1What a failed emissions test costs

The worked case in this paper takes an accredited 3 m semi-anechoic chamber day at 2,000 to 4,000 US dollars and a first radiated scan at two or three days (the bench and cost assumptions are collected in Table 3). The larger cost is the loop: the product fails at 240 MHz, a ferrite is fitted, and the chamber is rebooked three weeks out. Two loops put a launch about two months late.

A bench is not a chamber, and a bench scan does not produce a number a certification body accepts. It answers the three questions that set the loop count: which frequencies lie close to a limit, what is radiating them, and whether the latest change helped. None of the three needs an accredited site. All three need the measurement made as the standard defines it, because a scan at the wrong bandwidth is not conservative; it is wrong by a computable amount, and section 2 computes it.

2What the standard specifies

CISPR 16-1-1 describes a measuring receiver, not a spectrum analyzer, and it fixes what an analyzer leaves to the operator: band boundaries, resolution bandwidth, detector time constants and overload factor. Table 1 collects them, and Figure 2 draws the bands on one frequency axis.

Bandwidth enters a bench scan twice. CISPR specifies it at the −6 dB points; analyzers specify it at the −3 dB points, a narrower filter for the same number. And a narrowband carrier reads the same at any bandwidth wide enough to hold it, while a clock edge does not, because a broadband emission follows bandwidth in volts rather than power:

Figure 1 plots the error against bandwidth, and the error is signed. Table 1 gives the settings the standard specifies, band by band, with the 3 dB equivalents an analyzer without CISPR bandwidths is set to.

The highest frequency a radiated scan reaches is set by CISPR 32 from the highest frequency generated or used inside the product, as Table 2 lists; the bands of Table 1 say how each decade is measured, and Table 2 says where the scan stops.

The bench assumed for the worked examples

Every number in the worked examples that is a property of the bench rather than of the standard is stated once, in Table 3, and used from there. None is a published ICX-FieldHawk specification except the analyzer noise figure, which is derived from one.

From an indicated level to a field strength

An analyzer displays the level at its connector; a limit is field strength at a distance. Everything between is a signed correction:

The conducted case has the same shape, with the mains network's voltage division factor replacing the antenna factor (Figure 3). Three of the terms vary with frequency, so a mid-band constant is decibels wrong at the band edges.

3Why a scan finds nothing and reports it as a pass

The second failure has nothing to do with amplitude: the emission was not present while the analyzer was tuned to it. A stepped scan visits each cell for the fraction of the sweep set by bandwidth over span:

Across 30 MHz to 1 GHz at 120 kHz with the one second search sweep of Table 3, t_d = 1 s × 120 kHz / 970 MHz = 124 µs, so p = (124 µs + 5 ms) / 2 s = 2.562 × 10⁻³. One sweep finds a 5 ms commutation burst a quarter of one percent of the time. Figure 4 draws the geometry and Figure 5 computes its consequence.

4Requirements for a pre-compliance bench

None of the above names an instrument. Table 4 converts it into six requirements, each with the number and the equation or table it comes from.

5Meeting the requirements, and what it costs

The ICX-FieldHawk is a general-purpose real-time analyzer. Two routes meet R1 on it, and which one applies depends on the model.

R1 on the ICX-045, ICX-060 and ICX-090. Option 74, EMI pre-compliance measurement, is published on the handheld datasheet for these three models: CISPR 16-1-1 Peak, Quasi-Peak, Average and RMS detectors run in parallel, so one scan serves every detector; the CISPR resolution bandwidths by band, 200 Hz, 9 kHz, 120 kHz and 1 MHz; a preloaded limit-line library covering FCC Part 15 Class A and B, CISPR 11, 22 and 32 and IEC limits with automated margin and pass or fail against the selected line; segmented sweeps with a one-touch sequence of initial peak scan, automatic peak finding and quasi-peak and average final measurement, which SpectraCore runs and flags against the selected limit line; and export of the raw and pass or fail data as CSV. With option 74 the bandwidth of R1 is the CISPR value itself, equation (1) is not needed, and the quasi-peak reading is made by the detector the limit is written for. Figure 6 draws the published sequence.

R1 on the ICX-095, ICX-200 and ICX-400, the ICX-FieldHawk-R and the modules. Option 74 is not published for these models. Their datasheets publish a swept resolution bandwidth of 1 Hz to 10 MHz, so the 3 dB equivalents of 141 Hz, 6.4 kHz, 84.9 kHz and 707 kHz fall inside the range. The scan sets the nearest available bandwidth to each equivalent and records it, and equation (1) gives the correction for the ratio used; a bandwidth within 10 percent of the equivalent is within 0.8 dB on a broadband emission. A quasi-peak detector is not published on these models, so the scan uses the peak detector, and section 7 says what a peak reading establishes.

