby steel too: enclosure panels, support brackets and gland plates all heat up without carrying load current.
Joint heating and hot spots under distortion
A bolted joint is a local resistance in an otherwise continuous conductor. Raise the current through it and the heating rises with the square.
That makes joints the first place distortion shows up thermally. A joint that was marginal under linear load — slightly under-torqued, lightly oxidized, or assembled on a burred hole — becomes a hot spot under a distorted one.
The failure loop is self-reinforcing. Heat accelerates oxidation, oxidation raises contact resistance, and higher resistance produces more heat.
For a clearer and more complete explanation, this page goes into far more depth than we can cover in a short text.
Mechanical vibration and audible noise side effects
Current in parallel conductors produces force between them. That force varies at twice the frequency of the current producing it.
Under distortion the force spectrum gains components at higher frequencies, and if any of those coincide with a mechanical resonance of the bar or its supports, the assembly buzzes.
Audible hum from a switchboard is worth investigating rather than tolerating. It is often the cheapest available symptom of a distortion problem, and it points at support spacing as well as at the loads.
These loss mechanisms are amplified by how the bars sit inside the enclosure. The design considerations for LV switchgear busbars cover the clearance, support and geometry choices behind them.
Busbar overheating and neutral overloading due to harmonics
Most switchboard failures trace back to busbar overheating harmonics produce, and a THD reading alone will not reveal it. Busbar harmonics and power quality problems concentrate in this section.
Triplen harmonics (3rd, 9th, 15th) and why they add in neutral
Fundamental currents in a balanced three-phase system are 120 degrees apart and sum to zero in the neutral. Triplen harmonics are not.
The 3rd, 9th and 15th are zero-sequence components — in phase with each other across all three phases. They add arithmetically instead of cancelling, so neutral current from the 3rd harmonic alone is three times the per-phase 3rd harmonic current.
In harmonic-rich LV installations, triplen harmonics in neutral conductors commonly produce 150% to 170% of phase current, with a theoretical worst case near 1.73 times. Note what that means: the neutral becomes the most thermally stressed conductor in the circuit while every phase reads within rating.
Neutral conductor and busbar sizing implications
There is a published threshold worth committing to memory. IEC 60364-5-52 Clause 524 addresses this directly. Where the 3rd harmonic content of phase current exceeds 33%, neutral current exceeds phase current, and the neutral becomes the basis for conductor selection.
Annex D of the same standard gives a derating methodology for four-core cables. The same logic transfers to a busbar system, even though busbar trunking is dimensioned differently.
One point deserves emphasis because it changes the risk category. Neutral conductors are generally not protected by overcurrent devices. An overloaded phase trips a breaker; an overloaded neutral simply gets hot.
Warning signs (discoloration, insulation aging, smell, trips)
Neutral overheating harmonics announce themselves before they fail, but the signs are easy to walk past.
Discoloration on insulation or on the bar itself near a joint. A faint hot-plastic smell in a panel that has no obvious fault. Insulation that has gone hard or brittle at one location while the rest of the run is fine. Breakers that trip below their nominal setting with no fault current recorded.
Thermal imaging finds all of these faster than inspection does, and it is the one measurement that connects an analyzer reading to a physical location.
Neutral loading also depends on how the neutral is terminated and distributed inside the panel. Connection practice at terminal bus bars decides whether that current meets clean joints or marginal ones.
Symptoms of poor power quality in switchgear
Harmonic distortion in switchboards rarely presents as a harmonic problem. It presents as unexplained equipment behavior, and gets diagnosed as something else.
Nuisance tripping and protection misbehavior
Thermal-magnetic breakers respond to heating, and distorted current heats more than its fundamental component suggests. A breaker can trip at what the metering reports as 80% load.
The reverse error is more dangerous. An averaging-type instrument reading a distorted waveform under-reports current, so a genuinely overloaded circuit looks comfortable. True-RMS instruments are not optional in a distorted system.
Residual current devices are affected too, through high-frequency leakage via equipment filter capacitance.
Transformer heating and capacitor stress
Transformers suffer twice under distortion: eddy-current losses in the windings rise roughly with the square of harmonic frequency, and triplen harmonics circulate in delta windings without appearing on the line.
The result is a transformer running hot at a load the nameplate says is comfortable. K-factor rated units exist precisely for this.
Capacitors have the opposite problem. Their impedance falls as frequency rises, so they act as a sink for harmonic current, which is where the resonance risk begins.
Voltage distortion, flicker, and sensitive loads
Once THDv rises, the problem stops being confined to the circuit that caused it. Every load on that busbar sees the distorted voltage.
Symptoms are diffuse and get blamed elsewhere. Control electronics resetting, drives faulting on DC bus ripple, metering disagreeing between devices, and lighting flicker where large intermittent loads share the board.
That diffuseness is why measurement has to come before mitigation.
If you would like to explore this subject further, you can read more about it here.
Left unaddressed, these symptoms escalate into damage rather than nuisance. Documented busbar failures show how often thermal warning signs preceded the event.
