Saturday, August 27, 2011

One Nuvistor on 40m CW

"Tube manufacturers have unveiled, in recent weeks, drastically new concepts and techniques aimed to keep them in the race with the transistor industry. Smaller than a thimble, more rugged and efficient than present tube designs, and particularly suited for mechanised production, the 'Nuvistor' represents a radical departure in the electronic-tube concept."
April 1959, Electronics Design

Some time ago I salvaged a pair of RCA type 8056 Nuvistors from the vertical amplifier board of a defunct Tektronix TV waveform monitor. These are medium-mu triodes, specially designed for low anode voltages.

Last week I decided to place one of my Nuvistors on the 40m CW band. The elements of this single triode are relay-switched between the receiver and transmitter cicuits. My setup is a vacuum tube implementation of W2UW's, now classic, "FET-1" design. G3XBM made a spendid job of Glenn's original circuit; the results of which may be seen here.

The circuitry that I settled on dates to the 1920's. For example, the regenerative receiver is a standard, tickler-feedback affair. I can just detect my 1uVrms weak-signal generator through a 20dB attenuator with this receiver. I suspect that a 0.5uVrms signal would provide comfortable copy under quiet band conditions. Out of curiosity, I learned the receiver will continue to function as a self-oscillating detector with as little as 5Vdc on the anode (for this test the feedback coupling was raised to four turns). 

I found the crystal-controlled Colpitts configuration produced a slightly higher transmitter efficiency as compared with a Miller-type layout. Although the Miller circuit has the advantage of a grounded-cathode (switching would only be required at the grid and anode) I decided to use the Colpitts arrangement. With 24Vdc on the anode the RF output power is 24mW. 

My receiver and transmitter circuits are shown below. I didn't bother to include the receive/transmit switching details in my drawing. I use a pair of 12V-coil DPDT sub-miniature relays (the coils are wired in series for 24V operation). One pole each is used for switching the grid, cathode and anode. The fourth pole switches the antenna between the receiver and the transmitter. A hand-thrown SPST switch activates the relay coils to enable operation of the transmiter. Of course an 8PDT manual switch would achieve the same end with considerably less current consumption on the 24Vdc supply line.

Also not shown in my drawing is the Nuvistor heater supply connections (the 8056 requires 6.3V at 135mA). I notice a mistake in my receiver schematic; C4 is shown as a fixed-value capacitor when it should be drawn as a variable capacitor. In fact, this is the "regeneration" control.

Component List

C1: 5pF
C2, C3: 100pF
C4: Improperly shown as fixed; should be ~15 to 100pF variable
C5: 18nF
C6: 470uF @ 50V
C7: 20pf
C8: 150pF
C9, C11: 0.1uF
C10: 75pF
C12, C14: 390pF
C13: 680pF
CV1: 5 to 25pF (main tuning)
R1: 1MegOhm
R2: 6.2k
T1: 3.5 to 5.5uH, 16 turn primary, 1-turn coupling
T2: Same as above, except 3-turn secondary
L1,L2: 1mH RF choke
L3, L4: 1.4uH
X1: 40m quartz crystal
HP: High-impedance headphones
V1: RCA Nuvistor, type 8056

I had my first QSO with this radio shortly after completing the final connection, when VE3DTJ/3 answered my CQ from his lake house north of Toronto; a distance of 303m/488km. Barry gave me a 579 and we kept a ragchew going for nearly twenty minutes.

When I came up to the house for lunch I sent a message to the online QRP-L group, telling of my operation. VE3DJX answered my call that same afternoon. Jim reported that I was 539 in Smiths Falls, Ontario (170m/274km). Despite my weak signal, Jim appeared to have a fairly good copy for the duration of our ten minute QSO.

I had an email message that evening from WA9ETW. Mark said that he had roughly a 20% copy of my signal in Monticello, Wisconsin (846m/1362km); not good enough to reply on the air, but he did manage to copy my complete callsign. 

Two receivers in the Reverse Beacon Network made a number of captures of my signal in the two days that I operated last week. K3MM is located in Damascus, Maryland (402m/646km). WZ7I is located in Pipersville, Pennsylvania (280m/451km).

I'll close with an excerpt from a letter that W2UW sent to Roger, G3XBM.

By the way, you might be interested in knowing that I have made 451 QSOs with my little rig. I have not met my goal of working all of the states east of the Mississippi river. HI I still have 4 to go (Fla., Ala., Geo., and Miss.)....

I am 81 yrs. old and still think radio is "magic". Hi
Glenn, W2UW

That is simply wonderful!

