NorCal QRP Club - NB6M SMK-1 Modifications

In the photo above, some of the changes made in order to make the SMK-1 a viable 20 Meter transceiver can be easily seen.

A One Watt Mod has been added above the PA section of the main board. A 10 DB RF preamp for the receiver is built on the inside of the left panel of the PC board case. The circuit on the underside of the lid is an active audio filter. Difficult to see in this photo is the surface mount Tick keyer chip that is epoxied to the underside of the stereo paddle jack.

In addition to the RF Gain, Rx Tune and Tx Tune pots on the front panel are the toggle switch on the left, for switching the audio filter in and out of the receiver, and a push-button switch on the right for programming the keyer.

A pair of holes was drilled in the right side panel, even with the bottom of the new PA transistor, and two holes drilled in the lid, to allow for airflow around the 1 Watt PA.

I had wanted for some time to give the little SMK-1 a try on 20 Meters, and having experimented with and implemented several "mods" that have improved the usefulness of the little rig, I decided it was way past time to see what it would do on the "Big Boys’" band.

Sure, the SMK-1 has long been "out of print", so to speak, as the kit has been unavailable for some time. And, its simple transmitter and receiver circuitry cannot compare to a full-featured rig. But, it is very surprising how well one can do in making contacts with stations both near and far with this little rig, and doing that with something as simple as this is much more exciting and rewarding to all of us in QRP.

Since the SMK-1 is nothing more than a combo of the Tuna Tin II transmitter and MRX40 receiver, and those seem to be timeless in terms of their ability to provide a lot of fun for very little effort and cost, and since there are a lot of SMK-1s out there that may not have been on the air in a while, perhaps a band change or update to a more satisfying frequency range, better receiver sensitivity, or higher transmitter power output would make operating that little rig even more exciting than when it was first offered as a kit.

My objectives in putting the SMK-1 on 20 Meters were to achieve a useful frequency range in both the transmitter and receiver, hopefully the same or near the same in each, that covered the 14.060 Mhz QRP calling frequency; as stable a transmitter signal as possible with enough output to guarantee many contacts; enough sensitivity in the receiver to hear QRP signals well; and to make the band change by using as many of the parts that were shipped with the kit as possible. These goals were met.

Constructed with those guidelines, the 20 Meter SMK-1, using single crystals in the transmitter and receiver VXOs, covers approximately 14053.5 Khz to 14061.5 Khz. Signal reports on the air have been very favorable in terms of stability, and the transmitter, with a hot 2SC799 in the "easy one Watt Mod", puts out almost two Watts.

To help stabilize both the transmitter and receiver oscillators, and maintain the low parts count, 6 Volts regulated was routed from the on-board 78L06 IC to the tuning circuits. In order to achieve a transmitter tuning range equal to the receiver's, the transmitter oscillator was changed to the configuration outlined in my previous work on extending the transmitter's tuning range. It is necessary to cut two PC board traces in order to accomplish the change.

A 10 DB RF preamp was incorporated into the receiver in order to boost its sensitivity to an acceptable level on 20 Meters, while still retaining the same tuned input circuit and diode T/R switching so as to keep the operation of the little rig smooth and simple.

An added benefit of using the already available 6 Volts Regulated to supply the tuning circuits is that only one additional PC board trace needs to be cut, and one extra wire is run from the top of C15 to pin 1 of the 78L06 IC. A comparison of the tuning range in my unit between using 6 volts or 9 volts on the tuning circuits gave surprising results. Very little difference was noted.

A wider tuning range can probably be realized by using larger value inductors in the two oscillator circuits than the 4.7 uH units that I used in this first conversion. I had two surface mount inductors of this value available, even though there is only one in each kit, because of my modifications to other SMK-1s. Further increases in the inductance values would certainly increase the tuning range, but may pull the upper frequency limit below the desired coverage of the 14060 Khz QRP calling frequency.

