Midwinter calling, day 10

Wave on, Wave off

Kari is horrified and ashamed by what we did last night in front of everyone, and it is of course my fault that I did not stop her, for which I just accept the blame. If you are right, you lose. If you are wrong, you lose. Might as well just accept that and move on. I promise to really try to avoid something like that happening again in front of the others - except for the start of our wedding night. That improves Kari's mood. Which honestly feels good to know.

I've tried to avoid too public intimate moments with just one or two of Iselin, Ciara and Kari, since I don't want anyone to feel left out. Sure, I didn't really chose to have a relationship with Ciara or Kari, but it's not like I dislike them, and I will try to make our relationships work. But it's hard, and I'm not the most caring or thoughtful man.

It can actually be very easy to build a primitive radio transmitter with the right parts, and a receiver is easier. A primitive receiver for Amplitude Modulated radio - ie AM radio - can be built as a diode radio, also commonly called a crystal radio, or foxhole radio when made by soldiers during World War II. The receiver consists of just a few part like a really long antenna wire, and an earthing nail pushed into the ground. Metal wire wound as a coil and a variabel capacitor works for tuning. That capacitor might be improvised, like two metal cans insulated with paper and stacked one inside of the other, and changing the overlap is the tuning. A rectifier was often improvised by a rusty or blued razor blade with a safety pin or piece of graphite pencil that sprung against the surface of the blade. And finally a pair of high impedance headphones or rather piezo headphones. That's all. No battery or other components, but it is a bad receiver and requires a strong AM transmitter for the radio energy is what you hear directly in the headphones.

A really primitive radio transmitter is almost as easy, and my spark gap transmitter is actually quite complicated compared to what was initially used on Midgård, but they did not understand how important resonance and tuning is and a lot of people like Marconi though the 'whip crack' of a spark was required for radio to work. The easiest transmitter is just a pieces of cable connected to a battery that you tap against the other pole so it sparks, and do it with a long antenna wire on one side, earthing on the other and a higher voltage battery, and it is basically a spark transmitter. A pure spark gap is inefficient and with an 'ugly' dampened wave 'DW' signal, which without tuning, is basically a jammer that all radios hear regardless of frequency setting.

But spark gap transmitters work and if a few kilowatts of high voltage are used they reach far and was used to transmit across the Atlantic to primitive diode / crystal receivers. Powerful spark gap transmitters with more than 100kW was made in the early 1900s. It is possible to make them frequency tuned - which is what I've done with my spark gap transmitter - and it is also possible to make a fast pulsating spark gap with an electric motor and a rotary gap, which gives a better and more melodic signal. Higher spark rate is easier to hear. It was a variant of an electromechanically tuned transmitter with rotary spark gap that the Titanic and similar ships had, and that radio had a few kilowatts of output in the then common range of 0.5-1MHz.

It is also possible to build a more mechanical transmitter of Alexanderson model in the so-called Very Low Frequency 'VLF' range of about 15-200kHz, but it requires precision machines to manufacture something large that rotates fast enough with the fine tolerance needed without destroying itself. And lots of power. But it is possible to build electromechanical DW, CW and AM transmitters of several hundred kilowatts. The only preserved Alexanderson type transmitter in Midgård is the 'Grimeton' station in western Sweden a bit south of Gothenburg, and it is damn fascinating. I have visited and heard it several times, because when standing beside it in operation you want to use hearing protection, which the museum lends out. The mechanical and electrical problems that need to be solved when working at that level are impressive, and I respect the engineers that had to solve all those problems. Having an idea and sketching it on a paper is one thing, actually building it is often the real problem. The museum starts the transmitter up on a few occasion, but transmit just once each summer, and it can still be heard around the world. It's VLF signal is so low at just 17200Hz that it can be received far away with just a coil plugged into the microphone port of a computor. That is all kinds of impressive when you think about it.