Saturday, August 20, 2011

A Point-Contact Transistor Superhet

In his booklet, Rundfunkempfang ohne Röhren, dating from 1959, H.G. Mende presented an AM broadcast band regenerative detector circuit using a point-contact transistor (PCT).

I breadboarded this circuit using a Western Electric 2N110 transistor; one of several that were given to me by Jack Ward of the Transistor Museum. My regenerative detector came to life with a minimum of fuss. As a quick test I swapped this circuit in place of the 2N107 alloy-junction based regenerative IF detector in a recently made 40m superhet receiver

The minimum detectable signal strength with the PCT regenerative stage increased to 1uVrms. As expected, the 2N110 produced conisiderable self-noise. Of the three devices on hand I selected the least noisy unit (date-coded November of 1956). I could still hear "eggs frying" in the phones with the antenna disconnected, but the level was now low  enough that I could live with it. I found the regeneration could be smoothly adjusted using the variable resistor located in the emitter-ground return. I only noticed pulling (frequency-injection locking) on the strongest signals. 

The DX came rolling in with this setup during the early evening hours on 40m. HB9CVQ and EI7KD both had big signals here. A couple of interesting signals were heard from Brazil. It's not often that you hear an 18 year-old girl working CW on 40m, but Lia, PU2AIL, was heard knocking off the DX in great style. Also heard was PQ8XB operating from the Reserva Biológica do Parazinho

The results were good enough that I decided to attempt to build a 40m simple superheterodyne receiver exclusively from 2N110 point-contact transistors. However, I was unsure whether or not these devices would oscillate in the vicinity of 7MHz, given their datasheet-listed 3dB-down frequency is only 1.5MHz.

Fortune smiled on me the next evening when I found that one of my two remaining 2N110s would not only oscillate on 40m, it would generate as much as 20mW under crystal-control! 

Well, I couldn't resist connecting this oscillator to my Czech military surplus straight key and calling a few CQs. This, despite the fact that changes in the thermal equilibrium were pulling the crystal-locking frequency around more than one could expect to get away with. It amounted to a slow, but very pronounced chirp on the keyed signal. Hearing no replies, I ceased transmitting and returned to thinking about the superheterodyne design.

However...the next morning I was surprised to find an email message from K8GU/3. Ethan is a scientist engaged in studying the ionosphere down in Silver Springs, Maryland. 

Good morning, Mike,
I've been spending a lot of time listening with an SDR
(SoftRocks+Rocky) lately and last night I noticed your curious-looking
signal in my waterfall.  It took a moment for your signal to come up
on the QSB, but I had pretty good copy on you.  I started rewiring the
station so I could give you a call, but my wife handed me our
7-week-old son for a diaper change.  The best I could do was hit
"Print Screen" as I whisked him off.  So, you'll have to live with a
screenshot for now...

Please click here to see his screen-shot of my 20mW signal. That is so cool...thank you Ethan!

Getting back to the superhet receiver; I decided not to try to place a 2N110 transistor directly into the 7MHz signal path. I opted instead to use a germanium diode single-balanced mixer in conjunction with a 2N110 crystal-controlled oscillator. The variable IF output signal feeds a second 2N110 working as a self-excited, regenerative IF amplifier/detector/audio frequency amplifier. Here's the resulting circuit

Please notice the lack of an explicit RF feedback path in both the oscillator and regenerative detector. I'm making use of a unique characteristic of these early point-contact transistors. Under certain conditions a negative resistance appears at the base of these devices. Unlike a tunnel diode, for example, the value of negative resistance produced can be easily controlled by adjustments made at emitter and the collector. For example, the detector regeneration level is adjusted by RV1 in the circuit shown above. 

Components list

T1, 2: 3.5 to 5.5uH adjustable, 16-turn primary, 2-turn coupling
T4: ditto above, except 1-turn coupling
T3: 1:4 balun (Mini-Circuits T4-1, etc.)
T5: 455kHz shielded IF can (yellow-slug)
L1: 470uH molded RFC
C1, C2: 114pF 
C3; 5pF
C4: 10nF
C5: 120pF
C6: 470uF
C7: 18nF
C8: 75pF
R1: 3.3k
R2: 2.2k
R3: 330 Ohms
R4: 1.5K
D1, 2: germanium point-contact diodes (1N34a, etc.)
CV1: 20 to 420pF air-variable capacitor (main tuning)
RV1: 5k
X1: 7376.4kHz quartz crystal from junkbox
HP: high-impedance magnetic (mine are 1950's Czech army surplus)
Q1: 2N110 point-contact transistor (date-code March 1957)
Q2: 2N110 point-contact transistor (date-code November 1956)

The total DC current on the -22Vdc supply is -4.87mA (107mW power drain). 

