HFA3600 Data Sheet May 1999 File Number 3655.4 Low-Noise Amplifier/Mixer Features The HFA3600 is a silicon Low-Noise Amplifier with high performance characteristics allowing the design of very sensitive, wide dynamic-range 900MHz receivers with minimal external components. • LNA - Low Noise Figure . . . . . . . . . . . . . . . . 2.3dB at 900MHz - High Power Gain . . . . . . . . . . . . . . . . 12.8dB at 900MHz - High Intercept . . . . . . . . . . . . . . . . +12.8dBm at Output The LNA, Mixer RF, and LO inputs are internally matched to 50Ω. The Mixer IF output is open collector allowing flexibility in choosing the IF output impedance, with 1000Ω operation fully characterized. The mixer performance is optimized for low LO drive (-3dBm) applications. • MIXER - Low Noise Figure . . . . . . . . . . . . . . . 12.1dB at 900MHz - High Power Gain . . . . . . . . . . . . . . . . . 7.0dB at 900MHz - High Intercept . . . . . . . . . . . . . . . . . .+3.2dBm at Output - Low LO Drive . . . . . . . . . . . . . . . . . . . . . . . . . . . - 3dBm Power consumption is kept to a minimum, making the device ideal for battery-powered hand-held communication equipment. An integrated power-down feature maximizes battery life and eliminates the need for external shut down circuitry. Although fully characterized under 5V single supply, the HFA3600 is operable down to 4V with slight performance differences. The HFA3600 is part of a complete solution including application circuit schematics, S-parameters, noise figure, third-order intercept characterization data and PC board artwork. Evaluation boards are also available through local Intersil Sales offices. Applications • Portable Cellular Telephone (AMPS, IS-54, GSM, JDC) • Wireless Data Com. (ISM, Narrowband PCS) Ordering Information PART NUMBER • LNA + MIXER - Low Noise Figure . . . . . . . . . . . . . . . 3.97dB at 900MHz - High Power Gain . . . . . . . . . . . . . . . . 19.8dB at 900MHz - High Intercept . . . . . . . . . . . . . . . . . . -16.7dBm at Input - Low Operating Power . . . . . . . . . . . . . . . . . . 5V/11.3mA - Low Shutdown Power . . . . . . . . . . . . . . . . . . . 5V/250µA - Small Package: 14 Lead SOIC (Plastic, Small Outline Package, 150 Mil Width, 50 Mil Lead Spacing) • UHF and Mobile Radio Receiver TEMP. RANGE (oC) PACKAGE PKG. NO. HFA3600IB -40 to 85 14 Ld SOIC M14.15 HFA3600IB96 -40 to 85 14 Ld SOIC in Tape and Reel • 900MHz Digital Cordless Telephone (CT-2, ISM) • Wireless Telemetry Block Diagram Pinout LNA VCC 1 HFA3600 (SOIC) TOP VIEW 14 MIXER VCC GND 2 13 IF OUT LNA IN 3 LNA VCC 1 14 MIXER VCC GND 2 GND 4 13 IF OUT LNA IN 3 GND 5 12 GND GND 4 11 RF IN GND 5 10 GND LO BYPASS 6 IF RF LO LNA LO BYPASS 6 LO IN 7 12 GND 11 RF IN 10 GND 9 LNA OUT BIAS 8 POWER DOWN 9 LNA OUT 8 POWER DOWN LO IN 7 1 CAUTION: These devices are sensitive to electrostatic discharge; follow proper IC Handling Procedures. 1-888-INTERSIL or 407-727-9207 | Copyright © Intersil Corporation 1999 HFA3600 Absolute Maximum Ratings Thermal Information Supply Voltage . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . -0.3 to +6.0V Voltage on Any Other Pin. . . . . . . . . . . . . . . . . . . . -0.3 to VCC+0.3V VCC to VCC Decouple . . . . . . . . . . . . . . . . . . . . . . . . . -0.3 to +0.3V Any GND to GND. . . . . . . . . . . . . . . . . . . . . . . . . . . . . -0.3 to +0.3V Operating Conditions Temperature Range . . . . . . . . . . . . . . . . . . . . . . . -40oC ≤ TA ≤ 85oC Supply Voltage Range . . . . . . . . . . . . . . . . . . . . . . . . . . . 4.0 to 5.5V Thermal Resistance (Typical, Note 1) θJA (oC/W) SOIC Package . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 125 Maximum Package Power Dissipation at 25oC . . . . . . . . . . . . . . 1W Maximum Junction Temperature (Plastic Package) . . . . . . . .150oC Maximum Storage Temperature Range . . . . . . .