Calhoun: The NPS Institutional Archive Theses and Dissertations Thesis and Dissertation Collection 1997-06 The design of the radio frequency (RF) subsystem printed circuit boards for the Petite Amateur Navy Satellite (PANSAT) Lahti, Carl Andrew Monterey, California. Naval Postgraduate School http://hdl.handle.net/10945/8237 N PS ARCHIVE 1997- OG LAHTI, C. NAVAL POSTGRADUATE SCHOOL Monterey, California THESIS THE DESIGN OF THE RADIO FREQUENCY (RF) SUBSYSTEM PRINTED CIRCUIT BOARDS FOR THE PETITE AMATEUR NAVY SATELLITE (PANSAT) by Carl Andrew Lahti June, 1997 Thesis Advisor: Thesis L242363 Co-Advisor: Approved Randy L. Borchardt Rudolf Panholzer for public release; distribution is unlimited DL,rv ~x ' KNOX UBR* OOL MC DUDLEY KNOX LIBRARY AL POSTGRADUATE SCH< v'TFREY C -!01 Form Approved OMB No. 0704-01 i REPORT DOCUMENTATION PAGE Public reporting burden for this collection of information is estimated to average 1 hour per response, including the time for reviewing instruction, searching existing data sources, gathering and maintaining the data needed, and completing and reviewing the collection of information. Send collection of information, including suggestions for reducing this burden, to Davis Highway, Suite 1204, Arlington, 1 REPORT DATE 2. 3 . June 1997. REPORT TYPE AND DATES COVERED Engineer's/Master's Thesis TITLE AND SUBTITLE The Design of the Radio Frequency (RF) Subsystem Printed Circuit Boards for the Petite AUTHOR(S) 6. Amateur Navy Satellite FUNDING NUMBERS 5. (PANSAT) Lahti, Carl A. PERFORMING ORGANIZATION NAME(S) AND ADDRESS(ES) 7. PERFORMING ORGANIZATION REPORT NUMBER Naval Postgraduate School Monterey CA 93943-5000 SPONSORING/MONITORING AGENCY NAME(S) AND ADDRESS(ES) 9. or any other aspect of this for Information Operations and Reports, 1215 Jefferson VA 22202-4302, and to the Office of Management and Budget, Paperwork Reduction Project (0704-01 88) Washington DC 20503. AGENCY USE ONLY (Leave blank) . comments regarding this burden estimate Washington Headquarters Services, Directorate 10. SPONSORTNG/MONITORING AGENCY REPORT NUMBER 1 1 . SUPPLEMENTARY NOTES The views expressed in this thesis are those of the author and do not reflect the official policy or position of the Department of Defense or the U.S. Government. 12a. 1 3 . DISTRIBUTION/AVAILABILITY STATEMENT Approved for public release; distribution is unlimited. ABSTRACT (maximum 200 words) The Petite Amateur Navy Satellite 12b. (PANSAT) is DISTRIBUTION CODE a small digital communication satellite being developed by the Space Systems Academic Group and the Naval Postgraduate School. This thesis describes the layout of the three final flight printed circuit boards for the radio frequency (RF) subsystem for PANSAT. The to verify system design and a circuits and layouts are documented power budget provided in detail. A link analysis is performed for integration with other satellite systems. Printed circuit board design fundamentals and high frequency printed circuit board construction techniques are also described. 14. SUBJECT TERMS Printed Circuit Board, 17. SECURITY CLASSIFICATION OF REPORT Unclassified NSN 7540-01-280-5500 15. Low 18. Earth Orbit Satellite, Unclassified PAGES RF Communications SECURITY CLASSIFICATION OF THIS PAGE 1 9 . NUMBER OF SECURITY CLASSIFICATION OF ABSTRACT 110 CODE 16. PRICE 20. LIMTTATION OF ABSTRACT UL Unclassified Standard Form 298 Prescribed by ANSI Std. (Rev. 2-89) 239-18 298-102 11 Approved for public release; distribution is unlimited. THE DESIGN OF THE RADIO FREQUENCY (RF) SUBSYSTEM PRINTED CIRCUIT BOARDS FOR THE PETITE AMATEUR NAVY SATELLITE (PANSAT) Carl Andrew Lahti Lieutenant, United States B.S.S.E., United States Submitted Navy Naval Academy, 1989 in partial fulfillment of the requirements for the degree of MASTER OF SCIENCE IN ELECTRICAL ENGINEERING and ELECTRICAL ENGINEER from the NAVAL POSTGRADUATE SCHOOL June 1997 / AJC DUDLEY KNOX LIBRARY IE SCHOOL MONTEREY CA . ABSTRACT The satellite Petite Amateur Navy Satellite (PANSAT) is a small digital communication being developed by the Space Systems Academic Group and the Naval Postgraduate School. This thesis describes the layout of the three final flight printed circuit boards for the radio frequency (RF) subsystem for layouts are a documented in detail. power budget provided A link analysis is PANSAT. The circuits and performed to verify system design and for integration with other satellite systems. Printed circuit board design fundamentals and high frequency printed circuit board construction techniques are also described. VI TABLE OF CONTENTS I. INTRODUCTION SATELLITE COMMUNICATIONS 1 B. PANSAT 2 C. II. PROJECT OVERVIEW 3 BACKGROUND A. B. III. 1 A. 5 PANSAT SYSTEMS LAYOUT RF SUBSYSTEM DESIGN BACKGROUND RF SUBSYSTEM PRINTED CIRCUIT BOARD DESIGN A. PRINTED CIRCUIT BOARD EXTERIOR GEOMETRY B. RF SWITCHING PRINTED CIRCUIT BOARD 12 Circuit Description 17 3. Power Requirements 18 PCB Layout RF LNA PRINTED CIRCUIT BOARD Functional Description Description 3. Power Requirements 4. PCB 19 22 22 23 Layout RF HIGH POWER AMPLIFIER PRINTED CIRCUIT BOARD 24 24 25 Functional Description 25 2. Circuit Description 3. Power Requirements 4. Printed Circuit Board Layout 27 28 28 1. ANALYSIS AND CONCLUSIONS A. ANALYSIS B. 12 2. 2. Circuit IV. 11 Functional Description 1. C. 11 1. 4. B. 5 7 31 31 1. Link Budget 31 2. Power Budget 31 CONCLUSIONS 32 APPENDIX A. RF SWITCHING PCB DETAILED DRAWINGS 33 APPENDIX B. RF LNA PCB DETAILED DRAWINGS 49 RF HPA PCB DETALED DRAWINGS 59 APPENDIX D. PRINTED CIRCUIT BOARD DESIGN 69 APPENDIX C. vn AND MICRO STRIP CIRCUIT CONSTRUCTION APPENDIX E. STRIPLINE APPENDIX F. LINK BUDGET ANALYSIS LIST OF REFERENCES 81 87 99 INITIAL DISTRIBUTATION LIST 101 Vlll I. INTRODUCTION SATELLITE COMMUNICATIONS A. A satellite communications system consists of a ground stations on the earth. Modulated RF satellite in space linking multiple waveforms, either analog or digital, represent the information in the system. Satellite communications systems are "line of sight" due to their high operating frequencies. Signals at these straight line sight to the and the antenna on the satellite must physically have an uninterrupted ground station antenna. In analog typically acts as a space based reflector. from a ground station, amplifies The satellite communications, the satellite receives "dumb" space based reflector as in analog line of satellite an analog transmission and retransmits the signal for receipt by stations within the satellite footprint. In digital satellite act as a high frequencies travel in a all ground communications, the communications or it satellite may be may a regenerative repeater. Regenerative repeaters contain processing equipment to demodulate the incoming waveform, process the information and retransmit the processed information. Typically this processing involves error correction. Satellites exist in different orbits around the earth are in geostationary orbit at a height of 35786 an inclination of spinning about Geostationary its earth. km The satellites farthest above the surface of the earth and degrees. These satellites orbit the earth at the axis, thus satellites, from the same rate the earth is appearing stationary to an observer on the earth. with their great height, provide a large footprint where multiple ground stations can access the same satellite at the same time. Low earth km. orbit (LEO) satellites orbit the earth at heights km 1500 to A LEO satellite has a much smaller footprint than the geostationary satellite. A given ground station may only be able to access the as from 300 its footprint passes over the transmitted ground power than geostationary and ground station is signal from the satellite's satellites satellite antenna, LEO station. significantly less The RF subsystem of a LEO satellite for several minutes each day satellites require significantly less because the distance between the and the corresponding energy loss processes filters RF waveforms. It satellite is less. typically receives a and amplifies the signal and converts its frequency to a lower intermediate frequency (IF) for further processing and demodulation. The subsystem also receives a signal for transmission at the IF and converts transmission frequency, then amplifies the signal and sends it it to the RF to the antenna for transmission. PANSAT B. The Petite Amateur Navy Satellite (PANSAT) designed by the Space Systems Academic Group Monterey, California. in the 70 cm amateur radio band. The storage using the may PANSAT will provide AX.25 packet is at the store When the LEO satellite frequency of 436.5 encoded phase The satellite will in and forward communications to users user an orbital server with 4 is MB of in the satellite footprint log onto the satellite's computer and check for any files addressed to him. also upload files for other users. being Naval Postgraduate School satellite will act as protocol. a small, he He may transmit and receive at a center MHz using direct sequence spread spectrum (DSSS), differentially shift keying (DPSK) modulation. The communications data rate will be second with a chipping rate of 1.25 mega chips per second. The chipping sequence pseudo random noise sequence produced from a 7 stage satellite will also per second. The have a narrow band high data rate channel with a data rate of 78125 bits 997 from the space station The approximate dimensions in Figure shuttle ground 1, is planned for launch in through the hitchhiker program. Launch during a technical specifications for the International Telecommunication C. feedback rate channel is privileged and intended for rendezvous mission, could place the spacecraft degrees. linear Naval Postgraduate School for maintenance and software upgrades. PANS AT, shown with 1 a shift register. Use of the high data controllers at the is PANSAT are Union at an inclination of at documented late MIR space least 28 in the required filing to [Ref. 1]. PROJECT OVERVIEW This thesis serves to document the design of the printed circuit boards in the subsystem for PANSAT. It incorporates a detailed link budget analysis to verify system design meets the required specifications. in detail. The RF The electrical and physical designs are described basic process of printed circuit board design and stripline circuit construction are described in appendices. 