As you'd expect, the receiver is somewhat less sensitive on account of the gain-less mixer stage. I can't hear my 1uVrms signal-source. Aside from that and the afore-mentioned weaknesses, the all-PCT receiver appears to function normally. In one night of listening quite a few DX stations were heard along with several domestic QRP stations. Some stronger stations were copied with the headphones lying on the workbench. I did not find it necessary to re-adjust the regeneration setting from one end of the 40m CW band to the other. 

By the way, I came across an interesting passage in Chih-Tang Sah's, The Fundamentals of Solid-State Electronics, on the subject of these nearly forgotten PCT relics.

"In this first point contact transistor, the minority carrier pathway through the bulk or base of the single crystal Ge had probably contributed only a small percentage of the total curret. Instead, the surface channel path may have dominated the current making it a p-channel junction-gate field-effect transistor." p. 705

The author proposes that 90% of PCT action was due to field-effect and only the remaining 10% on account of (base-injected minority current) bipolar-transistor action.

The PCT was a manufacturer's nightmare (William Shockley refered to certain steps in their fabrication process as "witchcraft"). As if the wide device-to-device performance variability, dismal production yield and user reliability numbers weren't bad enough, they tended to be horribly noisy. Even the earliest junction-type germanium transistors exhibited a 20dB noise improvement! 

"People beyond AT&T were beginning to suspect that all the original enthusiasm over the transistor had been premature. 'Current Bell statements concerning the transistor are far more subdued and give an impression that it is under wraps,' noted a September 1949 article in Consumer Reports. 'Very little is said of immediate practical applications. Such transistor difficulties as high noise level are stressed.'" p191, Crystal Fire, Riordan and Hoddeson

The 2N110's used in my receiver described above first appeared in 1955. But as late as 1952, as this Electronics Magazine editiorial suggests, transistors remained at best, a tantalizing but unfulfilled promise. 


Saturday, August 13, 2011

1950's Transistorized Ultra-Simple Superhet

My interest in simple amateur radio gear made from early semiconductors recently led me to the March-April 1956 edition of GE Ham News (thanks to N4TRB). There I noticed a particularly simple plan for an AM broadcast-band regenerative detector using a 2N107 hobbyist-type germanium alloy-junction transistor. The circuit appears in the lower, left-hand corner of the image below (please click-on the photos to enlarge them).

Now this is a bare-bones circuit! For example, the base to collector leakage provides the working DC bias and the headphones double as a radio-frequency (RF) choke.

I had the circuit working on the breadboard in a matter of minutes using a 455kHz shielded IF "can" in place of their "loopstick." My 2N107 transistor is date-coded February of 1957 and my "4k Ohm" headphones date from the same time-period. 

As is typical with these early, low-gain bipolar transistor circuits, the regeneration adjusts very smoothly and without hysteresis. While one expects a fairly narrow peaked response (especially with a ~350kHz operating frequency), I noticed what appears to be a positive feedback loop for audio-frequency (AF). Increasing the overall feedback past the threshold of RF oscillation progressively narrows the audio bandwidth until the point of sustained AF oscillations is reached. While a small dose of AF peaking is not unwelcome for CW reception, too much results in an unpleasant "sea shell" effect, along with unnecessarily critical beat-frequency tuning. Fortunately, as it stands the AF bandwidth isn't excessively narrow at the optimum RF feedback setting.

With the aim of producing an ultra-simple 40m superheterodyne receiver, I next threw together a self-excited mixer, or converter, to precede the intermediate-frequency (IF) regenerative amplifier/detector. The circuit I chose was well-known in the late 50's and early 60's. For example, it appeared in the "Hints and Kinks" column of QST Magazine in 1960.

Here it is again in the March 1963 edition of DL-QTC; "Kristallgesteuerter Transistor-Konverter"

I chose to use a 2N78 in my 40m converter stage. This rate-grown, NPN, germanium transistor was introduced by GE in 1955.

“The General Electric NPN transistor type 2N78 is designed for RF and IF amplification in broadcast receivers. It employs a new and revolutionary manufacturing technique – the exclusive GE rate-growing process." General Electric History

Although the "alpha cutoff frequency" for the 2N78 is only 5 or 6MHz, it functions well enough as a self-oscillating mixer on the 40m band. The device I used is date-coded October of 1956. The snapshot below was taken from the 1956 Radio Shack Catalog. The 2N78 would cost $77.32 in 2010 dollars. 

Here's the circuit that I'm using.