-65oC ≤ TA ≤ 150oC Maximum Lead Temperature (Soldering 10s) . . . . . . . . . . . . .300oC (Lead Tips Only) CAUTION: Stresses above those listed in “Absolute Maximum Ratings” may cause permanent damage to the device. This is a stress only rating and operation of the device at these or any other conditions above those indicated in the operational sections of this specification is not implied. NOTE: 1. θJA is measured with the component mounted on an evaluation PC board in free air. DC Electrical Specifications SYMBOL ICC TEMP (oC) MIN TYP MAX UNITS Normal PD = 2V A 25 - 11.3 12.5 mA Shutdown PD = 0.8V A 25 - 250 375 µA PARAMETER CONDITION Total Supply Current at 5V ALL GRADES TEST LEVEL VIH Shutdown Logic High Normal Mode A 25 2 - VCC V VIL Shutdown Logic Low Shutdown Mode A 25 -0.3 - 0.8 V IIL Shutdown Input Current PD = 0.4V A 25 -200 -150 -100 µA IIH Shutdown Input Current PD = 2.4V A 25 -45 -24 -3 µA LNA Input DC Level Normal Mode A 25 - 0.79 - V Shutdown Mode A 25 - 0.0 - V Normal Mode A 25 - 4.9 - V Shutdown Mode A 25 - 5.0 - V Normal Mode A 25 - 0.79 - V Shutdown Mode A 25 - 0.0 - V Normal Mode A 25 - 2.1 - V Shutdown Mode A 25 - 0.0 - V B 25 - 10 - µs MAX UNITS VLNA-IN VLNA-OUT VMX-RF VMX-LO tOFF , ON LNA Output DC Level Mixer RFIN DC Level Mixer LOIN DC Level Shutdown On-Off-On Time AC Electrical Specifications SYMBOL All Characterization Results have been Obtained with the Use of a Standard Evaluation Board. PARAMETER TEST LEVEL TEMP (oC) ALL GRADES MIN TYP LNA (VCC = +5V, TA = 25oC, Test Figure 1 and f = 900MHz Unless Otherwise Noted In Characterization Curves) S21LNA LNA Gain B 25 11.8 12.8 13.8 dB S12LNA LNA Reverse Isolation B 25 - 23 - dB S11LNA LNA Input Return Loss B 25 6.0 7.3 - dB S22LNA LNA Output Return Loss B 25 10.0 13.0 - dB LNA Output 1-dB Gain Compression Point B 25 - -2.0 - dBm IP3LNA LNA Output 3rd-Order Intercept B 25 +11.2 +12.8 - dBm NFLNA LNA Noise Figure B 25 - 2.30 2.60 dB P-1dBLNA 2 HFA3600 AC Electrical Specifications SYMBOL All Characterization Results have been Obtained with the Use of a Standard Evaluation Board. (Continued) TEST LEVEL PARAMETER TEMP (oC) ALL GRADES MIN TYP MAX UNITS MIXER (VCC = 5V, TA = 25oC, fLO = 825MHz at -3dBm, fRF = 900MHz, fIF = 75MHz and Test Figure 1, Unless Otherwise Noted) PGC MIXER Power Conversion Gain B 25 5.9 7.0 8.1 dB S11RF MIXER RF Input Return Loss B 25 8.0 11.0 - - S11LO MIXER LO Input Return Loss B 25 18.0 26 - dB NFMIXER MIXER SSB Noise Figure B 25 - 12.1 13.9 dB P-1dBMIX MIXER Output 1-dB Gain Compression B 25 - -7.5 - dBm MIXER Output 3rd-Order Intercept B 25 +1.0 +3.2 - dBm MIXER IF Output Capacitance B 25 - 2.3 - pF GRF-IF MIXER RF-IF Isolation (Includes Matching Network) B 25 - 25 - dB GLO-IF MIXER LO-IF Isolation (Includes Matching Network) B 25 - 16 - dB GLO-RF MIXER LO-RF Isolation B 25 16 21 - dB Mixer LO-LNAIN Isolation B 25 42 50 - dB LNAOUT-Mixer RFIN Isolation B 25 35 40 - dB IP3MIX COUTMIX GLO-LNAIN GLNAOUT -RF (LNA + MIXER) VCC = 5V, TA = 25oC, fLO = 825MHz at -3dBm, fRF = 900MHz, fIF = 75MHz and Idealized Lossless External Filters CPGC Power Conversion Gain B 25 - 19.8 - dB CNF Noise Figure B 25 - 3.97 - dB CIP3 Input 3rd-Order Intercept B 25 - -16.7 - dBm NOTE: Test Level: A. Production Tested. B. Guaranteed Limit or Typical Based on Characterization. Test Circuits 2 x 0.01µF 4.7µF 1 VCC (+5V) 14 10µH LNA IN 1000pF 2 13 3 12 4 11 5 10 6 9 1kΩ 1nF RF IN 1000pF 1000pF LNA OUT 1000pF 7 LO IN 1000pF 8 PD(TTL) FIGURE 1. EVALUATION TEST CIRCUIT 3 390nH 3-10pF IF OUT 75MHz 50Ω HFA3600 Test Circuits (Continued) 2 x 0.01µF 4.7µF 1 DUPLEXER LNA IN 1000pF VCC 14 0.01µF 10µH 2 13 3 12 4 11 5 10 6 9 7 8 1kΩ IF FILTER (50Ω) 390nH IF OUT 1nF 3-10pF IF AMPLIFIER 1000pF 1000pF FROM TRANSMITTER IMAGE FILTER (50Ω) 1000pF LO IN 1000pF PD(TTL) FIGURE 2. TYPICAL APPLICATION CIRCUIT TABLE 1. TYPICAL CELLULAR FRONT-END CASCADED PERFORMANCE DUPLEXER LNA IMAGE FILTER MIXER IF FILTER IF AMP UNITS Noise Figure 3.0 2.3 3.0 12.1 8.0 3.0 dB Gain -3.0 12.8 -3.0 7.0 -8.0 20.0 dB 100.0 12.8 100.0 3.2 OUTPUT IP3 Cascaded Noise Figure = 8.55dB Not Applicable (Note) Cascaded Gain = 25.8dB dBm Input IP3 = -10.8dBm NOTE: Cascaded results are using 100.0dBm for IP3. Supply Characteristics 11.0 300 TICC TICC OFF 10.0 250 200 9.0 350 TOTAL ICC (mA) 350 TOTAL SHUTDOWN ICC (mA) TOTAL ICC (mA) 12.0 11.7 300 11.5 250 T ICC OFF 11.3 TICC 200 11.1 150 4.5 4.7 4.9 5.1 5.3 SUPPLY VOLTAGE FIGURE 3. TOTAL ICC vs SUPPLY VOLTAGE 4 5.5 - 40 - 20 0 20 40 60 TEMPERATURE (oC) FIGURE 