Statement of the Problem

The use of high power microwave transmitters in amateur radio is becoming more of a solution for crowded bands. As the use of cellular phones and other consumer communications increase, the bands below 1GHz become crowded. Portions of the amateur VHF a nd UHF bands are being sought after by commercial satellite providers. Commercial low-earth-orbit satellites operating below 1 GHz, so-called "little LEOs," have been presenting frequency sharing plans at informal work group meeting (IWG-2A) of a FCC ind u stry advisory committee preparing for the 1997 World Radiocommunication Conference (WRC-97). At the meeting, an industry representative seeking additional spectrum for little LEOs put forward a list of "candidate bands" for possible sharing that include d the amateur 144-148 and 420-450 MHz bands. The Amateur Radio Relay League (ARRL) has protested the inclusion of these bands on the grounds that amateurs use the bands for a wide variety of purposes, and sharing with a commercial service would impose una cc eptable limitations on our future flexibility to operate in the public interest (Sumner). The need for increased frequency space continues to grow as more Americans choose to stay in contact. Amateur radio operators are being forced out of bands that t hey previously occupied. However, they are permitted to use any mode of operation on 2300 - 2310 MHz and 2390 - 2450 MHz. Another key advantage to microwave operation unlike VHF and UHF is that all classes of amateurs are permitted to use any mode of oper ati on in microwave bands (ARRL). The high frequency transmitters and amplifiers that are on the market today are very expensive. Currently, one could spend $500-$700 building a microwave station

Background and History of the Problem

Experimenting in the microwave has been around almost as long as the amateur service has been in existence (ARRL). The popular perception about microwave communications is that it too expensive or difficult to construct. That is no longer true, loo k around at all of consumer devices above 900 MHz. In-home video transmitters, radar detectors, and microwave ovens are all examples of RF devices above 900 MHz. Microwave ovens! Microwave ovens are RF devices that operate around 2300 MHz.

High power RF amplifiers for frequencies above 1 GHz are not commonly available from amateur suppliers. Low level transverters can be built using Monolithic Microwave Integrated Circuits (MMIC) that convert signals to the microwave bands. Transverters interact with factory made transceivers in the HF or VHF range and are often home built. These units convert the transceiver transmit signal up to a higher frequency and convert the receive frequency down to the transceiver receive frequency. The resultin g performance and signal quality at the higher frequencies are enhanced by the frequency stability and the signal processing capabilities of the transceiver. High frequency transverters only have approximately 10mW of output power. With these low power l e vels an operator needs to live within driving distance of the contact station. The desired transmitter would need to produce at least 100 watts of RF power to be capable of Earth-Moon-Earth (EME) contacts. During EME contacts, the moon is used as a pass iv e reflector to communicate with stations as far away as 400 miles. Down East Microwave, a supplier of microwave components to the amateur community, offers their highest output amplifier at 2 watts for $295 (Down East). Another supplier, SSB Electronic s o ffers a 10 watt amplifier for $570 (SSB). Techlock Distributing is a commercial supplier of RF Microwave amplifiers; a +10dBm amplifier costs $595 (Techlock).

The goal of this project is definitely cost; a usable transmitter should be built for less than $200 with a minimum output power of 100 watts. The final design should be capable of transmitting or amplifying between 1.2 GHz – 10 GHz. Efficiency is not a concern with this project, many of these devices operate around 10% efficient. The complete transmitter will operate off of 120V and accept line level audio to be frequency modulated.

Possible Overall Approaches

There are several alternatives in the design of microwave amplifiers. Only one other design meets the 100 watt output power specification. A 250 watt amplifier referenced in the UHF/Microwave Projects Manual (Angle). This amplifier uses a tuned cavity amplifier. Unlike using inductors and capacitors at lower frequencies to tune the amplifier, this design uses a mechanically tuned cavity to provide the resonant circuit. The resonant frequency is set by mechanically adjusting the cavity. This design uses accurately machined plates and cavity rings, which would be difficult to manufacture by hand and quite expensive to pay a machine shop. The necessary parts include aluminum, copper, and Teflon rods. Due to heating effects, water cooling is needed to achi eve 250 watts. A complicated system of water fittings and jackets are used to circulate water trough a radiator. The water cooling system would possibly eliminate remote operations of the amplifier. Remote operations have always been an interest to amate u rs during contests and emergencies. The cooling system alone might kill the project. This circuit requires twelve watts of drive. If a no-tune transverter were used, a small solid-state amplifier would be need for preamplification.

The second possibility is the 80 watt solid-state amplifier. This is a forced-air cooled amplifier, which, unlike the water cooled amplifier, may be portable. This amplifier uses the SD 1870/TCC2223-18 transistor amplifier. The maximum output power of the amp is 45 watts so two amplifiers are coupled together to generate 80 watts (Hackford). The circuit board is very difficult to manufacture and expensive to purchase since it makes-up the tuning. This high cost is due to the on board third and four th order Chebyshev band-pass filters (Davey). Similar boards cost as much as $60 from Down East Microwave (Down East). The packaging used is milled from brass stock, another costly job for the machine shop. This amplifier is also not a good design for be g inning RF experimenters. Twelve watts of drive are needed to achieve the 80 watts output with two amplifiers.