C1: 120pF
C2: 3.1nF 
C3, C4: 10nf
C5: 10pF
C6: 22nF
C7: 18nF
C8 470uF
R1: 52k Ohms (optimum value depends upon transistor used)
R2: 100 Ohms
T1: 3.5 to 6uH slug-tuned; 16 turns with 3 turn coupling
T2 455kHz shielded "yellow-slug IF Can"
X1: 7.3764MHz quartz crystal ( everything else ;o)
L1: 5.6mH molded plastic inductor
L2: 3.3mH "gum drop" inductor
CV1: 15 to 360pF, 6992 to 7070kHz main tuning capacitor
CV2: 15 to 360pF, regeneration control
HP: high-impedance magnetic phones
Q1: 2N78, NPN germanium transistor
Q2: 2N107, PNP germanium transistor

Frequency stability is excellent although grasping the headphone cable originally produced a slight shift in the receive beat-note. The RFC and capacitor low-pass network connected between the 2N107 collector and the headphone (shown in the above schematic) eliminates the problem. I can clearly copy my weak-signal test generator/step attenuator at 0.2uVrms on this little receiver. 

The first station I copied was VE3BYR/QRP, operating north of Toronto with an excellent, S-7 signal. On the next evening I copied TY1KS in Benin at 559 among a dozen other DX stations. I easily copied OE5KC with the headphones lying on the workbench. Obviously he was 599! When N4NQ mentioned he was running 75w to a Windom antenna, I connected my step-attenuator between the antenna and the receiver with the 33dB pad switched in. He was a 439 copy; the same as he would have been if he were running 37.5mW with no added attenuation on my end. I was equally pleased at what I did not hear this SWBCI whatsoever! 

I swapped several 1950's vintage 2N169 and 2N170 transistors in place of the 2N78 with roughly the same results. 

Here's an old German transistorized regenerative detector circuit that's nearly as simple as the one shown above. This one uses a Siemens AC-151 junction transistor (ft ~ 1.5MHz) in the common-base configuration. I've lost track of where this circuit came from, although I suspect it was from one of Herbert G. Mende's books; late 50's to early 60's. (Yes, DL3PB confirmed that it's from
H.G. Mende's, Leitfaden der Transistortechnik. Thanks Peter!) 

Tuesday, August 2, 2011

The Ancients Have Stolen Our Inventions

In his August, 2001 QST Magazine article, "The WBR Receiver," Dan Wissell, N1BYT, writes

"The future of Major Armstrong’s namesake may be more open-ended, however, because a simple and effective solution to the coupling problem has been found. The method of coupling the antenna to the tank circuit described below is reminiscent of a Wheatstone Bridge circuit, and thus the receiver name,“Wheatstone Bridge Regenerative(WBR) Receiver.” I’m reluctant to claim that this is a “new” detector design, even though an extensive search hasn’t yielded anything similar. But with nearly 90 years of use, I’m sure every method of detector-antenna coupling has been tried at one time or another!"

It turns out that Dan was correct in assuming that our grandfathers must have first hit upon his imaginative technique for improving the reverse-isolation between the antenna and an oscillating regenerative detector.

A portion of N1BYT's 2001 circuitry appears at the left-hand side of the above image. On the right hand side Jackson H. Pressley's 1924 oscillating mixer is shown (the antenna connects between terminals 1 and 2-3). N1BYT's bridge-isolation circuitry is employed as part of a Q-multiplier -> detector type regenerative receiver, whereas Mr. Pressley's nearly identical circuit appears in the front-end of his superheterodyne receiver.

In fact, I thought that I was being clever when I adapted N1BYT's idea to a vacuum-tube regenerative receiver some years ago. Please click here to view my design for a 40m regenerative detector employing a 6AK5 pentode. Notice there is precious little difference between Mr. Pressley's circuit and mine!

Update 24 December 2011: Today I came across an excellent example of an early WBR regenerative receiver by the well-known, Frank C. Jones. This circuit appeared on pages 24 and 25 of the May, 1927 edition of Radio Broadcast magazine. Please click here to read the original article.

Chester Rice, the co-inventor of the moving-coil loudspeaker once exclaimed in exacerbation, "The ancients have stolen our inventions!" 

My feelings are quite the reverse. I find it positively charming that dwelling on some particular problem, human minds far removed in time, place and culture so often seize upon nearly identical solutions. 

Once upon a time I took up solving Sangaku problems as a past-time. The more difficult proofs would often require a week or so of intense concentration. Eventually arriving at what I surmised was a novel proof; only then I would peek at the given historical proof. Time after time I was pleasantly surprised to find that some ancient Japanese court mathematician, or perhaps a farmer of noble mind had hit upon precisely the same solution. 

I am a man: little do I last
and the night is enormous.
But I look up:
the stars write.
Unknowing I understand:
I too am written,
and at this very moment
someone spells me out.
Octavio Paz, Brotherhood