The simplest all-electric radio transmitter that actually makes a pure tuned signal, is an interesting but very primitive narrowband ultra-low power transmitter using a classic crystal and making a Zinc Negative Resistance transmitter. But its output power is measured in microwatts, so even with a really good antenna and sensitive modern receiver, the range from a mountain peak without any obstacles is 'only' a handful of kilometers. But such an ultra low power transmitter is neither practical nor usable, and I also lack any suitable classic crystal, and I sure won't try to make one any time soon. There is a big difference between a classic two-legged crystal and a modern 5V powered crystal oscillator with digital output. The transmitter in a car keyfob or similar short-range remote is a few milliwatts, so around 1000 times stronger, and everyone knows how short the range of those are, but it is hardly an optimal antenna for both the transmitter and receiver. In unobstructed line of sight, you can reach surprisingly far with very low signal strength, or if the signal is extremely narrowband with the right atmospheric conditions. Radio amateurs have succeeded with impressive feats. A simple cheap UHF PMR446 radio usually reaches only a few hundred meters in normal terrain or city, but go up a mountain and it is possible to reach 100-200km, or longer when atmospheric conditions are right.

Hmmm ... I actually have my car keyfob in the keychain. Can I do something with its high frequency at probably the standard ISM frequency 434MHz or 868MHz? I have no idea if I can reuse anything from Jane and Tom's rental Tesla keyfob, but it should have a couple of crystals as well, and a couple of push buttons. But again ultra high frequency. Everything modern is on even higher frequencies, and often at 2.4GHz.

The absolutely simplest transmitter I can build is to connect the antenna directly to the output of such a digital output crystal oscillator from my Wouxun radio and just supply it with 5V, but it is also not practical for my use even if it is possible to transmit both CW morse code and AM voice with it. The transmitter has too high a frequency, and I still need to build an output stage to increase the power to something usable. But it is a frequency stable fine narrowband transmitter. And tiny.

So to achieve my goal, I need to make better and more efficient radio designs for both the receiver and transmitter sections, and I need to build at least two complete radio units that will be relatively compact and energy efficient. Thank the gods I brought stuff with me that I can take modern amplifying semiconductors from, and it helps that the mansion's electricity is close enough to both 9 and 12V which is the usual voltage in most radio designs. But if the radio is to be mobile and use power banks, that does not work so well and the radio must be a 5V design. Or maybe just use two series connected power banks to get 10V.

Another problem for better radio designs? I have no integrated circuits, crystals, ceramic filters, prefabricated filters, ferrites, or even trim potentiometers and switches. So, any designs that require crystals, ceramic filters or integrated circuits is just a no-go. It will be a very analog design, although the semiconductors I have will make the radios performance good. Simple inductors, capacitors and resistors can be made, and I know that my power resistors works, because the ones I made for the wind turbine have worked fine for three months, and I my newer ones are better.

I borrow from different radio designs to put together a transmitter and receiver. In my phone and tablets radio subfolder in the electronics folder, I find several I can build straight off, as long as I replace and adapt components. I find a pair with a very simple design with a transistor and a mosfet that is switched between receiving or transmitting, but I can not build them because I lack suitable classic crystals, and it doesn't have much output power and would need an amplifier stage, although the regen receiver has good sensitivity. 'Regen' is an abbreviation for 'Regeneration', and is a design developed in the early 1920s to use as few expensive vacuum tubes as possible in a radio receiver, but still provide a sensitive receiver. Mosfets makes such a difference to input sensitivity and amplification that the regen design principle is improved. However, I want to try to avoid dealing with regen receivers, and even super regen. I just don't like them.

A lot of thoughts and designs to contemplate. The spark gap transmitter is basically finished so I make an improvised RF signal strength meter. A moving coil meter with a capacitor, diode and a tuning LC circuit, becomes a simple field effect meter, which helps me to calibrate the maximum output power from the spark gap transmitter, and more accurately determine the frequency as I can better calculate the meters LC circuit. The higher the voltage the moving coil show, the more powerful the transmission is, so by resting the RF meter in a suitable place and reading its value, adjust the transmitter, read the value, adjust, and so on, a good tuning can be done. Since that the antenna will also receive, the antenna should also be tuned for reception as well, although the input stage must also be adjusted to a weak transmitter.

As long as everything is tuned with each other, it doesn't matter that much what the frequency is, but it would have been nice to do it exactly, and it would be far more efficient. Just because it's as good as I can measure, doesn't mean it's as effective as it could be.

