…………………………………………………………………………………………

Intro to RFID

RFID stands for Radio Frequency Identification. This technology has been around since World War II, when it was used to identify aircraft to avoid friendly fire. Like many other military technologies, RFID has been reduced in cost and downsized to the commercial and consumer markets. An RFID system is made up of two basic components: readers and tags.

An RFID reader is about the size of a cable box and, depending upon the application, would be located under dock doors, exits, or shelving units. The readers vary in complexity, but are typically connected via a network to a server that processes and monitors the information the readers pick up. RFID tags have a wireless antenna, a chip, and some kind of delivery packaging, used for attachment to the object being identified. The chip can contain as little information as the items color or barcode, or as much as your pet’s complete medical history.

Tags can be either active or passive. Active tags include a power source, like a battery, and have greater range. An example of an active tag is a FasTrak or EZPass automated toll collection unit for your car. Passive tags do not have a power source; have a reduced range, and quite often a smaller package size. Passive tags can cost as little as a dime. An example of a passive tag is a chip implanted in the neck of your pet that can be used for identification.

Walmart has mandated that its top 100 suppliers use RFID by January 1, 2005. The Department of Defense followed closely on their heels, also with a January 1 deadline, and is in the process of developing guidelines for the implementation of RFID by their over 43,000 suppliers.

Amphenol RF offers a complete line of RF solutions to connect your RFID infrastructure. With RFID currently in commercial use from LF up to 2.5 GHz, and readers mounted in a variety of environments, Amphenol can provide a tailored solution that meets your performance and operational needs.

…………………………………………………………………………………………


Q: "How does Amphenol RF use TDR simulations to enhance the RF design process?"

A: In the last issue of The Amphenol RF Connection, I gave an overview of how we use ANSOFT HFSS to simulate the RF performance of our connectors. I presented an example and showed how after simulating the connector's performance, it met the specification requirement. In this issue, I will present an example of a connector that performed well up to the initial frequency for which it was designed, but did not meet the VSWR requirement when the frequency range was extended. I will describe the process by which we analyze the TDR results, make changes to the design, and then evaluate the impact of those changes in both the frequency and time domains.

The connector is an SMA Jack receptacle which originally had a VSWR requirement of 1.1+.01* Frequency (GHz) up to 8 GHz. At 8 GHz, this equals a VSWR of 1.18. The new requirement kept the same VSWR (1.1+.01*F), but increased the frequency to 12.4 GHz. (1.224 VSWR max @ 12.4 GHz)

Cross section drawing of the connector

The connector was modeled in HFSS and its VSWR was simulated up to 12.4 GHz
 

HFSS Model of the Connector

Initial VSWR Results

You can see that its VSWR is significantly over the specification limit above 9 GHz. A very powerful tool included in HFSS is the ability to perform a Time Domain Sweep. This allows us to see the reflections within the connector as a function of distance or time (distance=velocity x time). Looking at the cross section drawing, one would expect to see a capacitive discontinuity at the knurl and high impedances at the 2 compensation steps and that is exactly what the TDR (time domain reflectometry) plot shows us. The knurl is capacitve, resulting in a lower than 50 ohm impedance and the compensation steps are inductive, resulting in a higher than 50 ohm impedance. Normally, it is desirable to have inductive compensation steps in order to compensate for parasitic capacitances introduced at the transition in line size, but it appears from the TDR plot that they are too inductive and need to be reduced.

Initial Time Domain results

Another helpful feature in HFSS is one called Tuning. It allows us to conveniently vary the dimensions manually and see the effects of our changes. We varied the length of the compensation steps and chose the length that gave the best overall VSWR performance. This process could have been automated by setting up the problem using the OPTIMETRICS capability of ANSOFT HFSS. With OPTIMETRICS, you set variables for the dimensions you want to vary and enter the maximum and minimum values. HFSS solves for the best results and a short while later, you have the optimum dimensions.

Tuning the Compensation Step Length

The best results still do not meet spec, but are much improved over the initial design. The TDR was plotted again, and you can see the inductances are decreased, but the capacitance from the knurl is still significant.

Time Domain Plot with Best Compensation Step

We can't eliminate the knurl because we need it to prevent rotation of the contact. One way to compensate is to put a groove in the insulator to make the impedance 50 ohms around the knurl, but in this case, it would have made the insulator asymmetrical and would therefore complicate assembly. It was determined that we could shorten the length of the knurl by about 20% and still maintain the correct anti-rotation requirement without the need to modify the insulator. The final graphs show the improved TDR and that the VSWR is now well within specification up to 12.4 GHz.

By fully utilizing all of the capabilities of ANSOFT HFSS, we are able to optimize the performance of our connectors in the shortest time possible. Careful analysis of the time and frequency domain simulation results, along with the utilization of the tuning and optimization capabilities of the software, allows Amphenol RF to quickly deliver the best performing connectors.