- How much digital control? Will there be any hard controls, or will it all be software? I'm fine with having software behind the controls as long as it's invisible. Menus I want to avoid.
- Frequency range. With a digital VFO, DDS for example, an extremely wide frequency range is available.
- Modulation modes. Somewhat dependent on previous question. AM, SSB, and CW are desired, digital and FM modes have yet to be determined.
- Modularity. Basic module structure has yet to be determined.
- Power structure / distribution. 13.8VDC bus + (5V bus?)
- Structural considerations, aluminum vs. steel vs. plastic. What and where.
- External connection options. (Can it operate when closed for waterproofing?)
- Speaker(s) This probably won't happen, but a little internal horn loaded speaker would be fun...
- Controller. I'm inclined to go with a Raspberry Pi 2 for the flexibility and horsepower this would provide. 40 GPIO should be adequate, and I can always expand that if needed.
- Connectivity. I want this to be as versatile as possible.
- Weight. For those people who know me, weight is pretty low on the list of concerns. Right now, my collected assortment of parts in an ammo can weighs about 25 lbs.
Digital control: If I use a Raspberry Pi (2), I could potentially make it a complete SDR. This opens up a lot of possibilities. On the other hand, I have a instinctive distrust of computers...
I don't know. If I used the raspberry pi, I could have a display port, and show a waterfall display, frequency, etc. on an external display. I could also potentially do some fancy audio filtering, but that requires more knobs unless I want to do software menus (I don't!) and I'm already short on front panel space.
Another advantage to using the Raspberry Pi is that it can be programmed in python. This is supposed to be an open project, when constructed, all schematics and source code will be released. I might sell plans and/or parts though. Python is very easy (relatively) to get into, so if people want to modify it, that's a plus. I hope to learn C (+, ++, ?) next semester though, and that's a more common language, so that's an option too.
Frequency range: Modern DDS chips can practically go the quintessential DC to daylight. For receive I know I want at least from the AM broadcast band to approx. 30 MHz. 6 meters would be a bonus, but not something I would focus on much. 2M would be very nice, but that's totally different from HF. If it goes to 2M, it needs to be able to receive the FM broadcast band, AM air stations, and the weather channels. 440 I'm somewhat indifferent to.
Modulation modes: Fairly simply, if it only does HF, it only needs to do AM, SSB, and CW. I know 10M has FM, but it's not worth putting in for one band. (Unless it's software and would be really easy to do) If it does anything above 6M, FM (narrow and wide) is a requirement. Digital modes are optional. Again, if using a Raspberry pi, there are a lot of options, but digital is close to the bottom of the list on things I'll be working on.
Modularity: I need to develop a good interconnection system. I'm thinking that each module should have access to a power bus, (see next point) the VFO, a discrete line to the controller, the previous module's output, and the next module's input. I'm not sure how I want to handle RF connections yet. I like BNC, but I think it's too big for this application. There are smaller versions of BNC, but I haven't found one I liked yet. I don't like SMA, but it might be a good fit for this, it's pretty small, and readily available.
Power structure / distribution: My current plan is to use two nominal 12V, 7AH gel cell batteries. I plan to wire them in series when they are discharging, and parallel when charging. The reason for this is that DC-DC converters can be very efficient nowdays, and it's easier to build a converter that only has to downconvert instead of providing a regulated 13.8 VDC from a source that may be above or below the output voltage. Buck vs buck/boost. There will be a front panel meter to monitor voltage. I may make it switchable to monitor the battery voltage (should show around 24-28 VDC when discharging, 12-15 VDC when charging) I drew up a series/parallel switcher with a DPDT relay and a sense resistor, but I don't have a digital copy at the moment, I'll see if I can put that in later. For distribution power, I think one 13.8VDC bus would be adequate, but I'm also considering putting in a 5V bus. If I have a 5V bus I'll probably use a good switching converter, but if each module only has a 12V bus and has to internally convert to 5V, I'll probably just end up using a bunch of LM7805's.
External connection options: Ideally there would be external connections for a mic, speaker and an antenna, so that the radio could be sealed, waterproof, and still operate. Let's worry about that later. Maybe I could just make the whole thing water resistant when open. (Ammo cans are fairly waterproof when shut) That's probably the best/simplest plan.
Audio: Audio output will be via a 1/4" stereo jack. Audio will be mono (wait, maybe I could do stereo FM broadcast... hm...) anyway, I'm still trying to decide if plugging in a connection should automatically disconnect the internal speaker or if there should be a switch for that. Having a switch gives you more flexability, so I'll probably do that. If someone wants to have the speaker auto disconnect, that's easy to wire up. There will also be a switch to select whether you have connected a speaker or a pair of headphones. This radio should be very versatile. There will be an internal speaker, but you can only do so much with a given amount of space. A little horn loaded speaker would be fun, but I'm already tight on space, so I'll probably go with a simple direct radiator. Shouldn't be a problem to get a 10 or 20 watt 3" speaker or so. I think that should be adequate for most scenarios, and for anything else you'd need headphones.
Controller: The Raspberry Pi 2 is cheap, adequately powerful, well documented, and has good I/O. I even remember hearing something about it's PWM pins being able to output up into the FM broadcast band. Hm... with a little filtering, I might be able to use that... Now things are getting interesting. The Raspberry Pi (2) also has a display output, so I could possibly provide a front panel display jack for an external band display. Maybe a waterfall display like the Elecraft panadapter. It would be even cooler if I could mount a little display on the lid of the radio, so when you open it... I'm going a little overboard here, I'll back up.
Connectivity:
- Antenna: UHF (SO-239) and binding posts
- Speaker/headphone: 1/4" jack
- Mic: Undecided. May leave option open for people who already have Yaesu, Icom mics.
- Power: IEC C-14 AC connector, Binding posts: 12-36V in. Bridge rectified and fused.
- Potential: USB (Keyboard, software upgrades, etc.?), HDMI.
Weight. I'm not going to worry about it. This is designed to be rugged, reliable, and flexible, but not light. Which is too bad, as it would be a great backpacking radio otherwise, but I can only make so many tradeoffs. (Battery life, etc.) It might be a 20 or 30 pound radio, but the weight should be similar to an ammo can filled with ammo, so it's not too unreasonable...
In case anyone's wondering, yes, this is the sort of thing I think about in my free time for fun.
If anyone finds a flaw in my reasoning, or something that I should think about, feel free to comment!
I have a number of other considerations, schematics, etc., but those will have to wait for another day.