18 in. r*i 21 in. Figure 1 PANSAT - Side View. * II. A. BACKGROUND PANSAT SYSTEMS LAYOUT The electronics in the satellite consists of the following major subsystems in addition to the RF subsystem: electrical power (EPS), digital control (DCS), mass storage, temperature monitoring (TMUX). These systems work together to provide the functionality of the satellite operating under the control of the satellite's system controller. A wiring harness containing power and control signals connects the subsystems. Figure 2 shows the basic layout of the subsystems. The EPS controls the charging of the batteries via the solar cells and the distribution of power to the rest of the spacecraft. production of power. Two power The to the power subsystems of the digital control functions of the satellite. subsystem The is for the satellite. heart of the modem is the 80186 processor. The The mass uploaded to the all subsystem has two completely separate, modem and the control microprocessor. PARAMAX PA- 100 spread spectrum demodulator DCS DCS controls microprocessor. The control processor all satellite storage system provides the satellite DCS commands the EPS the computer that provides control for digital control functioning under direct control of the Intel The satellite. redundant systems. The subsystem consists of the The has 18 solar panels for the satellite banks of 9 batteries each store power. Each battery bank has the capability of providing complete to provide The by users of the an functions. memory needed satellite. It is for storage of the files contains two redundant systems, each Mass TMUX DCS Storage EPS Temperature Modem A Multiplexer Mass A Storage System Controller A A RF Peripheral Control Bus Subsystem Modem B System U Controller Temperature B Multiplexer B Figure 2 with 4 mega PANSAT System Diagram bytes of random access correct any potential errors in data memory. The memory coming from is self error correcting to single event upsets induced by space radiation. The TMUX is an analog multiplexer that directs which temperature sensor is being read by the to provide a DCS. There are temperature sensors monitoring capability. The DCS on all major satellite will periodically read the temperature various components and record the data in the mass storage system for later by the satellite's operators. components down on loading RF SUBSYSTEM DESIGN BACKGROUND B. Brown [Ref.2] performed initial design Brown's work was the initial work for the PANSAT RF PANSAT communications design for the subsystem. system. Brown designed a binary phase shift keyed direct sequence spread spectrum using parameters established during a preliminary link analysis. His RF system design as well as printed obsolete when the modem became spectrum demodulator integrated circuit RF board (PCB) layouts. Brown's design became based around the PARAMAX PA- 100 spread circuit. Building upon Brown's work, the work included a complete modem and Dawson and Eagle [Ref.3] conducted a redesign of subsystem. Their design goal was to reconfigure the system so any one lost low- noise amplifier (LNA), and high-power amplifier (HPA), and oscillator the use of either partial DCS. They also identified specific components for use working model using connectorized components, and obtained model. Figure 3, less the bandpass filters on the IF side of the mixer, would still allow and constructed a test data was from the the design produced by Dawson and Eagle. Figure 3 shows block diagram of the system consists of three separate PCB's, a HPA PCB. An additional PCB and the RF RF RF subsystem. The flight version of this switching PCB, a RF LNA PCB and a RF serves as an interface between the peripheral control bus subsystem, providing control and power to the system. A staff engineer at the Space Systems Academic Group of the Naval Postgraduate School designed the interface of the RF subsystem with the DCS as well as the PCB for the interface. LNA Oscillator- RXDCS-A SPDT Pin Diode Switch SPDT LNA Pui Diode Switch BPF E SP4T Pin - ' RPF ^- Mech Diode Relay HPA SPDT ^Pin Diode Switch SPDT SPDT Pin Diode Pin Diode Switch Switch Switch RX DCS-B TXDCS-A X< E BPF SPDT Pin Diode Switch Mech Relay HPA TX DCS-B Oscillator-2 Figure 3 RF subsystem Block Diagram Gericke [Ref.4] designed the housing for the RF RF subsystem. Figure 4 shows the subsystem housing. The unusual shape allows the subsystem to constraints of the spacecraft. Switching, The 4 pockets HPA, LNA and power and in the circuit boards are control interface EMI frequency integrity. PCB A top cover fits over all the shielding. fit within the A sophisticated computer aided design tool produced by Cadence Design Systems was used for the for high PCB. designed using an exterior geometry to housing designed by Gericke [Ref.4]. within the housing provide mounting areas for the pockets in the housing and provides protection and The fit circuit board layout. Design rules construction from several sources were used to assure signal The following chapter describes the circuit and PCB designs in detail. Figure 4 RF Housing. From Ref.[4]. 10 m. RF SUBSYSTEM PRINTED CIRCUIT BOARD DESIGN A. PRINTED CIRCUIT BOARD EXTERIOR GEOMETRY The RF subsystem PCB design was accomplished using electronic computer aided design tools created by Cadence Design Systems including the Concept schematic capture tool and the Allegro PCB layout tool. The unusual exterior geometry of the PCB's required special care in design. The housing for the RF subsystem was designed using the IDEAS computer design software from the Structural Design Research Corporation. The housing aided was manufactured at the Naval Postgraduate School using a numerically controlled milling machine. The control data for the milling machine came form the IDEAS drawings. aerospace engineer designing the housing created mechanical drawings of the RF subsystem PCB's using IDEAS. These outlines mirrored the inside of the housing's pockets, shrunk by 1/32" in all exterior dimensions. mounting surfaces were then added to the dimensional PCB model which fit PCB The The holes, connector locations PCB and drawing. This produced a three exactly into the housing. These drawings were then electronically imported into the Allegro layout program, where they became the PCB outline and references for mounting hole placement. This allowed precise control of the physical dimensions of the PCB's without the tedious, error prone process of transcribing data from conventional drawings. The completed PCB drawings from Allegro were then exported to the IDEAS program where they became three dimensional models. The components appeared as 11 boxes on the surface of the PCB. The PCB was then checked to manufacture, ensuring a precise and eliminating the need for redesign due fit for fit into the housing prior to a physical size mismatch. B. RF SWITCHING PRINTED CIRCUIT BOARD 1. Functional Description The switching section serves to up convert the outgoing IF from the and route the signal to the desired LNA. selected It also receives the incoming DCS RF from to RF the HPA and down-converts the signal to IF, routing the signal to the desired DCS. Figure 5 contains a block diagram of the detailed schematic diagrams of the A transistor-transistor logic RF RF switching switching (TTL) PCB. Appendix A contains PCB. level input on each individual switch controls Oscillator- LNA- RXDCS-A SPDT Pin Diode BPF LNA-2 SP4T Pin Diode E SPDT SPDT SPDT Pin Diode Pin Diode Pin Diode Switch Switch Switch Switch RX DCS-B Switch TXDCS-A HPA-1 BPF \ E SPDT Pin Diode Switch TX DCS-B HPA-2 Oscillator-2 Figure 5 RF Switch PCB Block Diagram 12 the positions of the pin diode switches. Figure 6 the TTL TTL inputs to the level present at the TTL is RF switching PCB. This buffering assures a sufficient signal input to the switches to prevent spurious switching caused by noise on the wiring harness The network for A 54HC245 integrated circuit, located on the RF switching PCB, switch inputs. buffers the shows the numbering scheme used coming into the RF switching PCB. half duplex, meaning that either transmission or reception can occur, but not both at the same time. The switching network is capable of selecting any combination of amplifier, oscillator and system controller. This grants fault tolerance allowing any amplifier, and any oscillator and any system controller to fail simultaneously, while still by allowing the system to function without a loss of performance. SW1 LNA-l RX DCS-A SW4 SW5 SPDT SW3 Pin Diode Switch Oscillator-1 LNA-2 SP4T RXDCS-B Pin SPDT SPDT SPDT Diode Pin Diode Pin Diode Pin Diode Switch Switch Switch Switch TX DCS-A HPA-1 SPDT Oscillator-2 Pin Diode Switch SW6-A TX DCS-B HPA-2 SW6-A Figure 6 RF SW2 Switching network 13 TTL inputs For both transmitting and receiving, primary paths have been designated, based on the active RF DCS. These primary control board, based possible states of the paths will be the default switch positions selected by the on the control board sensing the active DCS. Table RF the DCS to choose an alternate signal path, primary path be inoperative or the the TTL DCS RF should the 1 also inputs required on the signal lines to achieve the desired switch positions. switching power from the same +5 PCB. Whenever the RF subsystem have power continuously Designation Primary all DCS. desire to exercise other components. Table All the pin diode switches receive the shows switching system, including the primary paths for each The non-primary paths allow lists 1 Rx DCS-A Path LNA-1.0SC-1, DCS-A LNA-1,OSC-2, DCS-A LNA-1,OSC-2,DCS-B LNA-2.0SC-1, DCS-A LNA-2.0SC-1, DCS-B LNA-2.0SC-2, DCS-A DCS-B Primary Tx DCS-A powered on, the switches must to assure they are in the desired state. This justifies the use LNA-1.0SC-1, DCS-B Primary Rx is VDC power input to LNA-2,OSC-2,DCS-B HPA-1.0SC-1, DCS-A HPA-1.0SC-1, DCS-B HPA-1,OSC-2,DCS-A HPA-1,OSC-2,DCS-B HPA-2.0SC-1, DCS-A HPA-2.0SC-1, DCS-B HPA-2.0SC-2, DCS-A