Antenna testing kit to accelerate IoT designs

Article By : Kenneth Wyatt

The IoTest Antenna Test Kit includes several test antennas that can be compared for impedance match and overall performance for your wireless or IoT product.

In recent years, many of my clients have been developing products that incorporate wireless and cellular communications into IoT products. I’ve also been partnering with a local CTIA-approved wireless test lab, BluFlux in Louisville, Colorado, to help mitigate various EMC issues. In working with BluFlux, the major issue (beside antenna matching and placement issues) seemed to be self-generated EMI affecting the cellular LTE and GPS receiver sensitivity. One thing I realized during this collaboration is how sensitive wireless antennas are to position on the PCB or where they’re located on the enclosure in relation to the PCB. In addition, the position of the antenna and routing of the transmission line can affect the EMC performance and receiver sensitivity of the system, due to a multitude of high frequency energy sources (Reference 1).

EDN hands-on projectI had a chance to review a new 1-port vector network analyzer (VNA) from Copper Mountain Technologies (CMT). Their model R60 VNA was packaged with a couple dozen test antennas from PulseLarsen Antennas. CMT’s “IoTest Antenna Testing Kit” may be purchased at $7,999. Their R60 VNA may be purchased separately at $3,750. The VNA can measure return loss (S11) from 1 MHz to 6 GHz, using linear, log, or segmented frequency spans. Dips in the return loss plot indicate a good match at that frequency. Generally, a measurement of minus 10 dB, or better, indicates a good match. Directivity across this range is 46 dB.

photo of the Copper Mountain Technologies IoTest Antenna Testing KitFigure 1 The IoTest kit from Copper Mountain Technologies and Pulse/Larsen Antennas may be purchased at $7,999.

Custom software was supplied to help evaluate each sample antenna and compare the return loss with the ideal response. Sample antennas were provided that worked from 315 MHz to 6 GHz, with several designed for multi-band operation. Separate, more professional, software was supplied that allowed the full use of the VNA for other antenna general measurements. I’ll show examples of both the test antenna measurements as well as measurements from some of my own antenna collection.

photo of the Copper Mountain Technologies R60 VNA in the kitFigure 2 The CMT R60 VNA came with several sample antennas mounted to PCBs. There’s another layer of test antennas underneath this top layer.

A sampling of the antenna types is shown in Figure 3. Each sample board has the appropriate transmission line and antenna mounted with a standard PC-mount SMA connector. The IoTest Kit comes with a coax cable and short-open-load (SOL) calibration fixture. Each time a new frequency span is selected, a new calibration must be performed.

photo of IoTest Antenna Testing Kit sample antennas on lab benchFigure 3 Sample antennas were provided that worked from 315 MHz to 6 GHz.

Kit antenna measurements

When measuring the test antennas, the first thing to do is launch the custom test software, which will lead you step-by-step through the measurement procedure. The first step is to enter the specific antenna model from the kit. This will automatically set the frequency limits and load the ideal response. The next step is to calibrate the system prior to connecting the antenna board using the SOL calibration fixture (Figure 4). The instructions will lead you through this, connecting the coax cable to each port of the calibration ports one at a time.

photo of R60 VNA and SOL calibration fixture Figure 4 Use the short-open-load (SOL) calibration fixture prior to making an antenna measurement.

Next, connect the sample antenna board and compare the return loss to the ideal response expected. Depending on how well the measured response matches, you’ll see pop-ups indicating “great,” “excellent,” or “consultation needed.” For this example, we’ll select the Pulse model W3796-K antenna (embedded 2G/3G/4G composite antenna for frequencies of 698 to 2700 MHz), shown in Figure 5.

photo of Pulse W3796-K multi-band antenna Figure 5 A Pulse model W3796-K multi-band antenna was chosen for an example measurement.

