What a power analyzer measures, and why it is not a clamp meter

An ordinary clamp meter answers one question at a time: how many amps are flowing right now. That is adequate for confirming a motor is drawing roughly what its nameplate says. It cannot tell you why your municipal account shows a maximum demand charge far above your metered consumption, why a transformer runs hot at two-thirds of its rated load, or why a neutral conductor is warmer than the phases feeding it. Those questions are all about the relationship between voltage and current — phase displacement, waveform shape, imbalance between phases — and answering them requires simultaneous multi-channel sampling.

The practical consequence is that a power analyzer is a diagnostic and datalogging instrument rather than a spot-check tool. The three-phase Extech models write measurements continuously to a removable SD card in Excel format, so a fault that appears for ninety seconds at 04:00 on a Tuesday is captured and timestamped rather than missed. This is what makes the category useful for energy audits, tariff disputes, generator commissioning and intermittent-fault investigations, none of which can be resolved by standing at a distribution board with a handheld meter.

The three instrument types on this page

Three-phase analysers and dataloggers — the PQ3450 family, PQ3350-3 and 382100 — connect to all three phases and neutral at once, measure up to 35 parameters, and log to SD card over days or weeks. These are the instruments for energy audits, maximum demand studies and power quality investigations.

Power clamp meters — the PQ2071 and 380976-K — are handheld single-instrument tools that measure true power, apparent power, reactive power and power factor on single-phase and balanced three-phase loads. They are for the electrician who needs a power reading during a service call, not a week-long recording.

AC voltage and current dataloggers — the DL150 and DL160 — record voltage or current over time without deriving power. They answer questions about load profile, cycling behaviour and voltage stability at a fraction of the cost and complexity of a full analyser.

How to choose an Extech power analyzer

Four decisions determine which instrument is right. Work through them in order — each one eliminates part of the range.

1. How many phases, and do you need them simultaneously?

If the load is single-phase, or three-phase and genuinely balanced, a power clamp meter such as the PQ2071 is sufficient and far quicker to deploy. If the phases are unbalanced — which is normal in any installation with a mix of single-phase and three-phase loads — a clamp meter measuring one phase and multiplying by three will give you a materially wrong answer. Unbalance itself is often the fault you are looking for, and only a true three-phase analyser measures it.

2. What is your maximum current, and what shape is the conductor?

Match the probe range to your expected maximum rather than buying the largest available — a 3000 A probe used on a 40 A circuit wastes most of its resolution. Below 200 A, the PQ34-2 probes with their 19 mm jaw are the right choice. Up to 1200 A, the rigid PQ34-12 clamps bundled with the PQ3450-12 and the 382100's included clamps both work. Above 1200 A, or where the conductor is a busbar, a bundle of parallel cables, or simply awkwardly placed in a crowded panel, you need the flexible PQ34-30 probes supplied with the PQ3450-30 — a flexible coil wraps around what a rigid jaw cannot close over.

3. Are you measuring energy, or diagnosing power quality?

These are different jobs. An energy or demand study asks how much and when: log kW and kVA over a week, find the peak, calculate the cost. Any of the PQ3450 family or the 382100 does this well. A power quality investigation asks what shape the waveform is: harmonic spectrum, THD, transient dips and swells, unbalance factors. The PQ3350-3 is built around this, displaying harmonics and waveform simultaneously on one screen so you can see distortion and its effect at the same moment rather than switching between screens.

4. How long must it record, and does it need to be unattended?

The PQ3450 family and 382100 store 30,000 measurement sets on a removable SD card with sample intervals from two seconds to two hours. At a one-minute interval that is roughly three weeks of continuous unattended logging — enough to capture a full billing cycle's load pattern. If you only need voltage or current over time and not derived power, the DL150 (100,000 readings, one channel) and DL160 (256,000 readings, two channels) are purpose-built dataloggers that cost and weigh considerably less.

Extech power analyzer comparison

All three-phase models are rated CAT III 600 V, covering South Africa's 400 V phase-to-phase / 230 V phase-to-neutral supply at 50 Hz.

