The Norcal 2N2/40 CW Transceiver Kit

I’ve been wanting to build a kit for some time now.  I built the Softrock 40 a few months ago as my re-entry into the world of building radio gear. It was my first experience with SMT devices, and it went quite well. However, I really wanted to build a complete transceiver – something I can solder up, stuff into a case, and use to talk to someone on.  It’s been years since I’ve done that.

As a kid I spent countless hours with my nose buried in RadCom, Practical Wireless and various books from the RSGB, staring at the schematics and pictures of all kinds of wonderful homebrew projects.  I built a few too, but spent many more hours just gazing at this stuff.  As an adult, I haven’t changed much, except that we now have the internet, so the opportunities for hours of happy browsing are even more numerous.

I spent a lot of time familiarizing myself with the various QRP CW transceiver kits out there and I don’t think there has ever been a better time to be a kit builder, there are so many good kits available. I looked at the kits from Wilderness Radio – the SST series of rigs (very appealing because of their simplicity),  the Norcal 40A and the Sierra transceiver. Then of course, there is the early 21st century version of Heathkit – Elecraft, and their excellent K1, K2 and KX1 kits. Steve Weber’s PFR3, Weber Dual Bander and MMR-40 from Hendricks kits all received good long looks from me as well. There are some really interesting SDR transceiver kits out there too, but I wanted to build a more traditional little QRP radio this time around.

Phew – so many kits to consider, and the problem for an internet “window shopper” like me is that I often spend so much time reading up about things that I can get to the point where I don’t feel the urge to do them any longer. It has something to do with the art of delayed gratification, which I learnt a little too well as a kid.  The good side of it is that it does save me a lot of money.  The bad side is that I spend a lot of time indoors reading about things instead of doing them.

OK, apologies for the digression. There was one other design that caught my attention, and that was K8IQY’s 2N2 series of transceivers.  The design was K8IQY’s response to Wayne Burdick N6KR’s challenge to design a transceiver that used no more than 22 2N2222 transistors.  No PNP transistors, voltage regulators or IC’s were allowed – the 2N2222 was the only active device to be used.  Dayton 1998 was where the judging was held, and Jim Kortge K8IQY took the prize. For about 10 years, a lot of folk built these neat little transceivers manhattan style until the Norcal QRP Club came along to help those of us who still want to build things, but like to have them in the form of a ready-to-assemble kit. The Norcal 2N2/XX series of kits included a double-sided PCB and to many, this was just too good to resist.

The original run of 500 kits looked like they were close to sold by the time I arrived on the scene.  The club had suspended all online ordering (and as of today, I notice that they have decided to suspend all order processing for a year in order to give their volunteers a good break.)  I lucked out and found an amateur on the Yahoo Norcal 2N2 Group who had a spare unbuilt 2N2/40 kit to sell.  40 meters was my preferred band,  so this was an exciting find for me – thanks Cameron!

While waiting for the kit to arrive, I built an RF probe.  I don’t own an oscilloscope, so the next best thing was an RF probe.  This would allow me to test each stage as I proceeded. The instructions are here. I built mine in an old 1/4″ jack plug.  The body was metal so that the probe would be properly shielded.  The test tip was a piece of stiff wire soldered to the end of the plug, with a piece of heat-shrink tubing placed over the joint:

Then I slid the insulation over the whole shebang, screwed on the body of the plug, added some leads, and voila – an RF test probe for around $2:

The kit was well packed and arrived intact.  Cameron, who had bought it from Norcal months earlier, and hadn’t had time to build it, also included all the replacement parts and updates that Norcal sent out.  A view of the entire kit laid out on my deck:

and a close-up  of the update packets that Norcal sent out (ignore the transparent packet of trimmer capacitors – that was part of the original kit):

From finish to end, it took me about 10 days to build this great little radio. I have no idea how many hours I spent on it.  Some days, I worked for a couple of hours, and then went off to meet a friend for the rest of the day, coming back to put in another couple of hours in the evening.  Other days I worked on it while entertaining company at home, and then there were the marathon sessions where I started in the evening and worked until the wee small hours of the morning when sleep got the better of me.  One time I started work at around 8pm and after many hours of toiling away with the soldering iron, realized that it was 6 o’ clock the next morning! Funny how times just melts away when you’re engrossed in a project.

Here’s a view of the board after I had built and tested the power protection, the receive active power decoupler,  receive main audio amplifier,  receive/transmit keying,  receive mute,  receive audio pre-amp,  receive local oscillator,  receive product detector, receive post IF amp crystal filter, and the receive IF amplifier. That sounds like a lot, but some of the stages were literally just a few components:

Same board, same point during the construction, different view:

While we’re at it, here’s a picture of my workbench/operating position. The candle isn’t evidence of any kind of gothic leaning  – I was using it to help strip the enamel from the ends of toroids after I had wound them:

Speaking of toroids, there are a fair number to wind in this project.  Some folk love winding toroids, more seem to dislike the process.  I take what I think of as a more “zen” approach, which is the same approach I take to all kit building. I take my time when I’m building; I don’t feel any rush to get it done and get the project completed.  Part of the enjoyment for me is in building the kit, so I take my time doing it.  While I’m doing it, whatever needs to be done gets done.  Some things are simple, like stuffing resistors into holes and soldering them, while some things, like winding toroids, take more time. While building, I may be listening to the radio – either some CW on 40 meters or my local college station KALX, which comes from the Berkeley campus.  I take frequent coffee breaks, and may also have company over.  The point is that the most detailed laborious task (such as winding a toroid) can be made more enjoyable by becoming engrossed in the process, or at the very least having something interesting playing in the background! The main point of kit building for me is the journey rather than the destination. Having said that, my toroids, though perfectly serviceable and not all bad looking, are still nowhere near as beautiful as K8IQY’s. He uses a #2 crochet hook to pull the wire through the toroid, and also mounts the core in a small vice while winding.  I hold the small toroid cores in my hand while winding and don’t use any other tools to help the process, so I may think about trying Jim’s technique for my next project.