Although I used all Silver Mica capacitors in the transmitter's output filter, a variety of high quality, 100 Volt rated capacitors can be used with good success. A prime example of why good quality capacitors must be used in the output filter is that in my unit, when I was using lesser quality caps in the output filter while experimenting with it, and while using the 6 Volts regulated to supply the tuning circuits, I experienced a small amount of transmitter chirp, and at first thought that I was going to be forced to use a better regulator than the 78L06, because when I switched to a 9 volt battery to supply the tuning circuits, the chirp disappeared. Additionally, the transmitter output was lower than normal. However, when I went to all Silver Mica caps in the output filter, not only did the chirp disappear but the output came up to where it should be.

This seems to be a reminder that tuning circuit voltage stability and transmitter PA impedance matching and stability work hand in hand in this rig in affecting the transmitted signal note. I am sure that PA heating also has a negative effect on stability.

As mentioned in my prior work on putting the SMK-1 on 80 or 160 meters, converting the rig to another band requires just a few parts changes in the transmitter's output filter network, the receivers input tuning circuits, the crystals and perhaps the feedback capacitor values in both oscillator circuits. Because of the fact that I did not have on hand any surface mount capacitors of the required values to replace the capacitors in the transmitter's output circuit, and wanted to put a toroid coil in place of L5 in order to support the higher output of later modifications, I used all full sized parts in the transmitter output network.

Full sized parts were also used wherever needed in making the band conversion, extended VXO mod, R14 in the PA emitter circuit, now 2.2 Ohms, and in the entire receiver RF preamp, which I built on the inner copper surface of the left side of the PC board case.

The eight parts needed for the RF preamp are:

34T on T37-2, tapped at 4T 1
.1 bypass cap 1
100 Ohm 1/4W resistor 2
MPF102 or equivalent FET 1
34T on T37-2, with 3T secondary 1
15-50 (or similar value) Trim Cap 1
.01 bypass cap 1

20 Meter SMK-1 Receiver RF Preamp and connections

Here is a picture of the RF Preamp and the receiver mixer portion of the board

A stage-by-stage description of the conversion procedure follows:

Make the following changes in the Receiver input circuit:

L1, remove and save for later use.

Build the receiver RF preamp circuit next. It can be built on any available surface of a PC board case, or on a small piece of PC board attached anywhere, if another type of case is used. "Manhattan" style is quick and easy.

Solder a short piece of insulated hookup wire between the L1 pad closest to R20 and the top of the input coil of the Receiver RF Preamp.

Solder a short piece of insulated hookup wire between the top of the output link of the Receiver RF Preamp and the L1 pad closest to the edge of the board.

C1, remove, save for future use, and replace with a 120 pf NP0 cap.

C2, remove, save for future use. Remove C24, 390 pf, and install it at C2.

L2, remove and save for future use. Remove L5, 1 uH, and install it at L2.

Make the following change in the mute circuit:

R5, across Q2, remove and do not replace. Save the 2 Meg removed for future use.

20 Meter SMK-1 Receiver Mixer and Local Oscillator changes

Make the following changes in the Receiver LO circuit:

C4, remove, save for future use, replace with a 47 pf NP0 cap.
X1, remove, save for future use, replace with a 14.060 Mhz crystal.
L3, remove, save for future use, replace with 34T on T37-2, or a 4.7 uH inductor.

20 Meter SMK-1 Transmitter Output Network changes

Make the following changes in the Transmitter output filter:

C24, removed in a previous step, replace now with a 200 pf, Silver Mica or equivalent.
C25, remove, save for future use.
C26, remove, save for future use. If you do not care to put the easy one-watt mod into your rig, replace with a 240 pf, Silver Mica or equivalent.
If you want to add the one Watt Mod, I suggest adding another section of output filter, which will involve the addition of one more T37-6 toroid with 13 turns, and a 470 pf cap will be used to replace C26.
Because I added the second PI section of output filter, I cut the trace between the Ant output connection and C25, ran a short wire from the Ant output connection to the junction of C25 and C22, and connected the added T37-6 toroid between the rear pad of C26 and the top of the added 240 pf capacitor.
This was done in order to move the receiver’s RF pickoff point to the input of the TX output filter, so as to make better use of its low pass filter characteristics.
If you choose not to do the One Watt Mod, the extra toroid coil is not needed, and the 240 pf capacitor will be used to replace C26.
L5, remove, replace with 13T on T37-6.