4. TOTAL ICC vs TEMPERATURE 80 TOTAL SHUTDOWN ICC (mA) 400 11.9 HFA3600 LNA Characteristics 20 20 2dB/DIV 2dB/DIV 25 5.5V MAGNITUDE (dB) MAGNITUDE (dB) 5.0V 4.5V 0 800 900 85 0 800 1000 1000 900 FREQUENCY (MHz) FREQUENCY (MHz) FIGURE 5. LNA S21 vs FREQUENCY AND VCC -40 FIGURE 6. LNA S21 vs FREQUENCY AND TEMPERATURE 0 0 2dB/DIV 5dB/DIV 85 25 MAGNITUDE (dB) MAGNITUDE (dB) 25 -40 85 -40 -50 -20 800 800 1000 900 FREQUENCY (MHz) FIGURE 7. LNA S11 vs FREQUENCY AND TEMPERATURE 900 FREQUENCY (MHz) 1000 FIGURE 8. LNA S12 vs FREQUENCY AND TEMPERATURE 0.00 0 2dB/DIV MAGNITUDE (dB) -1.00 P1dB (dBm) 85 25 900 FREQUENCY (MHz) 1000 FIGURE 9. LNA S22 vs FREQUENCY AND TEMPERATURE 5 -3.00 -4.00 -40 -20 800 -2.00 -5.00 800 900 FREQUENCY (MHz) 1000 FIGURE 10. LNA OUTPUT 1dB COMPRESSION vs FREQUENCY HFA3600 LNA Characteristics (Continued) 1.0 2.318 f = 900MHz 0.0 -1.0 NF (dB) PO 1dB (dBm) 2.309 -2.0 2.300 2.291 -3.0 2.282 -4.0 - 40 - 20 20 0 40 60 80 800 900 TEMPERATURE (oC) 1000 FREQUENCY (MHz) FIGURE 11. LNA OUTPUT 1DB COMPRESSION vs TEMPERATURE FIGURE 12. LNA 50Ω NF vs FREQUENCY 15.0 2.9 f1 = 900.5MHz f2 = 899.5MHz IP3 OUT (dBm) 14.0 NF (dB) 2.7 2.5 13.0 12.0 2.3 11.0 2.1 - 40 - 20 20 0 40 60 80 10.0 800 850 TEMPERATURE (oC) 900 950 1000 FREQUENCY (MHz) FIGURE 13. LNA 50Ω NF vs TEMPERATURE FIGURE 14. LNA OUTPUT IP3 vs FREQUENCY 13.5 f1 = 900.5MHz f2 = 899.5MHz FREQ IP3OUT (dBm) 13.0 12.5 12.0 11.5 11.0 - 40 - 20 20 0 40 60 S11 S21 DEG S22 dB DEG MHz dB DEG dB 800 -6.7 153 13.7 11.4 -11.9 -170 -23.8 -41 850 -7.0 143 13.3 1.5 -12.0 171 -23.1 -48 900 -7.3 133 12.8 -7.7 -13.0 155 -23.0 -56 950 -7.4 123 12.6 -18 -12.0 137 -23.1 -65 1000 -7.6 113 12.2 -27 -11.8 120 -22.8 -70 80 TEMPERATURE (oC) FIGURE 16. LNA S-PARAMETERS FIGURE 15. LNA OUTPUT IP3 vs TEMPERATURE 6 S12 dB DEG HFA3600 Mixer Characteristics 9.0 POWER GAIN (dB) POWER GAIN (dB) 8.0 7.0 6.0 5.0 8.0 7.0 6.0 -6 -2 -4 0 +2 - 40 +4 - 20 FIGURE 17. MIXER PG vs LO DRIVE NOISE FIGURE (dB) NOISE FIGURE (dB) 80 13.0 13.0 12.0 12.0 11.0 10.0 11.0 -4 -2 0 +2 - 40 +4 - 20 0 20 40 60 80 TEMPERATURE (oC) LO DRIVE (dBm) FIGURE 19. MIXER NF vs LO DRIVE FIGURE 20. MIXER NF vs TEMPERATURE f1 RF = 900.5MHz f2 RF = 899.5MHz 5.0 OUTPUT IP3 (dBm) 15.0 NOISE FIGURE (dB) 60 FIGURE 18. MIXER PG vs TEMPERATURE 14.0 -6 40 20 0 TEMPERATURE (oC) LO DRIVE (dBm) 14.0 13.0 12.0 11.0 4.0 3.0 2.0 1.0 50 75 100 125 FREQUENCY (MHz) FIGURE 21. MIXER NF vs IF FREQUENCY, RF = 900MHz, FLO < FRF 7 150 -6 -4 -2 0 +2 LO DRIVE (dBm) FIGURE 22. MIXER OUTPUT IP3 vs LO DRIVE +4 HFA3600 Mixer Characteristics (Continued) - 6.0 - 6.0 P- 1dB (dBm) - 5.0 P- 1dB (dBm) - 5.0 - 7.0 - 7.0 - 8.0 - 8.0 - 9.0 - 9.0 - 10.0 -6 -4 -2 0 LO DRIVE (dBm) +2 +4 - 10.0 - 40 - 20 0 20 40 60 80 TEMPERATURE (oC) FIGURE 23. MIXER 1dB COMPRESSION vs LO DRIVE FIGURE 24. MIXER 1dB COMPRESSION vs TEMPERATURE 4.0 OUTPUT IP3 (dBm) OUTPUT IP3 (dBm) 4.0 3.0 2.0 3.0 2.0 1.0 1.0 - 40 - 20 0 20 40 60 80 FIGURE 26. MIXER OUTPUT IP3 vs RF FREQUENCY 0 0 2dB/DIV 5dB/DIV -40 MAGNITUDE (dB) MAGNITUDE (dB) 1000 FREQUENCY (MHz) FIGURE 25. MIXER OUTPUT IP3 vs TEMPERATURE - 50 700 900 800 TEMPERATURE (oC) 25 85 85 -40 25 - 20 850 FREQUENCY (MHz) FIGURE 27. MIXER LO S11 vs FREQUENCY AND TEMPERATURE 8 1000 700 850 FREQUENCY (MHz) FIGURE 28. MIXER RF S11 vs FREQUENCY AND TEMPERATURE 1000 HFA3600 Isolation Characteristics 0 0 -100 700 25 10dB/DIV MAGNITUDE (dB) MAGNITUDE (dB) 10dB/DIV 85 -40 850 FREQUENCY (MHz) 25 85 -100 700 1000 FIGURE 29. LNA OUT TO MIXER RF ISOLATION vs FREQUENCY AND TEMPERATURE -40 850 FREQUENCY (MHz) FIGURE 30. MIXER LO IN TO LNA IN ISOLATION vs FREQUENCY AND TEMPERATURE 0 5dB/DIV MAGNITUDE (dB) -40 25 85 -40 700 850 1000 FREQUENCY (MHz) FIGURE 31. MIXER LO TO RF ISOLATION vs FREQUENCY AND TEMPERATURE 9 1000 HFA3600 LNA Noise and Gain Characteristics 4.0 0.5 MINIMUM NF (dB) 1 2 3 3.0 10.0 2.0 NF 5.0 5 1.0 10 900MHz 100MHz 1 2 600 5 900 FREQUENCY (MHz) FIGURE 32. LNA GAMMA OPTIMUM vs FREQUENCY FIGURE 33. MINIMUM NOISE FIGURE AND ASSOCIATED GAIN vs FREQUENCY 1 0.5 2 3 2.5dB 2.3dB 10 2.2dB 1 0 11.5dB 2 5 -10 13.5dB -5 14dB -3 -0.5 -2 -1 FIGURE 34. LNA NOISE AND GAIN CIRCLES AT 900MHz 10 0 1200 ASSOCIATED GAIN (dB) 15.0 GAIN HFA3600 Evaluation Board Layout Information Component List: C1, C6 Cap, fixed.01µF R1 Res, fixed 1kΩ C2 Cap, fixed Tantalum. 