Midwinter calling, day 11

Radio ga ga

The tests on the finest inner diameter glass tubes have given very positive results so useful mercury thermometers are possible, and I also try my two lab ovens for the first time and test how they work. They get really hot, and I heat up the best insulated one to 500 Celsius and make sure it's at that temperature for an hour to tests who slow it then cools down. Both seem to work great in any case and I'm glad I chose the two gas thermometers with the highest scale for the ovens, and I'm also glad that the workshop window can be opened a bit to let out the waste heat. There is also the issue of having a fire indoors, and carbon monoxide/dioxide poisoning. I will need to do a couple of proper top range and cooling tests by lighting the fires in the morning and setting alarms in my mobile phone every hour and recording the temperature of each oven as it cools down during the day and creating a spreadsheet. But the ovens need to be really thoroughly warm first.

In the afternoon, there are a few fairly small electronic modules on my workbench, as I've also soldered together a preamplifier for a microphone and made a simple receiver with only two mosfets to be in my office. But it's enough for an official test, and I get Jane to help, which she is happy to do. I leave most of my equipment in the workshop with clear instructions to Jane what to do, and what not to do, with written notes and arrows. I have also made an ugly mod so I can record on the tablet using the microphone input. We shout instructions between us, with Caecilia in the middle as a repeater. After we changed places and I fiddled a bit with the equipment, I can clearly hear and record the morse code, the 'Shave and a haircut' signal and Jane's voice coming from the speaker.

WO! HO!

Of course, everyone else is wondering what the hell we are doing, and as I shout to Jane to come up, there are many curious people. I already have Kari and Ciara sitting in here, who looks very confused as they've heard Jane's voice in here, when she is clearly not, and it doesn't seem to be a recorded video. They've both learned what that is. So I use the tablet's built-in speakers and Jane hears what she just said, and she gives me a careful hug and rewards me with a: "Congratulations Robert!".

I'm very proud of my build and gladly accept it, even though I'm a little uncomfortable when she does that. I also notice Caecilia's quick look on my pants turn disappointed.

When it comes to the radio design, I cheat like hell by partially using high tech components, especially irreplaceable semiconductors, but to actually be able to build something that works with primitive soldering tools, not optimal components and something I improvised...

Yeah, I'll admit I'm proud of this. This is so incredibly important.

The others see we're excited to hear the recording of Jane's voice, but they don't really understand. So I ask Ciara and Kari to go out and whisper something secret to Jane, so Jane can run back down. Jane starts giggling and laughing in the corridor and I hear her thunder down the stairs, and I hope they didn't say anything too embarrassing. I start recording, turn up the volume so that the static is clearly heard from the speaker, but all of a sudden the static disappears and Jane's clear voice cuts in, and after a quick giggle;

"Jane speaking in the workshop to everyone listening in the office. Kari said; 'I love Robert'." Jane switch to English: "Robert, you really have to think about what you say to Ciara because she took it literally that she should tell me a secret!" Jane starts laughing over the radio, and I facepalm while I wonder what secret. In a teasing voice Jane sings: "I know there is a secret cellar! I know there is a secret cellar! Sneaky bastard! Now you have to show me!"

Jane continues to laugh and happily sing 'We are the champions' through the radio before she says she's coming back up. Everyone is staring, looking at each other and the stuff in front of me, and some are looking out through the corridor in the workshops direction. Total confusion, not least considering what Jane said and how she laughs.

"Jane talks in my workshop in what I built on the bench, and we hear her here. It's powerful and incredibly advanced technology. Ciara, it's my fault that I was vague and I'm not mad at you, but I did not mean that you should literally telling something that's secret to Jane."

Ciara realises her mistake and instantly turns horrified and tear flowing sad, so I wave her to me and say it's okay while I hug her in my lap. She did nothing wrong; it was me who was unclear. Kari and Alith gives me and Ciara annoyed looks. And they're probably wondering what Ciara said.

Anyway, the radio test is a great success, even if the others doesn't really understand how important this is. But I need to build another transmitter and two improved receivers to really be able to test, and I need to add the final stage amplifier, and better match the output. But the oscillator seems very frequency stable, and it is possible to set frequency about 10% plus-minus, and CW (Continuous Wave) for morse and AM (Amplitude Modulation) for voice works. I need to make the radio units matching in adjustment, and make all the switches, housings, proper antennas, etc etc. But it works. Chances are quite high that we just made the first real voice radio transmission in Alfheimr.

