Maxim MAX9993 Manual 

tion of the MAX9993 UMTS, DCS, and PCS base-station downconversion mixer. It is fully assembled and tested at the factory. Standard 50Ω SMA connectors are included on the EV kit for the input and output to allow quick and easy evaluation on the test bench. This document provides a list of equipment required to evaluate the device, a straight-forward test procedure to verify functionality, a description of the EV kit circuit, the circuit schematic, a bill of materials (BOM) for the kit, and artwork for each layer of the PC board. Contact MaximDirect sales at 888-629-4642 to check on pricing and availibility for these kits. Features o Fully Assembled and Tested o +23.5dBm Input IP3 o 1700MHz to 2200MHz RF Frequency o 1400MHz to 2000MHz LO Frequency o 40MHz to 350MHz IF Frequency o 8.5dB Conversion Gain o 9.5dB Noise Figure o Integrated LO Buffer o Switch-Selectable (SPDT), Two LO Inputs o Low 0dBm to +6dBm LO Drive o 40dB LO1 to LO2 Isolation Evaluates: MAX9993 MAX9993 Evaluation Kit ________________________________________________________________ Maxim Integrated Products 1 19-2706; Rev 0; 11/02 For pricing, delivery, and ordering information, please contact Maxim/Dallas Direct! at 1-888-629-4642, or visit Maxim’s website at www.maxim-ic.com. Ordering Information PART TEMP RANGE IC PACKAGE MAX9993EVKIT -40°C to +85°C Thin QFN 20-EP* (5mm x 5mm) DESIGNATION QTY DESCRIPTION C1 1 4.0pF ±0.25pF, 50V C0G-type ceramic capacitor (0603) Murata GRM1885C1H4R0C C2, C6, C7, C9, C10 5 22pF ±5%, 50V C0G-type ceramic capacitors (0603) Murata GRM1885C1H220J C3, C5, C8 3 0.01µF ±10%, 50V X7R-type ceramic capacitor (0603) Murata GRM188R71H103K C4 1 10pF ±5%, 50V C0G-type ceramic capacitor (0603) Murata GRM1885C1H100J C11, C12, C13 3 150pF ±5%, 50V C0G-type ceramic capacitors (0603) Murata GRM1885C1H151J J1–J4 4 PC board edge-mount SMA RF connectors (flat tab launch) Johnson 142-0741-856 L1, L2 2 470nH ±5% wi r e-w ound i nd uctor s ( 1008) Coilcraft 1008CS-471XJBC DESIGNATION QTY DESCRIPTION L3 1 10nH ± 5% w i r e- w ound i nd uctor ( 0805) Coilcraft 0805CS-100XJBC R1 1 523Ω ±1% resistor (0603) R2 1 383Ω ±1% resistor (0603) R3, R4 2 7.15Ω ±1% resistors (1206) Digi-key 311-7.15FCT-ND R5 1 200Ω ±5% resistor (0603) R6 1 47kΩ ±5% resistor (0603) T1 1 4:1 transformer (200:50) Mini-Circuits TC4-1W-7A TP1 1 Large test point for 0.062in PC board (red) Mouser 151-107 TP2 1 Large test point for 0.062in PC board (black) Mouser 151-103 TP3 1 Large test point for 0.062in PC board (white) Mouser 151-101 U1 1 MAX9993ETP-T** *EP = Exposed paddle. SUPPLIER PHONE WEBSITE Coilcraft 800-322-2645 www.coilcraft.com Digi-key 800-344-4539 www.digikey.com Johnson 507-833-8822 www.johnsoncomponents.com Mini-Circuits 718-934-4500 www.minicircuits.com Murata 770-436-1300 www.murata.com Component Suppliers Component List **The exposed paddle conductor on U1 must be solder attached to a grounded pad on the circuit to ensure a proper electrical/ thermal design.

