AMMP-6408 6 to 18 GHz 1 W Power Amplifier in SMT Package Data Sheet Description Features The AMMP-6408 MMIC is a broadband 1W power amplifier in a surface mount package designed for use in transmitters that operate in various frequency bands between 6 GHz and 18 GHz. At 8 GHz, it provides 29 dBm of output power (P-1dB) and 20 dB of small-signal gain from a small easy-to-use device. This MMIC optimized for linear operation with an output third order intercept point (OIP3) of 38 dBm. • • • • • • Pin Connections (Top View) 1 2 Specifications (Vd = 5 V, Idsq = 650 mA) • • • • 3 8 4 7 6 PIN 1 2 3 4 5 6 7 8 FUNCTION Vgg Vdd DET_O RF_out DER_R Vdd Vgg RF_in 5 x 5 mm Surface Mount Package Wide frequency range 6-18 GHz Highly linear: OIP3 = 38 dBm Integrated RF power detector ESD protection (50 V MM, and 250 V HBM) Input port partially matched (For narrowband applications, customer may obtain optimum matching and gain with an additional matching circuit.) Frequency range 6 to 18 GHz Small signal gain of 18 dB Return loss: input: -3 dB, Output: -9 dB High Power: @ 8 GHz, P-1dB = 29 dBm Application • • • • 5 PACKAGE BASE GND Microwave radio systems Satellite VSAT, DBS Up/Down Link LMDS & Pt-Pt mmW Long Haul Broadband wireless access (including 802.16 and 802.20 WiMax) • WLL and MMDS loops • Commercial grade military Attention: Observe precautions for handling electrostatic sensitive devices. ESD Machine Model (Class A) ESD Human Body Model (Class 1A) Refer to Avago Technologies Application Note A004R: Electrostatic Discharge, Damage and Control. Note: This MMIC uses depletion mode pHEMT devices. Negative supply is used for the DC gate biasing. Absolute Maximum Ratings[1] Symbol Vd Vg Id PD Pin Tch, max Tstg Tmax Parameters[1] Positive Supply Voltage Gate Supply Voltage Drain Current Power Dissipation CW Input Power Maximum Operating Channel Temperature Storage Case Temperature Maximum Assembly Temp (20 sec. max.) Units V V mA W dBm °C °C °C Value 6 -3 to 0.5 900 4.6 23 +155 -65 to +155 +260 Notes note 2 note 2,3 note 2 note 4,5 Notes: 1. Operation in excess of any one of these conditions may result in permanent damage to this device. 2. Combinations of supply voltage, drain current, input power, and output power shall not exceed PD. 3. When operating at this condition with a base plate temperature of 85°C, the median time to failure (MTTF) is significantly reduced. 4. These ratings apply to each individual FET. 5. Junction operating temperature will directly affect the device MTTF. For maximum life, it is recommended that junction temperatures be maintained at the lowest possible levels. DC Specifications/Physical Properties Symbol Id Vg Rqjc Tch Parameters and Test Conditions Drain Supply Current (Vd = 5 V, Vg set for Id Typical) Gate Supply Operating Voltage (Id(Q) = 650 (mA)) Thermal Resistance[6] (Channel-to-Base Plate) Channel Temperature Units mA V °C/W °C Value 650 -1.1 20 150.6 Note: 6. Assume SnPb soldering to an evaluation RF board at 80°C base plate temperatures. Worst case for the channel temperature is under the quiescent operation. At saturated output power, DC power consumption rises to 4.26 W with 1.14 W RF power delivered to load. Power dissipation is 3.11 W and the temperature rise in the channel is 68.4°C. In this condition, the base plate temperature must be remained below 86.6°C to maintain maximum operating channel temperature below 155°C. RF Specifications[1,2,3,4] TA = 25°C, Vd = 5 V, Id(Q) = 650 mA, Zo = 50 Ω Symbol Freq. Gain Parameters and Test Conditions Operational