DCS-B HPA-2,OSC-2,DCS-B u "0"=TTL Low "1"=TTL High X"=don't care Primary Tx Table 1 of a SW1 SW2 SW3 SW4 SW5 SW6-A SW6-B 1 X 1 1 X 1 X 1 1 X 1 X 1 X 1 1 X 1 1 X X X X X X X X X 1 1 1 1 1 1 1 1 1 RF Switching Paths 14 1 1 1 common power supply for all the switches, as they could not operate independently and maintain system functionality. The +5 the VDC switch power supply also supplies power to 5HC245. Each oscillator has a separate PCB. This allows PCB and under +12.5 individual control of each oscillator. command from the DCS, The RF switching board provides temperature sensor largest is power consumption and The RF switching signals into will housing from The board locations for two temperature sensors. One PCB uses to the oscillators this have the PCB. The temperature SMA connectors and coaxial cables to route IF and satellite to to and from the DCS is made SMA connectors mounted on the top of the HP A and LNA sections are made through the bottom of SMA connectors mounted on the bottom of the PCB. also contains board. Table 2 and Table 3 RF The produce the most heat on two 1 5 pin connectors that supply signals to the board as well as the signals screws secure the RF control power to the RF subsystem. and out from the board. The IF connection PCB. The RF connections RF all the TMUX, will monitor this temperature. through the equipment plate of the the supplies The EPS, through physically located near each oscillator. sensors, under control of the RF VDC power supply input to the RF switching PCB control from the temperature sensors located on the show the pin connections switching power and for the 15 pin connectors. Capture into its housing. This places the in intimate contact with the housing. 15 bottom of the PCB Pin Number 1 Function +12 Volts (OSC#l) SW6-A SW6-B 2 3 SW5 SW4 SW1 SW3 SW2 4 5 6 7 8 Ground Ground 9 10 11 Ground 12 Ground Ground 13 14 15 Temp Temp Sensor A Sensor B Table 2 Connector #1 Pin Description Pin Number Function 4 Ground Ground Ground Ground 5 Ground 6 7 Ground Ground 8 +5VDC 1 2 3 9 Temp 11 Sensor A Ground 10 Temp Sensor B 12 Ground 13 +12VDC(OSC#2) 14 Ground 15 Ground Table 3 Connector #2 Pin Descriptions 16 Circuit Description 2. The circuit to accomplish frequency conversion The RF consisting of an oscillator and a mixer. the IF (fjf) is 70 MHz. The is a simple heterodyning circuit carrier frequency oscillator frequency (fLO ) is set at is (/c ) 366.5 436.5 MHz and MHz. The switching circuit provides no amplification, allowing any radiated noise associated with the frequency conversion to stay within the housing for the PCB simplifies layout and switching PCB. This also power requirements. During the receive operation, the 436.5 RF MHz MHz + 366.5 = 803 fc - fto = atsc — ~,c<:c = -7A 436.5 366.5 70 ,/rr oscillator output to the IF stage will The band pass • be on the IF side of the mixers filters provide for rejection of the 803 MHz signal, providing in excess of 40 dB attenuation. The desired RF input frequency to the oscillator is MHz. If the input frequency will also is fL0 - fIF = produce the 70 signals at two 366.5 different carrier frequencies. rejection filter, attenuating the filter at U band pass frequency. 366.5 + 70 = 436.5 366 5 _ 70 = 296 5 filter at the The RF 436.5 image frequency, the section the input to the oscillator RF of the receiver removes the section serves as an image image frequency. During the transmit operation, the = , , MHz IF signal. This will result in the simultaneous reception of image frequency. The band pass fio ± MHz - 70 = 296.5 fL0 + fIF MHz MHz the output of the RF The modem up-converts ' oscillator output to the The fre( l uenc y selectivity of the section attenuate the the 10 RF will be HPA and me unwanted 296.5 MHz MHz in the PA- 100 demodulator to the 70 17 MHz IF. The band pass filters on the IF side of the mixers now serve unwanted frequencies from the conversion Table 4 lists total loss is 12.5 transmission lines in the internal a 50 components on the dB. This assumes the oscillator modem any modem. in the the signal losses for the to attenuate is RF switching PCB. The dBm output power. producing +7 and amplifiers are matched to a 50 The Q impedance, as are mixer input and output. This necessitates matching the IF and RF signal paths to Q impedance to minimize return loss. 3. circuit. Power Requirements The power supplies for the switches and oscillators are regulated outside the The power is filtered immediately after it is launched onto the RF switching PCB. This provides protection for the components from spikes in the power supply and reduces emitted electromagnetic radiation. to ground, reducing the effect ferrite core bead to reduce A series of capacitors short the DC power input lines of power spikes. The input power EMI. This from reaching the integrated circuits filtering prevents is coupled through a any noise on the power supply and any noise generated on the PCB from getting back into the power supplies. The oscillators and switches receive their power through a Component SPDT Pin Diode Loss Switch 1.3 SP4T Pin Diode Switch PIF-70 Band Pass Filter 1.1 1.1 Mixer Table 4 5.1 dB RF Switching PCB 18 dB dB dB (conversion loss) Signal Losses power plane. This provides a further reduces low inductance, high-frequency power supply noise and distributed capacitance that eliminates the need to locate a capacitor next to each of the switches. Table 5 requirements. lists The the total components on the power required RF for the switching board and their power RF PCB switching operation is 950 mW. This assumes that both oscillators are turned on simultaneously. PCB Layout 4. The RF switching PCB consists of 6 signal layers separated by FR-4. Detailed drawings and photoplots are contained in Appendix A. Figure 7 shows the layer by layer buildup of the RF switching PCB. The bottom 3 layers of the stripline structure for circuit construction. layer provides high frequency transmission The bottom power to all Pin Diode switching Quantity PCB Appendix E discuses layer and the ground layers are identical. components and the top Component SDPT lines. RF form a stripline The power layer routes control signals. Voltage Supply Power Current Required Required (ea) (total) 5 +5 VDC 10mA 1 +5 VDC 10 mA 50 VDC VDC 20 mA mA 500 250 mW Switch SP4T Pin Diode mW Switch 366.5 TTL MHz Oscillator Buffer Table 5 2 +12.5 1 +5 5 RF Switching Circuit Components 19 mW mW 25 ^ —— All Layers are 31 milFR-4 31 milFR-4 5 mil FR-4 ounce copper. 1 Top Layer hh^ mmh»-— Power Layer mmmmmmmmmmm rf Ground Layer - 31 mil FR-4 31 mil FR-4 mmhm^— - Figure 7 The RF switching PCB RF Switching PCB Ground Layer Bottom Layer Cross Section contains 2 ground planes in the center. The lower one number of layers provides the top of the stripline structure. The upper one creates an even in the PCB. This helps the layers are roll The top bonded PCB manufacturing process to prevent the for mounting. lip mount components mount components The plane serving as the interface with the because the oscillators power are small and require DCS. These plane lid from warping as filtering no holes and control drilled through SMA connectors areas, at ground, are placed there of the housing comes into contact with the surface of the PCB. This allows the housing to remain The power for areas on the top layer surround the of an opening in the PCB, and makes an easy PCB together. layer contains surface signal routing. Surface the in the Signal Layer "RF tight," provides shielding for the structure to assemble. layer contains 3 separate planes supplying and the 54HC245 buffer. The power 20 is power to the switches, the routed, from the connectors, through a filtering network on the top with the power planes. The From layer. left and the filtering network, vias provide the connection right planes in Figure each oscillator. The center plane supplies power to all 24 supply individual power the switches and the to 54HC245 integrated circuit. The RF signal layer routes RF and IF between components. geometry, as discussed in Appendix E. The trace width for a 50 impedance is calculated below using Equation (2) It requires a critical Q characteristic and Equation(3); with t = 1 .4 mil, b= 31 mil, and sR =4.6, the relative dielectric constant for FR-4.: cf - 0.0885(4.6) 1 + In n i-H 1- -1 1 14 In 1.4 1- 1 , 31 ^ 31 31 14 31 = 0.19601/7/ I cm w = 31 f 94.15 0.19601 I V50V4.6 0.0885(4.6)7 The components mounted on are in hermetically sealed metal cans The bottom layer also serves as a mounted against flanges in the \ — 1.4, J 1- the bottom handle the mounted RF signal processing. in contact with the mounting point for the HPA and LNA sections. The PCB is V 1- They bottom ground plane. SMA connectors going to the bottom, in the area surrounding the SMA connectors, bottom of the housing which provide an "RF tight" enclosure. 