Rock-bound No Longer!

by Wayne McFee, NB6M

About the AmQRP Club
Frequently Asked Questions
QRP Forums

SMK VFO - Rock-bound No Longer!

I have had so much fun with the little SMK-1 that I couldn’t help thinking how much more fun I might have with it if only it had a VFO. I thought “why not?” So, I got started thinking about what all I would have to change or add in order to couple a VFO to the receiver and transmitter circuits in the SMK-1.


First of all, I would have to remove both VXO crystals and their associated circuitry. Then, I would have to modify the transmitter’s oscillator circuit so that it would act as an amplifier instead of as an oscillator. Not only would I need to build a stable VFO, with sufficient tuning range to cover most of the CW portion of the band, I would need to provide for transmitter frequency offset, and, since the receiver would no longer have its own LO, a means for providing sidetone would have to be added to the rig.

After a little thought, and some experimental work, the result is a very nice VFO, with RIT and transmitter frequency offset a la Roy Lewallen, W7EL, a tuning range of from just below 7.000 MHz to 7.115 MHz, NO expensive tuning capacitors or even tuning diodes to buy, and only one tuning coil and two toroids to wind. The rest of the VFO, including the RIT circuit, is all built with cheap, easy to come by (Radio Shack) parts.

In perusing the Internet for any and all QRP related info, I had seen a circuit on WA6OTP’s website for a Permeability Tuned Oscillator which used a small diameter, air wound coil, with a 6-32 brass screw threaded into the center, as a tuning element. That got me really excited. That tuning element was just what I was looking for. The tuning coil was small, would cost almost nothing to make, and would provide not only the inherent stability of an air-wound coil, but, with the brass tuning screw, would provide plenty of tuning range. Also, with the use of brass as the varying element, the VFO frequency would increase, as the tuning knob was turned clockwise, just like radios “normally” tune. Nice.

As it turns out, at the low end of the tuning range, one full turn of the tuning knob moves the VFO output about 8 KHz. At the upper end of the tuning range, one full turn of the tuning knob moves the VFO output about 13 KHz. Not quite linear, but not bad at all.

Considering the fact that I wanted to eventually be able to build the VFO right inside the small confines of my SMK-1 case, which puts space at a premium and almost precludes adding a lot of shielding around the VFO oscillator, I wanted to put the VFO on 3.5 MHz and employ a frequency doubler circuit in order to arrive at 7 MHz. This would minimize any inter-action between the transmitter signal and the VFO, preventing any pulling or chirping.

Since removing the VXO circuitry for both the receiver and transmitter would mean that I now had a couple of spare 10K pots, I thought why not remove the Receiver’s VXO pot completely and use that position as the eventual location for the tuning coil, and why not use the Transmitter’s old VXO pot as an RIT control, since by providing an RIT circuit, a la W7EL, I would not only have transmitter offset but would enjoy RIT as well.

The RIT circuit, with the values used, provides about 1.5 KHz of tuning range. Just about right for a 600 to 800 cycle tone with the RIT control centered in its range and with the transmitter tuned for zero beat on an incoming signal. The “Spot” switch makes tuning signals really easy. Simply hold the Spot switch in, tune the VFO for zero beat on the incoming signal, release the Spot switch, and adjust the RIT control for the desired tone pitch.

As for providing sidetone, a couple of routes came to mind. A sidetone oscillator circuit could be built in, but since I really like and enjoy the little Tick Keyers, I thought why not put in a Tick chip and keying transistor, and run the audio from the Tick to the audio chain as sidetone. This route has been used successfully in several rigs, including the Tixie, and works very well. The choice is yours.


Now that I had a plan, the work could begin. As a very fitting start, at a NorCal meeting at California Burger, in Pleasanton, I kept the drinking straw from my soft drink to use as my coil form. It was a nice, substantial one of 5/16” diameter. WA6OTP had used old potentiometer housings as a building platform for his coils, but I elected to do it an even easier way. Using an electric grinder, I rounded the corners off a brass 6-32 nut, rounding it to a size that the drinking straw would just fit over.

Then, after drilling a clearance hole in the front panel of the VFO, which is double sided PC board, I threaded the rounded nut onto a brass 6-32 screw, set the screw in the hole, with the rounded nut on the inside surface of the front panel, and soldered the rounded brass nut to the inside surface.

