John Harvell has pointed out that Corfu, at the entrance to the Adriatic, was a well-defended major Venetian naval base, making it a secure place to put an important radio station. George Iconomou drew attention to the strong Venetian positions on Cythera, Anticythera, Tinos, Kotor, and especially Crete. That raises the possibility of a ground wave relay chain running from Venice to Corfu, then Corfu to Crete with some loss crossing Greece, and possibly on to Tinos. Iver Cooper suggests a look at Malta. Malta wasn't a Venetian possession, but it was under the control of the Knights of Malta, who had even more reason than the Venetians to oppose Ottoman expansion. While there isn't an all-water path from Corfu to Crete, there is one from Corfu to Malta, and from Malta to Crete. So an alternate route for round-the-clock ground wave relay service is Venice to Corfu to Malta to Crete to Tinos. Without Malta, we run into the complication that the best place on Crete to cover the eastern Mediterranean is at the eastern end, but the only path to Corfu that doesn't cross a long stretch of land is from the extreme southwest corner. (Also, Crete is long enough to lose a lot of signal strength on a lengthwise path, so the island would need two stations for optimum results in all directions. We'll analyze all these paths and see what the numbers tell us.)
Taken together, stations at those places would give the navy and Venetian interests full-time radio communication over a large part of the Mediterranean.
From Corfu: the Adriatic, the Ionian Sea, the "toe" of Italy and the southern Tyrrhenian Sea, and the central Mediterranean between Sicily and Greece across to Libya
From eastern Crete: the Aegean and nearly all of the eastern Mediterranean except the region beyond Cyprus
From Tinos: stronger communication into the northern Aegean; the ability to hear weaker ship transmitters
From Malta: Sicily, Sardinia, Tunisia, and the passage between Sicily and Tunisia
From Venice itself, not only the whole Adriatic, but limited coverage in the northern Tyrrhenian after land losses crossing Italy
Let's look at the distances involved, and do the path loss and power calculations for the longest and shortest paths of interest. We'll omit the calculations for some intermediate distances, since they're bracketed by other calculated cases:
Corfu to Malta, 636 km
Malta to western Crete, 819 km
Crete to Tinos, 262 km
The Corfu to Crete calculations are more complicated than the rest; they're mostly over water, but the middle part is over land. The losses in the successive legs of a mixed path can't be simply calculated as if they all started at the transmitter, because the path loss is the product of two different mathematical functions, which vary differently with distance and frequency. The geometric spreading factor follows an inverse square law modified by the earth's spherical shape and is independent of frequency, while the resistive energy loss follows a decaying exponential function and degrades rapidly with frequency. Instead, each leg's loss must be determined from the loss curve for the specific type of surface it crosses, over the specific portion of the path it covers between the transmitter and the receiver.
The path between Corfu and Crete that crosses the least land is from points on the extreme southwest of each island, from near Chalikounas to Elafonissi, just skimming the west coast of Greece for much of the way and occasionally crossing a short bit of projecting land.
On the other hand, the best place on Crete to cover the eastern Mediterranean and up into the Aegean by ground wave is on the eastern tip near Kato Zakros, but crossing much more land on the way across Greece.
We'll calculate both paths and see what we get.
We won't consider shore stations in the Aegean, because all the shores of that sea are in hostile territory. Defending a shore station there would be impractical.
Clearly, there's no difficulty hearing a coastal station at either end of the Adriatic, anywhere in that sea. Further, at this distance there's no need to resort to the 500 KHz band and the tall transmitting antenna it requires, if all we want to use is ground wave over salt water. In fact, if that band were used, a short antenna with its reduced gain and efficiency would be more than sufficient, without requiring a powerful transmitter.
If ground wave is all we're interested in, all of these paths could reasonably be covered on 2 MHz, avoiding the need to build anything taller than a ship's mast.
