On frequency!

Finally!

I put a 68 pF capacitor in series between the tuning capacitor, C12, and existing 68 pF capacitor, C10. That seems to have dragged the operating frequency sufficiently high to hear some signals. The oscillator circuit now looks like this:

Truth be told, to get up to the phone portion of 40 meters with this oscillator, I had to crank the T2 tuning slug ALL the way up, perhaps more than I really should, but it works. Currently, I'm covering 7,150.5 kHz up to 7,242.5 kHz according to my Si5351 signal generator (hooray for homebrew test equipment!).

I'm thinking I'll replace the two series 68 pF capacitors at some point with a single capacitor, perhaps in the 10 pF to 20 pF range to get the same coverage without putting the tuning slug up against the stops. There's also another reason to replace them: this receiver is drifty as hell right now. I couldn't figure out why it wouldn't stay on frequency, as I was seeing several tens of Hertz drift per minute. After a good bit of head scratching, I looked over my parts list and it turns out that all of the other oscillator caps are NP0 or COG, except for my 68 pF ones. D'oh!

The section on oscillators in "Experimental Methods in RF Design" needs some reviewing this evening. I'm not satisfied with the limited 92 kHz band spread that I'm getting out of this oscillator; it seems like I should be able to get more range of tuning. Without replacing it with an Si5351, I want to try and experiment with changes to the oscillator to get more spread, and I'll write up what I find.

Another area I'd like to experiment with is the "filter" between the mixer and the audio section. There's an awful lot of signals close together out there, and tuning or discerning one from the other is a pain. I ran a quick LTspice model on the components playing at being a filter, and it would seem that they do nothing of the sort. In fact it would seem that they actually attenuate audio below about 3 kHz. Not, I think, what I need for selectivity.

For anyone interested, I put together a short video of me sweeping the band. The audio isn't terribly great, and I think there was entirely too much interference, but I'm pleased to hear some QSOs out there today.

The first signals I heard were a round table of sorts on 7,185 kHz, with KK6TS, K6DNS, and W6NYW. Thanks to them for helping me, unknowingly, to validate my new receiver. I also got to spend some time listening to what I believe the participants called the "Do Nothing Net" on 7,222 kHz.

Let me know what you think...I really would appreciate the feedback.

Oscillator thoughts

I'm still noodling around with the Neophyte, which is still oscillating terribly off frequency; about 1.2 MHz low from the 40 meter band target for this design.

Today, I tried to adjust the 42IF123 primary coil slug to see how "adjustable" the oscillator might be. It will adjust the oscillator output frequency but not by very much. Here are the results.

T2 slug maximum counter-clockwise (all the way "up"):

• Top end: 6,801.2 kHz
• Bottom end: 6,559.0 kHz

T2 slug maximum clockwise (all the way "down"):

• Top end: 4,993.2 kHz
• Bottom end: 4,822 kHz

A surprisingly large tuning range; about 2.0 MHz of tuning or about 1.8 MHz of shift, although the slug is entirely too coarse to be useful for tuning. As you can see, I still won't be able to get this on the 40 meter band as built.

That got me to thinking about the oscillator. I've poked around the components and construction layout; everything is as it should be. So one wonders: If the device is built properly, could the problem be in the design? Hubris, arrogance, pride, take your pick, but it's an honest question. 

As 40 meter direct conversion receiver, an ideal oscillator should percolate in that band and range across the entire band; 7,000 kHz to 7,300 kHz (I did say "ideal").  The original design from the QST article shows the standard Colpitt's oscillator layout (the amplifier transistor is internal to the NE602 chip and is omitted from my drawing).

WA3RNC's original Neophyte oscillator

Since a Colpitt's oscillator operator frequency is determined by a parallel LC tank circuit, bear with me while I render the complex WA3RNC tank circuit to its reduced form. Note that C12, a 4-365 pF air variable capacitor is set to a mid-range 180 pF for the purposes of this reduction. First, we'll reduce C10, C11, and C12:

Which is then added to the transformer's internal 47 pF capacitor. Simultaneously, I reduce C7, C8, and C9 on the right side of the transformer to yield the following:

Since capacitors in parallel are a simple sum, resulting in the equivalent parallel RC tank as:

Final equivalent tank circuit for WA3RNC's oscillator

Using the equation for parallel resonance, I get a resonant frequency of about 5,700 kHz. Strange, that's just about where my Neophyte operating frequency lies. Is it at all possible that the values for the oscillator NP0 capacitors from the QST magazine article is wrong? I have a spare 68 pF capacitor on the table, so if I add that in series between the C10 and C12, that raises the tank resonance to 7,360 kHz, which may be close enough for me to adjust the span using the transformer tuning slug so that I can cover 7,000 to 7,300 kHz.

Someone out there tell me if I'm flying too high for an unpressurized cabin, or if I'm on the right track, please! Comments?

The Neophyte is working...

...for certain values of the word "working."

Over the last few days I've been hacking away at building and Arduino based signal generator using an Si5351 clock generator board from Adafuit. In another post I'll document that ordeal, but suffice to say that the signal generator is working!

Well, the very first thing I want to try is to test the Neophyte build with the output of the generator; haven't heard anything but AM broadcast breakthough so far.

So, I fire up the receiver and boot up the signal generator, and start tuning the generator starting at 7.000 MHz. Nothing, nada, bupkis, zilch. Neither a chirp nor a burp all the way to 8 MHz. How can this be? Now I'm no further along than I was before; I can't tell if the receiver is not receiving or if the signal generator is not generating. Oh, the frustration.