21 RF LNA PRINTED CIRCUIT BOARD B. Functional Description 1. The RF LNA board is used to provide the antenna. Figure 8 shows a block diagram of the the antenna states is and sends it to RF switching amplification of the signal after the LNA section. It receives the signal from one of two duplicate low noise amplifiers. Table 6 shows the of the pin diode switch and the required sent to the first PCB for frequency TTL inputs. After amplification, the signal down conversion and routing to the A detailed schematic of the LNA circuit is provided in Appendix B. LNA-l RF Switch Section SPDT Pin Diode Mechanical Switch Relay LNA-2 RF Figure 8 Switch Section RF LNA PCB Block Diagram LNA Selection Switch TLL Input LNA-l LNA-2 Table 6 1 RF LNA PCB States 22 Switch DCS. The RF LNA PCB requires +5 VDC power for each LNA, +5 VDC diode switch and +5 switch and the RF switching VDC for the 54HC245 are combined. PCB. Table 7 buffer. The power supply Two temperature for the pin diode sensors are provided as in the shows the connector pin assignment for the RF LNA PCB. Circuit Description 2. The low noise minimum 54HC245 for the pin gain of 28 between an amplifiers used in the circuit are Avantec dB with a maximum noise UTC-554. They have a figure of 3.0 dB. SMA connector and the pin diode switch. Each Each LNA is placed LNA has a separate power connection in the 15 pin connector allowing individual control of the power to each The power is filtered through a capacitor network and Pin Number 1 2 3 ferrite Function +5 VDC LNA-1 Ground +5 VDC LNA-2 4 Ground 5 +5 6 Ground 7 VDC Pin Diode Switch 8 Ground Ground 9 Temp 10 Ground 11 Temp Sensor ASensor A-2 12 Ground 13 Temp Sensor B-l Temp Sensor B-2 TTL Switch Control 14 15 Table 7 bead on lead as RF LNA PCB Connector Pin Assignments 23 in the LNA. RF switching PCB. The the RE control signal for the switch also passed through a is switching PCB. The loss through the pin diode switch gain for the 54HC245 is 1.1 buffer as in dB. This yields a total LNA PCB of 26.9 dB. Power Requirements 3. Each LNA requires 40 mA at +5 VDC for a total power of 200 mW each. diode switch required 50 The pin mW and the 54HC245 requires 20 mW. This yields a total RF LNA PCB power requirement of 470 mW if both low noise amplifiers are operating at the same time. PCB 4. The RF layers. It Layout LNA PCB is composed of 4 layers with an uses stripline circuit construction as in the RF FR-4 switching contains the detailed photoplot and assembly drawings for the shows a cross section of the RF The stripline circuit the Switching same as in the switching planes to minimize PCB between the PCB. Appendix B RF LNA PCB. Figure 9 LNA PCB. geometry for the PCB. This means dielectric the 50 PCB. This LNA PCB is identical to the geometry for Ohm characteristic impedance trace width is the layout was completed without the use of power complexity and reduce manufacturing costs. Additional power supply filtering and wide traces are used for the power supply connections to minimize the impact of not using power planes. 24 , Top Layer 31 milFR-4 Ground Layer RF Signal Layer Bottom (ground) Layer Figure 9 RF LNA PCB Cross Section RF HIGH POWER AMPLIFIER PRINTED CIRCUIT BOARD C. Functional Description 1. The HPA PCB serves to amplify outgoing transmission to the required level to ensure successful data link closure. Figure 10 shows a block diagram of the PCB. The HPA string contains 2 amplifiers, a approximately dBm and the final power preamp RF HPA to boost the signal to driver stage. Detailed schematic drawings are contained in Appendix C. The final output amplifier has an output of 5W. This power level prohibited the use of a pin diode switch for selecting the relay is HPA output for the antenna. A mechanical used. This also improves the isolation between diode switch. Due to the high gain HPA stages by 30 and power required by the may have power applied at any given time. Table 8 connector on the RF HPA PCB. 25 lists dB over HPA sections, a pin only one the pin connections for the HPA Pre-Amp HPA-l to antenna HPA Pre-Amp HPA-2 Figure 10 Pin RF HPA PCB Number Function 4 Temp Temp Temp Temp 5 +5 1 2 3 6 7 8 9 10 11 12 13 14 15 Table 8 Block Diagram Sensor ASensor A-2 Sensor B-l Sensor B-2 VDC HPA-2 Pre Amp to +5 VDC HPA-2 Control Ground +6 VDC HPA-2 Power +5 VDC HPA-l Power Ground +5 VDC Relay Coil Power to +5 VDC HPA-l Control Ground Ground +6 VDC HPA-l Power RF HPA PCB Connector Pin Assignments 26 Circuit Description 2. The HPA circuit is straight forward, routing to the amplifiers, the relay and to the output signals from input SMA connectors SMA connector. The preamplifiers are Avantec UTC-552 TO-8 can amplifiers. They use a +5 internally capacitively decoupled at the input VDC power supply and are and output and matched to a 50 impedance. The power supplied for the preamplifiers is filtered at the Ohm power input to the amplifiers using capacitors. The final power amplifiers have 4 internal stages. They require a special external decoupling network for the power supply due to the high gain of the four stages. The decoupling network uses surface mount components to achieve a compact design and prevent conventional component leads from becoming transmission lines and coupling together. The decopuling network is similar to one recommended by the manufacturer [Ref.5]. Power control output power from in the RF power and network. the It is is accomplished using a dBm to +37 dBm. SMA connectors. the testing of the RF +5 VDC control The power control control section. Figure placed in the to 1 1 signal is externally generated shows a "T" surface mount signal path immediately after its This network will allow fine tuning of the PCB. 27 signal. This will vary resistor entry on the RF PCB through signal level during Figure 11 Resistive "T" Network Power Requirements 3. The UTC-552 amplifier requires 18 The final produce 5 power amplifier is mA at +5 VDC for a total power of 90 mW. 45% efficient. The final power amplifier requires 11.1 W to WRF output power. The relay requires 20 mA at +5V when switched to the normally open position, for a power of 100 mW. The total power required for RF HPA PCBisll.3 W. Printed Circuit Board Layout 4. The RF HPA PCB, using micro strip circuit construction, with an FR-4 dielectric between the layers. Appendix construction. Figure 12 components are used The The final power amplifiers are RF HPA PCB. signal path. This required the use amplifiers, mounted with shown contacts the bottom ground plane of the is also fastened to the PCB. 28 layers strip circuit Surface mount of a micro mounting in Figure 13, require their leads perpendicular to the block of aluminum. The aluminum block composed of 2 E describes micro shows a cross section of the RF in the is PCB PCB strip circuit. to a heat sink. back to back on a with screws, so it The micro with/2 = width strip is calculated below, using Equation (6) and Equation (7) 31 mil: 4.6+1 4.6 -1f 0.11, 2 4.6 + 1V 4.6. A=-rrJ—T—+-—-1023 + —r\ 50 60 W V Se = 3l This is = 15577 1.5577 ,2(1.5577) -2 = 57 mil approximately double the width of a stripline trace of the same characteristic impedance. The HPA printed circuit board as well as the other printed circuit boards are designed to meet the functional requirements of PANS AT. The materials and construction techniques used represent those currently used in industry. chapter provides an analysis of the PCB The following designs with recommendations and conclusions. Top Layer 31milFR-4 Bottom (ground) Layer Figure 12 RF HPA PCB 29 Cross Section i OaCNSKMNGMeTOBWCMGPER ANSI 2 3 I IKW >9K CONTOOUJNG OUENSQN: MCH CHUENSOF TO CENTER OF l£AOS. REF MOCATES NON-COKTROUEO OMENSON FOR REFERENCE USE ONLY IOCS OM MN A 1760 e c 0245 1.370 017 | F G H 780 44 70 1390 0266 0023 0100 34 JO i29o esc K 0230 0300 1090 esc N P 0290 1190 8SC 0010REF 118 R s v * I T 132 0S35 05S5 044S 0465 0990 BSC esc 0790 esc 0390 esc PIN RF INPUT VSl VCONT 4 VS2 VS3 VS« Rf OUTPUT 6 7 CASE GROUND Ref.[5]. 14 10 11 15BSC 14 99 BSC 20 07 BSC 9 91 BSC 25 3 5 30 X 11 1 1 2 RF HP A Final Power Amplifiers. From 13 59 590 STYLE Figure 13 673 058 254 3 35 esc 4191 BSC 32 77 esc 676 7 11 584 752 27 69 BSC 3023 BSC 0.25 REF 300 3 3S 16S0BSC 0X6 4521 3SJ1 622 043 203 0080 0132 BSC J L MLUMETEM mt MAX MAJt 1 81 IV. A. ANALYSIS AND CONCLUSIONS ANALYSIS 1. The Link Budget PANSAT is budget link Appendix F shows the positive for both up link link computed margin for the and down link, RF in detail in Subsystem Appendix PCB F. design. Table The 1 1 link in margin is indicating that the communications link will be closed in both directions. The link budget would be improved by using a high gain directional antenna on the not possible because spacecraft. This is attitude control. The link PANSAT is a free tumbling budget assumes that PANSAT is horizon, in rain and at the highest data rate. This power may be reduced 2. if degrees above the a worst case scenario and transmit any conditions improve. requires the LNA PCB and the Switching PCB for operation. HPA PCB will remain without power to prevent any attempt to transmit during a receive operation. This first with no Power Budget The receive mode The is at 10 satellite would cause an overdrive of the stages to be destroyed, rendering control circuit will prevent the them F£PA section from LNA section is receiving power. The total During transmit operations power for the reason discussed above. useless. The LNA amplifiers and cause their A hard logic interlock in the RF receiving power required will at the in the receive same time the mode is 1.42 W. be removed from the low noise amplifiers HPA PCB 31 power will have power applied, with only one amplifier string in operation at a time. power requirement total reduced by transmitting is 12.1 W, The switching for a 5 PCB power. The will also require W RF output. This power requirement may be a lower power. at CONCLUSIONS B. The RF switching HPA PCB PCB has been manufactured and LN PCB and of the computer aided design tools art is in currently in testing. are awaiting contract approval for manufacture. The time required is critical to learn to use The The use of state working product. to achieve a Cadence Design System's tools is much longer than initially anticipated. The tools are very powerful, but assume the user has a D describes the basic steps of PCB design and background in PCB provides a of commonly used terms encountered list design. Appendix Printed circuit board design in the little RP detail must be gained from students pursuing work PCB frequencies written on practical steps for design and layout. the electromagnetic theory in The is design. almost a "black PCB design. in this area in the future is needed. 