A Popcorn QRP Regen Receiver and Lots More Air-Spaced Variable Capacitors

Many of us in the home-brewer community were really disappointed when Todd VE7BPO recently discontinued his popular and very through-provoking site “The QRP/SWL Homebuilder”. An archive of the site is available for download in pdf form and is a great resource. Though not quite the same as having the site, it’s great to have a copy of it for reference. Todd was having some issues with the site. He didn’t take it down due to lack of enthusiasm on his part and by way of proof, he’s back in style, with the new “Popcorn QRP – Scratch Homebrew Component-Level Radio Electronics” blog. It’s great – just like having his old site back. Todd combines circuit analysis and a theoretical approach with a strong leaning towards practical circuits that can be built by the home experimenter. Economy of design and performance are both considered, and from reading the accounts of Todd’s exploits, one suspects he is having a complete blast. If you haven’t visited Todd’s current blog or his previous site before, please note that the intent isn’t to provide the home-builder with complete step-by-step instructions on how to build a series of projects. This is circuit-level stuff, but if you’ve had a little experience at building circuits from schematics, all the circuits presented are tried and tested by our faithful protagonist, VE7BPO aka retired Professor Vasily Ivanenko.

All this is leading up to something, and that something is that our intrepid experimenter recently announced he had been revisiting the subject of regenerative receivers. He had built 4, and was sharing one of them with us on his blog. Happy, happy, joy, joy! Even better, this circuit was the same basic topology that I used in The Sproutie – a  circuit that separated the Q-multiplier and detector into separate stages. It’s an arrangement that is well-behaved and performs well. The regen control, a resisitive component, exhibits no hysteresis, and the addition of an RF preamp means that the circuit doesn’t suffer from common mode hum. If you build it well, it will be frequency-stable too, but that part is up to you 🙂

The blog-post is here. It is based on the regen circuit published by Makota 7N3WVM, whose website is a treasure trove of circuits for the home builder. You’ll notice great similarity to the circuit of Nicky’s TRF that first appeared in SPRAT Issue 70 in 1992, and N1BYT’s WBR from QST Aug 2001. What’s really interesting though, is that Mr Ivanenko has added a bipolar transistor (a 2N4401) to the J310 FET infinite impedance detector, to turn it into a hycas (hybrid cascode) pair in order to increase the level of audio from the detector, as well as the reverse isolation, which can never hurt, right? Great idea! His circuit includes a bipolar RF preamp in common base mode to add isolation of the oscillator from the antenna, as well as a little bit of gain, and a 2-transistor preamp after the detector that he has used in other “popcorn” designs. He also makes his argument for continued use of the LM386 as a final audio amp in this post. The description in the 2nd paragraph, of the folklore surrounding regens is mirth-inducing – and spot-on.

I know that at least a few builders are considering making their own Sproutie and if you are still at the planning stage, you might want to think about Todd’s version of the front end. I haven’t tried it (yet) but Todd ain’t no slouch 🙂  If I were to incorporate it into my next regen build, my current thinking is to use his front end, and feed it into a one-stage active audio filter using a 5532 op-amp or similar, and then into a nice, low-noise LM380 with it’s fixed amount of 34dB gain. If I ever do this, I can promise you two things –

a) it will be many months, maybe even a year before I do it, because I am very slow at these things and

b) I’ll show you the circuit of my AF stages so you can join it up with Todd’s (ahem, I mean Vasily Ivanenko’s) front end circuit


Todd just revealed 2 encouraging pieces of news –

a) He just ordered some black and red chicken-head knobs for his next project and

b) He has decided to spend another 2 weeks working on regens

If you don’t already follow Popcorn QRP, it’s well worth adding to your RSS reader, bookmarks bar or similar.


The rest of this post may be a little annoying to those who have limited bandwidth connections, though I imagine anyone who is still surfing the internet on a dial-up connection or a very slow mobile connection learned long ago not to come here 🙂 I recently acquired some more air-spaced variable capacitors and I’d like to share some photos of them with you for no other reason than short-term visual gratification! As always, my preferred brand is Hammarlund. Not only are they of high quality, but I happen to like the way they look as well. They are drop-dead gorgeous.

