The MMMM Receiver

Prototype construction of the MMMM receiver. The, um, interesting position of the coils is explained below.

The 80m Madagascar Mighty Mite was suffering from “a tree falls in the forest but nobody hears it” syndrome. Eighty meters is a tall ask for Madagascar — there aren’t that many hams in the coverage area, and given local noise, I doubt any of them can hear well on 80m. It would be a long wait for a signal report about the on air performance of the MMM. Clearly, the thing to do was to create a mate for the MMM, the Madagascar Mighty Mite Mate (MMMM).

In keeping with the philosophy of back-to-basics rockbound simplicity, I decided to build an 80m version of the Sudden Receiver originally described by George Dobbs in SPRAT, and reprinted in 73 (October 1991, page 8, available online thanks to the Internet Archive).

The Sudden Receiver is a direct conversion receiver based around the NE602 mixer and LM386 audio amplifier. With these magic black boxes performing so much of the heavy lifting, only a handful of other components are required to constitute an upfront bandpass filter and serve as off-chip components of the NE602 oscillator.

In the original design, two TOKO-brand inductors were used to further simplify the design, but these are out of production. The GQRP club does have a current kit, the Limerick, and they use canned inductors as well. However, I did not have these on hand, so I had to make mine from scratch.

To parallel the design of the MMM, I thought the MMMM should also be based around a colorburst crystal (since this is the only crystal I have in the junque box that falls within the ham bands anyhow). I based my build on the version 3 schematic of the Sudden Storm Receiver. The frequency-determining component of the original Sudden receiver was based around a Colpitts Oscillator involving a fixed inductor and variable capacitor. The Sudden Storm uses a variable crystal oscillator, which is capacitively pulled by a reverse-biased 1n5818 diode serving as a varactor. The voltage on the varactor is adjusted by a potentiometer, which means that I don’t have to tie up a variable capacitor in the tuning circuit. The varactor range is limited, but since the whole point of this build is to hear the MMM, I wasn’t worried about that.

In this design, two inductors are required, a 39uH for the bandpass filter and 33uH for the oscillator. In the kit distributed by QRPme, I believe they use molded inductors. Not having any of these on hand, I wondered if I might wind my own. However, the number of windings required to make these on the usual iron-containing core materials (type 2 or 6) was not practical. However, it occurred to me that the A(L) of ferrites is about two order of magnitude greater.

I didn’t think it would be a great idea to use ferrites in these positions, but gave it a try anyhow: 9 turns on an FT50-43 for 35.6 uH and 5 turns on 50-43 (11 uH) plus 8 turns on 37-43 (22.4uH) for 33.4 uH. I figured that 35.6 uH would probably work for 39uH if I added more capacitance than in the original design. Somewhat surprisingly, this all worked, but probably not optimally. The oscillator did oscillate at the desired frequency and I was able to hear the plaintive tone of the MMM from the other bench. Connecting the receiver to an 80m antenna, I could hear noise, but couldn’t tell much about sensitivity because — not surprisingly — no one happened to be transmitting near 3.579 Mhz in the vicinity of Madagascar at the time of testing.

I ran this idea past the QRP-Tech list and received feedback that confirmed my gut feeling that ferrites don’t belong in these circuits. Aside from temperature instability, the main objection was the resistive loss in ferrites. So, while they make fine HF chokes, they are best left out of tuned circuits. Everyone agreed that an air-core inductor would be preferable, and since I have both air and magnet wire, that’s what I did: I used the “Professor Coyle” online calculator to estimate windings on a pill bottle core, secured the coils with nail polish, and glued the coils awkwardly to the side of the circuit board. Since I had originally built around toroids, I had not left enough room for large coils on the board; consequently, lead lengths were a bit long, but this did not prove to be a problem.

I did not hear any difference with regard to the MMM tone, but I had the impression that the background noise level was up with this build. Certainly, the band pass filter tuning was much sharper, presumably thanks to higher Q of the air core inductor.

Diagram of components on graph paper
Rough layout for Manhattan construction; not entirely to scale.

All told, this was a pretty limited design, but taken together, the MMM and MMMM constitute the Mighty Madagascar Multi-Module More-or-less Monofrequency (MMMMMM) *Transceiver*. Granted, no one may ever hear or be heard by it, but it was a fun build.

Alas, the life of the receiver is destined to be short. Rather than have a very frequency-limited receiver, it occurs to me that the core of this receiver should be put to more general use. The next project on the list is the LBS transceiver designed by N6QW and KK6FUT. In that project, the first step is to hook up a DDS oscillator to a balanced mixer and audio amplifier. Well, it occurs to me that’s exactly what I already have.

I’ll still build a “legitimate” LBS from basic components, but on the way to building the LBS Part One, I am going to detour slightly and hook up an AD9850-based oscillator to the MMMM in place of the varactor VXO. With different front-end filters, the MMMM could effectively be multi-band. So, next step: putting together a microprocessor-controlled AD9850 DDS.

1 thought on “The MMMM Receiver”

Leave a Reply

Your email address will not be published. Required fields are marked *