2 _______________________________________________________________________________________ Quick Start The MAX9993 EV kit is fully assembled and factory test- ed. Follow the instructions in the Connections and Setup section for proper device evaluation. Test Equipment Required Table 1 lists the equipment required to verify the opera- tion of the MAX9993 EV kit. It is intended as a guide only, and some substitutions are possible. Connections and Setup This section provides a step-by-step guide to testing the basic functionality of the EV kit. As a general pre- caution to prevent damaging the outputs by driving high-VSWR loads, do not turn on DC power or RF signal generators until all connections are made. This procedure is specific to operation in the U.S. PCS band (reverse channel: 1850MHz to 1910MHz), low- side injected LO for a 200MHz IF. Choose the test fre- quency based on the particular system’s frequency plan, and adjust the following procedure accordingly. See Figure 1 for the mixer test setup diagram. 1) Calibrate the power meter for 1700MHz. For safety margin, use a power sensor rated to at least +20dBm, or use padding to protect the power head as necessary. 2) Connect 3dB pads to the DUT ends of each of the three RF signal generators’ SMA cables. This padding improves VSWR, and reduces the errors due to mismatch. 3) Use the power meter to set the RF signal generators according to the following: • RF signal source: -5dBm into DUT at 1900MHz (this will be about -2dBm before the 3dB pad) • LO1 signal source: +3dBm into DUT at 1700MHz (this will be about +6dBm before the 3dB pad) • LO2 signal source: +3dBm into DUT at 1701MHz (this will be about +6dBm before the 3dB pad) 4) Disable the signal generator outputs. 5) Connect the RF source (with pad) to RF IN. 6) Connect the LO1 and LO2 signal sources to the EV kit LO inputs. 7) Measure loss in the 3dB pad and the cable that will be connected to IF OUT. Losses are frequency dependent, so test this at 200MHz (the IF frequen- cy). Use this loss as an offset in all output power/gain calculations. 8) Connect this 3dB pad to the EV kit’s IF OUT connec- tor, and connect a cable from the pad to the spec- trum analyzer. 9) Set DC supply to +5.0V, and set a current limit around 250mA if possible. Disable the output volt- age and connect the supply to the EV kit through the ammeter. Enable the supply. Re-adjust the sup- ply to get +5.0V at the EV kit. There will be a voltage drop across the ammeter when the mixer is drawing current. 10) Select LO1 by connecting LOSEL (TP3) to GND. 11) Enable the LO and the RF sources. Testing the Mixer Adjust the center and span of the spectrum analyzer to observe the IF output tone at 200MHz. The level should be about +0.5dBm (8.5dB conversion gain, 3dB pad loss). There will also be a tone at 199MHz, which is due to the LO signal applied to LO2. The amount of suppression between the 200MHz and 199MHz signals is the switch isolation. The spectrum analyzer’s absolute magnitude accuracy is typically no better than ±1dB. Use the power meter to get an accurate output power measurement. Disconnect the GND connection to LOSEL. It will be pulled high by a pullup resistor on the board, selecting LO2. Observe that the 199MHz signal increases while the 200MHz decreases. Reconfigure the test setup using a combiner or hybrid to sum the two LO inputs to do a 2-tone IP3 measure- ment if desired. Terminate the unused LO input in 50Ω. Detailed Description The MAX9993 is a highly integrated downconverter. RF and LO baluns are integrated on-chip, as well as an LO buffer and a SPDT LO input select switch. The EV kit cir- cuit consists mostly of supply decoupling capacitors and DC-blocking capacitors, allowing for a simple design-in. Supply Decoupling Capacitors Capacitors C2, C6, and C7 are 22pF (±5%) high-fre- quency supply decoupling capacitors necessary to keep EQUIPMENT QTY DESCRIPTION HP E3631A 1 DC power supply Fluke 75 Series II 1 Digital multimeter (ammeter) HP/Agilent 8648B 3 RF signal generators HP 437B 1 RF power meter HP 8482A 1 High-power sensor (power head) HP 8561 1 Spectrum analyzer 3dB Pad 4 3dB attenuators 50Ω Termination 1 50Ω (1W) termination Table 1. Test Equipment

Capacitors C3 and C8 are larger 0.01µF ceramics used for filtering lower frequency noise on the supply. DC-Blocking Capacitors The MAX9993 has internal baluns on the RF, LO1 and LO2 inputs. These inputs have almost 0Ω resistance at DC, and so DC-blocking capacitors C1, C9, and C10 are used to prevent any external bias from being shunt- ed directly to ground. Capacitors C12 and C13 are used to keep DC current from flowing into the trans- former, as well as providing the flexibility for matching. LO BIAS and IF BIAS Bias currents for the integrated IF output amplifier and the LO buffer are set with resistors R1 (523Ω, ±1%) and R2 (383Ω, ±1%), respectively. These values were care- fully chosen for best linearity and lowest supply current through testing at the factory. Changing these values, or using lower tolerance resistors, degrades perfor- mance. Current-Limiting Resistors Resistors R3 and R4 are used for current limiting at the supply. Resistor R3 dissipates 80mW and R4 dissi- pates 125mW. TAP Network The network at TAP formed by R5 and C5 helps to ter- minate the second-order intermodulation products at the RF input to balance the upper and lower side-band input IP3 performance. LEXT The 10nH (±5%) wire-wound inductor, L3, improves LO-to-IF and RF-to-IF isolation. If isolation is not critical, then this pin can be grounded. IF± The MAX9993 employs a differential IF output to offer increased IP2 system performance. The EV kit uses a 4:1 balun to transform the 200Ω differential output impedance to a 50Ω single-ended output for easy bench evaluation. Inductive pullups L1 and L2 provide DC bias to the IF output amplifier, using C11 for supply filtering and R4 for current limiting. Series capacitors C12 and C13 work in conjunction with the inductors and the 4:1 balun transformer (T1) to match the IF out- put for 40MHz to 200MHz operation. As the differential IF outputs are relatively high imped- ance (200Ω), they are more susceptible to component parasitics. It is often good practice to relieve the ground plane directly underneath large components to reduce associated shunt-C parasitics. LO_SEL The EV kit includes a 47kΩ pullup resistor for easy selection of the LO port. Providing a ground at TP3 selects LO1, and leaving TP3 open selects LO2. To drive TP3 from an external source, follow the limits called out in the MAX9993 device data sheet. Logic voltages should not be applied to TP3 without the +5V supply applied. Doing so can cause the on-chip ESD diodes to conduct and could damage the part. Layout Considerations The MAX9993 evaluation board can be a guide for your board layout. Pay close attention to thermal design and close placement of components to the IC. The MAX9993 package exposed paddle (EP) conducts heat from the device and provides a low-impedance electrical connec- tion to the ground plane. The EP must be attached to the PC board ground plane with a low thermal and electrical impedance contact. Ideally, this is achieved by soldering the backside of the package directly to a top metal ground plane on the PC board. Alternatively, the EP can be connected to an internal or bottom-side ground plane using an array of plated vias directly below the EP. Depending on the ground plane spacing, large sur- face-mount pads in the IF path may need to have the ground plane relieved under them to reduce parasitic shunt capacitance. Modifying the EV Kit The RF and LO inputs are broadband matched, so there is no need to modify the circuit for use anywhere in the 1700MHz to 2200MHz RF range (1400MHz to 2000MHz LO range). Retuning for a different IF is as simple as scaling the values of the IF pullup inductors up or down with fre- quency. The IF output looks like 200Ω differential in parallel with a capacitor. The capacitance is due to the combination of the IC, PC board, and external IF com- ponents. The capacitance is approximately 4pF (per output) to ground. This capacitance is resonated out at the frequency of interest by the bias inductors L1 and L2. To determine the inductor value use the following equation: The IF output is tuned for operation at approximately 100MHz, so a 470nH inductor is used. For lower IF fre- quencies (i.e., larger component values), maintain the component’s Q value at the cost of larger case size, unless it is unavoidable. f L x C IF = 1 2π Evaluates: MAX9993 MAX9993 Evaluation Kit _______________________________________________________________________________________________________ 3