Then, threading a new 6-32 brass nut onto the screw on the outside of the front panel, I tightened it down just enough to provide added thread support, but not tight enough to prevent easy turning of the tuning screw. Once that was done, I soldered the outside brass nut to the surface of the front panel.

While allowing the assembly to cool, I used a sewing needle to poke small holes close to one end of the drinking straw in order to anchor the end of the # 30 magnet wire used in the tuning coil, and wound 95 turns, close-wound, onto the 5/16” diameter drinking straw. Although the coil needs about three less turns than that in its final version, the extra turns provide plenty of room for error. It is quite easy, after the entire VFO circuit is finished, to remove coils one at a time to make the needed frequency adjustment.

The sewing needle was again used to poke small holes in the straw, next to the last turn, to anchor the end of the coil in place. The coil form (drinking straw) is easily trimmed, close to the anchored last turn, with scissors. Once the coil form is trimmed, use 5 minute epoxy to fasten one end of the coil form over the rounded brass nut, up against the inside surface of the VFO front panel. Use the tuning screw itself, a 6-32 brass screw about 1 ¼” long, as a guide to ensure that you have the coil on straight. The head is cut off the screw to facilitate installing a small tuning knob, and mine ended up being just over an inch long, including the portion that is inside the knob. Leave yours 1 ¼” long to begin with, and cut it as you like to give the upper frequency limit you wish.

In my experimental version, all the VFO, RIT and Keyer stages are built “Ugly” style, in the manner of Wes and Roger Hayward, W7ZOI and KA7EXM. The final version, which will be installed in my SMK-1 with the 5 Watt Mod, will probably be done in “Ugly” style as well, but with a great deal more thought given to parts placement and circuit miniaturization.

Since this VFO would do very well as a stand-alone module which could be used with your choice of QRP rig, I would suggest that you give some thought to building your version in a case of its own, particularly if you have put your SMK-1 into a minimally sized case. The VFO box would need to be supplied with 12 Volts (or could be battery supplied). A wire would be needed from the VFO to the rig’s keying line to provide for RIT (and transmitter frequency offset) disabling during transmissions, and a small coaxial cable would run the VFO output to the rig. As long as the rig had it’s own sidetone, no more cabling would be needed.

The two 22 pf caps shown on the VFO output would be installed inside the rig itself, one at the input of the transmitter amplifier and one at the receiver mixer. In that fashion, only a single VFO RF Output cable would be needed to connect to the rig.

In order to make the building process easier, I have provided additional circuit diagrams that break the main circuit diagram down into four parts, detailing the Oscillator, RIT circuit, First Buffer Amplifier and Frequency Doubler, and Tuned Buffer Amplifier. I have separated the parts list into four sections in the same manner, so that you can gather the parts for each stage individually. The building style is up to you. “Ugly Style” works very well. “Manhattan Style” would also be a good choice. You could even design a printed circuit board, if you wish. Mine is built “Ugly Style” and works very well.

I suggest that you build the oscillator circuit first. DO NOT coat the entire coil with epoxy or other fixative yet, as a few turns will have to be taken off in order to fine-tune the frequency range. Be mindful of the fact that the 4.7 K biasing resistor, the top lead of the primary of T-1 and the 47 Ohm resistor for the voltage supply to the Tuned Buffer Amp will all have to attach to the junction of the 100 Ohm resistor, 220 Ohm resistor, .01 Cap and 100 UF Cap, so plan accordingly in your layout. Also, NOTE that all the small value capacitors (values in pf) are NP0, Poly, or Silver Mica. Good chance to use some of those NorCal capacitors, hi.

Once the oscillator is built, apply 12 Volts and check for oscillation by listening for the output in a general coverage receiver, or by measuring the RF level at the Drain of the MPF-102 (about 3-4 volts RMS), or by looking at the waveform with an oscilloscope.

You will probably have to listen well below the 3.5 MHz Fundamental Frequency initially. At this point we are just checking for oscillation, so don’t make any frequency adjustments to the coil until after the amplifier and frequency doubler stages are all completed, as the loading effect of these stages will affect the frequency of the oscillator.