32 There is currently available texts provide and some general construction techniques. practical experience in art." Much of the design A good contact for APPENDIX A. RF SWITCHING PCB DETAILED DRAWINGS This appendix contains the detailed schematic, photoplot and assembly drawings for the RF reduced to switching PCB. The photoplot, assembly and physical dimension drawings are 64% of their original size for inclusion in this appendix. The following drawings are included: Figure 14 RF Switching Circuit Schematic Figure 5 RF Switching Circuit Schematic- Oscillator Section Figure 16 RF Switching Circuit Schematic - Amplifier Section Figure 7 RF Switching Circuit Schematic - Connector #2 Figure 18 RF Switching Circuit Schematic - Connector #1 Figure 19 RF Switching PCB Figure 20 RF Switching PCB - Top Assembly Drawing Figure 21 RF Switching PCB - Bottom Assembly Drawing Figure 22 RF Switching PCB - NC Figure 23 RF Switching PCB - Top Figure 24 RF Switching PCB - Power Layer Photoplot Figure 25 RF Switching PCB - Ground 1 Layer Photoplot Figure 26 RF Switching PCB - Ground2 Layer Photoplot Figure 27 RF Switching PCB - RF Figure 28 RF Switching PCB - Bottom 1 1 - 33 IF Section - Physical Size Drill Drawing Layer Photoplot signal Layer Photoplot (ground) Layer Photoplot K D O \ z a D a n tn U a tn X a \ u Q tn \ u a «n X X \ g\ X a £ Figure 14 T RF Switching Circuit Schematic - IF 34 a a $ 8 I I Section 3 I en cr o hCE _l u o en ® r i h '1 Jff ma ^ 3= T u r *S kS u Figure 15 RF 3 a (fl u 8 u 8 u I I Switching Circuit Schematic- Oscillator Section 35 a Si «& it I Figure 16 RF it £6 £6 Switching Circuit Schematic 36 - ? Amplifier Section S 8 I i """II "—I—tt— §!J ru # q: o iu u 3* ? J{| I _L IV.W.i . ^° -r-**g 2c -n-a? z z o u i* df B"**" Is-"* Figure 17 * wmi a »>OMDa>o-n«T» in RF Switching Circuit Schematic - Connector #2 37 -i—ii— T-I^i o^-» SI N& £ Oil l • <« 1 IH t-ii-5 «-? a o hu LJ Z Z O u Figure 18 RF Switching Circuit Schematic 38 - Connector #1 •— Figure 19 RF Switching PCB 39 - Physical Size Figure 20 RF Switching PCB - Top 40 Assembly Drawing Figure 21 RF Switching PCB - Bottom Assembly Drawing 41 Figure 22 RF Switching PCB - NC Drill Drawing 42 Figure 23 RF Switching PCB - Top 43 Layer Photoplot SifSipiiilimlfiBWi'lli' lk^^*j^i::i.i,!.:!j : 'laP^i i™piM»i z> o o D O O D © © © © 3 3 3 O Figure 24 RF Switching i CD PCB - Power Layer Photoplot 44 Figure 25 RF Switching PCB - Groundl 45 Layer Photoplot Figure 26 RF Switching PCB - Ground2 46 Layer Photoplot Figure 27 RF Switching PCB - RF signal Layer Photoplot 47 Figure 28 RF Switching PCB - Bottom 48 (ground) Layer Photoplot APPENDIX B. RF LNA PCB DETAILED DRAWINGS This appendix contains the detailed schematic, photoplot and assembly drawings for the RF LNA PCB. The photoplot, drawings are reduced to 64% assembly and physical dimension of their original size for inclusion in this appendix. following drawings are included: Figure 29 RF LNA Figure 30 RF LNA PCB Figure 3 1 RF LNA PCB - Top Assembly Drawing Figure 32 RF LNA PCB - Bottom Assembly Drawing Figure 33 RF LNA PCB - NC Figure 34 RF LNA PCB - Top Layer Photoplot Figure 35 RF LNA PCB - Ground 1 Figure 36 RF LNA PCB - RF Figure 37 RF LNA PCB - Bottom Circuit Schematic - Physical Size Drill Drawing Layer Photoplot signal Layer Photoplot 49 (ground) Layer Photoplot The - -I O -C BlslSl cu nj n GND cm GND GND GND m . ni GND - U4 23P hi s T s T GND in GND GND m U7 22P Q &—# 1 Ss U10 5P VDC (XI UTC554 ni ro cu GND GND (XI GND GND ni GND ,GND U12 10P "5^ 1 SI T (XI GND RFIN T RF SMP o RFOUl U8 GND 7P (xi UDC TOSW230 RF1 RF2 RFIN U2 8P SMA RF G- 4P TTL i — * i UTC554 US -3£_ (XI GND Si (XI GND 54HC245 TS GND ii RF 3FIN MX is. H> dir e GND U3 2P U9 IP YS I 44032 YS I 44032 TEMP SENSOR Figure 29 TEMP SENSOR RF LNA Circuit Schematic 50 RFOUl U13 9P SMPI -o o CO Figure 30 RF LNA PCB 51 - Physical Size Figure 31 RF LNA PCB - Top 52 Assembly Drawing Figure 32 RF LNA PCB - Bottom Assembly Drawing 53 Figure 33 RF LNA PCB - NC 54 Drill Drawing i| : Op 1 i 1 K # i © I-" ^ f *' o %a a • a a o a a , Figure 34 1i • RF LNA PCB - Top 55 Layer Photoplot RS Figure 35 HI • RF LNA PCB - Groundl 56 Layer Photoplot Figure 36 RF LNA PCB - RF signal Layer Photoplot 57 Figure 37 RF LNA PCB - Bottom 58 (ground) Layer Photoplot APPENDIX C. RF HPA PCB DETALED DRAWINGS This appendix contains the detailed schematic, photoplot and assembly drawings for the RF HPA PCB. The photoplot, drawings are reduced to 64% assembly and physical dimension of their original size for inclusion in this appendix. following drawings are included: Figure 38 RF HPA Circuit Schematic- Input Section Figure 39 RF HPA Circuit Schematic- Output Section Figure 40 RF HPA PCB Figure 41 RF HPA PCB - Top Assembly Drawing Figure 42 RF HPA PCB - Bottom Assembly Drawing Figure 43 RF HPA PCB - NC Figure 44 RF HPA PCB - Top Figure 45 RF HPA PCB - Bottom - Physical Size Drill 59 Drawing Layer Photoplot (ground) Layer Photoplot The ©-•wm^m D dm-T U3 2P <>e YS I 44032 ^7ft U4 IP TEMP SENSOR YS I 44032 TEMP SENS UCOIL HPP2-PUIR -< 2 = 2 = 1 - - 1 - - St l^fil-CONT ^7p. 25P HPftl-PWR — —as up _= 5 -us, I UTC552 s Figure 38 RF HPA 2 I s Circuit Schematic- Input Section 60 jf|T B.22l*<fc^ Hi! ^0 ^o 2T pl-pPH-PK? VCOIL * e U14 2P L^ s S: ^7 A Vft YS1 RTOUT VS2 YS3 MHW707-2 :£ «JS4 e. RFIN " bisufd MPPl-l 8 II UCONT S "vp t HPA1-CONT m& U12 R4P <fe uou ZZI I I -a- 29 I ? • 1 2 B.22UH<£-^ 5 s s rururu 0.0.0. <n l S S nisi *S? ^ MPft2-PW« II! 8- -re UOU IZZ aaa ^7ft I I ifTT I s U:3 := s =r i;?7P ni USl RFOUT VS2 US3 MHW707-2 US4 B.B18LF0 " RFIN HPft-2 S I I UCONT 12 5* HPA2-CONT L. a m ui n a d — ^A Figure 39 RF HPA Circuit Schematic- Output Section 61 Figure 40 RF HPA PCB 62 - Physical Size Figure 41 RF HPA PCB - Top Assembly Drawing 63 Figure 42 RF HPA PCB - Bottom Assembly Drawing 64 Figure 43 RF HPA PCB - NC 65 Drill Drawing 3 © © Lv£[ j Figure 44 RF HPA PCB - Top Layer Photoplot 66 if K*P ' Figure 45 RF HPA PCB - Bottom 67 iilli (ground) Layer Photoplot 68 APPENDIX D. PRINTED CIRCUIT BOARD DESIGN This appendix describes printed circuit board design and construction. of terms associated with PCB's is provided to help the reader understand the text as well PCB as speak intelligently to others in the be using a computer aided design A glossary (CAD) design tool for field. It PCB assumes the PCB designer will design. BACKGROUND A. A printed circuit board is a wiring structure used for the interconnection of electrical devices. It consists of a thin film of conductive material deposited on a non- conducting planar surface. The electrical devices are mounted to the planar surface with connected by the film of conducting material. This film their leads lines, called traces. The use of PCB's has allowed the number of holes, Oz. Copper - (1.4 mils thick) Aperture the size of the - A conducting material used in PCB of copper. complexity increases with PCB. with movement Anti-Pad is - construction consisting of thin film A solid plane of the copper weighs 1 ounce per square inch. A specially shaped hole, used by a photoplotter to pass light through to create a given shape on a film. pin number of layers and PCB GLOSSARY B. 1 the typically shaped into mass production of electrical systems and minimized the space taken by those systems. the is to make An aperture can be used without movement to make a shape or a line or arc. The area surrounding a pad which is void of all conducting material not connected. Typically appears as a "swollen" version of the pad. 69 when a Blind Via more - A via which does not go all the way through the PCB, just between two or internal layers and is invisible on the exterior of the PCB. Figure 46 shows a blind via. Buried Via - A via which goes partially through the PCB and is visible from the exterior of the PCB. Figure 47 shows a buried Dielectric DRC - - via. A non-conducting insulative material used between layers in a PCB. Design rule check, a CAD program may have the capability of checking the design database against a set of user defined rules to aid in determining design and manufacturing validity. Flash - A shape made by a photoplotter when the aperture is not moving. Footprint - The specific geometry of the pin holes and mounting holes together with the pads necessary for the mounting and connection of an electrical device to a PCB. G-10 Epoxy - A commonly used epoxy fiberglass dielectric for PCB construction. A low out-gassing material which can be used in space applications. Typical thickness' include Signal Layers Blind Via Figure 46 Blind Via 70 Signal Layers n v Buried Via Figure 47 Buried Via 5,10,15,31 and 62 mils.[Ref.6] FR-4 A flame retardant version of G-10 Epoxy. - GERBER A set of standard computer generated data formats used to describe the layers - PCB of a for the manufacturer. Name originates from the Gerber Scientific Company, an early pioneer in photoplotter manufacturing. [Ref.7] Layer A planar surface of conducting material. - Layers can be stacked on top of one another to form a multi-layer PCB. Layout Refers to the entire process of PCB design. The process of physically placing - component Net - The Netlist pins. A - footprints and interconnections logical connection An ASCII file on a PCB. Typically done using a CAD tool. between two pins. which contains the interconnection data between integrated netlist is typically extracted circuit from a schematic drawing by a computer aided design tool. 71 NC Drill File - An ACSII PCB data file with a numerically controlled containing the data required to drill. drill the holes in the Usually generated by a computer aided design program. NC Route File - An ASCII data file containing the data required to cut the final exterior shape of the PCB with a numerically controlled router. Usually generated by a computer aided design program, not required for simple Pad - The conductive PCB exterior geometry. material surrounding the plated-through hole for a pin which allows a connection between the pin and other traces or planes. Padstack will - A multi-layer build-up be used in each Plane - layer. of pads around a hole for a pin. Specifies which pads Figure 48 and Figure 49 show padstack composition. A thin film of conducting material which covers a large area of the PCB. are generally used to provide Planes ground and power connections because they provide a low resistance interconnect. Photoplotter - A plotter that writes using light. Just as in the case of a pen plotter, a photoplotter must be told which tool to use, were to go next and the path to use to the next point. For a photoplotter, a "tool" means an aperture. [Ref.7] Rats (Nest) - A display by some CAD programs to routing and can only be used in conjunction with a aid in netlist. connection of all the nets on a component, allowing the connections need to be made when component placement and The PCB placing the component. 