Ground wave ranges for ship transmitters
Next, we'll calculate how far ships could hear each other above the noise level in the Mediterranean. They're more limited in antenna space and transmitter power than shore stations.
In the non-canon vacuum tube timeline, it's estimated that a 25-watt transmitting tube good up to perhaps 10 MHz could go into pilot production some time in 1635. That would be suitable for a ship's transmitter running off batteries charged by a masthead wind generator, as depicted in "A Friend in Need." A simple 250-watt tube could be available not too long afterward; four of these in a push-pull/parallel combination would make possible a 1 kilowatt transmitter, which would be manageable in ships with engines.
So let's look at how far a marine transmitter could be heard at these two power levels. The following table assumes that:
On the 500 KHz band, transmission is with a short vertical.
On the 2 MHz band, the ship is large enough to mount a quarter-wave vertical; for example, a Fearless-class 100-foot escort schooner. Smaller vessels can use a short vertical, but we won't make that calculation.
On the 4 MHz band, all ships transmit with a quarter-wave vertical. (Higher vertical directivity might be possible on shipboard, but could actually be detrimental during severe rolling and pitching.)
Ships receive with their transmitting antennas.
Ereqd here is the calculated signal strength in db relative to 1 microvolt per meter, required to produce a +16 db signal-to-noise ratio in a 100 Hz bandwidth at the stated regional noise level, at the given frequency. With that, we can apply the ground wave curves and antenna gain adjustments to determine the maximum range at the stated power.
The major path distances of interest in these two seas are:
Comparing, we see that ships can communicate with each other the length and breadth of either of these seas with a kilowatt on any of these bands. Furthermore, they can do it with 25 watts on 500 KHz, and not a great deal more than that on 2 MHz where the antennas are simpler and more efficient.
The one real reason to equip ships in this theatre with a top-hat antenna and an antenna tuner that can reach down to the 500 KHz band is to give them night-time seven-league boots, so they can reach out to distant ships and shore stations and over land obstacles.
Again, these numbers assume a signal barely strong enough to copy during the noisiest time of day and season, with some effort by the operators. Authors can adjust with higher power, better receiving antennas, more favorable hours, or shorter distances for more solid communication, or they can choose greater distances, lower power, and worse antennas if they want lower speeds, multiple attempts to get a message through, or chancier efforts. Or they can afflict their characters with a powerful thunderstorm at a critical moment or make a tube expire.
Radio navigation
Communication isn't the only use for these radio frequencies. As we saw in "Storm Signals," radio signals can be used in many ways for navigation. For simple direction finding, two shore stations on bearings roughly 90 degrees apart as seen from the ship can provide a reasonably accurate position. The ship's navigator can take bearings on two or three beacon stations using a direction-finding loop, or a couple of manned shore stations can take considerably more accurate bearings on the ship's transmitter, undisturbed by pitching, rolling, and yawing.
A considerably better S/N ratio than +16 db would be desired for navigation, but as we can see, a 25-watt beacon or ship transmitter would be more than sufficient at the distances the Adriatic allows. Even 5 watts would put in a strong signal across the width of that narrow sea.
Two challenging cases: MF
/HF ground wave between fixed stations across land
Before we go on to sky wave, there are two ground wave paths worth analyzing to see what we'd be up against if we want 24/7 communication. These are the USE-to-Venice route, and the shorter path from Corfu to the Aegean. For each of these cases, we'll use the "poor earth" signal strength curves, but the same full-size fixed station antennas described above.
As canon has developed, there's a good working relationship between the USE and the Venetian Republic. Even without a cooperative effort with the USE navy, Venetian merchant interests could well have a large MF station on the air by 1635, handling overnight commercial message traffic with the USE, the Netherlands, the Kalmar countries, and possibly France.