VXO Changes:

20 Meter Transmitter VXO with extended VXO mod
Make the following changes in the Transmitter VXO circuit:

Remove R9, save for future use and cut the PC board trace between the rear pad for the removed R9 and X2, right at the R9 lettering.
Cut the PC board trace between X2 and C20, between the junction of the trace that goes to L4's pad and C20. We want the trace to still connect from X2 to L4's pad, but not to C20 or the collector of Q2.
C18, remove, save to replace C20.
C17, remove and save for future use. It will be replaced with a 33 pf NP0 cap after the next four steps.
C20, remove, save for future use, replace with the 100 pf removed from C18, but installed at a 45 degree angle, between the left pad for C17 and the rear pad for C20, as viewed from the front of the rig.
R10, remove and save for later use. It will be replaced after the next step by 1 K Ľ watt resistor.
Add a 47 pf NP0 cap between the base of Q2 and the left pad (closest to Q2) for R10. The base of Q2 is the contact of Q2 closest to X2, on the side of the transistor that has two contacts. This capacitor, along with the 33 pf cap that replaces C17 provide the feedback necessary for Q2 to oscillate.
Install a 1 K Ľ watt resistor in R10’s spot.
Install a 33 pf NP0 cap at C17.
L4, remove and save for future use.
R8, remove, save for future use, and replace with a 100 K Ohm, 1/4 watt resistor.
C16, remove and save for future use. Replace with 34T on T37-2 or a 4.7 uH inductor.
Solder a short piece of insulated hookup wire between the L4 pad nearest C19 and the C20 pad nearest Q2.
Solder a short piece of insulated hookup wire between the L4 pad nearest C20 and the R9 pad nearest X2.
X2, remove, save for future use, and replace with a 14.060 crystal.
You will note that the extra 9.1 Volt Zener and associated parts added to the VXO oscillator circuit in the original Extended VXO Mod are not used here, as it was determined that they did not add to the stability of the circuit.

Make the following changes to the tuning circuitry:

Cut the PC board 12 Volt trace between the printed copyright "C" and C15.
Solder a short piece of either insulated hookup wire or cut off part lead between the end of C15 closest to R8 and pin 1 of the 78L06, U1.

Here is the circuit for the One Watt Mod:

If you care to add the easy One Watt Mod, Gather the following parts, plus the 13T T37-6 and the 470 pf cap mentioned earlier in the changes to the TX output circuit:

two - .1 uf disc or monolithic, etc (bypass cap)
two - 100 ohm resistors
one - RF transistor of your choice
one - roughly 10 uh, RF choke, FT37-43 toroid with 5 turns # 24 (or similar gauge)
one - .001 bypass cap

Here is a picture of the One Watt PA and output filter
Some of the changes in the VXO section can be seen as well:

I did the mod "ugly" style with all leaded parts, soldering the necessary leads to the pads on the board.
Here is what you do:

