Out of all of the tools typically found in the Electromagnetic Compatibility (EMC) laboratory or engineering facility, the measuring receiver (spectrum analyzer or EMI receiver) is probably one of the most useful. Measuring receivers of one form or another are used to diagnose and repair EMI problems that very often arise during the early stages of the product development cycle. Later, after all EMI fixes have been implemented by the design team, measuring receivers are used to perform full-compliance certification testing prior to shipment of finished product to customers. In particular, spectrum analyzers and EMI receivers are most typically used early in the product development cycle to perform unofficial quick engineering scans after design changes have been made, later for more elaborate pre-compliance radiated and conducted emissions investigations to determine if the product will pass official compliance testing, and then finally at the end of product development cycle in order to perform fully compliant regulatory certification testing where compliance with FCC and CISPR requirements is confirmed on final production-intent products. If you’re in electronics product development, it’s essential that you understand the basics of measuring receivers and how to use them. The successful development and subsequent sale of highly complex electrotechnical products is nearly impossible without first utilizing either an EMI receiver or spectrum analyzer to confirm RF emissions are below mandatory and/or customer driven regulatory radiofrequency (RF) emissions limits.
EMI receivers or spectrum analyzers along with the correct cables and transducers (antenna, LISN, E/ H field probe, etc.) are used to measure the radiated and conducted emissions emanating from electronic devices. Like oscilloscopes, EMI receivers and spectrum analyzers are basic tools used for observing RF signals. However, where oscilloscopes look at signals in the time domain, the EMI receiver or spectrum analyzer looks at signals in the frequency domain. Thus, the EMI receiver or spectrum analyzer will display the amplitude of a RF signal on the vertical scale, and the frequency of the RF signal on the horizontal scale. The horizontal axis of the EMI receiver or spectrum analyzer is linearly calibrated in frequency with the higher frequency being at the right-hand side of the display.
The vertical axis is calibrated in amplitude. Although there is normally the possibility of selecting a linear or logarithmic scale, for most applications, a logarithmic scale is chosen for most measurements. This is because it enables signals over a much wider range to be seen on the measuring receiver and regulatory limits are specified in dBs (decibels). Typically, a value of 10 dB per division is used on the vertical axis. This scale is normally calibrated in dBm (decibels relative to 1 milliwatt) and therefore it is possible to see absolute power levels as well as comparing the difference in level between two signals. Similarly, when using a linear scale, it is often calibrated in volts to enable absolute measurements to be made.
Although spectrum analyzers and EMI receivers are very similar and often lumped under the single term “measurement receiver”, the two devices are not the same, and it’s important to understand the differences. The main differences between the spectrum analyzer and EMI receiver instrument types are described in CISPR 16-1-1:2010 Annex I. These differences are summarized below:
- Swept spectrum analyzers are scanning instruments which tune their local oscillator (LO) frequency continuously to cover the selected frequency range of interest. Some EMI receivers perform a stepped sweep, where the instrument is tuned to fixed frequencies, in defined frequency step sizes, to cover the frequency range of interest. The amplitude at each tuning frequency is measured and retained for further processing or display.
- Most swept spectrum analyzers do not have preselection. Preselection is extra filtering built into the front-end of the instrument just before the first frequency conversion mixer stage. This usually results in an inadequate dynamic range for measurements of low repetition frequency pulses with quasi-peak (QP) detection and may lead to erroneous measurement results under these circumstances.
- Swept spectrum analyzers with preselection are commercially available. These types of instruments may meet all requirements called out in CISPR 16-1-1 and, in case of full compliance with it, can be used without any restrictions to perform fully-compliant emission measurements in accordance with the CISPR 16-2 and other emissions standards such as EN 55011 and EN 55022.
- The specifications applicable to spectrum analyzers without preselection in regard to QP detection are less stringent and their use is conditional on the signals to be measured.
- Spectrum analyzers may not have a built-in preamplifier. EMI receivers tend to have a preamplifier built in after the preselection stage and therefore have a much lower noise floor figure. This results in EMI receivers having a much greater capability of picking up signals that would otherwise be hidden in the noise floor of less sensitive spectrum analyzers.
- The frequency selectivity criteria, defined in CISPR 16-1-1 may not be met by swept spectrum analyzers. Typically swept spectrum analyzers use Gaussian shaped filters that may not meet these requirements. “CISPR Compliant” swept spectrum analyzers must meet the selectivity requirements found in Clause 4.5 of CISPR 16-1-1:2010.