Here’s the board with the receive chain finished and fully operational. The controls are temporarily wired in for testing purposes. In the center of the board, instead of T6, there is a jumper wire.  This is because on 40 meters the RF receive amp isn’t used:

Wow.  Exciting! There are two crystal filters in this radio – a 4 pole (the main crystal filter) and a 2 pole filter just after the IF amp. The bandwidth of 500Hz sounds just about right.  I have not used a lot of narrow filters in my life, but compared to the 300Hz filter I have installed on my FT-817 (which is a little too narrow for everyday use) this one sounds like a good width for regular CW use.

By the way, in these pictures, the radio is sitting on a schematic of the actual radio being built.  As suggested in the excellent assembly instructions available on the Norcal website, I printed a large version of the circuit diagram to use as a reference while building.  It makes a pretty backdrop for pictures of the 2N2/40 as well.

Here’s the finished board with the temporary wiring for all controls removed, ready for installing in the case:

I’m breezing through the construction in this blog, but a look at Norcal’s assembly instructions will show you that there is a fair amount of work involved in putting together this transceiver. It’s definitely not for beginners, but if you’re good at soldering and have some circuit building experience, it’s not hard. You’ve just got to put in the time. After you’ve built each section, the instructions show you how to make a few measurements to verify that the stage is working as intended.  Don’t skip these tests.  The peace of mind you get from knowing that everything you’ve completed so far is working is well worth the time.

Here’s the board mounted in the case with all the controls attached. Is this thing beautiful or what?

I think it looks decidedly less pretty when fully ensconced in the case, but that’s only because I love looking at components and circuit boards:

A head-on view of the front panel, showing the lettering:

I was keen to see how stable the fully free-running VFO would be.  I haven’t made any measurements but as expected, the VFO does drift quite a bit in the first few minutes after initial switch-on.  If your main experience is with commercial radios controlled by PLL or DDS synthesized VFO’s, you’ll have to get used to the fact that you shouldn’t be switching the radio on and transmitting almost immediately.  I had intended to run some tests to determine the minimum amount of time I should leave the radio on before operating, but since finishing it, I’ve left it switched on nearly all the time.  It uses such little current (measured at around 135ma) that it’s convenient for me to leave it on, so if I wake up in the middle of the night (a regular occurrence) I can listen to the radio and immediately respond to any stations I might hear calling. I will say this – when the radio has settled down to a stable internal temperature, it is easily stable enough for serious CW operating.

The night I finished the radio, I went to sleep, woke up just before 3am, put the earbuds in and worked JM7OLW.  He gave me a 419 with the 4 watts from the 2N2/40.  Not a stellar report, but my first confirmed contact – and it was with Japan!

The sidetone is a little loud for me, so I’ll be adjusting the value of resistor R14 to reduce the monitored transmitted signal to a more comfortable level. The other change I’m going to make will be to add a digital frequency display (probably the KD1JV Digital Frequency Dial.) With these two changes made, the 2N2/40 will be a very serviceable TX/RX for everyday use in the shack.

A few more thoughts on this radio.  If you’ve come from a position of operating mainly commercially built rigs, you’ll probably go through a short period of adjustment when getting used to this radio (or many other similar types of designs.)  As previously mentioned, it has a free-running VFO,  so it needs to be switched on a while before you plan to operate.  It doesn’t have any AGC, so you may need to be a bit more nimble with the AF gain control from time to time to compensate.  The radio does need to be powered by a well regulated power supply, as it has no regulator of it’s own (due to the original design criteria.) Another difference I noticed was the inability to switch to a wider receive bandwidth.  As this is a CW only rig, you may not see the need to be able to set the receive filter something wider than 500Hz. I spend a lot of time at home, and often leave the radio monitoring 7030.  When doing so with my FT-817, I tune to about 7029 and leave the filter in SSB mode.  That way, I can hear anyone calling from about 7029 to a little over 7031KHz. On hearing a call,  I can then zero beat the other station and switch to a narrower filter if necessary before replying. I am not able to do this with the 2N2/40 and feel that I may be missing out on the occasional QSO because of it.

All this being said, it isn’t fair to perform these types of comparisons, as the 2N2 was designed to meet a strict set of criteria for the Dayton 1998 contest, the main one being that the only active device to be used should be the 2N2222, and no more than 22 of them. No “complex” active devices, such as IC’s or voltage regulators could be used either.

Should the above deter you from building this radio? If you specifically want a radio for backpacking and portable operations, there are others that are better suited.  If you’ll be using this at home though, it is a very worthwhile project with plenty of room in the case for you to add your own extras if you wish, such as keyers or frequency displays. I built it because the original design concept caught my imagination, and the changes that Jim has made to the design in the years since haven’t caused it to stray too far from the original idea.