How to measure harmonics and THD in distribution systems
Here is the diagnosis workflow, in order: measure, interpret THDi and THDv together, check neutral and thermal behavior, then decide on mitigation. Skipping straight to the last step is how filters get installed that do not fix anything.
What to log: THDi, THDv, harmonic spectrum, neutral current
Four data sets, and the third is the one most often skipped.
Log THDi and THDv at the same point. Log the individual harmonic spectrum, not just the totals. Log neutral current as a measured value rather than a calculated one. And log all of it over a period that covers the real duty cycle, not a quiet afternoon.
The reason the spectrum matters is decisive. An installation dominated by the 3rd and one dominated by the 5th and 7th can report almost identical THD while presenting completely different risks. Harmonic spectrum analysis is what separates them.
Recommended machine
HBC-A120
Cuts, punches and bends the exact copper and aluminium bar sizes in this guide — up to 120 × 12 mm, three stations at once.
120 × 12 mm
3 power packs
Cut · Punch · Bend
Where to measure (main incomer vs feeders vs critical loads)
Measurement point changes the answer, which is why a single reading proves very little.
The main incomer tells you what the utility sees and whether a compliance obligation is being met. Individual feeders tell you which loads are producing the distortion. The terminals of a critical load tell you what that equipment is actually enduring.
An internal bus close to a heavy nonlinear load can show distortion well above what the point of common coupling reports, and still be entirely within the standard.
Trending and thermal imaging correlation
A single snapshot catches a moment. Trending catches the duty cycle, and distortion is usually worst at a specific operating state.
Correlate the electrical trend with a thermal survey taken at similar load. A hot spot that appears only when a particular drive runs is a diagnosis; a hot spot recorded once at unknown load is an observation.
Record the load at every survey, so the next reading can be compared rather than guessed at.
Measured data is only useful if it feeds back into the design. Modern busbar design software lets you test a revised cross-section against real logged current instead of a nominal figure.
Harmonic mitigation methods
Harmonic mitigation in switchboards works at three levels: reduce distortion at the source, block it at the load, or absorb it in the system. Cost and effectiveness rise in that order, and so does complexity.
Active filters: where they shine
An active harmonic filter measures the load current, calculates the harmonic content, and injects an equal and opposite current in real time.
Its advantage is adaptability. It handles a changing spectrum, a changing load, and multiple harmonic orders simultaneously, and it cannot resonate with the system because it is not a tuned circuit.
The costs are capital and complexity: highest price per amp of correction, a device with control electronics that needs commissioning and maintenance, and its own small losses. Choose it where the load mix is varied or changes over time.
Passive filters: tuning, resonance risks
A passive harmonic filter is an LC circuit tuned to present low impedance at a specific harmonic order, giving that current a preferential path.
It is cheap, robust and has no control electronics. It also does exactly one job. Tune for the 5th and the 7th stays where it was.
The risk is that a passive filter is a tuned circuit connected to a network whose impedance changes with configuration. Change the transformer, add generation, or open a tie breaker, and the filter can end up amplifying rather than absorbing.
Recommended machine
HBC-C200
Cuts, punches and bends the exact copper and aluminium bar sizes in this guide — up to 120 × 12 mm, three stations at once.
120 × 12 mm
3 power packs
Cut · Punch · Bend
Line reactors and multipulse approaches
Both of these attack the problem at the source, which is where it is cheapest to solve.
A line reactor in series with a drive input smooths the current pulse the rectifier draws. Typical 3% and 5% impedance reactors reduce THDi substantially for very little money, and they protect the drive from supply transients as a side benefit.
Multipulse rectifiers go further by cancellation. A 12-pulse arrangement removes the 5th and 7th by phase-shifting two six-pulse bridges; 18-pulse removes more still. The trade-off is a phase-shifting transformer and the space it occupies.
Detuned capacitor banks and avoiding resonance
Do capacitor banks make harmonics worse? On their own, frequently yes. A capacitor bank and the supply transformer inductance form a parallel resonant circuit, and if that resonance lands near an existing harmonic, amplification follows.
Detuned reactors solve it by placing a reactor in series with the capacitor, moving the resonance below the lowest significant harmonic. On 50 Hz networks the common tunings are p = 7% at about 189 Hz and p = 5.67% at about 210 Hz. Where the 3rd harmonic dominates, p = 14% gives about 134 Hz.
The design rule is straightforward: put the resonance safely below the dominant harmonic frequency, typically under 90% of it. IEC 61642 covers industrial networks affected by harmonics, and IEC 60831 governs the capacitors themselves.
Capacitor decisions rarely stand alone from the reactive compensation strategy. Understanding power factor correction is what keeps a detuning choice consistent with the compensation target.
Standards and limits context
Compliance framing matters here, because the most widely repeated number in this field is being applied to the wrong place.
See our machines in action
Practical interpretation (project-dependent limits)
The familiar “5% THD” figure is not a universal rule, and under the current edition it is not even the LV number.
IEEE 519-2022 introduced a separate voltage class for systems at or below 1 kV. Individual harmonic voltage is limited to 5% and total harmonic distortion to 8%, both higher than the band immediately above. Current limits are different in kind: they scale with the Isc/IL ratio, because a stiff supply tolerates more harmonic current than a weak one.
