Carrier Frequency Generators: ----------------------------- Marconi XH-100 receiver with a crystal oven and crystals for 13.583 and 13.77770 MHz. The latter was installed as a new xtal in September and set to 13,777.700 kHz. By December 2003, it had drifted up by a mere 7 Hz and was reset. This corresponds to a long term stability of 0.5 ppm. Siemens Pegelsender signal generator set to 13.7777 MHz and monitored on a frequency counter. (Out of Service) Another XH-100 was used with xtals for 13.2 and 13.4 (difference = 137 kHz, variable). The HFO was mixed with the LO in the MC1496 DBM to produce a nice sine at 137 kHz, but its frequency wasn't stable enough. It was even less stable than the Pegelsender. (abandonded) Thanks to some fine gentlemen north and south of the border on the Left Coast, ie: W7OIL and VE7TIL it is hoped that the Pegelsender flexibility without its instability will soon be realised in a new carrier generator. A PIC programmed and provided by VE7TIL was used with a AD9850 DDS chip from W7OIL to produce a carrier on a wavelength of 100 metres (2997.92458 kHz +- a bit of course) This was multiplied by 16 with a CMOS 4020 frequency divider to 1600 metres (F=187.37 kHz.) An OXCO frequency synthesiser was very generously provided by F1AFJ. This can be programmed to operate anywhere in the 135-190 kHz range. Acting on the advice of G0NBD, a couple of boards with ad9850s were obtained in December 2012. These have been programmed via the usb port by a pc. Stability is about 1 ppm. Logic Circuitry --------------- The 13.8 MHz is fed to the divider/logic TTL board which is keyed by a CMOS diode matix ID circuit during beacon operation. This ID sends VO1NA (59 bits) then does nothing for 41 bits, after which it recycles. The logic board consists of 2 7490 decade dividers and a 7400. The keying logic precludes a DC output that would fry the MOSFET in the final amp. From the logic board, the keyed 137 kHz signal goes into an NPN switch, a PNP/NPN totem pole, across the room via twisted pair to the gate of the final amplifier. A BASIC program was written to send QRSS. This keys the logic board through a handshaking line on the serial port. Final Amps ---------- Class E amps designed using G3NYK spreadsheet based on Sokal's design equations. 15 watt P channel MOSFET. 13.8 volts at 1.5 amps to IRF9540 source. 125 watt N channel MOSFET. 45 volts at 3 amps to IRF640 drain from a modified choke input power supply. The choke buzzes very loudly when the tx is keyed. The power supply has a big (~30 cm long) 50 ohm bleeder across the output to improve the regulation and keeping the voltage peaks below Vds(max) 1200 watt Decca 5501A modified by G3LDO with 12 IRF250 MOSFETS. 10 watt IRF60 Class E for 160-190 kHz. Aerial and Impedance matching ----------------------------- RF is taken from a powdered iron loop core with 5 primary turns (50 uH) and 12 secondary turns (optimised for a 50 ohm load). This goes to a air-wound auto-transformer then to a 0.4-1.4 mH variometer. The RF current at this point is about 1.7 amps. The aerial comprised 2 parallel 100 metre wires 1 metre apart to the top of a 25 metre tower on the side of hill. The base of the tower is about 150 metres to the west and 20 metres above the shack. The wires are connected at the end near the house and fed at this point. There is no connection between the tower and the wires. As a result of nasty WX, only one wire remains. More inductance was needed to tune the single wire. 16 Jan 2005. A tuning helix and variometer were installed at the near end of the 100 metre wire. These are fed with coax. This was to keep RF (and high voltages) away from the house. There was little change in the aerial current and relative field strength as a result of this rearrangement. New Power Supply for the Decca 5501 ----------------------------------- 10 December 2005. Finally got a rectifier and lashed a big power supply together using Hammond type Q transformer Cat No. 93895 1.1 kVa 115V/55V max temp rise 80 degrees C. Hammond reactor type 195 G20 5 mH 20 A DC. Hammond xfmr 12.6V 20A, Capacitor 7,400 MFD 200 V. 25A 1000 V PRV rectifer. Vf=1.1 V at 12.5 A. No load 82 volts or so. 16 Ohm load 61 volts. 2.5 Ohm 55 volts. Hope to have ERP up a few dB or so soon! Using a 5 kW duct heater as a bleeder. 4 Feb 2006. Decca draws 14A at 60V with 2/3 PA's engaged. The original XH-100 (modified as an exciter) feeds the Decca amp via a 0.02 uF for an aerial current of 3.1 amps. Gain ~ 3.1^2/1^2-> 10dB ! Running the Decca with all PA's engaged (spring 2007) worked well, but the electric company had to provide a new pole transformer to correct voltage drops caused by the TX keying. RX Facilities ---------------- 16 turn RX loop 2m diam. 2004-2009 (destroyed by wx) miniwhip PA0RDT National LF 10 preselector CMC XH-100 with DDS060 local oscillator Yaseu FT-817 with Burhans preamp Short (525' beverage) 30 cm loop