Portable wireless transceivers have proliferated rapidly in recent years, for a wide range of applications including transmission and reception of voice, data, and video. One of the key requirements for wireless equipment is that the electronic circuits operate in the vicinity of other high-frequency radio transmitters, such as Bluetooth devices. Previous studies on interference have considered any source as a far-field effect and focused on interference collected by coaxial cables and metalized traces on printed-circuit boards (PCBs).1-4 However, the interference collected by package lead-frames and bond wires was considered negligible and was not investigated.5

In many portable wireless designs, the transmitter may be only inches from the unit's audio circuits. This can cause problems if the designer is not aware of the location of the antenna in the final product. Traditional fixes, such as added shielding, may not be suitable for today's extremely compact electronic products.

Radiated interference can be separated into far-field and near-field interference. Far-field interference is defined as coming from a distance beyond approximately 10 wavelengths of the frequency of interest. Based on the Bluetooth frequency of 2.4 GHz, 10 wavelengths would be 125 mm (4.1 ft). Thus, it is easy to see why earlier studies did not investigate this.

For consistency with previous studies, consider far-field radiation as conductive interference while nearfield radiation is considered nearfield interference. Conductive interference is energy from the modulated RF signal of a far-field source collected by coaxial cables, PCB traces, and external components. The energy is conducted into the input pins of the portable unit's audio amplifier. Near-field interference is a combination of conducted interference, from a near-field source, and interference collected by the lead frame and bond wires of an audio amplifier's package due to close proximity to the wireless product's antenna. To mitigate RFI effects in receiver circuits, previous studies under farfield conditions developed basic rules of thumb for designers. To evaluate RFI effects, these studies injected a modulated RF signal directly into a coaxial cable (Fig. 1). From these studies, several preventive measures to reduce RFI were established. This article will compare some of these earlier studies and farfield RFI solutions based on feedback resistors,6 RFI capacitors,7 and careful design of the input-stage configuration5 to typical near-field conditions endured by modern compact wireless products.

In order to minimize the effects of RFI on a wireless transceiver, it may help to first understand how the interference-in the form of a modulated RF signalcan make its way into a much lower-frequency audio amplifier and its supporting circuitry. Figure 2 offers a conceptual model of how the carrier frequency is stripped away from a modulated RF carrier, leaving a modulated lower-frequency interference signal. The process starts when an amplitude-modulated (AM) RF signal is conducted via the audio amplifier's input signal pins. The amplifier's low-bandwidth filters screen out the RF carrier, resulting in a demodulated signal at the output of the audio amplifier. Figure 3 is a behavioral model and equivalent circuit for an IC in close proximity of a high-frequency source. The model shows both the conductive and near-field paths to the input of the audio amplifier.

Basic antenna theory explains that a circuit trace of less than onequarter wavelength at the carrier frequency can form an efficient antenna for signals at that frequency. For the 2.4-GHz carrier signals of a Bluetooth device, PCB traces to 31.25 mm (1.2 in.) form efficient antennas. External components, such as capacitors and resistors, on an evaluation board also serve as excellent receiving antennas for RF signals.

Since it is clear that interference signals have more than a few paths to enter a wireless product's audio amplifier circuitry, it is also understandable that a great deal of research work would have been done on trying to prevent such interference. Several papers have been written about conductive interference from a far field antenna and the effects on RF demodulation by the circuit's operational amplifier (opamp).1, 4-6 Once again, these experiments injected the RF modulated signal directly into the amplifier's input pins.

Experimental results from these earlier studies indicated the following: 1. Increasing the values of the input and feedback resistors improves the RFI immunity of the inverting opamp circuit due to the increase in series resistance and parasitic capacitance. This resistive-capacitive (RC) combination forms a lowpass filter that prevents the RFI from reaching the input of the audio amplifier.5

2. Parasitic capacitances Cin (between the inverting and non-inverting inputs) and CRg (shunted across resistor Rg) cause the inverting opamp circuit to have better RFI immunity than the non-inverting opamp circuit.6

3. Metal-oxide-semiconductor (MOS) field-effect transistors (FETs) can be considered less susceptible to RFI than bipolar transistors, since the variation of collector current, induced by the RF signal, is higher than that induced in the drain current of an MOS transistor. As a matter of fact, FETs are inherently less susceptible to RFI than bipolar transistors because of their smoother nonlinearity.4 Also, most audio-frequency opamps are fabricated in large-geometry, higher-voltage CMOS processes that have much lower RF bandwidth than similar-voltage bipolar processes.

These previous studies determined that higher-value feedback resistors, the addition of RFI capacitors, and the use of inherently more linear MOSFET input devices reduced RFI. But these results should also be evaluated under near-field conditions in order to determine their validity for modern wireless designs with many different circuit subsections in close proximity, including the antenna.