4 _______________________________________________________________________________________ + - - + POWER SUPPLY 3-OUT, HPIB (AG E3631A) RF SIGNAL GENERATOR (HP 8648B) RF POWER METER (GIGA 80701A, HP 437B) RF HIGH- POWER SENSOR (AMMETER) 3dB 3dB 3dB 3dB LO2 RF IN LO1 IF OUT +5V GND MAX9993 U1 RF SIGNAL GENERATOR (HP 8648B) RF SIGNAL GENERATOR (HP 8648B) 1900.000MHz 1700.000MHz 1701.000MHz 5.0V 250mA (MAX) RF SPECTRUM ANALYZER (HP 8561x) BENCH MULTIMETER HPIB (HP 34401A) 202mA Figure 1. Test Setup Diagram

_______________________________________________________________________________________ 5 20 19 18 17 16 13 12 11 14 15 4 3 2 1 5 6 7 8 9 10 VCC RF TAP GND VCC LOBIAS LO1 GND GND LO2 L3 10nH L1 470nH L2 470nH J4 SMA LO2 J3 SMA LO1 TP3 LOSEL LOSEL LEXT IFBIAS R1 523Ω R4 7.15Ω 5.0V 5.0V 5.0V J1 SMA RF IN C11 150pF C2 22pF C3 0.01µF C13 150pF C10 22pF C9 22pF C12 150pF IF+ IF- GND VCC GND GND GND MAX9993 U1 R3 7.15Ω R2 383Ω R5 200Ω C5 0.01µF 5.0V C7 22pF C8 0.01µF C6 22pF C4 10pF C1 4.0pF R6 47kΩ T1 J2 SMA IF OUT 4:1 (200:50) TRANSFORMER 6 3 2 1 4 5.0V TP1 +5V TP2 GND Figure 2. MAX9993 EV Kit Schematic

6 _______________________________________________________________________________________ Figure 5. MAX9993 EV Kit PC Board Layout—Top Layer Metal 1.0" Figure 3. MAX9993 EV Kit PC Board Layout—Top Silkscreen 1.0" Figure 4. MAX9993 EV Kit PC Board Layout—Top Soldermask 1.0" Figure 6. MAX9993 EV Kit PC Board Layout— Inner Layer 2 (GND) 1.0"

implied. Maxim reserves the right to change the circuitry and specifications without notice at any time. Maxim Integrated Products, 120 San Gabriel Drive, Sunnyvale, CA 94086 408-737-7600 _____________________ 7 © 2002 Maxim Integrated Products Printed USA is a registered trademark of Maxim Integrated Products. Evaluates: MAX9993 MAX9993 Evaluation Kit Figure 9. MAX9993 EV Kit PC Board Layout—Bottom Soldermask Figure 10. MAX9993 EV Kit PC Board Layout—Bottom Silkscreen Figure 7. MAX9993 EV Kit PC Board Layout—Inner Layer 3 (Routes) Figure 8. MAX9993 EV Kit PC Board Layout—Bottom Layer Metal 1.0" 1.0" 1.0" 1.0"