Two further points decide most disputes. The limits apply at the point of common coupling, not at every internal bus. And an internal bus reading above the PCC figure can still be entirely compliant while damaging the equipment connected to it.
Documentation and acceptance testing mindset
Write the acceptance criterion before the equipment arrives, not after a problem appears.
A defensible specification names four things: the measurement point, the load condition under which the measurement is valid, the indices being assessed, and the standard edition being applied. Without the load condition, a compliant result at partial load can hide non-compliance at full load.
Treat the result as a record rather than a certificate. Distortion changes when the load mix changes, and every commissioning measurement is a snapshot of one configuration.
Harmonic limits sit alongside the construction rules the assembly must already satisfy. A working knowledge of busbar standards keeps the acceptance specification consistent across both.
Source, symptom and mitigation map
One table connecting what is on the board to what to do about it.
| Source | Harmonic signature | Symptom at the busbar | What to measure | Typical mitigation |
|---|---|---|---|---|
| VFDs, six-pulse rectifiers | 5th, 7th, 11th, 13th | Phase busbar heating, capacitor stress | THDi at feeder, spectrum, capacitor current | Line reactor, 12/18-pulse, active filter |
| UPS and IT loads (SMPS) | Strong 3rd, triplen | Neutral overheating, transformer heating | Neutral current, 3rd harmonic %, neutral temperature | Full or oversized neutral, zigzag, active filter |
| LED drivers, small SMPS | 3rd dominant | Neutral overload on lighting boards | Neutral current vs phase current | Neutral sizing, load redistribution |
| EV chargers | Mixed, load-dependent | Rising THDi as installed base grows | Trended THDi and spectrum | Active filter, source-side reactance |
| Welders | Intermittent, unbalanced | Flicker, voltage distortion, nuisance trips | THDv, flicker, trend under duty cycle | Dedicated supply, source stiffening |
| Capacitor banks (as amplifier) | Amplifies existing orders | Capacitor failure, fuse operation, hot bank | Capacitor current, resonance check | Detuned reactor at 189/210/134 Hz |
How the board is configured changes which of these risks concentrates where. Comparing busbar arrangements shows how layout redistributes both load and fault exposure.
Choosing between mitigation options
The decision matrix below is the honest version. No option is best in general; each is best under stated conditions.
| Option | Best when | Effectiveness | Resonance risk | Relative cost |
|---|---|---|---|---|
| Line reactor (3–5%) | Distortion originates at identifiable drives | Moderate, source-side | None added | Lowest |
| Multipulse rectifier (12/18) | New installation, large single drives | High for characteristic orders | None added | Moderate, needs space |
| Passive tuned filter | One dominant order, stable network | High at tuned order only | Significant if network changes | Moderate |
| Detuned capacitor bank | Power factor correction needed in a distorted network | Prevents amplification, does not remove harmonics | Eliminated by design | Moderate |
| Active harmonic filter | Varied or changing load mix, multiple orders | Highest, adapts in real time | None | Highest |
| Equipment selection (low-harmonic drives) | Specifying new equipment | High, avoids the problem | None | Varies, often lowest lifecycle |
Mitigation hardware is only half the specification; the bar itself is the other half. This selection guide covers the material, section and grade choices that sit behind it.
Commissioning checklist
Ten checks, in order. Work through them before specifying any mitigation equipment.
- Log THDi and THDv at the incomer, at the same point and time.
- Capture the individual harmonic spectrum, not just the totals.
- Measure neutral current directly — never calculate it from phase balance.
- Record the 3rd harmonic as a percentage of phase current, and check it against the 33% threshold.
- Identify which feeders carry the nonlinear load, and measure them separately.
- Log over a period covering the full duty cycle, including start-up and peak.
- Run a thermal survey at comparable load, and record the load with the images.
- Check every capacitor bank for current above its rating and for resonance risk.
- Confirm the neutral conductor and neutral busbar rating against measured neutral current.
- Verify that instruments used are true-RMS, and that the standard edition and measurement point are stated in the record.
Thermal findings should be checked against the assembly’s physical margins, not only its ratings. Verifying busbar clearances during the same survey turns a temperature reading into an actionable one.
Conclusion about Busbar Harmonics and Power Quality
Harmonics are not just numbers on an analyzer. They are extra RMS current in a phase bar, and arithmetically accumulating current in a neutral. They are frequency-dependent losses in copper and in nearby steel, and protection devices responding to a waveform their settings never anticipated.
That is why the spectrum matters more than the total, and why the neutral deserves a separate measurement rather than an assumption.
Measure first. Interpret THDi and THDv together. Check the neutral and the thermal picture. Only then choose between a reactor, a filter, a detuned bank or a different piece of equipment. The right answer depends on the spectrum, and the spectrum is something you have to go and read.
Nonlinear load will keep growing, so today’s spectrum is not tomorrow’s. Watching the future trends in busbar systems is part of designing for a moving target.
Rate this article
Choose a rating from 1 to 5 stars.
No ratings yet