4.7µF L1 Ind., fixed 10µH C8 Cap, var. 3pF to 10pF L2 Ind., fixed 390nH Cr1 Diode DL4001 C3, C4, C5, C7, C10, C11 Cap, fixed 1nF EVALUATION BOARD LAYOUT SCALE X1 TOP VIEW EVALUATION BOARD COMPONENT PLACEMENT GND VCC IF OUT CR1 L2 C1 C2 LNA IN L1 C7 C8 RF IN C6 R1 C6 C3 C4 C5 C10 8 C11 LNA OUT LO IN PD NOTE: See Evaluation Board testing information. 11 HFA3600 Pin Description Characterization Information LNA VCC The curves and data depicted in the Specifications Section are the result of the design characterization performed by the use of a standard evaluation board and a statistically significant sample procedure which reflects the INTERSIL UHF-1 process variation. Supply voltage for the Low Noise amplifier. LNA In LNA input. Requires AC coupling. Minimum coupling capacitor value of 100pF is suggested. This input is optimized for 50W match in the 800MHz to 1000MHz range. LO Bypass Mixer LO Bypass. Capacitor required to assure a good AC ground. Placement is critical. The bypass capacitance should be located close to the device with low ground impedance. Minimum coupling capacitor value of 100pF is suggested. LO In Local oscillator input. Requires AC coupling. Input is optimized for 50W match in the 700MHz to 1000MHz range. Minimum coupling capacitor value of 100pF is suggested. Power Down Power down control with internal pull up. A low TTL or CMOS level disables the bias network, shutting down both the LNA and the MIXER within 10ms. The internal pull up is provided for users that do not require the power down feature. Provided for Time Division Multiplex Systems and/or power savings. LNA Out Output of the LNA. Requires AC coupling. This output has been optimized for 50W match in the 800MHz to 1000MHz range. Minimum coupling capacitor value of 100pF is suggested. RF In The use of standard RF techniques have been employed throughout the characterization process with special emphasis on noise figures, gains and LO level performances. Special attention has been given to the Local oscillator signal purity and integrity throughout the low and high frequency spectrum. The use of low Excess Noise Ratio (ENR) noise sources have been employed to guarantee a good 50Ω noise source output impedance during the LNA noise measurements. The use of attenuators for most of the setups have assured output impedances of signals closer to 50W when the use of power splitters and filters with poor return loss were necessary. 50Ω environment measurements have been carried throughout the characterization process including the IF output from the MIXER. Device Description The HFA3600 is fabricated in the INTERSIL UHF-1 Bonded wafer, Silicon on Insulator process. ft characteristics of 10GHz and Power bandwidth product of 6GHz together with the robustness of the SOI process ensure high reliability for high frequency volume production. The process features low parasitic capacitances and very low leakages. LNA RF input to the MIXER. Requires AC coupling. Input optimized for 50W match in the 800MHz to 1000MHz range. Minimum coupling capacitor value of 100pF is suggested. IF Out Open collector output of the MIXER. Output capacitance is 2.3pF typical. The use of a RF choke maximizes the voltage output swing but is not mandatory. An output resistance controls the conversion gain as well as IP3 within the useful range of 300W to 1500W. It also affects the output impedance required for the next filter stage and facilitates any output matching network design requirements. Conversion gain is reduced upon use of low value resistors. Mixer VCC Supply voltage for the MIXER and the Bias Network. The LNA uses a single stage topology with a collector spiral inductor to improve the stability at lower frequencies and to optimize the power gain in the 900MHz range. Typical noise figure of 2.3dB, gain of 12.8dB and third order output intercept point of +12.8dBm are the main features. Bias currents are laser trimmed for optimum performances and for tight distribution among production lots. Under a 50Ω environment, the LNA input return loss is 