← Scroll to see full table →

Discontinued model: Extech PQ3350-1

Current clamp probes & accessories

The PQ3450 and PQ3350-3 are supplied without current probes so the range can be matched to the installation. Probe choice sets the analyser's maximum current and, just as importantly, its resolution at the low end — a 3000 A probe on a 60 A circuit will read, but poorly. Select the range closest above your expected maximum.

← Scroll to see full table →

Probes are ordered together with the analyser in most cases. Email sales@ecotao.co.za with your conductor size and expected maximum current and we can confirm the right combination.

Power analysis under South African conditions

South Africa's supply environment creates measurement requirements that generic international product literature does not address. Four situations account for most of the power analyser work done here.

Maximum demand tariffs are billed in kVA, not kW

Municipal and Eskom large-power tariffs charge separately for energy consumed and for maximum demand, and that demand charge is levied in kVA. The distinction is where money is lost. A site drawing 400 kW at a power factor of 0.75 presents 533 kVA to the meter; the same 400 kW at a corrected power factor of 0.95 presents 421 kVA. The work done is identical, the energy charge is identical, and the demand charge differs by more than a hundred kVA every month.

Establishing whether that gap exists requires logging kW, kVA and power factor together across a representative period — a full week at minimum, so weekday and weekend patterns are both captured — and reading the sustained peak over the tariff's integration window rather than the instantaneous maximum. The PQ3450 family and 382100 log directly to SD card in Excel format, so the analysis can be done in a spreadsheet rather than through proprietary software. This is also the dataset you need if you intend to query a demand charge with your supplier, since an argument backed by timestamped measurements carries weight that an estimate does not.

Load shedding, generator changeover and inverter transfer

Scheduled supply interruption has made transfer events routine, and transfer is where equipment damage tends to originate rather than during the outage itself. Voltage dips as the generator picks up load, frequency wanders while the governor settles, and phase unbalance can be markedly worse on generator supply than on utility supply because the source impedance is higher. Sensitive equipment that survives the outage sometimes does not survive the return.

An analyser logging across a full shedding cycle captures the transfer in both directions with timestamps, which turns a suspicion into evidence. With 30,000 measurement sets and a one-minute sample interval, the PQ3450 family covers roughly three weeks unattended — enough for several shedding stages plus the surrounding normal operation needed as a baseline for comparison. The transient capture capability of the PQ3350-3, which records dip, swell and outage events against a programmable threshold, is the better fit where the events themselves rather than the overall pattern are the subject of investigation.

Harmonics from solar inverters and variable speed drives

Rooftop PV, battery inverters and VSD-driven pumps and fans have proliferated across South African commercial, agricultural and industrial sites. All are non-linear loads: they draw current in pulses rather than smoothly, injecting harmonic currents back into the installation. The consequences are physical. Transformers and cables heat beyond what their RMS current rating would suggest. Neutral conductors carry triplen harmonics that can exceed phase current, which is a genuine hazard in older installations where the neutral was sized smaller than the phases. Capacitor banks installed for power factor correction can be driven towards resonance with the supply impedance.

NRS 048-2 sets compatibility levels for voltage harmonic distortion, unbalance and voltage dips on South African networks, so measuring total harmonic distortion is frequently a compliance exercise as well as an equipment-protection one. The PQ3350-3 analyses THD and displays the harmonic spectrum alongside the waveform, which is the measurement to take before adding capacitor banks or a large drive to an existing installation rather than after the resulting failure.

Phase unbalance on rural and agricultural supplies

Long rural feeders with a mix of single-phase and three-phase loads develop voltage and current unbalance as a matter of course. The effect on three-phase motors is disproportionate: a small percentage of voltage unbalance produces a much larger percentage of current unbalance, and the resulting negative-sequence current heats the rotor without contributing useful torque. Irrigation pump motors failing repeatedly on a farm supply are a common presentation of this, and it is regularly misdiagnosed as a motor quality problem.

Measuring it requires simultaneous three-phase readings — voltage and current unbalance ratios and unbalance factors, plus the calculated neutral current, which the three-phase analysers report directly. A clamp meter used on one phase at a time cannot detect unbalance at all, since the phases are not measured at the same moment.