R2. The three option 74 handhelds reach 4.5, 6 and 9 GHz, so the ICX-060 and ICX-090 cover every ceiling in Table 2 and the ICX-045 covers products whose highest internal source frequency is below 500 MHz. The ICX-095, ICX-200 and ICX-400 handhelds, the ICX-095R, ICX-200R and ICX-400R tablets and the ICX-200U and ICX-400U modules reach 9.5, 20 or 40 GHz, so the 20 and 40 GHz models cover band E to its 18 GHz edge for a standard that asks for it.

R3 and R5. On any model the scan can be built in a script. SCPI is standard, with C, C++, C#, Python, MATLAB, Qt and LabVIEW on Windows and Linux: the script steps center frequency in 60 kHz increments across band D, 300 MHz to 1 GHz, holds 100 ms and records the peak. Antenna Factor, Amplitude Offset and amplitude correction are published functions on every ICX-FieldHawk datasheet, so SpectraCore applies equation (2) point by point from the antenna-factor table and the measured cable loss, and Figure 7 shows the amplitude offset applied to a trace. At 100 ms a point, a confirmation pass over bands C and D takes 27 minutes. The one second sweep of Table 3 is the search; the 100 ms dwell is the confirmation of what the search found, and a confirmation pass run as a search costs the 27 minutes at every iteration of the loop.

R6. The published engine is an FPGA transform with no missing samples across the analysis bandwidth, and its relation is published with it: 100 percent probability of intercept equals 2 × N × D × 8 ns on the ICX-095, ICX-200 and ICX-400 family, or 32.768 µs at 2,048 points, and 2 × N × D × 16 ns on the ICX-045, ICX-060 and ICX-090 as standard, or 131.072 µs at 4,096 points. A 5 ms burst is 150 times the first figure and 38 times the second, so it is measured at true amplitude on the first pass with either. The window is 100 MHz wide on the ICX-095, ICX-200 and ICX-400 family and 50 MHz as standard on the other three, so band D, 300 MHz to 1 GHz, takes seven or fourteen placements. Put equation (3) to the seven-placement case: parked for one event period each, the placements cover the band in 14 seconds at a probability of one, against the 1,795 sweeps and thirty minutes the stepped scan needs for 99 percent. Figure 8 is a time-domain record of that kind, and Figure 9 shows the other case the bandwidth table does not cover: a narrowband emission, which reads the same at any bandwidth wide enough to hold it.

Worked example: what the bench can see

Referred to field strength, the noise floor is:

In band D at 240 MHz, 10 log₁₀(84,900) = 49.3 dB. With 2.2 dB of cable ahead of a 15 dB analyzer the system noise figure is 17.2 dB, so E_n = −174 + 17.2 + 49.3 + 107 + 17.5 = 17.0 dBµV/m against a 46 dBµV/m FCC Class B limit at 3 m: 29 dB of clearance. The assumed preamplifier, 3 dB noise figure and 30 dB gain, takes the system noise factor to 1.66 × (1.995 + 30.62/1000) = 3.362, or 5.3 dB, which is 11.9 dB better than the 17.2 dB bare figure. Put the preamplifier at the antenna instead of at the analyzer and the cable loss follows the gain rather than preceding it: 3.1 dB, which is 14.1 dB better.

It also spends headroom, which is R4. At +10 dBm output-referred 1 dB compression the preamplifier compresses at −20 dBm in, or 87 dBµV. Band C asks for 43.5 dB of overload factor, so the highest level the chain measures without compressing an impulse is 43.5 dBµV at the preamplifier input, or 63.2 dBµV/m at the antenna, 17 dB above the limit. A product failing by more than that is measured with the preamplifier out. That ceiling assumes the preamplifier at the analyzer; at the antenna it sees the extra 2.2 dB, so the ceiling falls to 61.0 dBµV/m. The antenna position gains 2.2 dB of noise figure and spends 2.2 dB of headroom, the same trade in both directions. Figure 10 draws floor, ceiling and limit across the radiated bands.

Figure 10 extends the two worked points across the bands. The same arithmetic in band E at 1 GHz gives 35.5 dBµV/m bare against a 54 dBµV/m limit, and the conducted noise floor is nearly 60 dB below the conducted limit. Sensitivity rarely sets the result of a pre-compliance bench built this way; R4 and R5 decide whether a reading can be trusted.