7.3dB and the output return loss is 13dB. Characteristics of the gamma optimum, which is shown in the specifications section, suggests that the optimum source impedance driving the LNA for minimum noise figure is located close to 50Ω. The trade-off between gain and noise figures at 900MHz are shown in the gain and noise circles representation of the specification section. Mixer The HFA3600 Mixer uses a single balanced topology. This topology features an open collector with an output capacitance in the order of 2.3pF. Bias settings are also laser trimmed for 12 HFA3600 optimum performance and tight distribution among production lots. The open collector output permits direct interface to moderate impedance IF filters as well as 50W input filters after a simple “L” impedance matching network. A collector resistor of 1K has been used throughout the characterization together with an impedance matching network for 50W load measurements. With a low -3dBm LO level, a typical SSB noise figure of 12.1dB, conversion gain of 7.0dB and a third order output intercept point of +3.2dBm are the main features. The LO input return loss is typically of 26dBm and the RF input return loss has a typical value of 11dB. Bias Network and Power Down • Total ICC: typical drop of 2.2mA • LNA Input Return Loss: degraded by 0.6dB • LNA Reverse Isolation: degraded by 1dB • LNA Output Return Loss: degraded by 1dB • RF to IF Isolation: no change • LOin to LNAin Isolation: improvement by 2dB • LNAOUT to Mixer RFIN Isolation: improvement by 0.2dB • Mixer LO to RF Isolation: no change The Bias Network is responsible for the accurate setting of both LNA and MIXER operating currents. The LNA operating current is accurately set to 5mA while the MIXER is set to 4mA. Laser trimming procedures and a temperature independent performance of the bias cell, assure the worst case operating current variation of the LNA and MIXER of 1% over the operating temperature range. The Bias network is powered by the Mixer VCC pin and has a built in feature of disabling both the LNA and the MIXER stages. The cell can be powered up and down within 10ms. Power down total current consumption is in the order of 250mA. The simplified schematic of the power down input circuit is shown below. MIXER VCC 15K PD 10K 100K FIGURE 35. ENABLE PIN INPUT CIRCUIT Low Voltage Operation Low voltage operation is possible with the HFA3600. The HFA3600 has been characterized with VCC of 4V and only moderate degradations have been observed compared to the AC performance at a VCC of 5V. The LNA gain shows a 0.8dB decrease and a 1.5dB degradation in the output intercept point with no measurable impact on noise figure. The MIXER behavior at 4V can be summarized with a degradation of conversion gain and output intercept point of 0.8dB and a slight improvement in noise figure of 0.6dB. 13 Other relevant 4V performance characteristics include: • Mixer LO to IF Isolation: degrades by 0.5dB • Mixer RF input Return Loss: degrades by 1dB • Mixer LO Input Return Loss: degrades by 0.3dB at 800MHz and 1dB at 700MHz Layout Considerations The HFA3600 evaluation board layout has been carefully designed for an accurate RF characterization of the device. 50Ω microstrip lines have been provided to permit the connection of the LNA and MIXER independently and facilitate the user interface for testing. Top ground planes were used to assure adequate isolation between critical traces. The HA3600 package pinout has been laid out for best isolation and overall device performance which also permits the placement and connection of ground planes at pins 2, 4, 5, 10 and 12. Pin 4 and Pin 5 assure a low impedance ground return for the LNA and also helps the isolation between the LNA input and the LO input. The LNA output pin is isolated from the RF input port with a good ground connection between the top and back ground planes terminated at pin 10. A series of plated through holes resembling a stitch pattern are sufficient and important for the LNA-OUT and RF-IN ports isolation, so the designer can rely on the full characteristics of rejection of the image filter. Similar isolation pattern is drawn and terminated