72 rats nest is a straight line designer to see what Pad Anti-Pad Hole Figure 48 Padstack - Top View External Pad Internal Pad Hole Figure 49 Padstack 73 - Side View Routing - The process of connecting the pins of components PCB on a to match the connections in a schematic. RS-274D - An ASCII file of vector GERBER data derived from the traditional numerical control tool languages. Requires a separate aperture RS-274X - An ASCII printer data formats. Surface file of raster Does not GERBER data derived from the traditional require a separate aperture Mount - A PCB mounting surface of the PCB list. strategy laser list. where the components are placed on the with no holes drilled through the PCB for mounting. This is a technique which allows a higher number of components per PCB newer because the components can be placed, on opposite sides of the board, above and below one another. Thermal Relief - A structure placed in a layer around the connection to prevent the heat generated component if it were fully of a pin to a plane from soldering elsewhere on the plane from damaging the connected to the plane. Figure 50 shows a thermal Plane Area Figure 50 Thermal Relief 74 relief. Through Mount - A PCB mounting strategy for components where the component leads extend through the PCB. This is an older technique which does not allow components to be placed directly above and below one another. Trace Via - - A line formed on a PCB, used to connect pins on components replacing a wire. A hole through a PCB which has been plated through with a conducting material to provide inter-connections between layers of the PCB. C. BASIC CIRCUIT BORD DESIGN STEPS The basic steps in PCB design follow. This description apply to any type of PCB using the commercially available software will have 1. The all all is made to be generic and CAD software. Not all CAD the capabilities discussed below. Footprint Creation PCB designer reviews the schematic and the the parts necessary. will give the physical of materials and determines A review of the manufacturer's data sheets for the components component size and pin spacing. The the standard libraries provided with his PCB designer then looks in CAD tool as well as his own previously built libraries for any component footprints which currently components to ensure their applicability. a. bill exist, comparing these with the Padstack Creation When a new footprint must be created or retrieved from a library. created, the required padstacks are first A padstack is created by first determining the physical dimension of the leads on the component. This can be done from the manufacturer's data sheet or physically measuring the leads with a micrometer. This dimension 75 is used to specify the hole drilled for the Components should but not loose fit enough component which will form the center of the padstack. snugly into the PCB, without the need for force or undue pressure, to fall out when soldered in place. A typical hole might be 4 to 6 mils larger in diameter than the exterior pin diameter [Ref. 6]. Pad size is determined by the component packaging type and pin spacing. A pad should be large enough to allow good electrical connection and not interfere with the adjacent pads. Pads must also not short pins to their case. Pads different layers and need not be circular. The anti-pad is may be different sizes in generally a "swelling" of the pad around the hole by 10-45 mils, but need not be the same shape as the pad. technique is to change the anti-pad shape on one pin of a component to allow easy determination of a reference pin on the the padstack. A typical Many PCB. A thermal relief may also be associated with CAD programs allow the creation and storing of padstacks in a library for future use. For surface mount components, padstack creation only exist on one layer and are typically rectangular. same width as the pin is simplified. The pads will A surface mount pad is typically the on the component and 10-30 mils longer. The anti-pad will extend around the perimeter of the pad. Thermal reliefs are not usually generated for surface mount pads because planes are not typically placed around surface 76 mount components. Pin Spacing Layout b. When the required padstacks are created, they are placed in the physical geometry of the component, allowing the component critical that the components failure to holes for component leads or the surface exactly. be placed on the PCB. mount pads match It is the A skewed lead could cause weakening of the lead and premature of the component package. At frequencies above 10 Mhz, a skewed lead may cause signal distortion. A component footprint is typically saved in a library following creation. footprints are usually many for fit PCB package and not the schematic part name because their components use the same packages and have the same different prudent named by The layout engineer will print a 1 :1 test plot footprints. The of the created component and check with an actual component. This prevents unnecessary rework after PCB manufacture. Physical 2. PCB size is PCB size should may be left up to the PCB engineer, at which point a be made. This guess may be made based on knowledge of the designer or by looking at other as connector or will drive and Constraints determined by the physical size of the finished product needed to a particular housing. This on Size PCB's of similar complexity. Any first fit in guess PCB other constraints such mounting hole location must be determined before layout starts, as this component placement. A CAD system may also allow design constraints and rules to be specified. default design rules are a good place to start and should be modified only 77 The after consulting PCB with the manufacturer as to the ability for the manufacturer to physically create the design using those rules. Constraints such as route keep-in/out and package keep-in/out may be specified to enhance the use of the CAD tool in creating a correct design the first time. The number of layers components and routing. in the PCB should also be specified prior to placing of The number of layers will be determined by the component density and routing difficulty, as well as any desire to use a PCB typical plane. The PCB should have an even number of layers and will prevent warping of the board in the if possible, especially if PCB on the footprints are placed performance. import a Some netlist, this will placing the parts. This first odd around a center core of dielectric roll PCB to CAD tools have algorithms to aid in placement for optimum allow the is allow for optimum routing and PCB CAD program has the ability to designer to see component interconnection as he a valuable tool in getting the components in the correct place time. The them is manufacturing process, as the layers are routing and will automatically place components. If the the it Component Placement Component is A together. 3. circuit plane. has 4 layers, a top and bottom for signal routing and a power and ground shaped. This allows the manufacturer to build the bonded power or ground first components placed should be those with physical constraints requiring in a specific location on the PCB. Typically connectors and have these constraints. Components with critical signal 78 large components paths are placed next. This will PCB requires the layout designer to have a good knowledge of the schematic and underlying circuit. The remaining components are placed by placing integrated circuits first, followed by discrete components. Components must be placed to enhance the "routability" of the board the , operation of the circuit, and the ability to physically Components should never be touching or so close novice PCB designer which has been done may want to make and assemble that the board. assembly becomes impossible. The consult an experienced engineer or look at other to get a feel for the proper work placement of components. Care must be taken near mounting holes, connectors and other structures to keep components an appropriate distance away. be used for PCB components The mounting and PCB designer should their size, particularly know what type of fasteners head size, are to before placing avoid have a conflict between a screw and a component package. to Routing 4. Routing is the physical interconnection of pins on components with traces and provides the electrical connections necessary to complete the circuit. Use of a "Rats Nest" allows the PCB designer to see connections between pins before they are routed and enhances the ease of routing. Some CAD programs will perform automatic routing when given a proper set of constraints. Connections between pins can be layer or span multiple layers The first possible with as by using the which should be as short as layer changes as possible. All other signals are then routed. and ground are provided last. If same vias. traces routed are the critical signal paths, few made on possible, the use of power and ground planes 79 is Power desired. These planes reduce noise and provide low resistance distribution networks. Trace width should be wide enough to carry the current required without overheating. Coombs [Ref.6] provides curves on the current carrying capability of traces. For high frequency signals, in excess of 1 Mhz, special circuit construction techniques must be used to ensure signal purity. This occurs because the signals are propagating as RF waves through a transmission line. Appendix E discusses these circuit construction techniques. 