Frequency Small-Signal Gain S21[3,4] Units GHz dB P-1dB Output Power at 1 dB[3] Gain Compression[2] Output Power at 3 dB Gain Compression[3] Third Order Intercept Point; ∆f = 100 MHz; Pin = -20 dBm Input Return Loss[2] Output Return Loss[2] Reverse Isolation dBm P-3dB OIP3 RLin RLout Isolation Minimum 6 17.5 (@ Freq = 8 GHz) 15.5 (@ Freq = 17 GHz) 28 (@ Freq = 8 GHz) 27 (@ Freq = 17 GHz) Typical Maximum 18 18 28.5 dBm dBm 29.5 38 dB dB dB 3 9 45 Notes: 1.Small/large-signal data measured in packaged form on a 2.4 mm connecter based evaluation board at TA = 25°C. 2.This final package part performance is verified by a functional test correlated to actual performance at one or more frequencies. 3.Specifications are derived from measurements in a 50 Ω test environment. Aspects of the amplifier performance may be improved over a narrower bandwidth by application of additional conjugate, linearity, or power matching. 4.Preassembly into package performance verified 100% on-wafer published specifications at frequencies = 7, 12, and 17 GHz. Typical Performances (Data Obtained from 3.5-mm Connector Based Test Fixture, and This Data is Including Connecter Loss, and Board Loss.) (TA = 25°C, Vd = 5 V, ID = 650 mA, Zin = Zout = 50 Ω) 40 0 -40 20 15 -60 10 RETURN LOSS (dB) 25 S12 (dB) -20 30 S21 (dB) 0 S21 (dB) S12 (dB) 35 -5 -10 -15 S11 (dB) S22 (dB) 5 0 2 4 6 8 10 12 14 16 18 20 -80 22 -20 2 4 6 8 FREQUENCY (GHz) Figure 1. Typical gain and reverse isolation 16 18 20 22 8 25 NOISE FIGURE (dB) P-1 (dBm), P-3 (dBm), PAE (%) 30 20 15 P-1 (dBm) PAE (%) @ P-1 P-3 (dBm) PAE (%) @ P-3 10 5 6 7 8 9 10 6 4 2 11 12 13 14 15 16 17 0 18 4 6 8 1000 Pout (dBm) PAE (%) Id (TOTAL) 25 800 20 15 700 10 5 0 -15 -10 -5 0 5 10 15 600 Pin (dBm) Figure 5. Typical output power, PAE, and total drain current versus input power at 8 GHz IM3 LEVEL (dBc) 900 Ids (mA) 30 12 14 16 18 20 16 18 20 Figure 4. Typical noise figure 40 35 10 FREQUENCY (GHz) Figure 3. Typical output power (@P-1, P-3) and PAE and frequency Po (dBm) and PAE (%) 14 10 FREQUENCY (GHz) 12 Figure 2. Typical return loss (input and output) 35 0 10 FREQUENCY (GHz) -20 -22 -24 -26 -28 -30 -32 -34 -36 -38 -40 -42 -44 4 6 8 10 12 14 FREQUENCY (GHz) Figure 6. Typical IM3 level vs. frequency at +20 dBm output single carrier level (SCL) 850 -10 800 -20 -30 750 -30 750 -40 700 -40 700 -50 650 -50 650 -60 600 -60 600 -70 550 -70 550 -80 500 -80 IM3 (dBc) Ids (mA) 4 6 8 10 10 14 16 18 20 22 24 28 850 IM3 (dBc) Ids (mA) 4 6 8 10 800 10 SCL (dBm) 18 20 22 24 28 500 Figure 8. Typical IM3 level and Ids vs. single carrier output level at 8 GHz 850 -10 800 -20 -30 750 -30 750 -40 700 -40 700 -50 650 -50 650 -60 600 -60 600 -70 550 -70 550 -80 500 -80 -10 IM3 (dBc) Ids (mA) -20 4 6 8 10 10 14 16 18 20 22 24 28 IM3 (dBc) 0 Ids (mA) 900 0 IM3 (dBc) 16 SCL (dBm) Figure 7. Typical IM3 level and Ids vs. single carrier output level at 6 GHz 900 850 IM3 (dBc) Ids (mA) 4 6 8 10 10 800 14 16 18 20 22 24 28 500 SCL (dBm) SCL (dBm) Figure 9. Typical IM3 level and Ids vs. single carrier output level at 12 GHz 0 Figure 10. Typical IM3 level and Ids vs. single carrier output level at 14 GHz 850 -10 800 -20 -30 750 -30 750 -40 700 -40 700 -50 650 -50 650 -60 600 -60 600 -70 550 -70 550 -80 500 -80 IM3 (dBc) Ids (mA) -20 4 6 8 10 10 14 16 18 20 22 24 28 IM3 (dBc) 0 Ids (mA) 900 -10 IM3 (dBc) 14 Ids (mA) IM3 (dBc) -20 900 900 850 IM3 (dBc) Ids (mA) 4 6 8 10 800 10 14 16 18 20 22 24 28 Ids (mA) -10 IM3 (dBc) 0 Ids (mA) 900 Ids (mA) 0 500 SCL (dBm) SCL (dBm) Figure 11. Typical IM3 level and