Once you have confirmed oscillation, build the next portion, the RIT circuitry. The RIT disabling transistor, Q-2, and its associated parts can be mounted on the RIT pot itself, if desired, which helps make for a neater installation. Also, you may want to hook up the RIT pot only temporarily at first, to ensure that you get the correct rotation. You want the RIT pot to tune the same direction as the main tuning knob, up in frequency with clockwise movement. The “Spot” Switch, a “Normally Open” push-button type switch, is mounted in a convenient spot on the front panel of the VFO. Test both the frequency offset function, by pushing the Spot Switch, and the RIT function, by turning the RIT control and listening to the output in your general coverage receiver. Remember that when you are listening to the 3.5 MHz fundamental frequency, the tuning range will only be about 750 Hz.

Once the RIT section is completed and tested, build the First Buffer Amplifier and Frequency Doubler. T-1 is a tri-filar winding of # 26 or # 28 magnet wire on a T37-61 core. Be sure to observe correct polarity when making its connections to Q-3 and the frequency doubler diodes.

Also, it will be important to measure the forward resistance of the diodes used, selecting two with the closest resistance measurements so as to balance the doubler, which will help ensure that the lowest amount of fundamental frequency energy possible will be felt at the VFO output.

If you have an oscilloscope, you can look at the waveform on both the input of the First Buffer Amplifier and the output of the Frequency Doubler to check their function. Alternatively, measure the RF output on the Collector of the First Buffer Amplifier, and then again on the output of the Frequency Doubler.

On mine, the approximate RF voltage readings are as follows:

Drain of Oscillator 4 Volts RMS
Base of First Buffer Amp .8 Volts RMS
Collector of First Buffer Amp 2.2 Volts RMS
Output of Frequency Doubler .48 Volts RMS

When the First Buffer Amplifier and Frequency Doubler section is finished and tested, build the Tuned Buffer Amplifier section. T-2 has a primary of 19 turns # 24 on a T50-2. The secondary is a 3-turn link spread over the center area of the primary.

A T37-2 could probably be used, to reduce the size, but a few more turns will be needed. A 21-turn primary would be a good start, but the 240 pf capacitor might have to be changed to 220 pf to get it to resonate. Remember that the 240 pf cap is either a NP0, Poly, or Silver Mica.

When the Tuned Buffer Amplifier section is built, now is the time to make frequency adjustments to the tuning coil. Hook an RF probe onto the top of the 3-turn link on T-2 and adjust the trimmer to see if the circuit will resonate at the second harmonic of the oscillator frequency, adjusting for highest output.

Alternatively, you can listen for the second harmonic on your general coverage receiver and adjust the trimmer for loudest signal. Although mine resonated both at the second harmonic of 6.880 MHz initially, and at the final second harmonic of 7.050, don’t be too upset at this point if your trimmer won’t bring the tuned amp to resonance at your initial, lower second harmonic. Just adjust for the most output you can get at this point.

To make the frequency range adjustment on the tuning coil, start by removing the tuning screw completely. This will ensure that you are at the lowest frequency possible to tune. Remove supply voltage from the circuit. Then, remove one turn from the coil, trim, scrape and tin the wire, and re-solder it into the circuit. Use your general coverage receiver, or a frequency counter if you have one, to check the new low-frequency end of the tuning range. NOTE how much the frequency changed when one turn was removed from the coil. If you want full coverage of the low end of the band, as I did, once you get the new low-end frequency within 20 KHz of 7.000 MHz, that is probably close enough.

However, it is your responsibility to ensure that you don’t transmit outside the band. The safest way would be to set the low frequency limit up just above 7.000 MHz, removing one turn at a time until you have the frequency set. Once the coil is adjusted for a low frequency limit inside the band, small amounts of capacitance (2 pf, for example) can be added from the Drain of the oscillator to ground to make fine adjustments downward, if needed. Remember, it is much more difficult to add turns, once they are cut off, so if the lower frequency is close, and you have tuning range to suit you, leave well enough alone. However, if you do make a mistake and cut one or two turns too many, don’t despair. Simply add a small amount of capacitance, say 2 pf, 4.7 pf, 10 pf, or 12 pf, NP0, from the Drain of the oscillator to ground, and that will bring the frequency back down.

Now that the oscillator is on the right frequency range, coat the winding of the tuning coil to fix the turns solidly in place. This is very necessary, in order to provide the best stability from the oscillator and VFO output. I used 5-minute epoxy, and coated the entire surface of the winding.