I’ll start with a piece that was acquired a few months earlier, before the current flurry of buying activity. This is an MCD-50-M and may well end up as the main tuning capacitor of my next regen, unless I happen to find another MCD-35-MX like the one that was used for tuning The Sproutie. MCD means that is is a double-ganged unit (unlike the single-ganged MC units), 50 is the maximum capacity of each section, and the M at the end refers to the fact that the offset rotor plates give it a “midline” capacity characteristic, keeping the rotation vs frequency characteristic reasonably linear. Nickel-plated brass vanes for a good temperature coefficient, and all mounted on a high-Q ceramic base. Perfection! Look at those bright and shining never-been-soldered-to-before terminals –

Leaping forward in time to a couple of weeks ago, this MC-200-M “midget condenser” came into my life sporting an older original box, in almost as good condition as the capacitor itself. I love the graphic design on these older boxes –

Then came my haul of just a few days ago. I bought them all from the same seller. The boxes are a bit beaten up, but the capacitors are in great condition. First off, here’s the group shot-

Among them were two of these HFA-100-A’s. If you’re as picky as I am, these are not ideal for maximum stability in a VFO or receiver tuning control, as the vanes are only supported at one end. It would be fine as an antenna trimmer where it’s placed in series with the antenna lead, or as a reaction control, in a regen where a variable capacitor is used for this function. I’m not sure what material the vanes are made of in these parts –

In the same vein (vane?), I also scored an HF140. What a great part. It offers two methods of mounting. If I possibly could, I’d use both of them for maximum stability. This variable cap could be mounted underneath a chassis so that the nut on the threaded shaft helps to hold the front panel to the chassis, while the mounting bracket was screwed to the underside of the main chassis for extra rigidity. Rigidity is a very good thing with regens 🙂 –

An MCD-100-M (2 x 100pF).  Nickel-plated brass vanes, a ceramic base (steatite actually), regular plate-spacing, and a “midline” capacity characteristic . Such a beautiful part. What else can I say? The US was once a manufacturing powerhouse –

Also in the haul were 3 x MC-20-S. I used one of these (from another buy) as the fine tuning control in The Sproutie –

Lastly, here’s an MC-50-MX. Single gang with capacitance swing of 10.5 – 53pF, midline capacity characteristic (for reasonably linear rotation-frequency relationship) and extra-wide plate spacing. This would be great for a VFO –


I also very recently acquired this National gear drive with 3 x 250pF variable capacitor attached. My relative lack of knowledge on National products caused me to misjudge. I discovered when it arrived, that the gearbox is a later model that is smaller than the older “classic” National gearbox drives. Also, it doesn’t have an eccentric bushing for driving the micrometer-style dial. I read somewhere that National stopped using the classic dial in their receivers in response to upgraded mil-specs that negated it’s use. It’s in good condition and turns freely, though the grease is old and it would probably benefit from a cleaning and re-greasing. I may well put this one up for sale –

There’s a little bit of surface dust on the plates and shaft but otherwise, it’s clean, and those terminals have never been soldered to. Amazing! –

No cracks in the insulators –

There’s only one detail that would require some attention from a builder, and that is that it looks as if the underside received a thwack at some point – either that, or it’s a manufacturing defect. One of the mounting holes is distorted and would need re-drilling and re-tapping if it is to be used. I think this is a solvable issue –

Another view showing this issue –

Old grease on the gears. If I were to use this, I’d want to clean out and re-lubricate, but that’s par for the course with these old yet still very serviceable gearboxes –


That’s it for now. Hope you didn’t mind me sharing all these pictures with you. As a parting thought, if you’re still planning to construct a regen and haven’t completely decided what to build, remember to take a look at VE7BPO’s Regen #4.




9 thoughts on “A Popcorn QRP Regen Receiver and Lots More Air-Spaced Variable Capacitors

    1. I try not to acquire too much useless stuff Steve, but sometimes these beautiful parts are hard to resist!


  1. Where do you find these things, they are gorgeous! I’ve been looking on eBay, but obviously not in the right place 🙂
    Great photos, and thanks for the links to Popcorn QRP – good stuff!

    1. I find most of them on eBay Mike, but I do a lot of looking, and pass on a lot of stuff – and I still occasionally get a duff one. Photographing them outdoors in bright shade helps to give them attractive lighting – and I’m lucky enough to have a nice lens that helps to make them look nicer than in the average eBay picture. eBay’s variable though. Sometimes there are great deals and not many others bidding against you, and other times there’s either nothing much to be found, or the competition is stiff. I have a few parts left to source before I can start thinking seriously about the next regen and I have no idea when I’ll find them – it could be next week or in 6 months. There’s a vintage radio fleamarket coming up in my area in about 10 days, so I’m hopeful that will yield something too! I just keep looking and looking………

      PS – I’ve said it before, but I’m looking forward to seeing how things go with your space-charge tube adventures!


  2. Hi Dave!
    I am starting the building of your lovely “Sproutie” receiver and will tell you soon how it goes.
    Tell me if I am wrong but I have found something weird in the electronic drawing. Drain and Source are inversed on both J310. I noticed that the photograph of your realization show them pluged in the correct way.
    I also got a question: have you tried classical coils instead of toroids? Are they less efficient? I am thinking of making a try of a classical coil for the beauty of it and also because it sounds easy to fit a plastic tube for wiring it around the octal bases!

