UPDATE May 7th 2021
In 2013 I suffered a stroke which severely limited my ability to carry out design work. Things that in the past would have been easy are now hard. As we approach the climb up from sunspot minimum, now is a good time to think about really simple gear for 10m. If you are able to continue this work, I should be very pleased. On 10m antennas are simple and small and not much power is needed. Ideally write an article and produce a simple, small, PCB that can be easily built. Use parts that can be easily be obtained.
Breadboarded TX AF amp, single balanced diode mixer, buffer and IRF510 FET PA. Power out currently about 1W pep. Will try a different PA device (2N3866) with the aim of reaching at least 2W pep although I think the IRF510 should be able to reach this target.
In the end I went for a FT37-43 toroid for the balanced mixer rather than a 28MHz tuned circuit. Currently the rest of the TX line-up is quite broadband and I may be able to trade some gain and Pout for bandwidth with some optimisation.
TX low pass filter
Still to breadboard:
The VFO. Current plan is a 32MHz crystal mixed with a 3.4-3.6MHz VFO although I may go for a 14MHz super-VXO and doubler.
The receiver. NE602 would be easiest but I am half tempted to avoid this and just use a single balanced diode mixer again.
A speech processor TX audio stage with double pre-emphasis, heavy clipping and strong lowpass filtering. Aim to make a 2W pep have the punch of a 10W radio.
Yes, you read it correctly: a simple sideband (not single sideband) rig for 10m (28MHz).
By simple I mean of course a double sideband supressed carrier design. This has been a project idea for too long, so now is the time to create this. What I have in mind is a simple, small rig to use from home or out portable to work local SSB stations and the occasional DX when conditions are favourable. As with many of my projects, this is a "work in progress", so come back regularly to see how this develops. Already I have breadboarded many of the subsystems needed, so putting these ideas together should not be difficult.
These are the outline requirements currently:
TX power ~2W pep (equivalent to 1W pep SSB). 5W pep would be better though, but higher power increases interference risk to others from the unused sideband.
0.5uV sensitivity direct conversion receiver
300mW AF ouput
Tuning range 28.4-28.6MHz (possibly VXO controlled over a smaller 25kHz range in a simpler version)
Possibly a CW option covering 28.0-28.1MHz
Small, single PCB design using commonly available parts. Initial version will be built "dead bug" style.
Simple, compact housing e.g. diecast box or aluminium box.
Target cost for ALL parts £20 ($32).
Easy to reproduce.
I'd like this design to become a classic and one that helps others get onto this great band.
Nothing too unusual here at all: standard design elements found elsewhere will be used (where already designed and available) from sources like the GQRP SPRAT magazine and the internet. I will try to make the design simple without over-design. This is a common mistake, I feel, in many published designs, especially for a single band rig. For example, why use wideband matching between TX stages when with narrowband designs a higher gain can be achieved with fewer stages?
For the receiver, simple approaches such as the NE602 integrated IC mixer is adequate for the intended use: most stations that will be worked will be using at least 10W, often 100W, so that sensitivity does not have to be fantastic, just adequate. Also, the IM2/IM3 performance of the NE602, although not fantastic, is fine for such a simple radio, especially if the input filtering is reasonable.
The advantage of the DSB approach is the design simplicity with no need for expensive SSB filtering. The downside is the waste of power in the unwanted sideband and the risk of some LSB interference to adjacent channel stations. In a later version, a phasing design may be considered to attenuate the unwanted sideband by around 20-30dB. Another problem is that receiving another DSB signal can be difficult, although not impossible.
A single balanced diode mixer followed by 2-3 stages of gain (2N3904), with either an IRF510 FET, 2N3866 or CB type bipolar PA.
Use of T37-6 (yellow) toroids for all tuned circuit inductors (small and self screened).
Discrete design using 2N3904 transistors.
Speech clipper/processor and LPF to increase "punch" with the QRP output without splatter.
Mixer/VFO with 32MHz crystal oscillator with 3.4-3.6MHz VFO. Could be based around an NE602. Stable enough for (short duration) sideband QSOs.
Direct Conversion Receiver
NE602 RFamp/mixer with 3kHz passive low pass filter and LM386 AF amplifier.
Able to be aligned with minimal test equipment such as a scope, wavemeter, DVM, diode probe and HF receiver.
BEWARE!! These links are very old and some may have expired. Remember that expired links can often be found on the archived pages at the Wayback Machine.
Here I will add any links from where ideas have been found and used. This should encourage others to look at these sources for their own versions.