The Novice Special: An Old Classic Revisited


By Gary Johanson, WD4NKA


Photos and images may take a little time to download
This project originally appeared in the ARRL Handbook of 1970. It was a very basic design built around a 6C4 pierce crystal oscillator and a 5763 pentode RF amplifier. The transmitter did not double at the oscillator output, being a strictly "Straight-thru" type of Master Oscillator Power Amplifier, or "MOPA". The power input averaged about 250v at 40mA, or ten watts input.

Click here to see the original schematic that i used, as published in the above mentioned 1970 ARRL Handbook.

Click here to see my adaptation, which really isn't all that different, but hey, it sure is a fine excuse to show off the "RSGB" 1960's style schematic symbols from the TubePad Library! TubePad 6.0 (Beta) can be downloaded free from this site. Schematic is "drawn" in Paint, and saved in .jpg or .gif format for posting.

In my case, as always, i decided to revisit this rig in the same manner i revisit all the classics: with the contents of my junque box. Which often means some departure from original design, but hopefully, not too far. In my case, i opted to begin with the power supply, having on hand a 360v c.t. transformer, a 6 Hy choke and a couple 450v filter caps to make a full wave rectifier with pi filtering. A 35k 40 watt resistor serves as the bleeder. This was actually quite a bit more power and voltage than required, but i thought i would try it anyway.

Power Supply As can be seen, the supply is outboard, built upon a 3"x5"x2" chassis, complete with a 0 - 100 mA meter on board. Note the filter choke just happens to have a handy carrying handle! The switch on the side is the power switch. The filter caps are mounted under deck, and cabling runs out the rear.
Inside Power Supply Interior View, Power supply

Here is an up close view of the rectifier and filter caps. These are mounted on a phenolic strip which is spaced from the metal chassis by two ceramic stand-offs. The two diodes which form the full wave rectifier are mounted on the phenol board, one on each side of the cap. All transformer and output wiring are attached hereto. It is a tight squeeze, figuring in the mA meter.

top View Transmitter, Upper Deck

The transmitter itself is housed in what used to be the cabinet of a piece of test equipment. The original chassis was stripped and re-used. Since the former unit was a hybrid, there was a huge gaping square were the printed circuit board went. This had to be filled with a piece of high-voltage rated Phenolic stock, cut to size. Most of the transmitter circuit is built upon this board. The tube sockets were fit into holes drilled to fit the pins, and then secured by 4-40 screws. Upper left is the 5763. To the right of it is the 6C4, and between them, slightly lower is an octal socket carrying two xtals which are switchable from the front panel. Lower right is the output indicator: a #47 pilot lamp. Next to it is the load control cap, around 150 pF, and next to it, the plate tuning cap, 75pF. Above it is the final tank coil, wound "Sutter" style, approx 2" diameter, and tapped from the load control cap. Just on top of the load cap is the opening for the oscillator grid cap, which "shapes" the xtal.

Bottom view Below Deck View

Just to orient you, the tube pins to the left are from the 5763, and center is the 6C4. The octal is the xtal socket. The red and black jacks are for the key. The thicker bare copper wire is the ground bus. Power is taken from the rear apron jack at the bottom, which is in reality a 7-pin tube base. Yes, that's right, a mini- tube base! These came with a pre-tinned bracket which soldered in place directly to the chassis' rear apron. No screw holes to drill! A miniature version of the old octal power receptacles. to the right of the power input is the SO239 coax jack. In reality, the circuit faithfully follows the original handbook version, except for the pilot lamp: in mine, it measures output. The original registers current dip. Since i have a mA meter on the PS, i figured it might be handier to toss a loop near the hot end of the tank, and make this Glow-Bug glow even brighter.

Transmitter front The Novice Special, 2004 version

Here is the completed rig. At center left is the indicator lamp bezel, next to it the load control, and then the tune. Below the lamp is the key jack. Next to this, the xtal switch. The cabinet is spray painted a dark gray hammertone, and the front panel is Hallicrafter light grey. This is pretty much the colour scheme of all my home brew rigs. This way they all match, and look pretty nice all together. At least, that's the idea. The knobs are from an Echophone EC-1. Harvey would be proud.

One thing i did like in the way this turned out: it's a later era rig, so i managed pretty much an early 1960's rounded-edge cabinet look. Now, if will work as nice as it looks, we'll be satisfied.

Rear View Rear cabinet view

Here you can see the power input jack and the antenna jack, with corresponding holes punched thru the cabinet to accommodate. I tried to make this rig smack a bit of Johnson-in-miniature. Thus, it is dubbed the "Mini BA", or Canoe Anchor. The bottom has felt feet which i swiped from an old Regency which was dozing off on the cabinet. Sorry, you snooze, you lose! In the process of assembly and repeated openings and closings, some of the paint has chipped, so there will be another session with the spray can.