It's a bit stressful to build and solder electronics when it's with inaccurate tools and I don't have the appropriate equipment to test it afterwards. I miss my workbench at home. In addition to my soldering irons and soldering station, I miss my signal generator, my oscilloscopes and different meters. Man, my yellow toolbox for 'fixing technology' would have been worth so much to have right now, but of course I left that in the car. I also have a red toolbox for 'mechanical stuff and various jobs', but it was left at home. But I am happy for what I have, and the soldering iron is a vast improvement over the copper rod.

A small problem is how I will determine the frequency of what I build, and thus be able to more easily calculate antennas and resonant circuits etc. It is possible to test and count backwards etc, or try with my RF field effect meter, but that is not particularly accurate without being calibrated to references. So the plan is to build a simple frequency standard with one of the crystal oscillators from the Wouxun radio, so I at least have something with a stable frequency to center everything at.

The crystal oscillators I have from the Wouxun outputs 29.25MHz, 14.4MHz and 32768Hz, where only 14.4MHz is practical to use. But for practical reasons it is still too high frequency, so the plan is quite logically to halve that in to stages, so from 14.4MHz I first get 7.2MHz and then 3.6MHz. A good frequency standard with basically sine wave output makes tuning and making filters so much easier. Another advantage is that 3.6MHz is in the middle of the 80 meter radio amateur band even if the frequency is a little too low compared to what I wanted. Since range isn't the priority for my spark gap transmitter, I can use 14.4MHz to tune my spark gap transmitter and receiver, and higher is better to protect the sensitive lower frequency radios.

So I need to build a transistor solution to lower the frequency, but it is at least an exact output frequency and not variable frequency, so easier to do. It is possible to make a simple flip-flop of two transistors plus a diode to halve the frequency or make it even lower, but I don't want a square wave or pulsed output because it is bad to test frequencies with and give unpleasant harmonic signals, even if it is possible to add LC filters and get a better and more sinus wave signal. But a tuned oscillator just over half the frequency that is triggered in a similar way to the flip-flop can give a better signal, and it can be built with just a transistor and a diode, even if it requires a little more tuning to work and get it right. But it can halve the frequency, and it is possible to trigger one with the output of another, and thus lower the frequency four times. I guess I just have to try to build and see what works best. If I make the divider frequency flexible I could use the 29.25MHz oscillator too, and get more references at 14.625MHz and 7.3125MHz, and there might be other usable oscillators in other stuff to get other frequencies.

Hmmm.... Decisions, decisions.

Ciara sits and press the button on my spark gap transmitter in the meeting room, so I can tune the diode receiver's sensitivity. While I sit and tune, I realise that this is probably the first real radio transmission made here in Alfheimr, but I need to make a contact diode and Alfheimr headphones to be able to say that this is really 100% Alfheimr tech. The headphones will have low impedance of just a handful of ohms, so it's probably best to try to make an output transformer that adjusts from the high ohm output on the diode receiver.

I have added a coherer in the same receiver. A coherer is a non-conductive tube with metal plugs that almost meet much like a spark gap. The space between is filled with metal filings, and I've used iron, but I wonder what is best. Copper? Iron? Tin? Gold? Silver? In normal resting state, the coherer has a high resistance, but when a weak electric signal appear over the plugs, the filings start to 'line up' and conduct electricity much better. They stay like that even when the signal disappear, so you have to tap the tube for the filings to 'come loose' and regain higher resistance. That's why the coherer should be suspended in some kind of springing mount, although the tap doesn't have to be that hard. An electromagnet that is feed current from a battery when the coherer conducts, can then tap the coherer tube, which causes it to stop conducting if the signal is gone. If the signal is still there, the coherer immediately starts to conduct again, and it works a bit like a bell and the tapping makes the receiver make a sound bell or buzzer like.

A radio receiver only needs to be a coherer connected to ground and a long antenna wire, with a battery and a 'bell' arm tapping on the coherer. However, it will react to anything that makes an electric field, so a tuning circuit is needed to make it frequency selective, and other things make the coherer receiver more practical.