With all the chaos and war elsewhere in the Italies, mainland station sites anywhere south of Venice are a chancy proposition. That's an issue for the mainline authors to wrestle with, however. But Corfu is ideally placed to punch a signal eastward across Greece to Velika on the Aegean coast and out to sea. The actual path is a short hop across water and then the long land leg, but nearly all of the loss will occur on the latter, so we'll just do the simpler calculation for that, to get a ballpark estimate.
Obviously, nobody in 1636 is going to be building a 2-megawatt transmitter. The 40 KW ganged alternator estimated to be necessary for the message from Vlissingen, Netherlands to Cantrell's squadron in the West Indies was pushing it. So forget 24/7 ground wave contact between Venice and navy headquarters in Lübeck, for relay to the fleet. Even from further south in the USE, the power requirements would be unreasonable for the first decade NTL.
From Corfu to the Aegean, though, ground wave looks like a very practical proposition, if a 500 KHz antenna can be built there in time for the action. Driven with a kilowatt, ships anywhere in the Aegean should hear it, even behind islands. A ship would need about twice the power to cover the same path as a land station with a full-size antenna, or about 320 watts. Corfu should be able to hear a ship transmitting with a kilowatt, well out into the Aegean. It certainly adds to the strategic importance of Corfu.
MF Sky Wave
And so we come to sky wave at 500 KHz, to connect Venice with the rest of Europe's growing radiotelegraph backbone―and USE Navy headquarters. As explained in the previous article, the low- and-medium frequency spectrum below about 700 KHz is where reliable nighttime sky wave will be found during the decades of the quiet sun. The 500 KHz region is something of a sweet spot for the mode: reasonably good propagation, moderate atmospheric noise levels, and near the practical limit for full-size vertical antennas supported by wood lattice towers. We can expect both the marine service and the fixed land communication service to make heavy use of that part of the spectrum in the late 1630s.
Sky Wave from Venice
There's no question that Venice's builders could put up a quarter-wave transmitting antenna for 500 KHz. Obtaining an unobstructed flat site big enough to accommodate the 300-meter diameter wire ground plane and defensible against pirate raids might complicate the task, of course. What's more of a question is how powerful a transmitter they could bring in by the time the trouble starts. 100 watts should be easily obtainable by the middle of 1636; a kilowatt tube transmitter might be harder to get and harder to supply with power. (An RF alternator like the one built in 1633 in "Canst Thou Send Lightnings?" is a possibility, but not an easy one to operate and maintain.) So let's see how far they could get out a usable signal, for each of the two vacuum tube cases.
As in the previous article, we have only the U.S. Navy curves for sky wave up to 200 KHz, so we must extrapolate to get an estimate for 500 KHz. One thing that helps is that whereas the published curves are for shortened vertical antennas, our hypothetical fixed station antennas at 500 KHz are full-size. The initial signal estimate in each case is for a 1 KW transmitter.
Why not do it at 200 KHz? That's possible, but it would mean a shortened transmitting antenna, which requires a top hat to optimize the current distribution on the radiating wire. That requires a minimum of two towers to support the ends; three to six would be better. The BBC at 198 KHz uses two; NAA at Cutler, Maine around 20 KHz uses seven per antenna. A ship normally has two or three, the fore and mizzen masts supporting the top hat and the mainmast carrying the radiating wire.
Interestingly, these cases all come out with just about 10 db more signal than is needed to achieve the target +16 db S/N. If we look further at the 100 KHz and 200 KHz curves to find the distance where the loss is 10 db greater than it is at 945 km, which would take us down to that +16 db target with a 1 kilowatt transmitter, we find 2044 km (1270 mi) at 100 KHz and about 2012 km (1250) mi at 200 KHz. That relative relationship probably holds at 500 KHz as well, because the decrease in signal strength with distance on a single-hop path is mostly due to geometry. The path is through air, which is nearly lossless, and only the ionospheric reflection factor varies with frequency.