This antenna is a multi-band design and the resulting measurement is shown in Figure 6. We can see that most of the measured response fits nicely within the ideal limit and that some bands are shown as great or excellent, while one area needs more work (consultation needed). Most wireless PCB-mounted antennas are very susceptible to mounting position and the software and antenna data sheets provide additional guidance on proper mounting and positioning of the antenna for best results. Of course, the user’s hand position can also greatly affect the response and this can easily be demonstrated by referring to the measured value, which has nearly instant responses.

performance graph for Pulse W3796-K antennaFigure 6 The performance graph shows the measured versus the ideal response for the Pulse W3796-K antenna.

The kit also includes short interface cables to fit some of the smaller on-board coax connectors. You can also use short pigtail coaxes to solder directly to the product under test to characterize the actual antenna performance.

Additional antenna measurements

I measured several of my own antennas using the more universal measurement software supplied. The first one was a favorite used for radiated emissions troubleshooting on the bench, the PCB antenna designed for 400 to 1000 MHz by Kent Electronics (Figure 7). Note that for accuracy, the SOL calibration needs to be performed at the antenna end of the connecting coax cable.

photo of Kent Electronics PCB antennaFigure 7 The antenna I use for evaluating radiated emissions on the work bench is designed to resonate from 400 to 1000 MHz and is available from Kent Electronics for $30.

The measured response is shown in Figure 8. The white box indicates the specification limits. It performs well from about 440 MHz through just over 1000 MHz (assuming a minus 10 dB return loss is “good”).

graph of the measured response for the Kent Electronics PCB antennaFigure 8 This graph shows the measured response (100 to 2000 MHz) for the Kent Electronics 400 to 1000 MHz PCB antenna. The performance was pretty good from the advertised 440 to 1000+ MHz.

I also own an impressive-looking Diamond model D220 multi-band antenna designed to work from 100 to 1600 MHz and used for interference hunting. I brought the R60 VNA and laptop out to the vehicle where it was mounted using a large magnet-mount on the roof. I removed the antenna and performed the calibration at the mount, then re-installed the antenna to make the measurement (Figure 9).

photo of the Diamond D220 antenna vehicle test setupFigure 9 The Diamond 100 to 1600 MHz multi-band antenna was mounted to the test vehicle.

The test results from 100 to 2000 MHz are shown in Figure 10 and markers were placed at each dip in the return loss plot.

graph of the Diamond D220 antenna return lossFigure 10 This graph shows the resulting return loss plot (100 to 2000 MHz) for the Diamond 100 to 1600 MHz multi-band antenna.

Since this antenna was partially designed for amateur radio use, we can see resonant points at 151 (near the 144 to 148 MHz band), 424 (near the 420 to 450 MHz band), 1091 (near the 902 to 928 MHz band) and 1310 (near the 1240 to 1300 MHz band). The bandwidths for most of these are wide enough to accommodate slight frequency offsets from the ideal. There are also resonances in the digital TV band (611 and 639 MHz) and cellular LTE band (809 and 860 MHz). The GPS channel at 1575.42 MHz (near 1639 MHz) should work as well.

Summary

The IoTest Antenna Test Kit from CMT and Pulse/Larsen Antennas is a handy all-in-one kit that includes several test antennas that can be compared for impedance match and overall performance for your wireless or IoT product. The R60 VNA is very fast at displaying the return loss (S11) and has an adequate number of markers for characterizing antenna match and bandwidth.

The custom software guides the user in all steps to make a measurement, so you don’t have to be an expert at VNA measurements to make use of this kit. For the more advanced engineer, the general-purpose software (also included) allows more freedom in selecting the frequency limits and display output. There are also several application notes and videos explaining how to use the measurement system and take a deeper dive into the antenna application and theory. The user interface for both versions of software is easy to understand and I seldom had to refer to the detailed user manual (PDF). All the software and documentation is installed on a thumb drive.

After spending an afternoon with the system, I was able to measure and characterize half the antennas in the kit and measure about a dozen “unknown” antennas in my collection. Overall, I’m impressed with the ease of use of this kit. This test kit would make a fine addition to any company that works with wireless or IoT products and for wireless compliance test labs. Highly recommended!

Kenneth Wyatt is president and principal consultant of Wyatt Technical Services.

References

  1. Wyatt, Platform interference, EDN, 2015

Related articles:

 

Leave a comment