Also, now that the VFO output is in the right frequency range, the trimmer on the output of the Tuned Buffer Amplifier will have to be re-adjusted. Tune the VFO to the center of your tuning range, and either use an RF probe or an oscilloscope to monitor the adjustment of the trimmer. Trim for best output and/or the cleanest waveform. The RF voltage reading at the top of the 3-turn secondary of T-2 is approximately .78 Volts RMS.

During my initial on-the-air trials of the VFO and SMK-1 combination, just to double check, I had a nearby ham listen for the fundamental frequency while I transmitted, and received a clean, nothing heard, report. Should you end up with too much fundamental frequency in your VFO output, it would indicate that the Frequency Doubler diodes were not matched initially, or could have been overheated when being installed, which resulted in their forward resistance characteristics changing.


Dual soldering irons are a must for this project. With those, the most interesting part of the exercise will be removing the removed part from whichever iron it sticks to. I recommend a damp pad of cloth to wipe the part off on.

As indicated, I removed both VXO crystals and their associated circuitry from the SMK board. This included X-1, VR-2, C-15, R-2, L-3, C-4, C-5, R-3, D-3 and D-4 in the receiver, and X-2, R-7, D-6, D-7, and C-16 in the transmitter. All the parts are taped down to a sheet of paper, and their values written next to them, one at a time as they are removed. Then the parts, with their identifying info, are appropriately filed away for later use.

If you have enough room to build the VFO inside your SMK case, cut the PC Board 12 Volt trace that goes to the VR-3 connection closest to the right hand edge of the board, cut the opposite, left hand VR-3 contact itself that goes to ground and unsolder and remove the cut off portion of the VR-3 contact.

The 12 volt trace going to the + side of VR-3 is cut just about even with the ground end of D-3. That leaves 12 volts going to the pad for C-15, which will allow us to pick up 12 volts for the base circuit of the 2N3906 RIT disabling transistor there, and will keep the trace going from the + side of VR-3 to the + side of VR-2’s connections going, so that we can put the 680 resistor for the RIT pot there, and run a wire from the 6 Volt side of C-6 to the R-2 pad closest to X-1’s location in order to supply the RIT circuit with 6 Volts regulated.

The 4.7 K Ohm resistor in the RIT circuit is run from the cut off contact on the VR-2 side of VR-3 to ground. The 2N3906 is soldered directly to the wiper and 6 Volt contact of VR-3, with the emitter going to the 6-volt contact and the collector going to the wiper. A wire also connects to the wiper to run to the junction of the 47 K Ohm resistor, .01 cap and the cathode of the 15 Volt Zener, which are all located on the oscillator board, close to the Drain of the MPF102. The 100 K Ohm resistor in the base circuit of the 2N3906 runs from the + pad of C-15’s location straight up, and the 10 K Ohm resistor connects to the top of the 10 K resistor and goes to the base of the 2N3906. The .01 cap is connected at the junction of the 10 K and 100 K resistors, and goes to the ground pad for D-4.

The 1N914 goes from the junction of the 10 K, 100 K and .01 to the key line side of R-16, with the cathode towards R-16. The spot switch is connected from the junction of the 10 K, 100 K, .01 and 1N914 to ground and is located in a convenient spot on the front panel.

That places all the RIT control and switching circuitry on the SMK board, at or near VR-3’s location, leaving fewer components on the VFO board. The main VFO board could then be integrated into the top and left side of the SMK case, leaving the right side for the 1 Watt or 5 Watt Mods, if desired, or could be stacked below the main SMK board, with the VFO board near the SMK board and the components of the VFO board facing away from the SMK board and the tuning screw protruding through the front of the case.

If you are going to build the VFO as an outboard item, VR-3 is removed. An easy way to remove VR-2 and VR-3 is to cut their contacts at the top surface of the SMK-1 PC Board. This still leaves enough contact remaining to solder to and use one of them as an RIT control.

If you are planning on making your VFO an outboard item, the two 22 pf caps shown on the output of the VFO are installed on the SMK-1 board. With all the VXO related parts removed from the board, the two 22 pf caps can go quite nicely in the X-1 and C-16 locations. A wire soldered to the L-3 pad closest to the X-1 location, the R-7 pad closest to Q-2, and running to a jack on the back of the rig, for VFO input, would then complete the VFO-to-Rig RF connection inside the SMK-1.