5. When Manufacturing Data Creation layout and routing are complete, data files for the be created. The layers are typically described using the requires a separate PCB is GERBER file. odd shaped, The holes PCB manufacturer must GERBER data format. are described using Each layer NC drill data, and if the NC route data should be provided to the manufacturer. CAD programs will create the required data manufacturer as to the particular data files. file The PCB designer should check with the format the manufacturer supports. All manufacturing data should be checked prior to submission. If possible, a separate any CAD program from the one used for design should be used. CAD system biases from entering the data. 80 This will prevent APPENDIX E. STRIPLINE AND MICRO STRIP CIRCUIT CONSTRUCTION This appendix describes basic stripline and micro techniques are used when signal frequency on a PCB is strip circuit construction. so high (above 1 Mhz) These that transmission line techniques must be used to ensure signal purity. Design equations for characteristic A. impedance of traces are also provided. STRIPLINE CIRCUIT CONSTRUCTION A stripline transmission line consists of 2 parallel ground planes separated by a dielectric with a signal plane between the ground planes. Figure 5 1 shows an example of a stripline transmission line. Stripline construction encases the signal within ground planes. This allows the signal to be more immune construction technique requires the use of through surface mount components as the to noise exterior to the PCB. This mount components. does not support It ground plane must remain unbroken. Ground Plane Signal Trace Figure 51 Stripline Transmission Line. After Ref[7], 81 The most important When a single characteristic of the impedance stripline is used, the stripline is its characteristic is real impedance. and given by the following equation [Ref.8]: 94.15 Zo=T (1) w ~b C/ + 0.0885^ t_ Where: 0.0885^ c C, = and n t,b and and width of the stripline is 1 + In t t between equal thickness is shown dielectric l stripline, ~i spacing between ground planes in Figure 51. Eqn. and the ratio j w/b is 1 applies when the greater than 0.35. 94.15 Z JTR 0.0885s* 4 (3) a more useful form of Eqn.l because typically the dielectric thickness and conductor thickness characteristic of the (2) w yields: w=b Eqn.3 1- w refer to the thickness stripline respectively, as Solving Eqn.l for ln 1 is fixed and the designer impedance. 82 is left with the stripline width to control When make a stripline is required to change directions on a sure the impedance of the stripline remains prevent a high return loss. [Ref.8]. The match is best made with is operating frequencies to a mitered 90 degree corner or a small bend. showing the The recommended technique area where there at the must the designer A 90 degree right angle bend is mismatched at all frequencies [Ref.8] presents experimental data stripline. matched PCB, for tight more room, a curve with a effects bends Howe of various bends on matching in a is the 90 degree mitered corner. For radius of 3 line widths or greater is A curve with this radius has almost no difference in impedance compared recommended. to a straight line. When striplines are run in parallel other. This is the principle transformers. This coupling is When the is on a PCB, they capacitively couple used in the design of stripline not desired when filters, to each couplers and designing with discrete components. Capactive prevented by not running high frequency striplines in parallel on a PCB. traces must be physically close, such as entering a component, they should approach from 90 degree angles. B. MICROSTRIP CIRCUT CONSTRUCTION A micro strip transmission line consists of a conductor placed over a ground plane separated by a dielectric material. Figure 52 shows a micro transmission line structure. more A microstrip circuit has no upper ground plane, requiring careful shielding for the top As with impedance strip of the PCB to prevent EMI. the stripline transmission line, the micro strip line characteristic is critical. The electromagnetic fields 83 surrounding the micro strip structure from the differ stripline because of the lack of the upper ground plane. The following equation describes the micro strip characteristic impedance [Ref.9]: \20tt Z°~ W (W — + 1.393 + 0.667 In — + 1-444 (4) ^ \h h where £-\( S,+\ ** = + h\~2 1 2 2 + 12 l (5) FJ and h and Prefer to the dimensions given in Figure 52. The following equations micro [Ref.9] are used in design for selecting the width of the strip line: 8^ W = h-^-^ 2A -: (6) e where A Z IF." £ +1 =—-»-— —+£ o 60 V 2 -1 £ + < 0.1 0.23 1 V (7) £r J The same construction techniques which apply micro more strip circuits. critical in a metal 1^ + to stripline circuits also apply to The mounting on PCB's containing micro because of the exposed box with a shielded metal fields strip above the PCB. The lid. 84 transmission lines PCB is should be mounted Dielectric w Material t= Micro Strip Conductor V Ground Figure 52 Micro Strip Transmission Line 85 Plane 86 APPENDIX F. LINK BUDGET ANALYSIS This appendix contains a derivation of the link analysis for the Petite Amateur Navy from Satellite Ha (PANSAT). The [Ref.10], component values derivation starts with the general satellite link equations and reduces them to a set of PANSAT specific equations. The actual are then substituted into the equations determined. The worst case values are used for all and the link margin is data to provide an estimate of the worst case link margin. A. DATA LINK BIT ERROR PROBABILITY A link analysis is a power budget used to determine signal to noise ratio at various points in the satellite link. The required performance 5 a probability of bit error (Pb) of 10" for the data link. repeater because it demodulates the regenerative repeater is PANSAT data link specifies for the PANSAT is a regenerative digital data prior to retransmission. given by Sklar [Ref.l The Pb satellite for a 1]: Pb = Pu +Pd -2Pu Pd Where P u for the is (8) the probability of-bit error for the up link and Pd down link. PANSAT uses the same and the users can be assumed to be at the is the probability of bit error frequency for both the up link and maximum usable range from the similar weather conditions. This allows the assumption that P u and Pd down link, satellite in are equal. Therefore Equation (8) becomes: Pb =2PB -2(PB f 87 (9) Where P B = P u = Pd- Substituting P b =10' 5 into Equation (9) yields 6 P B = 5(10" ). ENERGY PER BIT TO NOISE POWER SPECTRAL DENSITY B. The use of direct sequence spread spectrum has no error for coherently detected differentially effect encoded binary phase on the probability of bit shift keying in the presence of additive gaussian white noise [Ref.12]. Therefore the probability of bit error is the same as without spreading and PB =2Q —7- Where is is given by Sklar [Ref.l I 2£„ W \-Q N.J K 2£? (10) KoJ the dimensionless ratio of energy per bit to noise ( Substituting 1]: P B = 5(10 ) into Equation (10) yields v power spectral density. \ VNJ = 10.425 = 10.181 dB LINK ANALYSIS EQUATIONS C. The basic link equations for the up link and down link are taken from Ha [Ref.10]: ^ =EIRPSAT - 20 \og^(**/A + — ^NJ„ J v c J GL y -L -101ogU)-101og(5)-5O, G -201og[^^j + y-101ogW-101og(5)-5O -r =£/^, w £) i Where: U,u D,d — indicates term pertains to the up link indicates term pertains to the down NJ = Carrier to noise ratio (dB). 88 link. (11) (12) EIRPsat = Carrier effective isotropic radiated power (EIRP) wave tube amplifier at ground station (dBW). to saturate traveling f = Carrier frequency (Hz). d = Slant range distance to satellite (m). c = Speed of light, 2.99792458 m/s. G u = Satellite antenna gain. Tu = Satellite system noise temperature (K). G = Ground station antenna gain. T = Ground station noise temperature. 23 k = Boltzmann's constant, 1.380658(10' ) J/K. B = Noise bandwidth of satellite channel (Hz). BOj,B0 = Required traveling wave tube amplifier back off (dB). L = Sum of antenna tracking loss and atmospheric attenuation for up link (dB). L' = Sum of antenna tracking loss and atmospheric attenuation for down link (dB). The general link equations are modified for use in the PAN SAT uses transistor amplifiers in both the back off terms, BOj and B0 , are not required. ground station PAN SAT link budget. and the satellite, The dependence of the EIRP terms on may also be removed because of the amplifiers in the satellite and ground station. The up link traveling wave tube the same, 436.5 both at the amplifier saturation MHz. PANSAT same is transistor and down link frequencies are capable of either transmitting or receiving, but not time. Therefore the distance to the satellite and the atmospheric attenuation and antenna loss will be the The modified parameter on the so the satellite link satellite [-} and same for the up link and down equations are given below. Where S link. indicates a G indicates a parameter on the ground station. =EIRPc -20\og[—^-J+^L -\0\og(k)-\0\og(B)-L C) — = EIRP NJ„ * s D —— (ATifd -20\og\ °\v c L J ' 89 -lOlogM-lOlog^)-! +-f z J C (13) (14) The following equation can be used bit to convert carrier to noise ratio, to energy per noise density ratio [Ref. 1 0] £-«(S Where Tb (15) the bit duration in seconds. Expressing Equation (15) in decibels yields: is -^- = ioiog(r4 )+ioiog(5) + f-^)w (16) VN Substituting Equation (16) into Equation (13) and Equation (14) yields the final link equations. It) ' ' D. =™^-W^^J+f^lOlogM-Z / (17) -20log[^^j+~^-\0\og(k)-L + \0\og(T ) (18) A E = EIRPs b FREE SPACE LOSS Free space loss station + lOlog^) v is the attenuation in the signal between the satellite and the ground caused by the propagation of the electromagnetic wave through space over the distance of the link. Free space loss is (Anfd\ - 201ogl represented by the term 1 in the link equations. The slant range to the satellite can be determined by the following equation [Ref. 10]: ( d= \(R e + Hf + R] - 2R e (R e + #)sin 90 £ + sin — R.—- cos{E) Y e ' KR P +H V (19) E= elevation angle to satellite in degrees. The height above the earth's surface for platform and launch mission. at a maximum height of 408 angle up the free space loss is kilometers (408000 meters). The minimum usable link and down link frequency of 436.5 elevation = 1461047. MHz and the above distance, 148.45 dB. RECEIVER SENSITIVITY E. Receiver sensitivity Gain is ratio is a dimensionless I number and temperature is expressed in degrees Kelvin (K). The normally expressed in decibels. The system noise temperatures are calculated for components up modem. to the 70 MHz intermediate frequency (IF) input to the modem. 