Ids vs. single carrier output level at 16 GHz Figure 12. Typical IM3 level and Ids vs. single carrier output level at 18 GHz 0 25 20 -10 S21 (dB) S11 (dB) -5 -15 S11_20 S11_-40 S11_85 -20 -25 0 5 10 15 S21_20 S21_-40 S21_85 10 15 20 5 25 4 6 8 FREQUENCY (GHz) 10 12 14 16 18 20 16 18 20 FREQUENCY (GHz) Figure 14. Typical gain over temperature Figure 13. Typical S11 over temperature 32 0 30 -5 P-1 (dBm) S22 (dB) 28 -10 -15 -25 24 S22_20 S22_-40 S22_85 -20 0 5 10 P-1_85 deg P-1_20 deg P-1_-40 deg 22 15 FREQUENCY (GHz) Figure 15. Typical S22 over temperature 26 20 25 20 4 6 8 10 12 14 FREQUENCY (GHz) Figure 16. Typical P-1 over temperature Typical Scattering Parameters [1], (TA = 25°C, Vd =5 V, ID = 650 mA, Zin = Zout = 50 W) Freq S11 S21 S12 S22 [GHz] dB Mag Phase dB Mag Phase dB Mag Phase dB Mag Phase 6 -3.83 0.64 -7.36 18.46 8.37 -45.38 -49.36 3.41E-03 59.85 -9.89 0.32 -112.35 7 -4.33 0.61 -37.59 22.06 12.67 -160.68 -47.90 4.03E-03 -10.90 -24.54 0.06 -97.72 8 -4.35 0.61 -57.25 21.82 12.33 105.82 -55.02 1.78E-03 -87.02 -12.59 0.23 -116.00 9 -2.87 0.72 -67.80 20.57 10.67 30.27 -58.31 1.21E-03 155.08 -11.66 0.26 -123.36 10 -2.18 0.78 -81.97 19.45 9.38 -34.10 -56.32 1.53E-03 87.15 -9.47 0.34 -111.81 11 -1.88 0.81 -99.66 19.28 9.21 -91.39 -50.78 2.89E-03 36.92 -8.10 0.39 -107.66 12 -2.85 0.72 -125.26 20.24 10.27 -154.70 -48.77 3.64E-03 5.73 -8.11 0.39 -96.60 13 -5.02 0.56 -151.04 20.41 10.49 130.30 -45.72 5.17E-03 -42.89 -5.74 0.52 -95.19 14 -6.38 0.48 -177.19 19.28 9.20 56.72 -45.56 5.27E-03 -90.74 -5.64 0.52 -116.87 15 -6.79 0.46 167.29 18.74 8.65 -8.92 -46.62 4.67E-03 -134.99 -6.02 0.50 -158.25 16 -8.64 0.37 129.42 19.07 8.98 -83.27 -47.25 4.34E-03 -179.47 -8.44 0.38 163.63 17 -14.40 0.19 34.52 19.99 9.99 -174.68 -45.92 5.06E-03 31.89 -12.65 0.23 142.26 18 -4.82 0.57 -87.84 18.06 7.99 61.47 -42.49 7.50E-03 -86.04 -12.88 0.23 -156.34 19 -3.86 0.64 -142.34 9.17 2.88 -58.13 -50.94 2.84E-03 -115.71 -5.42 0.54 127.81 20 -19.84 0.10 171.38 -6.42 0.48 -160.84 -39.18 1.10E-02 -92.64 -6.15 0.49 -6.31 21 -4.51 0.60 -70.79 -16.88 0.14 -164.95 -42.22 7.74E-03 -168.17 -2.48 0.75 -89.99 22 -1.76 0.82 -104.56 -24.20 0.06 137.84 -64.23 6.15E-04 172.50 -1.13 0.88 -122.31 23 -1.30 0.86 -129.94 -33.63 0.02 69.70 -46.41 4.78E-03 -96.28 -1.28 0.86 -144.94 24 -1.04 0.89 -159.31 -42.07 0.01 -90.70 -41.89 8.05E-03 -130.89 -1.01 0.89 -164.03 25 -0.57 0.94 176.91 -50.13 0.00 109.52 -50.58 2.96E-03 122.04 -0.82 0.91 173.03 26 -0.12 0.99 158.94 -44.39 0.01 -58.10 -41.20 8.71E-03 -50.18 -0.33 0.96 155.22 Note: 1. Data obtained from an ICM test fixture with TRL calibration. Reference planes were defined at RF I/O on the package. Biasing and Operation The recommended quiescent DC bias condition for optimum efficiency, performance, and reliability is Vdd = 5 volts with Vg set for Idd = 650 mA. Minor improvements in performance are possible depending on the application. The drain bias voltage range is 3 to 5 V. A single DC gate supply connected to Vg will bias all gain stages. Muting can be accomplished by setting Vgg to the pinch-off voltage Vp. A simplified schematic for the AMMP6408 MMIC die is shown in Figure 17. The MMIC die contains ESD and over voltage protection diodes for Vg, Vd1, and Vd2 terminals. In a finalized package form, Vd1 and Vd2 terminals are commonly connected to the Vdd terminal. The package diagram for the recommended assembly is shown in Figure 18. In finalized package form, ESD diodes protect all possible ESD or over voltage damages between Vgg and ground, Vgg and Vdd, Vdd and ground. Typical