In order to change the transmitter oscillator on the SMK-l board into an amplifier, you will need to remove R-8 and C-18 from the circuit. Then, move R-9, the 4.7 K Ohm resistor, to R-8’s old position (changing R-8’s value from 47 K to 4.7 K) and install a leaded 1500 Ohm resistor in R-9’s place.

How you provide for a keying line for the Frequency Offset/RIT circuit will depend on whether you choose to build a keyer into your rig or not. If you do, then you will have to provide a jack and cable for that line to the VFO. If you use an outboard keyer and an internal sidetone oscillator, one way to provide the keying line to the VFO would be to use a one plug to two jack adaptor in the key jack on the back of the rig, and plug both the keyer and the key line from the VFO into that. Those decisions are for you to make.

Also, if you elect to install a Tick Keyer in order to provide sidetone (not to mention the benefits of a keyer, apply the audio, using the circuit components described in the Tick circuit diagram provided with the chip, to the input of U-2 at pin 3, and make resistor value adjustments as necessary to adjust the sidetone level to your liking.


Yes, I know this project is a bit more complicated than adding the 5 Watt Mod to the SMK-1. However, it is a very worthwhile project that can be used with a variety of small, formerly rock-bound, QRP rigs, and, taken in stages, is not difficult at all to build. Just imagine being able to tune the whole CW portion of the band, with RIT thrown in for good measure. And, with a 1.5” long tuning screw, the VFO will tune the entire 40 Meter band, allowing you to listen to your favorite 40 Meter SSB activity as well as doing your regular CW operation on the lower part of the band.

I have worked Japan and other far away places on the low end of 40 Meters with one watt out of other QRP rigs, at sunrise in the morning, and I am eager to try it with my VFO-controlled SMK-1 with its stock 300 MW output.

Should you wish to use the VFO for another band, the oscillator, with the 120 pf cap removed, and with the number of coil turns adjusted downward appropriately, is quite stable certainly up past a 7 MHz fundamental frequency, which would give you 30 and 20 meter capability. I suspect that, with care, it might even be stable enough for a 10.5 or 14 MHz fundamental, which would provide for operation through 10 Meters. It would certainly be worth experimenting with.


Wayne NB6M




Q-1	MPF102, 2N3829, 2N4416, etc				1
D-1	6.2 V, 1 W Zener					1
	100 Ohm						1		
	220 Ohm						1
	100 K Ohm					1
	270 pf NP0					1
	120 pf NP0					2
	150 pf NP0					1
	.01 Disc Ceramic					2
	100 UF 15 Volt Electrolytic				1
	22 uh RF Choke					1
	Home-Made Tuning Coil				1	

Q-2	2N3906						1
D-2	15 V, 1 W Zener					1
D-3	1N914, 1N4148, etc					1
SW-1	NO miniature push-button				1
RIT Ctl	10K Linear Pot					1
	100 K Ohm					1
	10 K Ohm						1
	560 Ohm						1
	4.7 K Ohm 					1	
	1 M Ohm						1
	47 K Ohm						1
	.1 Disc Ceramic					1
	.01 Disc Ceramic					1
	4.7 pf NP0						1

Q-3	2N2222, 2N3904, 2N4401, etc				1
D-4, D-5	1N914, 1N4148, etc					2
 T-1	10 Tri-Filar Turns # 26 or # 28 on T37-61			1
	4.7 K Ohm						1
	470 Ohm						1
	100 Ohm						1
	22 pf NP0						1
	.01 Disc Ceramic					1
	22 uh RF Choke					1


Q-4	2N2222, 2N3904, 2N4401, etc				1
T-2	Primary, 19 Turns # 24 on t50-2				1
	Secondary, 3 Turn Link, spread on
	Center of Primary
	47 Ohm						1
	6.8 K Ohm						1
	470 Ohm						1
	100 Ohm						2
	.01 Disc Ceramic					2
	240 pf NP0					1
	60 pf Trimmer Capacitor				1
	22 pf NP0						2


Material and concepts presented on The American QRP Club (TM) website is Copyright 2003 by The American QRP Club, Inc.
These pages are designed and maintained by George Heron, N2APB
Page last updated:  April 15, 2004