I suggest you read all the instructions first, to get an understanding of the mod, before proceeding with the actual work.
First, remove C22 from the board. (I used two soldering irons, quick and easy)(save it)
Cut one lead of one of the .1 caps so as to leave about Ľ", bend a 90 degree angle about 1/8" from the end of that lead, and solder that lead to the C22 pad closest to the edge of the board. Leave the other lead long for a moment.
Cut one lead of the other .1 cap so as to leave about Ľ", bend a 90 degree angle about 1/8" from the end of that lead, and solder that lead to the C22 pad closest to Q3. Leave the other lead long for a moment.
Cut one lead of a 100 Ohm resistor so as to leave about 1/4", bend a 90 degree angle at 1/8" from the end of the short lead, and solder that lead to the ground pad of R14.
The idea is to have the resistor standing just about straight up, maybe leaning just a little bit towards Q3. By the way, the ground end of R14 is the pad right at the edge of the board.
Now, using short leads on all three parts, solder the other 100 Ohm resistor between the free lead of the first 100 Ohm resistor and the free lead of the .1 Cap that comes from the C 22 pad closest to Q3.
What we are doing here is providing a coupling capacitor from the output of T1 to the base of the Final Amp we are going to install. The two 100 Ohm resistors form a divider network so as to provide about 2 Volts RF on the base of the Final Amp. The 4 Volts or so right from the output of T1 is way too much drive for the Final Amp and output network we are using, and if we drove it with that much RF, it would overheat and probably self destruct in a short while.
So, at this point you should have a .1 from the C22 pad closest to Q3, connected to a 100 Ohm resistor, which is connected to another 100 Ohm resistor, which is connected to ground.
Position your RF transistor above the existing PA transistor, with the emitter lead towards the edge of the board, the base lead towards the rear of the board, and the collector lead towards the center of the board.
We want the emitter lead to be as short as possible, but the other leads, the body of the transistor and any heat sink used will need to clear the other parts. Remember, the body of the transistor, and therefore the heat sink as well, are connected to the collector, and have both 12 Volts and RF on them. We can’t allow the heat sink or transistor body to come in contact with any other parts or with ground. With this in mind, take note of how long the emitter lead will need to be in order to solder it to ground at the ground side of R14. Cut the lead to that length.
Bend the base and collector leads out to their respective sides of the transistor, close to the transistor body. Leaving the emitter lead pointing pretty much straight down, only slightly bent out towards its edge of the transistor body, bend an eighth of an inch of the lead over at a 90 degree angle, and solder the emitter lead to ground at the pads of R13 and C18 closest to the edge of the board.
Cut the base lead so that it will reach the junction of the two 100 Ohm resistors, and solder it there.
Wind 5 turns of #24 (you could use small insulated wire for this if you don't like scraping the ends of magnet wire) on the FT37-43 core. Scrape about 1/8" of wire clean of insulation at the end of each coil lead.
(This is meant to be about a 10 uh RF Choke. If you don't have an FT37-43, you could use any combination of core and number of turns that gives you roughly that figure. Suggestions are: 35 turns # 22 on T68-2, or 43 turns # 28 on T50-2, to name two.)(Or use a 10 uh RF choke)
The RF choke for the new final amp sits right above R15 and R19 on the board. It should be oriented so that the toroid is at a 90 degree angle from T1 in order to minimize coupling between the two. The two leads from the RF choke should be about 3/8" long.
Before you install the RF choke, solder the .001 bypass cap in place between the 12 volt side of R15 and the ground side of C19. The proper pads of these parts are those that are closest to the front of the PC board.
Then, solder one lead of the RF choke to the 12 Volt side of R15 (front pad). This is just a convenient spot to pick up 12 volts.
Solder the other lead of the RF choke to the Collector of your RF transistor. Check the lead lengths of the two and adjust accordingly before soldering them together.
Now, the free lead of the the .1 uf Cap already soldered to the C22 pad closest to the edge of the board is soldered to the collector of your RF transistor. Check for appropriate lead length, cut the lead, and solder it in place.
You should place a heat sink on your new final RF Amp transistor.
I added a second PI section to the output filter, by replacing C26 with a 470 pf cap, glued a Manhattan style pad (with 5 Minute Epoxy) to a convenient surface of the rear of the PC board case, unsoldered the antenna lead from the center of the antenna jack on the case, soldered that lead to the Manhattan style pad, soldering one lead of a 13T, T37-6 toroid to the Manhattan style pad, the other end of that toroid to one lead of a 240 pf cap, the other lead of that cap to ground, and connected the junction of the 240 pf cap and the added toroid to the antenna connector on the case of the rig.
The amount of transmitter output will vary, depending on the type of RF transistor used. I got .85 watts out with a 2N3053, and with a hot 2SC799 got almost 2 Watts out. Your mileage will vary.