in pin 12 to isolate the RF-IN from the IF-OUT port. A ground pad has been laid down beneath the package with a series of plated through holes to minimize the inductance to the ground plane and improve the device gain characteristics. All device grounds must be connected as close to the package as possible and the same applies to both VCC inputs and all VCC bypass capacitors. A small 4.7µF tantalum capacitor at the VCC line will prevent supply coupling to the bias network if the device is subjected to strong low frequency interference signals. A protection diode has been added to the demonstration board for extra protection and is not needed in an actual application. HFA3600 Evaluation Board Testing Information Cascaded Evaluation The following paragraphs contain information related to the evaluation of the HFA3600 LNA/Mixer noise figure and common errors encountered during individual and cascaded performance verification. A simple cascaded arrangement using a simple Π network as an intermediate filter is included. The cascaded evaluation of the HFA3600 demo-board must be carried out with a filter network between the LNA and the mixer when noise figure or sensitivity measurements are made. Any bandpass/highpass implementation must be utilized to function as either an image or noise rejection filter. Poor isolation from the RF input to the IF output results in direct amplification (not only frequency translation) of undesired signals at the RF input port. For example, any noise within the IF passband generated by a previous active system block (LNA or any other amplifier) is directly transferred and amplified to the IF output. This lack of isolation can considerably degrade the translated signal to noise ratio of the IF output. An image filter placed before the mixer RF input port can solve the problem. Image filters are normally implemented as narrow bandpass filters which are tuned to pass only the desired (LO+IF) or (LO-IF) frequency of interest. Consequently, the role of rejecting noise at frequencies within the IF passband is accomplished. Poor isolation from the LO input to IF output can also slightly degrade the translated signal to noise ratio of the IF output in two distinct ways: the noise generated by the local oscillator at the IF frequency band is directly coupled to the IF port, and the noise at the RF and image RF passbands (LO SSB noise) gets translated to the IF passband and appears in the IF output. To overcome these problems, the use of a band pass filter is recommended between the local oscillator and the LO input for optimization of the mixer noise figure. The lack of isolation from the LO input port back to the RF input port can cause constructive or destructive interference at the RF port which can affect noise and conversion (translation) gain performance. LNA SMA 1000pF 3.5pF 10nH SMA 1000pF IF RF LO 10nH Π COMPONENTS SHOWN ARE FOR 900MHz RF A “T” FILTER CAN ELIMINATE THE 1000pF COUPLING CAPACITORS FIGURE 36. HFA3600 HIGH PASS FILTER IMPLEMENTATION Tuning of the Π network, if necessary, is done by changing the value of the 3.5pF capacitor. This low value of capacitance may be dependent on the rider layout. The value may be optimized for low insertion loss and, therefore, for optimum cascaded noise figure. Figure 37 and Tables 2 and 3 illustrate the overall performance of the HFA3600 in a cascaded form at 915MHz RF input and 75MHz IF frequency: TABLE 2. SSB MEASUREMENT SET UP (BANDPASS INPUT FILTER) (NOTES 1, 3) NF (dB) GAIN (dB) COMMENTS Saw, 3dB Loss 5.1 16.0 Gain reduced by the filter loss Short/No Filter 14.4 N/A NF degrades due to the IF noise from the LNA Π Filter, No Loss at the RF Frequency 5.2 19.0 Note the increase in cascaded gain IMAGE FILTER 14 RFIN Active single balanced mixers are low cost, low power dissipation devices which require low local oscillator levels to operate. As single balanced mixers lack high isolation from the RF and LO input ports to the IF output and operate with moderate feedthrough from the LO input to the RF input, special precautions must be taken when evaluating these devices with test set ups, specifically filtering, and cabling hook