Transmitter and supply The Complete Transmit Station

Here she is, complete and ready to tune up into the dipole with that Kenwood antenna tuner. Actually, it does well enough with it's pi circuit alone. Pretty nice grouping, but all was not well.

In my past, i have gotten away with running 6L6 rigs well into the 50 watt arena on class C. That's all fine and good with the bull-dog 6AG7 pentode oscillators and hefty 807s and large envelope octals, but on these little minis, i discovered something: Bad News!

I was running well over 100v higher than tube rating, achieving a dip at 65-70mA at 400 plus volts, danged near 30 watts input, and lighting a 25 watt bulb full brilliance! And you had to dip very, very quickly or else the plate of the 5763 lit up like a Christmas tree! And the poor 6C4 just could not find stable footing. I managed to stabilise things for the most part, but tuning the final greatly pulled the xtal, there was no decoupling the oscillator from the final without the final losing the grid current and again lighting up the plate cherry red-hot! I could not tame the oscillator, who wanted to see 200v, not 325v dropped thru a mammoth voltage divider!

Finally, i tossed in the towel and resigned to the only solution: Another, lower volt power supply.

New Power Supply The New Supply

This was not gonna be easy, because i didn't have really another decent chassis to build on. And i didn't want to wait till Christmas to finish this project. As i cast about, my eye fell on an old computer power supply culled from an old 386 machine. I knew i had the proper transformer: i wonder if it would fit? So i tore out the innards, mounted a thick phenolic deck in the former computer power supply cabinet, and rested the transformer atop this deck. Yes! It just cleared! This transformer had a high voltage secondary of 130v, so i wired a voltage doubler, finding a couple 20uF/ 450v axial electrolytics in the junque box, and another pair of diodes. Taking this mess to work, i drilled and wired the supply together. The result is not near as pretty as the former supply (which will now be paired with another, future MOPA) but much more palatable to the transmitter! Gone, however, is the nice mA meter. And the handy carrying handle. But, not bad considering it all fit in a nice compact box. You can see i left the fan in. I have to add a switch, fuse and Bleeder yet.

But the radio is much, much tamer, and has a far more stable note, and the drift factor is not near what it was, in fact, there is no drift to speak of after warm up for a few minutes. She handles FT-243 xtals fine, but really likes the HC6-U types especially!

Rear Rear View with New Power supply, and Cabinet removed

Here you can see the power supply hookup. Since the Kenwood antenna tuner is in use, the loading cap is fully unmeshed, the tuner taking over the output control, essentially. Without the tuner, load and plate tune come to rest at half mesh each, just the way i like it! The output is now a bit less hair-raising, about 10 watts. The plate voltage to the final (5763) is now just under 300 volts, dipping at about 45 mA. I have a 15 watt lamp for a dummy load, and it lights up, only not to full brilliance, but close enough to let me know that RF is definitely there!

Mission Accomplished! Now to get that dipole up in the air where it was before the hurricanes!

Full Rig Completed Transmitter set-up, less receiver.

Well, here's the parting shot. Far left is the Champion, and next is the (yuck) new power supply that's definitely gonna get a face lift --- next, the Novice Special, cabinet off for final touch-up, next to right, the antenna tuner, and sitting atop it, the 15 watt lamp. It's all temporarily sitting on a board on a stool. You can just see the HT-37 on the upper right trying to get into the picture. I plan to use the Q5'er with this, but for now i am using my HW-101 as the receiver, and switching an antenna switch back and forth for sending and receiving. See ya on 7031 or 7050!

Full 6V6 Rig And a Happy Ending

....Ahh, yes, the poor bride left stranded at the altar! Well, she found a new friend, the trusty, dusty, slat-board 6V6er. So all is well in the kingdom, and they all lived happily ever after! Power supply, far left, then the 6V6er and Pi tuning adapter, and to the right of that, the antenna tuner with the Champion on top.
The Sutter Coil A Close-up of the "Sutter Coil"

Do you remember me referring to the "Sutter" type coils? Here is an example of one atop the Slat board 6V6er. Solid core copper wire wound thru Bakelite, phenolic or lexan strips to hold shape. This particular coil is 3" in diameter. It was originally wound as a tank with antenna link, the wire antenna base clipped to the hot end of the link, and the counterpoise to the cold end. You can see the link leads poking straight up. But right now i have it set up for pi output tuning. If carefully made in the manner Fred Sutter outlined in his 1938 QST article "The QSL-40", absolutely beautiful coils can be made to rival Air Dux or any other commercial product.
Over the first weekend of November (2004) i took my "Novice Special" out on the road, setting up in a fourth floor hotel room in Fort Lauderdale. My receiver was a genuine Q5'er. I took this get-up because to me, it was a great representative of an actual Novice homebrew setup one might find in a Novice Shack of 1960 - 1970. The receiver was indeed one of my Novice projects. The Q5'er is simply a BC-453 with a converter front end. Two methods were commonly employed, the most traditional being a tunable converter, using the BC-453 (which tuned 190 - 550kc) as an IF strip fixed at 455 or 460 kc. This unit has a final IF of 80kc, and is capable of decent single signal reception. It makes a great cw rig, good on sideband, way too narrow for AM, but useable nonetheless.