ESD diode current versus diode voltage for 11-connected diodes in series is shown in Figure 13. Under the recommended DC quiescent biasing condition at Vds = 5 V, Ids = 650 mA, Vgg = -1 V, typical gate terminal current is approximately 0.3mA. If an active biasing technique is selected for the AMMP6408 MMIC PA DC biasing, the active biasing circuit must have more than 10-times higher internal current that the gate terminal current. An optional output power detector network is also provided. A typical measured detector voltage versus out- put power at 18 GHz is shown Figure 20. The differential voltage between the Det-Ref and Det-Out pads can be correlated with the RF power emerging from the RF output port. The detected voltage is given by, V = (Vref – Vdet) – Vofs where Vref is the voltage at the DET _R port, Vdet is a voltage at the DET _O port, and Vofs is the zero-input-power offset voltage. There are three methods to calculate Vofs: 1. Vofs can be measured before each detectore measurement (by removing or switching off the power source and measuring Vref – Vdet). This method gives an error due to temperature drift of less than 0.01 dB/50°C. 2. Vofs can be measured at a single reference temperature. The drift error will be less than 0.25 dB. 3. Vofs can either be characterized over temperature and stored in a lookup table, or it can be measured at two temperatures and a linear fit used to calculate Vofs at any temperature. This method gives an error close to the method #1. The RF ports are AC coupled at the RF input to the first stage and the RF output of the final stage. No ground wired are needed since ground connections are made with plated through-holes to the backside of the device. Vg DQ Vd2 Vd1 50 50 50 800 µm DET_O 800 µm 6.5 µm 50 10K 200 1K RFIN RFOUT 50 800 µm 10K 800 µm 6.5 µm 50 200 50 Vg Vd1 Figure 17. Simplified schematic for the MMIC die Vd2 50 DET_R DQ DET_O 1 RF INPUT 2 3 8 RF OUTPUT 4 7 6 5 DET_R 5V 50 Ω -0.8 V 1 µF 100 pF 100 pF PIN 1 2 3 4 5 6 7 8 1 µF FUNCTION Vgg Vdd DET_O RF_out DER_R Vdd Vgg RF_in Figure 18. Schematic for recommended assemble example 20 0.45 18 12 10 8 6 0.25 0.15 0.10 2 0.05 5.5 6.0 6.5 7.0 7.5 VOLTAGE (V) Figure 19. Typical ESD diode current versus diode voltage for 11connected diodes in series 8.0 0.1 0.20 4 0 5.0 0.30 0 5 10 15 20 25 30 Pout (dBm) Figure 20. Typical detector voltage and output power, freq. = 18 GHz 35 0.01 DET_R – DET_O (V) 0.35 14 DET_R – DET_O (V) DIODE CURRENT (mA) 0.40 |Icomp (I_METER.AMP1.0)| (mA) Diode_current 16 1 Recommended SMT Attachment for 5x5 Package The AMMP Packaged Devices are compatible with high volume surface mount PCB assembly processes. The PCB material and mounting pattern, as defined in the data sheet, optimizes RF performance and is strongly recommended. An electronic drawing of the land pattern is available upon request from Avago Sales & Application Engineering. Figure 21. Suggested PCB Land Pattern and Stencil Layout Figure 22. Stencil Outline Drawing (mm) 10 Figure 23. Combined PCB and Stencil Layouts Manual Assembly • Follow ESD precautions while handling packages. • Handling should be along the edges with tweezers. • Recommended attachment is conductive solder paste. Please see recommended solder reflow profile. Neither Conductive epoxy or hand soldering is recommended. • Apply solder paste using a stencil printer or dot placement. The volume of solder paste will be dependent on PCB and component layout and should be controlled to ensure consistent mechanical and electrical performance. • Follow solder paste and vendor’s recommendations when developing a solder reflow profile. A standard profile will have a steady ramp up from room temperature to the pre-heat temp. to avoid damage due to thermal shock. • Packages have been qualified to withstand a peak temperature of 260°C for 20 seconds. Verify that the profile will not expose device beyond these limits. 