Full-sized EMI antennas can be large and expensive, so let's explore smaller options for radiated emissions pre-compliance testing.
Full-sized EMI antennas can be relatively large and expensive, so the question many companies that want to bring radiated emissions pre-compliance testing in-house ask is, “Isn’t there something smaller and less expensive?” Of course, there are many possible options, including the Aaronia antenna I reviewed earlier. In this article, I’d like to compare two others to a full-sized EMI antenna.
The three antennas are the full-sized Chase CBL6111A, the new Tekbox TBMA1 biconical, and the York ARA01 active antenna. None of these are directly comparable, but each are unique in interesting ways and all may be used (with some precautions) for in-house pre-compliance testing for radiated emissions.
I’ll be using a benchtop spectrum analyzer from Siglent (model SSA 3032X) to make manual measurements and a Tektronix (model RSA306B with EMCVu pre-compliance software) for more automated measurements. All range measurements will be taken in a rural outdoor test range, but without a reflective ground screen (Figure 1). I’m hoping this will simulate the situation faced by most manufacturers when setting up a temporary test site.
Figure 1 Here is the typical 3-m measurement setup for these tests. Both the Tekbox comb generator and antenna are about 1.2 m above ground level and are horizontally-polarized.
EMI antennas: The reality
No practical EMI antenna is as efficient as a full-sized half-wave dipole antenna. In addition, we ask the EMI antenna to be as efficient as possible across a wide frequency range. We rate the sensitivity of EMI antennas by antenna factor (AF), which is simply the ratio of E-field (V/m) in the far field illuminating the antenna elements to voltage measured at the antenna terminals. The units for AF are dB/m.
AF is dependent on a number of factors; antenna physical size, beam width, impedance mismatch, etc. This AF is added to the measured value from the spectrum analyzer (or EMI receiver), along with other loss/gain factors (described below) and yields the amount of E-field illuminating the antenna as measured at a certain test distance of typically 3 or 10 m. For example, at 30 MHz (the lower limit of the commercial radiated emissions band), a full-sized half-wave dipole wire antenna is 5 m from end-to-end and is relatively efficient.
My full-sized Chase CBL6111A antenna with “bowtie” low-frequency elements is designed to cover 20 to 1500 MHz and the lower-frequency “bowtie” element is only 1.4-m wide. Because it is significantly shorter than a resonant dipole at 30 MHz, it isn’t nearly as efficient or sensitive at that frequency and the AF will be relatively large. As the size of antennas is reduced further, that AF (a penalty in the equation for E-field) only increases further. The AF also tends to increase as you go higher in frequency past resonance, unless antenna gain acts as an offset to that.
Because the impedance at the antenna port varies a great deal, we also tend to insert a 6-dB attenuator between the antenna port and coax cable to level out the impedance match. Of course, we also need to add/subtract all other system losses or gains (ex: attenuator, coax loss, and preamp gain) for the actual measurement. This value for E-field (dBμV/m) can then be compared directly to the test limits as specified in the appropriate product standard. The equation for calculating E-field is shown below.
E-field (dBμV/m) = spectrum analyzer measurement (dBμV/m) + antenna factor (dB/m) + attenuator (dB) + coax loss (dB) – preamp gain (dB)
For example, at 100 MHz the spectrum analyzer measured 25 dBμV/m, the AF was 15 (dB/m), the coax loss was 2 dB, the preamp gain was 20 dB, and we’re using a 6-dB attenuator at the antenna port:
E-field (dBμV/m) = 25(dBμV/m) + 15(dB/m) + 6(dB) + 2(dB) – 20(dB) = 28 dBμV/m
The bottom line is that these smaller-sized calibrated EMI antennas will tend to have significant AF on each side of their resonance frequency. Many of the smaller antennas I’ve tried have resonances in the hundreds of MHz and are relatively inefficient below 200 to 300 MHz (as compared to full-sized EMI antennas). Bottom line: when evaluating EMI antennas, you want to consider both the low-frequency and high-frequency ends of the AF chart – lower numbers are better.