Another way to use it was as a tunable detector with a crystal front end converter. This way, you just change xtals to change bands, and re-peak the front end. This is what i used. The converter uses a 6700 kc xtal, tuning 300 - 500 kc for 7.0 - 7.2mc. Another xtal, 6900 kc will cover 7.2 - 7.4mc, and so on. I also use a colourburst xtal to catch the phone activity on 75m, which is why i have that funny looking dial on the front of the '453: it is calibrated for use with the colourburst xtal. The regular dial is used for 40m.

I made some more mods to address the still existing yoop in the Novice Special, one of them was removing the "Sutter Coil", using a 1.25" plastic 35mm film canistre wound coil instead. Resonance was checked with a GDO. I also removed the grid variable capacitor in the 6C4 because it was damaged, which might have caused part of the problem. I replaced it with a fixed 31pF cap. I also removed the bleeder resistor from the supply. The slight regulation it was supposed to provide actually made matters worse! Instead, i replaced the load resistor for the 6C4, originally 3900 ohm/ 1 watt to 27k/2w., and what was originally a .001 cap from the oscillator to the grid of the 5763 became a 300pF cap. This resulted in approximately 170v on the plate of the 6C4, which did wonders for ironing out the chirps and yoops. Loading of the final still has an effect on the xtal frequency, but keying has very little effect. You can see the entire rig that i took to Fort Lauderdale, plus some close-ups of the modified upper deck of the Novice Special, and close-ups of the Novice Q5'er below. If you are interested in the schematic of the Q5'er converter, see my link to the "Q-dyne" or click here. Note that the xtal oscillator is not shown: use any type you wish. I used a 6C5 in a xtal pierce circuit. The schematic does show where to inject the L.O. See the schematic for the Regenerodyne for the exact xtal L.O. used. Note that the converter is regenerative, which really helps the overall volume and Q, hence the name Q-dyne.
Upper Deck Compare this with the previous upper deck shot. Note the coil change. The #47 pilot lamp is now showing relative output via a single link, shown using green insulated wire, wound on the top end of the new coil. The film canistre is white. Note also the xtal loading cap is now missing. The hole remains for a future trim cap to replace the fixed 31pF now soldered in the place of the damaged trimmer i removed.
Novice Station Here is the entire station which i took down to Fort Lauderdale. From left to right: the "control head" housing the power supply, converter, and BC-453 outputs. The little bolt-on box on the upper left side of the unit is the front-end "Q" regeneration adjustment. There is also a trim cap knob on the back of this box to finely adjust the front end when using high feedback settings. If you haven't yet noticed, the cabinet for this unit is the shell and chassis of a stripped out Johnson Messenger One "White Face". I am also using it's original speaker. Next is the BC-453 itself. To the right of this is the Novice Special power supply, and then the Novice Special itself. Finally, the key. In Fort Lauderdale i used my Vibroplex Champion, but at home where these images were shot, i use an old RAF 8-amper.
Q5'er Here is a close-up of the Q5'er. You can see the controls: the bolt on regen unit, which was an afterthought. I had some image problems with maritime NMR, which needed attenuation, hence the addition. On the "White Face" left of speaker is the band select, below the front end tune. The two toggle switches to the right of the speaker are the CW/AM detector switch and Standby toggle. Below these is the audio gain. Speaker jack is below the speaker, lower centre. I usually run 600 ohm headsets, but 8 ohm stereo headphones do fine. The BC-453 heater voltages are converted to 12v. The accessories are run from the rear. The output of the converter is fed via thin coax to the BC-453's antenna input. I should mention that the BC-453 has a fairly underpowered audio section in it. The use of a Q multiplying mixer like the Q-dyne greatly enhances the audio, requiring much less gain on the part of the BC-453's audio section. I seldom need to run the regeneration control beyond it's half-way setting. Especially now that NMR no longer exists (although i wish sometimes it still did!)
Novice Special This really needs no explanation. It's the 1970 ARRL Handbook "Novice Special" revisited, with minor mods. I understand that the original chassis and cabinet just might be still available, but i have not found exactly where. But it is irrelevant to me: as novices, we built our rigs in whatever we could find. In my case, i used a broken piece of test equipment's chassis and cabinet plus dark hammertone grey and light "hallicrafter" grey from Wally World. Both are Krylon products.

The crystals i am using, except for two, come from Bry, AF4K's Excellent Amateur Radio Site.

Give this rig a try! On my next one, i will use much shorter lead lengths for the point to point wiring. Long wire lengths due to the use of the PC board may have aggravated the chirp/yoop problem originally. The next rig will be built into a much smaller container!