So if Venice gets a full-size transmitting antenna and a kilowatt to drive it, what else is within 2000 km, and should be able to hear its transmissions with a simple receiving antenna? Vaasa, Finland. Trondheim, Norway. The Shetland Islands. Lisbon, Portugal. St. Petersburg, Russia. Ankara, Turkey. All of the USE, the Netherlands, Bohemia, the Austro-Hungarian Empire, England, Scotland, Ireland, France, Spain, Switzerland, the Italies, Poland, Lithuania, Belarus, and Ukraine. Nearly all of the Mediterranean and a good stretch of North Africa. In short, Venice gets plugged into the European commercial message system and can manage its business affairs everywhere the net reaches.
Now, it also happens that Corfu is 1730 km from Lübeck. So if there is a fleet unit operating in the Aegean, with 24/7 ground wave contact with Corfu, Corfu will be able to relay message traffic direct to USE Navy headquarters as soon as night comes, without needing a relay through Venice.
Now, what if the transmitter puts out only 100 watts? Then, according to the sky wave propagation curves, the coverage zone at +16 db S/N is a ring, starting at about 200 km and ending at about 1000 km. (Sky wave skips over nearby terrain, because there's very little power radiated off the end of the antenna, and because ionospheric reflection works best at shallow angles.) This tells us that to play in the sky wave arena at 500 KHz at all, anything under 100 watts is useless for reliable commercial service―unless we're transmitting to a mature station with much more sophisticated receiving antennas, meaning large and expensive. That's possible, but not likely by 1636.
Sky Wave Offshore
There's one other interesting case to think about. What if there isn't a shore station in Venice, but the USE does have a steamship in the northern Adriatic? The shortened 500 KHz ship antenna isn't as good as a full-size antenna on a well-constructed ground plane ashore, but it isn't dramatically worse. If we think a Venetian coastal station could reach Lübeck at night with 100 watts, a ship should be able to do it with 200 watts. A kilowatt should provide plenty of margin. So even in this case, the fleet could communicate at night with Navy headquarters.
Tactical voice communication
Up-time CB units would probably be hard to find by this time, even for the Navy. New-build short-range tactical voice transmitters and receivers wouldn't strain Grantville's tube production; it's more a question of whether the industry's very limited technical manpower would have time to design and debug all the necessary components in time for the Mediterranean campaign. This could lead to voice radio equipment being in short supply and temperamental, forcing the fleet to fall back to 4 MHz Morse Code for urgent battle coordination. It's another variable authors could manipulate for story purposes.
Summing up
With the equipment coming out of Grantville's industries by the time the fleet heads for the Mediterranean, they'll be able to communicate. Within the theatre of operations, full-time Morse Code communication is easily feasible. Overnight communication is possible with home, at levels of difficulty that could be adjusted for story needs. Night-time communication back to the USE for ships in transit off the Atlanti
c coasts of France and Portugal may be more difficult, but should be possible. One-way transmissions from the high-powered station at Vlissingen, Netherlands could be heard far out to sea.
Notes from The Buffer Zone: The Past is Another Country by Kristine Kathryn Rusch
Lately, when I look at science fiction, I find myself looking at the past.
Some of that is my age: I’m now old enough that the world of my childhood is so far distant from this world that it actually needs explanation to anyone under the age of forty. I remember when World War One vets marched in the Veteran’s Day parade—when old people were people who were born in the nineteenth century, not the twentieth. I knew some of the pioneers of the modern age of science fiction. Some were my instructors and a few became my friends.
I met others at a conference, and got to shake their hands. And a few others I watched from afar, too nervous to even try speaking to them.
Another reason that I’m looking to the past is because it’s being forgotten. In September, an anthology I edited Women of Futures Past appears. It shows how women science fiction writers have always been with us and, more importantly (at least to me), how they influenced what’s being written today. Early next year, I’ll be Kickstarting some anthologies that will reprint lost stories from sf’s past—award-winning and award-nominated stories by authors whose names you might not have heard of.
Grantville Gazette, Volume 66 Page 19