I let people go down to my workshop and say something that the others can hear, and they run back and forth for a while, and Jane is smugly responsible for making sure no one is touching anything and handle the PTT (Push-to-talk) button. I also let the staff try, and I take a couple of photos as memories. A good one of Alith talking into the microphone, and another where Kari and Gunhild attentively listen. I record them too, but don't play it for them - there will just be too much explanation that you don't sound like you hear yourself and so on.

Jane will sketch a whole pile of these pictures and add it to the chronological books. It is not possible to create photo albums, so that is the closest, and Jane is good with a quill. I've already whispered to very smug Jane that I'll show her the secret tomorrow.

This is a day to celebrate, and the staff is preparing a small feast while we go outside to really test the spark gap equipment in nature. I want to test how much long antennas with grounded transmitters and receivers affect the range compared to two dipole antennas suspended in trees, and a simple loop antenna. Ciara's questions about helping like she did yesterday, leads to a fun reaction from our company, and I again comfort Ciara that yesterdays test wasn't secret. It is an interesting test that shows that the house wasn't close to the maximum range, and the spark gap transmitter is overkill to just show that it works. I explain why these devices that are so much simpler isn't something that I will continue building on. However, Jane is very impressed that I've actually made 100% Alfheimr radios, and with some duplication of components and lead acid batteries, 'short-range' morse 'walkie talkies' is possible to make. I point out that a simple remote control would also work, and there were simple toys made in the 1950s that used similar systems, as the coherer's electromagnetic reset can power an electric motor, or activate a cam system to cycle through functions.

It is annoying that the most difficult thing about building electricity and electronics here - beside component deficiency - are insulators. Wood is not good enough in many cases, and insulators are needed everywhere. In all sockets and switches. In variable capacitors to separate between the moving part and the fixed part, and to prevent contact with the hand to parts that are a part of the radios circuit, for the human body is a large bag of water and can affect the tuning. All cables needs to be insulated, and there is no plastic or rubber here. Well, not enough of it. Add coils where the copper wire is up against the next turn. It's just one problem after the other. Air as an insulator only works sometimes, and everything can not be done in ceramics, glass, lacquer, paper and cloth.

What solves a lot of the more complicated insulator problems is that Jane and I are modern people, and we have lots of plastic things with us, even if everything isn't useful. A lot of plastic can be heated, melted and reshaped. Thank goodness for waste plastic from credit cards, headphones and so on that I can use for some insulation problems. But plastic is just a very limited solution, and I really need those earthenware insulators.

The lower frequency a radio use have, the larger the value the coils and capacitors need to be, and this means a variable capacitor with many more or larger metal surfaces that should lie stacked close to each other, the closer the better, but not touching each other. This will be for low voltage so air works as an insulator as long as there is no condensation on it. A coil needs to have many windings and it often has to be insulated, but I have lacquer paint for that, and the copper wire I bought is painted. I can not use any of the small surface mounted coils from the VHF and UHF parts of the radio, since they are nano or micro Henry. I need larger values, a thousand to a million times higher inductance in the millihenry area, but luckily that don't always mean a million times bigger. So I will wrap my own coils without ferrite core because I lack that too, and using lacquer, paper, cloth or glass.

I need to make some kind of iron core transformers in the future, and test the frequency response and see if I can improve the iron core with some method. What exactly is a ferrite core? My best guess is some kind of 'ceramic' made from iron or rust along with more exotic elements, but I know that they don't conduct electricity. So I'm probably boned on making that, but I'll probably need to find ores and just try. That's going to take time and be very annoying work that I hopefully can pay someone to do.

Each problem with components must be solved separately. For example, I can tune in with a modded trim potentiometer until it works well and seem close to ideal, then measure and recreate that resistance. But changing the type, size and construction of the component effects the performance and need, so I have to test for the same performance. So possibly with several positions via switched resistors where the design needs it. Not optimal. So I choose designs who avoid potentiometers, although for things like Volume and Squelch multi-position rotary switches works fine. I just have to think about everything failing gracefully when there is a problem, so the highest value is the default, and minimizing any spark problems.

I've known about this problem for months, and it's a good idea to standardize and mass produce exactly the same multi-position rotary switch, and make it so that the switch can be changed to different configurations. Like the common solution by limiting the steps via removable blocking pins. One of the switches I've designed and sent templates and sketches to the potter, is a 12 position rotary switch. The design have a variable amount of contact plates so the same design can be changed from 12 position 1-pole switch to 6 position 2-pole, 4 position 3-pole, 3 position 4-pole, or a 2 position 6-pole. Just change the contact plate, connect the inner parts as needed and move the blocking pin. One and the same basic switch can then become a part that is modified depending on the function, instead of having to have several different parts made. Which is also great for spare parts. Mass production also allows me to streamline production with jigs and templates so production goes faster and easier. In addition to all possible combinations, a 12 position rotary switch also allows the volume knob to go to 11.