That’s not to say these smaller antennas are unusable for the purposes of pre-compliance testing, just that there are some physical realities and accompanying limitations, which I’ll point out as we go. For example, one important specification, as I alluded to, is the AF chart that accompanies all calibrated antennas. The more sensitive antennas will have lower AF, which means lower loss factors added to the spectrum analyzer reading and the ability to more easily observe lower-level harmonics as compared to the system noise floor.
Another limitation for the smaller-sized antennas will be they’ll need the product under test to be closer, rather than farther away, for easily-observable signals. That is, 3-m test distances will work better than 10-m test distances. This is actually not a bad thing, because most manufacturers will prefer this shorter test distance anyway, due to the more limited space they have to set up this test.
Antennas to be compared
This is a full-sized EMI antenna and a combination of a “bowtie” dipole and log-periodic broadband antenna calibrated from 20 to 1500 MHz (Figure 2). The dipole elements may be removed easily for storage. It’s relatively large and heavy and requires a sturdy tripod to hold it. It is available from Ametek/Teseq and their distributors for $3,840.
The Tekbox TBMA1
This is a newly-released miniature biconical dipole antenna calibrated from 30 to 1000 MHz (Figure 3). It is just 0.43m wide and weighs 0.46 kg (1 pound), and retails for $589 through Saelig Electronics in the U.S. and several other distributors worldwide.
Figure 3 The Tekbox TBMA1 biconical antenna is calibrated from 30 to 1,000 MHz. Source: Tekbox Digital Solutions
This is an active antenna (receive, only) and comes with two different-length dipole elements; the shorter DAE01 (3-cm long, calibrated from 30 to 1000 MHz), shown in Figure 4, and the longer (9-cm long, calibrated from 30 to 300 MHz). These longer elements are more sensitive (lower AF) than the shorter elements. It is powered by a standard 9V battery, includes a battery-monitoring LED, and retails for $2,998 (kit with shorter dipole elements) and $3,393 (kit with both long and short dipole elements) from Reliant EMC in the U.S..
As I mentioned, each of these are unique in different ways with their own advantages and disadvantages. The full-sized Chase antenna will serve as the basis for comparison and should represent the performance of many similar antennas. The Tekbox antenna is a reduced-size biconical, similar to the much smaller Aaronia active biconical reviewed earlier. Because the Tekbox antenna is larger, it should capture more signal with lower noise than the Aaronia with preamplifier. The York antenna is purely an active antenna and has more sensitivity with smaller AF than even the full-sized Chase antenna. Let’s see how they compare.
Comparison of antenna factors
The following AF charts (figures 5-7) are all provided by the respective manufacturers. In addition, you’ll receive a tabulated list of factors versus frequency that you can enter into your favorite pre-compliance software.
Figure 5 The AF chart for the Chase CBL6111 full-sized antenna shows two resonant frequencies of 60 and 200 MHz, and relatively high AF at each end of the frequency range. This is pretty normal for these type antennas. Source: Ametek/Teseq
Figure 6 The AF chart for the Tekbox TBMA1 biconical antenna shows a relatively smooth curve, which makes it nice for pre-compliance software to perform interpolations between AF measured data. Note, however, the 10 to 20 dB increased AF in comparison to the full-sized Chase antenna. This limits the usefulness unless used with a broadband low-noise preamplifier, which can effectively lower this by 20 dB, or so. Source: Tekbox Digital Solutions
Figure 7 This is a comparison of the AF chart for the York active dipole antenna with the two types of dipole elements, and the full-sized Chase CBL6111C antenna (in green). The AF at the low-frequency and high-frequency band edges are very comparable to the full-sized Chase antenna, but it’s less sensitive below 100 MHz, as expected, due to the smaller physical size. However, from 100 to 700 MHz, it’s significantly more sensitive. Source: Eurofins/York.
[Continue reading on EDN US: Comparison tests performed]
This article was originally published on EDN.
—Kenneth Wyatt is president and principal consultant of Wyatt Technical Services.