ups. These constraints, although important during the evaluation of the device, are not major issues in the design of the overall system. To remove the IF noise being generated or amplified by the LNA, a low cost Π or “T” high pass filter can be utilized. This simple high pass filter can be used for a cascaded noise evaluation of the HFA3600. Although this implementation does not remove the image signal nor the image noise being generated by the LNA, this filter gives an overall cascaded performance that closely approximates the results obtained by calculation. The large contribution of the LNA gain at the IF frequency (from a white noise source at its input and its own IF noise), to the overall noise figure measurement is practically eliminated by the high pass filter. Figure 1 shows an implementation of a high pass filter network used to filter out the incoming IF noise from the LNA. A rider board can be built to connect the LNAOUT and the RFIN SMA connectors of the demo-board. The 1000pF decoupling capacitors are included in the demo-board. LNAOUT Background HFA3600 LOW NOISE LO FILTER BROADBAND FILTER A NOISE SOURCE HP8970A NOISE FIGURE METER HP346B LNA TUNED AT THE RF FREQ HFA3600 FIGURE 37A. SSB NOISE FIGURE MEASUREMENT FILTER BROADBAND NOISE SOURCE LOW NOISE LO HP8970A NOISE FIGURE METER HP346B LNA HFA3600 FIGURE 37B. DSB NOISE FIGURE MEASUREMENT TABLE 3. DSB MEASUREMENT SET UP (NO INPUT BANDPASS FILTER) NF (dB) GAIN (dB) Saw, 3dB Loss 5.1 16.0 Short/No Filter 1.8 31 Π Filter, No Loss at the RF Frequency 3.6 19.0 IMAGE FILTER COMMENTS Equivalent to SSB Measurement Invalid Measurement Note 3 NOTES: 2. The single side band input filter (filter A) loss is accounted for and removed in the Noise figure and gain values. 3. The difference of a DSB to a SSB noise figure is theoretically 3dB. The expected value of 2.2dB NF for a DSB measurement is degraded to 3.6db due to a small attenuation of the Π filter at the image frequency. 4. The cascaded results presented in the AC Specifications Table of the data sheet are calculated assuming the use of an ideal image filter (no loss) and a SSB measurement. HFA3600 Mixer Evaluation Notes The evaluation of the HFA3600 mixer by itself is facilitated by the demo-board design which provides access to the 3 ports by SMA connectors. As discussed before, RF to IF feedthrough and LO to RF/IF ports moderate isolation can cause errors during noise measurements. The inherent RF to IF feedthrough of the single balanced mixer mandates that noise measurements be single side band only (with an appropriate band pass filter at the RF frequency of interest). Because of this lack of isolation, the incoming energy located at the IF passband from a 15 broadband noise source for example, will feedthrough and cause significant noise figure measurement errors. As noise measurement equipment often makes use of broadband noise sources with energy covering a wide spectrum, SSB measurements are made using a band pass filter in front of the RF port. The role of the band pass filter is to prevent the image and IF noise energy from being fed to the mixer. However, band pass filters exhibit poor return losses at frequencies outside their passbands. Because a moderate amount of power from a local oscillator is transferred back to the RF port in many active mixers, and this returned LO signal is outside the passband of the SSB filter being used, the signal will get reflected back again to the RF port due to impedance mismatch between the filter and the RF port. This impedance mismatch occurs at the LO frequency and these multiple signal reflections can affect gain and noise performance of the mixer. This situation, although not a problem for the actual receiver design, can become a source of error during mixer noise measurements. To minimize the problem, the simplest method is to provide a short connection (well below λ/4 of the LO frequency) between the filter and the RF port. In case a coaxial cable connection is required, it maybe necessary to provide a length of cable which assures minimum degradation to the noise figure reading. Long cables above 3 feet can provide the required standing wave