- Swept spectrum analyzers may not have a QP detector built in. CISPR 16-1-1 requires spectrum analyzers to meet the stated specifications in Clause 4.4 of CISPR 16-1-1 for QP detection. This is not necessarily a bad thing if you’re only performing engineering scans and pre-compliance work. If you’re doing full-compliant testing, then not having this capability is an issue.
- Swept spectrum analyzers may not have the proper response to intermittent, unsteady and drifting narrowband disturbances as described in Clause 6.5.4 of CISR 16-1-1. In order to be considered fully CISPR Compliant, swept spectrum analyzers must also meet this requirement.
Compliance requirements, engineering needs, budget and ease of use often drive selection of which measuring receiver type to purchase (spectrum analyzer or EMI receiver). As alluded to earlier in this article, the main standard that specifies the requirements for EMI receivers is CISPR 16 (Part 1-1: Measuring apparatus). If you need to perform fully compliant emissions testing then you will want to purchase a fully CISPR 16 compliant EMI receiver (along with a fully compliant 3 or 10 Meter chamber). These types of receivers come with the correct intermediate frequency (IF) filter bandwidths (6 dB), a normal ±2 dB absolute amplitude accuracy, detector functions (peak, quasi-peak, and average), dynamic range, and specified input impedance with a nominal value of 50 ohms; deviations specified as VSWR (voltage-standing-wave-ratio) required for proper measurement of RF emissions according to measurement standards. Naturally, this type of EMI receiver is higher in price than a spectrum analyzer that does not fully meet CISPR 16 requirements.
Perhaps all you need is a spectrum analyzer that is suitable for conducting engineering scans and pre-compliance measurements in-house and on a test bench instead of needing to perform emissions testing in a fully compliant and expensive EMC chamber. The intent with this type of scenario is to quickly gain confidence that your design will pass full compliance testing before sending it out-of-house for expensive and time consuming full compliance certification work using an EMI receiver. If this is the case then you can get away with purchasing an analyzer that is missing some of the full CISPR 16 requirements and is therefore less expensive to obtain. These types of spectrum analyzers typically only include the peak detector function and leave off the QP and average detectors. You shouldn’t really care about only having the peak detector capability because the intent with pre-compliance testing is to quickly determine if the product will pass the full compliance test without first having to setup your equipment under test exactly the way that is called out in the applicable emissions standards. The peak detector reading is quickest and provides a worst-case reading. If your design is below the established emissions limits using the peak detector function then it should be good-to-go when the time comes to perform full compliance testing and everything is setup in strict conformance with the applicable standards.
Filter bandwidths (6 dB bandwidths) are specified at 200 Hz from 9 to 150 kHz, 9 kHz from 150 kHz to 30 MHz and 120 kHz from 30 to 1000 MHz, and 1 MHz above 1 GHz. The frequency response of these filters is also found in the CISPR 16-1-1 standard.
You’ll want to know what the highest frequency generated or used in your product is before purchasing either an EMI receiver or a spectrum analyzer. It’s not a show-stopper if you end up with a spectrum analyzer that doesn’t go high enough in frequency, but it certainly does make your life just a little more difficult. The FCC’s requirements for the highest frequency in which to test radiated emissions are shown in the table below:
|Highest frequency generated or used in the device or on which the device operates or tunes (MHz)||Upper frequency of measurement range (MHz)|
|Above 1000||5th harmonic of the highest frequency or 40 GHz, whichever is lower.|
Again, the greater the capability, the greater the cost, especially when you’re talking about fully compliant CISPR 16 EM receivers.
Even if you only use a spectrum analyzer and its peak detector function during engineering work and pre-compliance testing you will still want to know the differences between peak, quasi-peak, and average detectors such as when and how they are applied during fully compliant RF emissions measurements.
As the name implies, the peak detector responds almost instantaneously to the peak value of the signal applied and it discharges fairly rapidly. It’s also known as the envelope detector because its output will follow the envelope of the signal. Of the three detector types, it is by far the fastest one to use and provides the highest amplitude reading.