The third possibility is a 1296 MHz linear power amplifier (Olson). This amplifier uses the Mitsubishi, M67715, and power amplifier module. The major drawback with this design is it only puts out 3 watts. This amplifier would be great to get on the air at 1296 MHz using a low-level transverter to generate the high frequency RF. Unfortunately, it falls short of being capable of EME contacts because of the low output power. This is a very simply, easy to build amplifier that requires no cooling syste m.

The Proposed Solution

The approach to be implemented was chosen for several reasons. First and foremost, it is inexpensive due to the abundance of microwave ovens. The 2M189A magnetron is common to several ovens such as: Hotpoint, Whirlpool, and Tappan (Pacholok 55). The microwave oven should cost less than $100. Since low cost of new units has driven many consumers to discard ovens with a good magnetrons. The only parts of the oven that will be used in the project are the high voltage power supply and the magnet ron. Using a 900 watt oven should give output power around 250 watts.

These cost factors dominated the design. This circuit could have been done with solid-state components or a tuned cavity. Either of these options would have required expensive machine shop costs. The 10 - 12 watts of drive needed for other circuits would have also been expensive. Converting low frequency RF signals to microwave signals is the common way of driving these amplifiers. Transverters, as they are called, usually only deliver 10mW of output power and can cost as much as $300 (Down East). The accurately designed and manufactured high-frequency circuit boards alone cost as much as $60. The tuned cavity amplifier requires a 1.9kV power supply for anode - collector voltage.

Specifications

120VAC @ 9 amps

Minimum Output: 200 watts

Output Impedance: 50

 RF Output Connector: N

 Output Frequency: 2430 MHz

 Frequency Modulated Video

 Maximum Humidity: 100%

 Operating Temperature Range: 0C - 75C

 Weight: 10 kg

Overall Block Diagram

Figure 1. Microwave Transmitter Block Diagram

Principals of Operation

Low Voltage Power Supply

From the block diagram above, it can be seen that the microwave transmitter has nine blocks. The first block is the low voltage power supply this is for the control and modulator circuits. The low voltage power supply provides 12.6 VAC for the filament s, 5 VDC, ± 17 VDC, and 12 VDC. The ± 17 VDC power supply is un-regulated. The 5 VDC and 12 VDC supplies are regulated with 3 terminal regulators. The 12 VDC power supply will be tested by loading it w ith a 12W resistor and measuring the output voltage to be within 5%. The 5 VDC power supply will be tested by loading it with a 1kW resistor and measuring the output voltage to be within 1%. The 5 VD C is used for a reference voltage in the modulator circuit.

High Voltage Power Supply

The second block is the high voltage power supply. This is a 5 kVDC power supply that is capable of about 350mA. The high voltage transformer and rectifier diodes will be salvaged from the donor microwave oven. The filter capacitor will be made up of t welve 400 VDC electrolytic capacitors. All cabling for the high voltage will be rated at no less than 5 kVDC. All high voltage terminals will be adequately insulated with silicon compound. Refer to my Safety Plan on the last page for safety precautions. F or safety precautions this block will not be tested.

Control & Interlocks

The third block is the control and interlocks block. The control devices allow the user to activate the high-voltage on the transmitter. The high-voltage relay (K4) can be energized by activating the "Power On" push-button (PB4) this will be sealed by a set of normally open contacts on K4. The relay will be de-energized by either an interlock or the "Power Off" push-button (PB5). The interlocks are safety devices that will not allow the high voltage power supply to turn on if there is a safety conditio n. There are three interlocks, cabinet safety switch, high magnetron current, and over magnetron temperature. All of the interlocks are normally open devices and they have individual indicator lamps. The cabinet safety switch will be mechanically mounted on the inside of the high-voltage power supply cabinet. This safety switch will not allow the high voltage power supply to be activated if the cabinet is open. The high magnetron current detector will operate a normally open relay to shut down the high v o ltage supply in the event of excess magnetron current. A test point on the modulator circuit is fed to a comparator to trip the high current interlock. The reference voltage is 8.1V which is equal to 300 mA of current through the 27 ohm resistor . A NPN bi-polar transistor will be used to drive the relay. The thermal overload device is a normally open device that will activate if the temperature of the magnetron exceeds normal operating temperature. All of the interlocks will be lat ching circuit s. The condition will have to be corrected and the reset button activated to reset the interlock. The transmitter will then have to be activated again before the high-voltage power supply is applied. There will be three indicator lights that will illumina te after an interlock has been activated until it is reset. Normally closed contacts from relays 1-3 will be in series with the 12 VDC supply for the high-voltage relay. The relay circuits will be tested by removing the individual sensors and replacing th em with a short circuit. Under a short circuit condition, the indicator lamp should illuminate and the high-voltage relay contacts on relay four should open. The sensing circuits will be tested by removing them from the relay circuit and conn ecting them to an ohm meter. The cabinet interlock will be tested by simply opening the cabinet enclosure. The over-temperature sensor will be tested by heating the thermocouple with a lighter and observing continuity on the ohm meter. The high current se nsor will b e tested by simulating a high current condition and observing continuity on the ohm meter. Alternatives for this block may have been TTL logic or a microprocessor. Either of these designs would have been far too expensive and complicated for th is block. Using relays keeps this block simple since it is only a supporting block for the entire project.