300 TEMPERATURE (°C) The most commonly used solder reflow method is accomplished in a belt furnace using convection heat transfer. The suggested reflow profile for automated reflow processes is shown in Figure 22. This profile is designed to ensure reliable finished joints. However, the profile indicated in Figure 1 will vary among different solder pastes from different manufacturers and is shown here for reference only. PEAK = 250 ± 5°C 250 MELTING POINT = 218°C 200 Ordering Information 150 AMMP-6408 Part Number Ordering Information 100 50 RAMP 1 0 A properly designed solder screen or stencil is required to ensure optimum amount of solder paste is deposited onto the PCB pads. The recommended stencil layout is shown in Figure 21. The stencil has a solder paste deposition opening approximately 70% to 90% of the PCB pad. Reducing stencil opening can potentially generate more voids underneath. On the other hand, stencil openings larger than 100% will lead to excessive solder paste smear or bridging across the I/O pads. Considering the fact that solder paste thickness will directly affect the quality of the solder joint, a good choice is to use a laser cut stencil composed of 0.127 mm (5 mils) thick stainless steel which is capable of producing the required fine stencil outline. PREHEAT 0 50 RAMP 2 REFLOW 100 150 COOLING 200 250 300 TIME (SECONDS) Figure 22. Suggested lead-free reflow profile for SnAgCu solder paste Part Number Devices per Container Container AMMP-6408-BLKG 10 Antistatic bag AMMP-6408-TR1G 100 7” Reel AMMP-6408-TR2G 500 7” Reel Package Dimensions 0.114 (2.90) 0.011 (0.28) 0.018 (0.46) 1 2 3 3 2 0.014 (0.365) 1 * A 8 AMMP XXXX YWWDNN 0.126 (3.2) 4 8 4 0.059 (1.5) 0.016 (0.40) 0.100 (2.54) 0.012 (0.30) 0.029 (0.75) 7 6 5 5 A B FRONT VIEW SIDE VIEW 6 7 0.016 (0.40) 0.028 (0.70) 0.100 (2.54) 0.93 (2.36) SYMBOL A B MIN. 0.198 (5.03) 0.0685 (1.74) MAX. 0.213 (5.4) 0.088 (2.25) DIMENSIONS ARE IN INCHES (MM) 11 DIMENSIONAL TOLERANCE: 0.002" (0.05 mm) BACK VIEW Carrier Tape and Pocket Dimensions 4 mm 12 mm AMMP XXXX AMMP XXXX AMMP XXXX 4.00 ± 0.10 SEE NOTE #2 ∅1.55 ± 0.05 2.00 ± 0.05 B R 0.50 TYP. Ao 1.75 ± 0.10 5.50 ± 0.05 12.00 ± 0.10 Bo A Ko Bo A B 8.00 ± 0.10 SECTION B-B ∅1.50 (MIN.) Ko Ao 0.30 ± 0.05 SECTION A-A A o: Bo : Ko : PITCH: WIDTH: 5.30 5.30 2.20 8.00 12.00 Ao Bo Ko 5.20 5.20 2.10 NOM. 5.30 5.30 2.20 MAX. 5.40 5.40 2.30 MIN. NOTES: 1. Ao AND Bo MEASURED AT 0.3 mm ABOVE BASE OF POCKET. 2. 10 PITCHES CUMULATIVE TOLERANCE IS ± 0.2 mm. 3. DIMENSIONS ARE IN MILLIMETERS (mm). Note: No RF performance degradation is seen due to ESD up to 250 V HBM and 50 V MM. The DC characteristics in general show increased leakage at lower ESD discharge voltages. The user is reminded that this device is ESD sensitive and needs to be handled with all necessary ESD protocols. For product information and a complete list of distributors, please go to our website: www.avagotech.com Avago, Avago Technologies, and the A logo are trademarks of Avago Technologies Limited in the United States and other countries. Data subject to change. Copyright © 2007 Avago Technologies Limited. All rights reserved. AV02-0243EN - June 28, 2007