Figure 4 shows an evaluation board that was used to investigate near-field interference along with the placement of the antenna. Details on how to assemble a test platform from standard equipment found in most high-frequency analog test laboratories can be found in Application Note AN1299 from Intersil Corp. (http://www.intersil.com/data/an/AN1299.pdf). The test platform generates a swept-frequency RF signal with 1-kHz modulation. The 1-kHz modulation signal is used to track the source of RF entry and the signal path to the audio amplifier's output.

The antenna was terminated in 50 ohms and the end of the antenna loop is bent to have a width approximately equal to the width of the integrated circuit (IC) package. Figure 5 shows the placement of the antenna and layout symmetry of the external components, with near- and far-field conditions shown graphically in Fig. 6. Both amplifiers (channels) of the dual ISL28291 audio amplifier were configured for differential gains of 10, so that the impedance at both inputs were identical. Channel "A" has 5-kohm/500-ohm resistors and Channel "B" has 500-kohm/50-kohm resistors, two orders of magnitude higher-value resistors.

Experimental results, based on swept-frequency measurements from 100 kHz to 6 GHz, showed concentrations of interference in the 1.4-to-2.8-GHz range and in the 3.8-to-5-GHz range (Fig. 7). The location of the antenna during the frequency sweep is shown in the lower right-hand side of Fig. 7. Note that the antenna is directly above the parts package for the initial sweep. The following tests were performed with a single carrier frequency within the above-mentioned concentrations of interference. The results from these tests are as follows:

1. Higher feedback resistor values can lower interference versus lower feedback resistor values. Placing the antenna directly over the higher-value resistors resulted in a lower level of interference than when placed over the lower value resistors. The higher the frequency the lower the level of interference. This observation is in agreement with the results previously reported by Ghadamabadi.5 Placing the antenna over the IC resulted in minimal interference for both sets of resistors.

Continue to page 2

Page Title

2. Adding RFI capacitors can hurt rather than help. Placing the antenna directly over the higher-value resistors resulted in lower level of interference than when placed over the lower-value resistors. The higher the frequency, the lower the level of interference was observed. This observation is in agreement with the results previously reported by Ghadamabadi.6 However, placing the antenna directly over the IC package resulted in much higher levels of interference at the outputs of both amplifiers, regardless of resistor values. Figure 6 shows the signal path for both far-field and near-field antennas. For the case of the far-field antenna, the series resistance of the cables, PCB traces, and external components form a lowpass filter with the RFI capacitor. In this case, the rule of thumb to add the RFI capacitor is effective in removing the RF signal before it gets into the amplifier. For the case of the near-field antenna, there is little to no resistance from the lowpass filter and the RFI capacitor actually results in higher interference at the output of the amplifier.

3. MOSFET input amplifiers are less susceptible to RFI than bipolar transistors.

Placing the antenna directly over the die or resistors showed the MOSFET input to have much less interference than the bipolar input amplifier. This observation is in agreement with the results previously reported by Fiori.4

In short, this study showed that the old rule of thumb to add RFI capacitors for interference control could actually result in an increase in interference depending upon the placement of the antenna. A system designer should become aware of the antenna's location in a wireless product design before using RFI capacitors as a possible RFI solution. The design rules of thumb for higher-valued feedback resistors and MOSFET input amplifiers are still effective for improving a circuit's immunity to RFI under nearfield conditions.

REFERENCES

1. Muhammad Taher Abuelma'atti, "Radio interference by Demodulation Mechanisms Present in Bipolar Operational Amplifiers," IEEE Transactions on Electromagnetic Compatibility, Vol 37, No. 2, May 1995.

2. Robert E. Richardson, Jr., "Modeling of Low-Level Rectification RFI in Bipolar Circuitry," IEEE Transactions on Electromagnetic Compatibility, Vol. EMC-21, No. 4, November 1979.

3. Don LaFontaine and Bob Pospisil, "Measuring RF Interference in Audio Circuits," Application Note AN1299, Intersil Corp., www.intersil.com/data/an/AN1299.pdf.

4. Franco Fiori, "Integrated Circuit Susceptibility to Conducted RF Interference," Compliance Engineering, November/December 2000, www.ce-mag.com/archive/2000/novdec/fiori.html.

5. Hamid Ghadamabadi, James J. Whalen, R. Coslick, C. Hung, T. Johnson, W. Sitzman, and J. Stevens, Department of Electrical and Computer Engineering, "Comparison of Demodulation RFI in Inverting Operation Amplifier circuits of the same gain with different input and feedback resistors values," www.ieeexplore.ieee.org/iel2/161/6451/00252748.pdf?arnumber=252748.

6. Hamid Ghadamabadi and James J. Whalen Department of Electrical and Computer Engineering, "Parasitic capacitances can cause demodulation RFI to differ in inverting and non-inverting operation amplifiers circuits," IEEE 1991 Electromagnetic compability, 1991, Symposium record.

7. Robert E. Richardson, Vincent G. Puglielli and Robert A. Amadori. "Microwave Interference Effects in Bipolar Transistors" IEEE Transaction on Electromagnetic Compatibility, Vol. EMC-17, No. 4, November 1975.