Modulator

The forth block is the transmitter. Connector J1 is the output and must be terminated. The output of the magnetron is a TE₁₀ waveguide connector, an adapter converting that to an N connector will be used to couple the transmitter to the load . A reference voltage of 5 volts is adjusted by R5 and R6, this controls the magnetron plate current. This voltage is applied to the non-inverting input of U1 which drives source follower Q1. The output of Q1, plus R9 and R7, provide negative feedback to U1 in the ratio of 5.7:1. At equilibrium, Q1’s drain-source current produces a voltage drop across R11 that equals 5.7 times U1’s non-inverting voltage. To increase stability, R3, R14 and R15 are added to help prevent parasitic oscillation in U1 and V1. T he magnetron was chosen for use in this block for several reasons. Unlike the previous tuned cavity option, the magnetron will require only forced air cooling. This is huge advantage for the amateur radio community in that mobile or portable operation i s still possible. Magnetrons are also very inexpensive if not free because of the abundance of microwave ovens. During operation a small dish may be used as the antenna. During testing I will use an air-cooled dry 300 watt dummy-load. This unit will be on l oan from Byrd Trueline Corp. in Ohio. During testing the transmitter will be modulated with color bars. A low-noise block converter will be used to downconvert the signal to L-Band where a standard satellite receiver can be used for demodulating the vi deo .

RF coupling

The fifth block is the magnetron RF coupling. The magnetron RF energy must be coupled to an RF connector to allow for transmission. The magnetron is mounted in rectangular waveguide using a TE₁₀ launch. The energy is released into the co oking chamb er through a rectangular hole. The radome/splatter cover and the field stirrer blades were removed. Using a small piece of copper stock the output will be shorted. The waveguide is then shortened to the desired .5l g. This is ac complished by screwing five brass screws through the broadwalls of the waveguide. An RF probe will be used to launch the RF out of the waveguide. The probe was deliberately shortened below .25l g. This will cause a reactive mism atch and the frequency will be pulled 25 MHz down from its design frequency, allowing legal operation within the amateur band. The probe was built on a Type N RF connector using brass tubing. The probe is built to allow tuning for maximum power while oper ating.

Waveguide Cavity

The sixth block is a second piece of tuned waveguide. This piece of waveguide will be used to sample the high power RF signal for measurement purposes. This piece of waveguide will have three Type N RF connectors. The waveguide will be constructed usin g stock copper sheet metal and brass tubing. Port 1 will be the input port. Port 2 will be the attenuated signal output. Port 3 will be the un-attenuated output port. The device will be swept from 2.35 GHz to 2.5 GHz and tuned for minimum loss between po r ts 1 and 3 and -20 dB of attenuation on port 2. -20 dB will result in a signal 1/100 of the transmitted signal, 300 watts will produce 3 watts on port 2 which can then be measured with an RF power meter.

Works Cited

American Radio Relay League. The ARRL Handbook for Radio Amateurs. 74^th ed. Newington, CT: ARRL, 1996.

Angle, Chip. "Quarter Kilowatt 23-cm Amplifier." The ARRL UHF/Microwave Projects Manual (1994): 8-71.

Davey, Jim. "No-Tune Transverter for 2304 MHz." The ARRL UHF/Microwave Projects Manual (1994): 3-24.

Down East Microwave, www.downeastmicrowave.com.

Hackford, David. "A 2304-MHz 80-Watt Solid-State Amplifier." The ARRL UHF/Microwave Projects Manual (1994): 8-87.

Pacholok, David. "ATV Transmitter From a Microwave Oven." 73 Amateur Radio July 1989: 54-59.

SSB Electronics, www.ssbusa.com.

Sumner, David. "More on Little LEOs." QST February 1997: 9.

Techlock Distributing, http://www.erinet.com/kenny/divider1.html.

High Voltage Safety Plan

I have many hours of experience with high voltages working as a maintenance technician at WLFI TV-18. The 5kV power supply will be salvaged entirely from the donor microwave oven. The new package will be a metal, electrical-grade enclosure. The enclosure will be grounded using the ground wire in the three-conductor ac power cord. The enclosure will completely contain the high voltage. Safety interlocks will be built into the cabinet to ensure that high voltage is not applied if the cabinet i s open. The high voltage will not be handled or touched in any way during testing or operation. A high voltage indicator will be added to the unit to alert individuals of the danger. This unit will never be left unattended with or without power to protec t unsuspecting by-standers. A spotter will always be used during testing to increase the safety of myself and those around me. A microwave leakage detector will also be used during testing to protect against RF radiation.