I'm a fan of Bill Meara's SolderSmoke podcast, and common refrain I've heard about troubleshooting is to take a step back and some time off. So, do I take his advice? Nah! I considered it, but where's the fun in that? But still, I'm at a loss as to how to proceed...no test gear, no other receivers, no clue.

I've been spending my commute time listening to older episodes of SolderSmoke; I think I'm back at about episode 60 right now, about when he was getting ready to depart London for Rome. One element of the podcast that was much more common in earlier episodes was "the bandsweep." Bingo! There is my next step.

My signal generator uses a rotary encoder to change the output frequency at the user's desire, so I could manually tune across the entire HF band to find a tone. Ugh, are you kidding me? I'd have to do it at 10 kHz steps in order to hear the chirp as I sweep around. For the arithmetically inclined, that's about 3,000 clicks of my 18 detent encoder, or almost 200 full rotations. Not really feeling the love for that task.

Instead of doing it with the hardware, I opted to beat it with software. The code for the generator was mangled to sweep the entire band and display the sweeping frequency on the LCD screen. When I hear a chirp, I narrow the sweep range. It took about four iterations of this to pinpoint my receiver's frequency. A little zero-beating and moving the Neophyte tuning capacitor, and I've got it.

Ladies and gentleman, I've managed to construct a not-quite-60M (52M or thereabouts) receiver using the plans and parts for a 40M receiver. I will now pause for a golf clap. The little bugger ranges from 5780 kHz to 5986.3 kHz, with a nice clear sound. Not much useful there, although according to NTIA the upper portion of this range is dedicated to broadcast radio so perhaps the "breakthrough" I was experiencing  was  actually bond fide radio signals. The range ~200 kHz is narrower than I expected, but that may be due to the oscillator working at a much lower frequency than I had planned.

Either way, I'm a lot happier than I was this morning: the signal generator works and the receiver works, albeit on a very wrong frequency than I built for. Now, just what the hell did I get wrong in the oscillator circuitry? Time will tell...

Derp!

I was doing some account cleanup on G+ and managed to make private the repository holding all the images for this blog. Hence the weird "place holder" images that looked like a giant minus sign enclosed in a circle. Sorry for derpiness, all should be fixed now.

Just 'cause it looks pretty...

...and it does look pretty, it just still isn't working.

I was thinking on my problem from last evening on the commute home this evening, particularly the symptom wherein the peaking capacitor on the RF input tuned circuit seemed to have no effect on the received signal. What I hadn't mentioned in the last post was that the tuning slug in the 10.7 MHz IF transformer did change the received signal, albeit not enough to copy any useful signal.

What, I thought, would be the effect if I had soldered the leads for the peaking capacitor to the wrong transformer leads? The front end is balanced feed into the NE602 inputs on pin 1 and pin 2, but oddly enough the peaking capacitor which supplements the transformer's internal 47 pF capacitor is specified to be from the mixer pin 1 to ground (vice across the balanced inputs). One could assume that putting the variable capacitor from pin 2 to ground would work as well, no? It turns out, not so much.

It may be difficult to make out in the photograph, but the green/black pair of leads, which run from the peaking capacitor, runs to the transformer pad on the lower right of the can, which is then connected (by the green jumper) to the NE602 pin 2 RF input B. A few short minutes to heat up the soldering iron, and it's now on pin 1, RF input A, as the design called for in the first place.

Now that that is properly placed, the front end has positive tuning ability via the peaking capacitor, and I now understand the purpose of the potentiometer on the primary side of the input transformer...the audio output of the receiver is now quite loud, louder than I would like to have jacked into my headphones for any significant length of time. The potentiometer was omitted during construction to simplify my prototyping and to have one less thing dangling from leads off the board (currently I have the big tuning cap, the peaking cap, the headphone jacks, the antenna/ground leads, and the battery pack). After I pack the munchkins off to bed later tonight, that potentiometer will be soldered into place.

This still hasn't fixed the problem of receiving useful signals here, but that may have been due to me having the antenna wadded up under the workbench in the garage with fluorescent lights running overhead. I'll take it out to the patio later for some real testing with a real-ish antenna.

Incidentally, does anyone know how many turns the 42IF123 tuning slug is capable or moving?

I Hear Voices...

No, not like that.

I finished the basic construction of the Neophyte this evening. No mounting or case yet, just the basic board in a cardboard box with the tuning and peaking capacitors dangling from twisted wire.

There's a good bit of background noise and as I tune, there is a lot of non-signal audio that shows up, but I can definitely hear some radio signals. I'm relatively sure that at least one of the signals is some AM broadcast blow-through; there's some serious fire-and-brimstone stuff on that channel. None of the other signals have been strong enough or clear enough to pick out actual intelligible conversations or, better yet, call signs, but it's a start.

Strange thing: The peaking capacitor on the tunable front end has zero impact. Have to check to make sure that I have it wired up correctly and that I haven't somehow killed the capacitor.

Also need to get a better quality antenna rigged. I found two bits of 30 foot long bare wire (actually gardening wire one would use to tie back plants to a trellis) and am using them as a bare ground and end fed wire antenna. I think I'll wander down to the hardware store and pick up a roll of something small gauge. There's a small pass through in the wall next to my desk, so I think I'll run the wire out that and down a line of palms that line the driveway; perhaps about 100' of wire would work. Through the same hole, I'll drop more wire down to the utilities connection point directly below where I sit.