Tick Keyer

A surface mount Tick Keyer IC was epoxied to the underside of the paddle jack. Small gauge stranded hookup wire was soldered directly to the IC pins in order to connect it to the paddle jack and supply voltage circuit. The keying transistor was mounted right in the PC board holes for the "key" connections and the rest of the keyer’s circuitry was built, Manhattan style, on the inside surface of the rear of the case.

A 1 K resistor and .1 cap in series route the sidetone audio from the Tick keyer to the earphone jack.

The Tick installation can be partially seen in this photo:

A 1N914 was placed in series with 6 Volts regulated taken from U3 in order to drop the voltage to 5.1 Volts to supply the Tick chip. A 10 K resistor was run from the cathode of the 1N914 to ground in order to stabilize the supply voltage at 5.1 Volts during the keyer’s "sleep" periods. The voltage dropping circuit can be seen on the inside of the front panel, in this photo:

The drawing below shows the schematic for the Active Audio Filter and for the Tick Keyer, and the necessary connections to the rig. The Active Audio Filter is switched in and out of the circuit by means of a DPDT toggle switch.

Tuneup:

When the rig is powered up again, TC1, the trimmer in the RF Preamp output, and TC2 will need to be retuned for highest received signal level. With the values given, very definite peaks should be had on each trimmer. They may need to be retouched a couple of times to ensure they are properly tuned.

Summary:

The drawings included show the circuit details of the added receiver RF preamp, and all the parts changes necessary to convert the SMK-1 to 20 Meters. At first glance, it might seem like quite a bit of work to make the conversion. However, each phase is easily accomplished, and the results are quite pleasing and well worth the effort. As always, two low-wattage soldering irons and a damp cloth are essential for removing and saving surface mount parts. I use a jeweler's headband type magnifier and strong light whenever I am working with any surface mount project.

The receiver's ten DB RF preamp brings signal levels up nicely, and is not found to overload the SA612 mixer. No switch was provided for removing the preamp from the circuit as the RF gain works quite well to reduce receiver gain whenever appropriate. R5, the sidetone pathway across the receiver's muting transistor, Q1, was removed and not replaced. With the RF Preamp in place, there is enough leakage of the TX signal across Q1 that the sidetone is quite audible at any reasonable gain setting, and is actually just a little too loud at the maximum RF gain setting. A 4.7 Meg Ohm resistor was put in R5's spot as an experiment, but was not sufficient resistance to lower the sidetone to a comfortable level at high RF gain settings.

I have added an 800 Hz active audio filter to the output of the LM386 in this rig, which can be switched in or out as desired, and have been very pleased with the available selectivity. It is particularly useful in helping the receiver reject nearby strong signals, as well as getting rid of the high pitched QRM of signals further away and elevating the chosen signal above the noise level.

On the air results with the little rig, even at the half-watt level, have been very satisfying. The first QSO with the little rig was with Ricardo, XE2RN/QRP, in La Paz, Mexico, who was running 5 Watts and was Q5 and a good strength 7. I was especially pleased with the contact, as La Paz is one of my favorite ports in the Sea of Cortez, which I visit regularly in my sailboat.

Another QSO of note was with Don, K5QK, in Louisiana who reduced his transmitter power to 500 Milliwatts to match the output of the SMK20, and his signal was still Q5 and solid copy during the remainder of the contact, which lasted several minutes longer. This speaks well of the sensitivity of the receiver.

Since those first contacts, I have added the easy One Watt Mod to the rig, and now have over 1.5 Watts of output, which is enough power to guarantee contacts any time the band is open.

Further experimentation is in order using dual crystals in the VXO, and perhaps changes in the inductance values used in series with the crystals, in order to optimize the frequency range and band coverage.

The one future addition that would further enhance the "user friendliness" of the SMK20 would be the 5 Watt Mod. For now, I am enjoying operating the rig at the one watt, plus, level and have continued racking up the contacts.

As mentioned above, any or all of these modifications to the SMK-1 can easily be applied to the Tuna Tin II and similar transmitters, as well as the MRX40 and other types of receivers, with corresponding benefits.

I hope to see you one the air around 14.060 Mhz.