6Measurement practice

1. Start with the conducted scan. A mains network, a 10 dB pad and 150 kHz to 30 MHz find much of what fails a mains-powered product. For a mains-powered box with no radio and one short cable, that scan and a limit line may be all the pre-compliance needed.
2. Use a current probe before an antenna. Cable common mode dominates radiated emissions below 1 GHz. Rearranging the short-radiator relation, I_CM = 2rE / (μ₀fL): at 200 MHz on a 1 m cable against the Class B limit of 150 µV/m at 3 m, that is (2 × 3 × 150 × 10⁻⁶) / (1.2566 × 10⁻⁶ × 2 × 10⁸) = 3.6 µA, or 11.1 dBµA. Through the 15 dBΩ transfer impedance of Table 3 that is 26.1 dBµV, well clear of the noise floor. Four microamps of common mode on a one meter cable fails Class B, and a clamp finds it.
3. Hold the near-field probe still. Loop field falls as the cube of distance, so 5 mm to 10 mm changes the reading by 60 log₁₀(2) = 18.1 dB, more than any margin being chased. Near-field probing ranks sources; it does not measure them.
4. Take the dwell from Table 1 and the max hold from Figure 5, with the peak detector for both. Thirty minutes if anything intermittent is suspected. Quasi-peak never exceeds peak, so a peak scan that clears the limit clears it under any detector. Keep one scan with the equipment off, and give the analyzer its published 10 minutes at 25 °C first.
5. Record the bandwidth, detector, dwell, corrections and limit line with every scan. Two scans are comparable only when all five match, and a change between them is the first thing to look for when a reading moves.

7What a bench result cannot establish

Uncertainty separates a bench result from a certificate; Table 5 compares the budgets of two benches.

Limitations

• A bench gives no legally defensible number, because the site is not validated. The certificate comes from an accredited site, whatever instrument the bench carries.
• A 3 m result does not reproduce a 10 m open area result by arithmetic. Extrapolating by 20 log₁₀(10/3) = 10.46 dB assumes a far-field single-source geometry a real product lacks below 100 MHz.
• 5.9 dB of expanded uncertainty means a reading 4 dB under the limit is not a pass. Class A equipment has 10 dB more room, so a bench of this uncertainty serves an industrial product better than a consumer one.
• Option 74 is a pre-compliance mode, not a CISPR 16-1-1 receiver certification. Its detectors and bandwidths are published for the ICX-045, ICX-060 and ICX-090 only. On the ICX-095, ICX-200 and ICX-400, the ICX-FieldHawk-R and the modules, no quasi-peak detector is published, so a scan there uses the peak detector, and a peak scan bounds a quasi-peak result rather than replacing it: a peak reading under the limit passes under any detector, and a peak reading over it is a candidate, not a failure.
• The overload factor of the analyzer itself is not published. The headroom arithmetic of section 5 is set by the preamplifier, which is the element that compresses first in the assumed chain; the analyzer's maximum continuous input is published at +23 dBm with the preamplifier off.

Field evidence

No chamber correlation has yet been scored against a bench built as described here, so this paper reports no measured agreement between a bench prediction and an accredited result. Every requirement is derived from CISPR 16-1-1, every instrument figure is a published one, and the arithmetic of sections 2 to 4 does not depend on any option.

8Summary and selection

Nothing above is a compliance measurement. The scan is shaped like CISPR's, the bench is not an accredited site, and the certificate still comes from one. Table 6 collects the published values with their conditions, by model family.

Table 7 is the selection guide. For a bench, the ICX-045, ICX-060 or ICX-090 handheld with option 74 carries the CISPR detectors, bandwidths and limit lines on the instrument; by Table 2, the ICX-060 reaches the 6 GHz CISPR 32 ceiling and the ICX-090 reaches 9 GHz. For band E beyond 6 GHz, which CISPR 16-1-1 defines to 18 GHz, the ICX-200 and ICX-400 handhelds, the ICX-200R and ICX-400R tablets and the ICX-200U and ICX-400U modules reach 20 or 40 GHz with the 3 dB-equivalent bandwidths of Table 1 set by script. The modules run the scripted scan beside a host in a rack; the handhelds and the ICX-FieldHawk-R tablet take the same script to a product in the field.

9Definitions, symbols and further reading

Overload factor

The margin in dB by which an impulse peak may exceed the indicated level without any stage compressing. CISPR 16-1-1 specifies it per band for a measuring receiver.

Quasi-peak detector

A detector with defined charge, discharge and meter time constants, so its reading depends on repetition rate as well as amplitude. It never exceeds a peak detector.

Pre-compliance

A measurement made as the standard defines it, on a site that is not accredited, to find and rank emissions before the accredited test.

Probability of intercept

The minimum event duration an analyzer measures at its true amplitude, as distinct from merely detecting it.

Further reading

• CISPR 16-1-1, Measuring apparatus, the source of Table 1; CISPR 16-2-3, Radiated disturbance measurements; CISPR 16-4-2, Measurement instrumentation uncertainty; CISPR 32, Emission requirements for multimedia equipment, the source of Table 2.
• 47 CFR Part 15 Subpart B, §15.107 and §15.109, with ANSI C63.4 as incorporated; H. W. Ott, Electromagnetic Compatibility Engineering, Wiley, 2009, chapters 12 and 18.
• The ICX-FieldHawk handheld datasheet for option 74; BNC-AN-104, Commissioning and maintaining a satellite earth station, for the detection-probability arithmetic applied to interference.

If you are building a pre-compliance bench and want the bandwidth, dwell and headroom arithmetic worked against your own product, our application engineers would be glad to help.