1. on the The IF signal to noise ratio available at the input to This differs from most link analyses which use demodulator all components up to the input. Ground The ground which of antenna gain to system noise temperature. —J is the ratio section will be separately tuned based the expected to be launched from the Space Shuttle 10 degrees. Substituting these values into Equation (19), yields d is meters. Using the all PANSAT is PANS AT is determined by the launch Station Receiver Sensitivity station in this analysis is the station at the Naval Postgraduate School will serve as the controlling station for the satellite. This very high gain antenna and powerful amplifier. a. Antenna Gain 91 ground station contains a Antenna Gain a. The ground polarized antenna. station antenna The gain (G) on the is KLM 435-40CX high gain circular a main axis 15.2 is dB and the antenna with a computer driven tracking system to keep the main axis pointed to the is equipped satellite. Antenna Noise Temperature b. Antenna noise temperature has contributions from "spillover of radiation, blockage, and, in general, various inefficiencies" [Ref.12]. The antenna noise temperature can be determined from the following equation [Ref.12]: TAC = K,TS + K 2 TS + KJg (20) Where: TA c = antenna noise temperature for a clear sky (K). Ts = sky noise temperature under clear T g = ground temperature (K). K] = efficiency of antenna main beam. l-K. The ground conditions (K). station antenna efficiency (Ki) is that "half of the spurious radiation is directed to the allows T g to temperature. be approximated as Ts is Tg=To =290K 0.75. Fthenakis [Ref.12] sky and half to the ground". This [Ref.12], where T is the ambient obtained from Fthenakis [Ref.12] for a frequency of 436.5 elevation angle of 10 degrees, T = 6.5 s assumes MHz and an K. Substituting into Equation (20) yields TAc = 41.94 K. Equation (20) is for clear sky conditions. noise temperature and this is Rain has the effect of raising the system accounted for in the antenna noise temperature calculation. 92 Ha [Ref. 10] states that rain attenuation PANSAT will operate at 436.5 is only a problem MHz. The below 10 MHz. Due data, rain attenuation in this calculation will is be assumed to be 1 GHz. Crane model, for to the unavailability of dB. Total antenna noise given by the following [Ref. 12]: TA 1 Where T A Substituting frequencies above 10 rain models, such as the predicting rain attenuation are not accurate temperature at is total 1 'i-O T (21) antenna noise temperature and Lr T A c and Lr = c. =TAC^+ 1 dB into Eqn. A. 14 yields TA = is loss due to rain. 101 .6 K. Composite Receiver Noise Temperature Sklar expresses the composite receiver noise temperature of a receiver having n stages as [Ref.l 1]: C,=r+7~+77r+"H- rr Where T° is [Ref. 11]. (22) the effective noise temperature of stage n in the receiver the gain in stage n of the receiver network. — where L ' is The the amplifier. Table 9 is gain, G, of a loss stage is considered to be the loss of the stage. For a lossy line in the network, T° For an amplifier stage, T° network and G„ ={F- 1)290 [Ref.l 1], shows the contribution of the ground where F station =(L- 1)290 is the noise figure of components to the composite receiver noise temperature in Equation (22). Adding the contributions of the ground station components yields T°omp = 135.58 K. 93 Receiver Component L(dB) 5 Feet Coaxial Cable G(dB) F(dB) Stage Noise 0.1 Landwehr Preamplifier 16.0 1.1 Temp Contribution to (K) Tcomp 8.812 8.812 83.592 86.132 20 Feet Coaxial Cable 0.5 36.887 0.955 K+L Bandpass 5.0 627.061 18.294 2.5 225.701 20.823 6.4 975.896 0.568 Filter Avantec UTC-51 7 Amp 22.0 Mini-CktsZFM-1W Mixer 5.4 Table 9 Ground Station Component Contribution to (K) System Noise Temperature System Noise Temperature d. The total system noise temperature of the ground station is given by the following equation from Sklar [Ref. 11]: (23) Substituting, antenna o _ and composite noise temperature into Equation (23) yields T$ = —7- 237. 18 K. Calculating receiver sensitivity for the ground station yields V 2. Satellite because will Receiver Sensativity Satellite PANSAT is antenna gain is -2.7 dB at the worst null. This value is used a free tumbling satellite and no specific point of the antenna pattern be pointing to the earth b. = -8.55dB. . Antenna Noise Temperature a. The in at any given time. Antenna Noise Temperature Antenna noise temperature is 300 K. This is based on the satellite having a nearly omni-directional antenna and receiving noise from the sun, earth and other galactic sources. 94 Composite Receiver Noise Temperature c. Using Equation (22), the composite receiver noise temperature is Table 10 shows the contribution of the various components of the calculated. receiver to the noise temperature. components yields Adding the contributions of the ground satellite station 7^ = 2890.95 K. System Noise Temperature d. Substituting antenna and composite noise temperature into Equation (23) yields T° ( r = 3 1 90.94 K. Calculating receiver sensitivity for the ground station yields \ \TrJ = -17.14 dB. Receiver Component L(dB) Impedance Matching Network Band Pass Filter RF-300 Relay TOSW-230 Pin Diode Switch Avantec UTC-554 G(dB) F(dB) Stage Noise Temp (K) Contribution to Tcomp 1 75.088 75.088 5 627.061 789.422 0.05 3.358 13.369 1.3 Amp 28 3 101.199 407.547 288.626 1567.962 TOSW-425 Pin Diode Switch 1.1 83.592 0.720 TOSW-230 Pin Diode Switch 1.3 101.199 1.122 891.403 13.337 16.477 0.798 SRA-1 Mixer 5.1 PIF-70 Bandpass 6.1 Filter 0.24 TOSW-230 Pin Diode Switch 1.3 101.199 5.178 TOSW-230 Pin Diode Switch 1.3 101.199 6.985 TOSW-230 Pin Diode Switch 1.3 101.199 9.423 Table 10 Satellite Component Contribution 95 to (K) System Noise Temperature EFFECTIVE ISOTROPIC RADIATED POWER F. The effective isotropic radiated and transmitted power as shown power (EIRP) l t is the transmitted the product of the antenna gain following equation [Ref.l in the EIRP = P G Where P is power measured 1]: (24) l at the input to the antenna and G t is the antenna gain. This means the power out of the amplifiers must be reduced by any losses between the amplifier and the antenna. Ground 1. Station EIRP The peak output power from the ground The output power antenna gain 2. is reduced by 5.6 dB 15.2 dB. Therefore is Satellite is gain -2.7 dB. Therefore reduced by 7.35 42.18 W (47.78 dBm). dBm output. The dB the satellite amplifier circuit loss, resulting in is 5 29.65 W (37 dBm). The output dBm output. The antenna EIRP S = 26.95 dBm. LINK MARGIN The link amount required (17) 60 EIRP power G. circuit loss, resulting in is EIRPg = 57.38 dBm. The peak output power from is station amplifier margin is the excess to close the link. and Equation power available The link in the satellite link beyond the margin may be calculated by using Equation (18): 96 MARGINU =£7i^G -201og[^^J+|L -10IogW-£ + 101og(rJ-(^-J MARGIN D = EIRPs -20\og The data also contains a -^- +-f -lOlogU)- L + 101og(7;)- —- rate for the spread mode of communication narrow band, non-spread mode for is 9600 bits/sec. maintenance of 78125 The bits/sec. (25) (26) satellite The higher data rate will be used in the calculations to be more conservative. This yields a Tb of 12.8(1 0" 6 ) seconds. Additional losses (L) will be assumed to be 1 dB. Table the contributions to the link margin in Equation (25) and Equation(26). Component Uplink EIRP (dBm) Free Space Loss (dB) Receiver Sensitivity (dB/K) Boltzman Constant (dBm/K-Hz) Additional Losses (dB) Downlink 57.38 26.95 148.45 148.45 -37.74 -8.55 -198.60 -198.60 1.00 1.00 -48.92 -48.92 (Eb/No) Required 10.18 10.18 LINK MARGIN (dB) 9.69 8.45 Bit Duration (dB-sec) Table 11 Link Margin 97 1 1 shows 98 LIST OF REFERENCES 1 International Telecommunication Union. Circular 1960/22.01.91. Special Section amended 2. in Radiocommunication Bureau. Weekly AR1 1 /A/676. Satellite Network PANSAT. As Weekly Circular 2212/30.01. 1996. Special Section AR1 l/A/676 MOD-1. Brown, A.O. Ill, Communications Subsystem for the Petite Amateur Navy Satellite, Master's Thesis, Naval Postgraduate School, Monterey, CA, September 1993. Pete, "Redesign of the PANSAT RF Subsystem," EC-2990 Design Project, Naval Postgraduate School, Monterey, CA, September 1994. 3. Dawson, David and Eagle, 4. Gerkhe, Olaf, Design and Analysis of the Housing of the Communication Payload of the Petite Navy Amateur Satellite (PANSAT), Master's Thesis, Naval Postgraduate School, Monterey, RF Device Data. 5. Motorola. 6. Coombs, Clyde 1979. 7. 9. Book Printed Circuits Handbook, McGraw-Hill Co., New York, . World Wide Web Page, http://www.apcircuits.com/grdef.htm, September 1996. Howe, Harlan, Jr., Stripline Circuit Design, Artech House, Inc. Norwood, MA, Gonzalez, Guillermo, Microwave Transistor Amplifiers, Analysis Prentice Hall, 10. F., Motorola Incorporated, 1996. Alberta Printed Circuits Co., Basic Photoplotting Principles, URL: 8. CA, September 1995. Ha, Tri Englewood Cliffs, T., Digital Satellite 1974. and Design, NJ, 1984. Communications, McGraw-Hill Inc., New York, NY, 1990. 1 1 Sklar, B., Digital Communications Fundamentals and Applications, Prentice Englewood NJ, 1988. Clifs, 12. Fthenakis, E., NY, Manual of Satellite Communications, McGraw 1988. 99 Hill, Inc., Hall, New York, 100 INITIAL DISTRTJBUTATION LIST No. of copies 1. Defense Technical Information Center 8725 John 2. Kingman J. Ft. Belvoir, Rd., Ste. VA 22060-6218 Dudley Knox Library Naval Postgraduate School 2 411 DyerRd. Monterey, 3. C A 93943-5101 Chairman, Code EC Department of Electrical and Computer Engineering Naval Postgraduate School Monterey, CA 93943-5121 4. Prof. Randy L. Borchardt, Code EC/Bt Department of Electrical and Computer Engineering Naval Postgraduate School Monterey, CA 93943-5121 5. Rudolf Panholzer Code 07 Dean of Engineering and Computational Science Naval Postgraduate School Monterey, CA 93943-5003 6. Chairman, Code SP Prof. Space Systems Academic Group Naval Postgraduate School Monterey, CA 93943-51 10 7. Code SP/Rd Space Systems Academic Group Mr. David Rigmaiden , Naval Postgraduate School Monterey, C A 93943-5 110 LT Carl A. Lahti, USN 30 Teal Lane Groton, CT 2 0944 06340 101 JJUOLFY KNOX LIBRARY M JJJ"' 'GRADUATE SCHOOL CA 8394 3 2768 00339207 7