The quasi-peak (QP) detector is like the peak detector except it has a weighted charge and discharge time constant (charge rate much faster than the discharge rate). This detector type was developed a long time ago to correct for the subjective human response to pulse-type interference. The higher the PRF (pulse repetition frequency) of the measured input signal, the higher amplitude it will read out on the EMI receiver or spectrum analyzer’s display. For continuous wave (CW) signals, the peak and the QP response are the same. In order to obtain an accurate reading, the QP measurement must dwell on each frequency substantially longer than that required if using peak detection. The longer dwell time of the QP detector is not a good attribute to have when you’re trying to quickly assess pass/fail of a design change during product development. It is a good thing to have if you find yourself in a situation where your product just barely fails full-compliance RF emissions scans while using the peak detector function and by employing the QP function your device now has several dB of margin below the establish limit line. QP detector readings will always be less than or equal to the peak detection readings.
The average detector simply measures the average value of the input signal. The average value of a CW signal will measure the same value as that which is obtained with a peak detector. Impulsive types of signal will measure lower than peak when using the average detector. The average detector isn’t used much for engineering scans for a couple of reasons. The first is that it takes longer to conduct an emissions scan using average detection versus peak. The second is that it is used only during conducted emissions measurements where limits are established 10 to 13 dB lower than the QP limits.
Hints and Tips
There are several hints and tips for using a spectrum analyzer or EMI receiver to its best effect.
Beware of input level: In order to ensure the optimum performance of the system, the input is normally connected to the primary mixer with only the input attenuator control, often labelled RF level, between them. Accordingly, RF can be applied directly to the mixer with no protection. It is therefore very important to ensure that the input is not overloaded and damaged. One major and expensive cause of damage on spectrum analyzers is the input mixer being blown when the analyzer is measuring high power circuits.
Determine if spurs are real: One aspect of using a spectrum analyzer that will often be encountered is that spurious signals are often viewed. Sometimes these may be generated by the item under test, but it is also possible that they can be generated by the analyzer. To check if they are generated by the item under test, reduce the input sensitivity of the analyzer by 10 dB. If the spurious signals fall by 10 dB then they are generated by the unit under test, if they fall by more than 10 dB then they are generated by the analyzer and possibly as a result of overloading of the input.
Wait for self-alignment: When a spectrum analyzer is first switched on, not only does it go through its software boot-up procedure, but most also undertake a number of self-test and calibration routines. In addition to this, elements such as the reference oven-controlled crystal oscillator oven need to come up to temperature and stabilize. Often manufacturers suggest to wait fifteen to thirty minutes before the instrument can be used reliably. The crystal oscillator may take a little longer to completely stabilize. Refer to the manufacturers handbook for full details.
Power measurement: Making power measurements with a spectrum analyzer is not as accurate as making power measurements with a power meter in terms of absolute accuracy. It should be remembered that both test instruments make slightly different power measurements. A power meter will make a measurement of the total power within the bandwidth of the sensor head (it will measure the power regardless of the frequency). In contrast, a spectrum analyzer will make a measurement of the power level at a specific frequency. In other words, it can make a measurement of the carrier power level without the addition of any spurious signals, noise, etc. While the absolute accuracy of a spectrum analyzer is not quite as good as that of a power meter, they are improving all the time and the difference in accuracy is generally small.
Spectrum analyzers and EMI receivers are very similar devices that are used to measure the amplitude versus frequency of RF emissions emanating from electronic devices. Knowing the similarities and differences and when and where to use one device over the other is very important to anyone performing product development work. References are provided so that readers can dig a little deeper into this important subject.
- CISPR 16-1-1:2010, Specification for radio disturbances and immunity measuring apparatus and methods – Part 1-1: Radio disturbance and immunity measuring apparatus – Measuring apparatus.
- Electronic Code of Federal Regulations, e-CFR data is current as of June 14, 2018, Title 47, Chapter I, Subchapter A, Part 15, Subpart A, §15.33.
- Williams, T., EMC For Product Designers, Fifth Edition, Newnes, 2017.
- Ott, H., Electromagnetic Compatibility Engineering, John Wiley& Sons, 2009.
- Andre, P., Wyatt, K., EMI Troubleshooting Cookbook for Product Designers, SciTech Publishing, 2014
EMI Measurements, Test Receiver vs. Spectrum Analyzer by Rohde & Schwarz, downloaded from https://www.ieee.li/pdf/essay/