A 60 meter SSB transmitter Nick Kennedy WA

A 60 meter SSB transmitter Nick Kennedy, WA 5 BDU Ozark. Con, Joplin MO April 29, 2006

Introductory inspirational quotes Thoughts from JF 1 OZL’s homebrew web page: Some of the successful projects was born from the failed idea. We have the proverb , " Failure is the mother of the success. " in Japan. You may write your notebook not to record your successful projects, but raise your new idea. Why I make my rigs by myself: Sometime I receive the question " Why do you make your rigs by yourself? ". I can not answer the question. The man making such question thinks as follows. It cost only about $500 to buy cheap rigs. It cost about $1000 to buy some test equipment. I can understand what he says. But I want to say as follows. The purpose of Amateur radio is to spend money and to spend times. It makes no economical profit to the station manager. To make high tower costs very high. To make shack on the mountain costs high. To go to expedition cost very high. And also to make my original rig costs very high. It is the way of Amateur life. From Frank Harris’ on-line book: Crystal Sets to Sideband, Chapter 15: The Nobel Prize for Sideband: Getting on 60 meters SSB - don't bother! Before the 60 meter SSB frequency became available to American hams on July 4 th 2003, I thought it would be fun to get on the air before the commercial gear was on the market. I naively thought that homebrewers would own the frequency for at least a little while. What I didn't know was that some of the newest transceivers can be reprogrammed to operate on any HF frequency just by pushing some buttons. Anyway, The band was instantly clogged with commercial transceivers.

Verbal Block Diagram • • Speech amp, then audio BW shaping Carrier oscillator (VXO crystal oscillator) Balanced modulator MC 1496; 9 MHz DSB out Crystal filter produces SSB DDS for local oscillator “VFO” (PIC control) Transmit mixer NE 602, moves to 5. 3 MHz Several stages of linear amplification We’re done, take a break!

electret microphone NE 5532 Audio BPF LM 387 40 d. B speech amp & clipper MC 1496 DBM 5 -p crystal filter DDS Controller 9 MHz carrier osc PIC 16 F 84 A 6 MHz LPF Class A TN 2219 A 20 d. B to +9 d. Bm Class A RCA 4013 15 d. B to 100 m. W out Class AB 2 N 3866 push-pull to 1 watt Class AB 2 SC 2075 push-pull to 5 watts 60 meter SSB transmitter block diagram NE 602 transmit mixer 101101 AD 9850 DDS 5 element LPF; Fco = 5. 75 MHz

Microphone preamp/limiter • • • Two stage LM 387 with 1 st stage clipping Gain of 10 per stage for 40 d. B total Balanced modulator needs about 1 Vrms Gain of 1 st stage set to clip at ~maximum Single supply design leads to some complexity & limits flexibility. • Bias voltage provided at input for electret type microphone. Also RF filter at input.

Microphone preamplifier and limiter

Audio bandwidth filtering • NE 5532 LPF and HPF stages in series • Voice BW filtering more difficult than CW; hard to achieve in a single stage • Gain of 1; target BW 500 to 2200 Hz • Two pole Salen-Key 40 d. B/decade roll-off; (TI “ Single Supply Op-Amp Circuit Collection”) • Attempted to use quality film or poly capacitors in audio signal train

audio bandpass filter

9. 000 MHz Carrier Oscillator • Colpitts with additional amplifier stage • Uses a crystal from the filter set • Series L & C may require juggling to hit the required frequency • Adjust ~500 Hz beyond edge of filter BW • MC 1496 needs about 160 m. V rms drive • Reliable circuit; requires minimal fuss to work. Ref. EMRFD fig 6. 103

9 MHz Carrier Oscillator

Balanced modulator • MC 1496 similar to NE 602; Gilbert cell • DC bias: R 1, R 2, R 3 are a voltage divider to bias pins 6/12 2 V above 8/10; 8/10 2. 7 V above 1/4; 1/4 2. 7 V above 5 • R 7 sets I 5 current into pin 5 at (12 -0. 7)/(6 k 8+500) or 1. 6 m. A • I 5 * R 8 = 1. 6 volts, sets peak allowable audio level at 1. 6 volts, or 1. 12 Vrms.

• R 8 sets gain; 1 k is typical per data sheet • R 9 sets current source differential output Z at 2 k 2 and T 1 matches 2 k 2 to the 200Ω crystal filter. • RF input for best sideband suppression 160 m. V rms @ 10 MHz per datasheet figure 22, in same ballpark with EMRFD value of 300 to 500 Vp-p. • R 4 & R 5 isolate RF & AF inputs from bias nodes at AC ground.

balanced modulator MC 1496

MC 1496 Schematic NE 602 (SA 612) equivalent schematic Comparison of MC 1496 & NE 602

5 -pole 9 MHz crystal filter • Passes desired sideband of DSB output from balanced modulator • Voice bandwidth is a greater challenge due to parallel resonance in crystals; compensated with parallel inductors. • XLAD, AADE & other software available • Actual BW usually wider than calculated • 200 ohm Z for better capacitor values

5 -pole 9 MHz crystal filter

Narrow view of inductance compensated filter BW (LTSpice simulation)

Wide bandwidth look at inductance compensated filter (LTSpice simulation)

Narrow look at filter without compensation (note LSB shape) (LTSpice simulation)

Wide view of filter without compensation-- good wideband attenuation (LTSpice simulation)

DDS “VFO” • Above or below 9 MHz to produce 5 MHz? – Above allows feedthru out of mixer to be filtered by LPF. But it inverts the sideband. • Output of NJQRP AD 9850 is adequate for NE 602 requirements with margin. • Discrete channels of 60 meter band simplifies user interface to 1 button & annunciation. • PIC 16 F 84 is more than up to the task.

DDS VFO with PIC controller

Transmit mixer: NE 602 • 6 VDC supply • R-pad matches 200 ohm filter to 1 k 5 input & gives 15 d. B attn of 9 MHz SSB to pin 1 • From DDS, resistive divider drops DDS output to about 630 m. Vp-p @ pin 6 • Transformer output for balanced 3, 000 output (pin 4, 5) to 200 filter • 4 -element LPF at output

Transmit mixer using NE 602

NE 602 3 rd order intercept plot from data sheet

Two class-A amplifier stages • Post-mixer amplification needed to get from -11 d. Bm to +20 d. Bm power level • Class-A used where possible for lowest distortion, plus easy & predictable • TN 2219 A first stage & RCA 4013 second • EMRFD figure 2. 57 primary resource; added 4: 1 xfmr at input to match 200 • Standing current 20 m. A 1 st; 40 m. A 2 nd.

1 st post-mixer amplifier stage: 20 d. B gain to +6 d. Bm

2 nd post-mixer amplifier stage: 15 d. B gain to +20 d. Bm

1 -watt push-pull driver • Class AB push-pull 2 N 3866 • Temperature compensated bias control • Classic design with these wrinkles: – Emitter degeneration, AC & DC – No shunt or xfmr feedback – Input swamping resistors AC coupled • Gain over 20 d. B; added some input attenuation.

1 -watt push-pull driver stage

Push-pull 10 w final amp / filter • Classic push-pull topology – Temperature compensated amplified bias – Balun binocular cores, brass tube primary winding on output. – Emitter degeneration; no shunt feedback • 5 -element output filter

Push-pull 10 watt final amplifier circuit

Output filter and bias circuit for final amplifier

Completed transmitter --- but where is it? In ten separate pieces ; nine boxes