The contact plates can also change the function from break before make, to make before break. A switch that changes current or functions needs to disconnect before the new contacts are connected to prevent a short circuit, while for example, a volume control and squelch should make contact before the previous contact is disconnected so that the volume doesn't jump to 0 at the actual switching moment. For those two examples it's not actually important, but for other future use it might be very important. If I use the rotary switch as a mechanical position encoder, that will matter. For rotary position encoders, 'gray code' is often used.

The manufacture of capacitors is in any case quite easy because I can both calculate them, and measure the capacitance with the multimeter, and the metal is less critical, rigid, solderable and easy to file. I am not limited to Leyden jars. Hopefully the stackable earthenware insulators for the M4 and M6 screws will be good, but it must be the absolute smallest and weirdest earthenware that potter Ruskva has ever made. And I want many dozens of them.

My final choice for general use and testing in the 3.6MHz range works well as a starting point, and it is likely that I will add dual band function around 7.2MHz as well, and maybe triple band to get 1.8MHz too. It is not particularly difficult to do with the design I have chosen, and I worry that I will need more frequency options since the ionosphere and sun will cause issues. I want a more reliable radio contact and for that I expect I need options. So there will be space inside and on the front. However, it still feels unnecessary and a complication if it doesn't turn out to make a significant difference in radio contact reliability or range, and there is all the issues with tuning and being on different frequencies. Standard frequencies, a schedule and watches will be needed. Maybe I should just make a radio with several fixed frequencies and fine tuning?

It's likely that I'm the only one who will be building radio in this world during my lifetime, and the Q factor on the parts, ie how narrow range the radio's filter has, is probably 'not good', but is not likely to be a huge problem and makes tuning easier. Narrowband antenna affects, but we should in principle use only one frequency. I have a good IF filter I can try to use on one radio, but it feels like a bad idea making a special solution when I can not repeat it on other radios. Although it might be valuable with the most sensitive and selective receiver as the mansions permanent radio if I build several more radios in the future, which is very likely. I will have components for it and radio is very useful. A couple of walkie-talkie style radios would be nice, and I can certainly find a use for all radios I can manufacture. Hell, that spark gap walkie-talkie would be very useful.

I find suitable wiring diagrams for 3.6MHz in my radio directory because 3.5-3.8MHz is a radio amateur band, but I also have plenty of diagrams and examples of 1.8MHz, 7MHz and 14MHz, and a few others far higher up in frequency on radio amateur frequencies, and those bands are generally in the following double or triple frequency. Harmonics then interfere with other radio amateurs instead of commercial traffic, and there are frequency and antenna advantages. But it is often easy to tweak a circuits frequency range, if you understand how the circuit work. Being a bit of compulsive 'always saving stuff, since it might be useful or might have gone offline' is now turning out to be useful. The antennas gets a little unmanageably large, especially if I need multiple antennas on the ship which is a little too small for it. A good thing is that the ship's antennas will be over salt water which is a good conductive surface and helps, but it is a future problem.

I will finish the frequency standard and give it a nice case, and also make my simple RF field effect meter into a practical well-built measuring instrument. Making a sensitive moving coil meter is a lot of work, and important components, so I should probably make a dip oscillator meter out of it as well to get more functions. It's quite easy even if I make the instrument cover a few different ranges from 1MHz up to 30-40MHz or so, but without good calibration references it won't be exactly correct. I can't measure inductance and just a limited range of capacitance, but what I can do is work from one known component that resonate on a known frequency, and calculate the unknown components value. That can then be verified and used as a reference for another. I will get errors, but it's better than just assuming. I might mix crystal oscillator frequencies to get interesting sums that I can filter out and use as a reference for calibration.

I let Ciara spend the evening and night with me so she knows that the mistake was my fault, and I make sure she knows that she didn't break the concubine agreement's rule of secrecy, but Ciara is very sorry and really tries to make up for her mistake, so it's a nice ending to the day.

Radio now exists in Alfheimr, and I'll make much better radios. Even if I'm cheating. Making vacuum tubes will be a true pain in the butt if I even ever succeed, and I have so many machines and tools to make until then. The lathe has only been slightly improved but is basically useable, and I have barely done anything on the other machines, and then there is all the material testing.