    73++ F6GMQ Henri

    1. Henri –

      I’m really glad that you have decided to build a Sproutie. I still listen to mine – it’s a solid little performer. About the J310’s – they are symmetrical devices, meaning that the source and the drain can be interchanged without affecting the performance of the device. In other words, as long as you find the correct lead for the gate, it doesn’t matter which way round you connect the other two leads. When you say that drain and source are inversed on both J310’s, are you referring to the way I drew it on the schematic? Most schematic symbols show the source at the bottom, as in this datasheet – http://www.g1sle.com/files/downloads/j310.pdf I do occasionally see someone draw it with the source at the top and although that’s incorrect, it doesn’t actually matter in practice, because if you connect it to the +ve supply, it becomes the drain. I hope that makes sense.

      I thought about winding some traditional-style coils too, but haven’t gotten around to it. It would be interesting to compare the performance. I like the coils wound with toroids though, for the following reasons. The toroid is completely contained within the tube base so if you drop it, or otherwise manhandle it, there is less chance that the windings will be moved, and the frequency coverage/calibration of the coil altered (make sure to hot-glue the toroid core once you’re sure you have the correct number of turns, to keep it stable.) Another advantage (I think) of having the toroid surrounded by the tube base is that it is partially shielded from quick temperature variations caused by sudden gusts of cooler or hotter air in the room. Also, when plugging and unplugging the coils, my hands never come into contact with the windings, which may not necessarily happen with traditional coils. The self-shielding properties of toroids mean that you don’t have to be as concerned with keeping the coil as far away from metal objects, which is one less thing to think about when building. I have wondered whether a traditional coil might be more stable due to having air as a dielectric instead of the material the toroid is made of. However, I remember Todd VE7BPO’s experiments on his older QRP Builder website in which he built an LC VFO for 7MHz that drifted less than 5Hz/hour, using a toroid wound on a T68-6 core. The coils I wound for The Sproutie are stable enough for my purposes, so I’m happy with them. It would be interesting to try traditional coils though – and they have a great look!

      Best of luck building a Sproutie, and please let me know how it goes!


  3. Hi Dave,
    Ooops, I was not aware that the J310 is symetrical. I now understand that my remark was not relevant!
    I have been lucky with my first coil: the receiver works great with a toroid coil tuned between 5500 and 7200 kHz.
    So I have been able to listen to some CW and SSB QSO as well as some broadcast stations. Sensitivity and selectivity are great respect to the simplicity of the circuit.
    I will stay and toroids which are effectively very convenient to manipulate.
    I have particularly appreciated the nice features of the AF board (the LP filter is a must) and the receiver is behaving exactly as you mentionned it.

    So, thanks again for the nice description of your realisation. Great OM spirit!

    73 Henri

    1. Fantastic Henri – that’s great that your Sproutie seems to be working well. Several people have told me they are building one, but you are only the second person I have heard from who was successful in finishing it. Congratulations! I am currently experimenting with a series of switched active audio filters in my next regen. It is a lot of extra circuitry and wiring, and I’m not sure that it is justified – the simple adjustable LPF in the Sproutie provides only a gentle roll-off but in a regen, that may be all that is needed. I also tried reducing the value of the 39pF capacitor that connects the top of L1 (the main tuning coil) to the base of the Q-multiplier transistor to 18pF to see if the reduced coupling would make it less susceptible to pulling of the oscillator on strong SSB signals. It’s working fine, but I haven’t been able to determine whether it’s an improvement. It does, however, make quite a big difference to the frequency coverage of the coils, so you can’t easily change that value once you have several coils wound – unless you want to wind them all over again.

      Aaah, regens – so many things to experiment with, but the design of this one does seem to work rather well as it is.

      If you ever have pictures of your regen, I’d love to see them. You can e-mail me at aa7ee@arrl DOT net

      73 for now!


  4. I’ve got a large collection of those Hammarlund (and similar) including quite a few of the dual units. Also two rather weird National variables that go from min to max in 270 degrees of rotation. These also have shafts that are insulated from the bearings, the ground connection is via a flexible wire braid that runs through the shaft and the rear bearing (insulated) and is then soldered to the frame.

    This type of capacitor was used in Alfred P. Morgan’s 1H4G regenerative receiver from ‘The Boy Electrician’ book (1950’s edition). That cap, and the specific vernier dial he used (bakelite clone of a National), were the final parts I needed to build that set as it appeared in the book. The breadboard 4 pin and Octal sockets were sourced from Leeds Radio in lower Manhattan years ago, and 1H4G’s were a dime a dozen (almost literally!). Also found the Bakelite coil forms (National) some time back at a ham flea market.

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s