dissipation for measurements in the 800MHz to 1GHz range. Note that long cable losses must be taken into account for the purpose of noise figure measurements. Adjustable line stretchers or isolators at the RF input port could also be used to optimize noise figure readings as an option for the mixer evaluation. And finally, the recommendation of filtering the local oscillator signal before applying it to the LO port is important for accuracy of noise measurements when evaluating the mixer by itself, due to the typical LO to IF feedthrough in single balanced mixers. HFA3600 LNA Evaluation Notes The evaluation of the LNA is straightforward. SMA connectors are provided in the demo-board. There are no recommendations for evaluating the LNA block other than using typical RF amplifier test techniques. Final Note The cascaded evaluation of the HFA3600 LNA and mixer blocks including an image rejection or high pass filter is the best method to obtain accurate results. The gain and noise performance contribution of the LNA and filter to the cascaded results surpass considerably the performance contribution of the mixer. The data collected by cascading the blocks together reflects the performance at the system level which includes the filter of choice for a required design. HFA3600 Small Outline Plastic Packages (SOIC) M14.15 (JEDEC MS-012-AB ISSUE C) N INDEX AREA H 0.25(0.010) M 14 LEAD NARROW BODY SMALL OUTLINE PLASTIC PACKAGE B M E INCHES -B- 1 2 3 L SEATING PLANE -A- h x 45o A D -C- α e B 0.25(0.010) M C 0.10(0.004) C A M SYMBOL MIN MAX MIN MAX NOTES A 0.0532 0.0688 1.35 1.75 - A1 0.0040 0.0098 0.10 0.25 - B 0.013 0.020 0.33 0.51 9 C 0.0075 0.0098 0.19 0.25 - D 0.3367 0.3444 8.55 8.75 3 E 0.1497 0.1574 3.80 4.00 4 e A1 B S NOTES: 1. Symbols are defined in the “MO Series Symbol List” in Section 2.2 of Publication Number 95. 2. Dimensioning and tolerancing per ANSI Y14.5M-1982. 3. Dimension “D” does not include mold flash, protrusions or gate burrs. Mold flash, protrusion and gate burrs shall not exceed 0.15mm (0.006 inch) per side. 4. Dimension “E” does not include interlead flash or protrusions. Interlead flash and protrusions shall not exceed 0.25mm (0.010 inch) per side. 5. The chamfer on the body is optional. If it is not present, a visual index feature must be located within the crosshatched area. 6. “L” is the length of terminal for soldering to a substrate. 7. “N” is the number of terminal positions. 8. Terminal numbers are shown for reference only. 9. The lead width “B”, as measured 0.36mm (0.014 inch) or greater above the seating plane, shall not exceed a maximum value of 0.61mm (0.024 inch). 10. Controlling dimension: MILLIMETER. Converted inch dimensions are not necessarily exact. MILLIMETERS 0.050 BSC 1.27 BSC - H 0.2284 0.2440 5.80 6.20 - h 0.0099 0.0196 0.25 0.50 5 L 0.016 0.050 0.40 1.27 6 N α 14 0o 14 8o 0o 7 8o Rev. 0 12/93 All Intersil semiconductor products are manufactured, assembled and tested under ISO9000 quality systems certification. Intersil semiconductor products are sold by description only. Intersil Corporation reserves the right to make changes in circuit design and/or specifications at any time without notice. Accordingly, the reader is cautioned to verify that data sheets are current before placing orders. Information furnished by Intersil is believed to be accurate and reliable. However, no responsibility is assumed by Intersil or its subsidiaries for its use; nor for any infringements of patents or other rights of third parties which may result from its use. No license is granted by implication or otherwise under any patent or patent rights of Intersil or its subsidiaries. For information regarding Intersil Corporation and its products, see web site www.intersil.com Sales Office Headquarters NORTH AMERICA Intersil Corporation P. O. Box 883, Mail Stop 53-204 Melbourne, FL 32902 TEL: (407) 724-7000 FAX: (407) 724-7240 16 EUROPE Intersil SA Mercure Center 100, Rue de la Fusee 1130 Brussels, Belgium TEL: (32) 2.724.2111 FAX: (32) 2.724.22.05 ASIA Intersil (Taiwan) Ltd. 7F-6, No. 101 Fu Hsing North Road Taipei, Taiwan Republic of China TEL: (886) 2 2716 9310 FAX: (886) 2 2715 3029