13

Even as far back as the early 1900s, telegrams transmitted wirelessly could reach hundreds of miles. For instance, the Titanic communicated with Canada, 400 miles away, with relatively low-powered equipment. Given that telegraphs are very simple, how could these pulses have traveled so far?

And would these pulses still travel that far today with the same equipment?

And doesn't this mean that there couldn't have been very many people using the systems, since operators within hundreds of miles would all be jamming the airwaves? It seems this would produce a lot of cross-talk. Or were there multiple frequencies available for wireless telegraphy?

InterLinked
  • 373
  • 2
  • 16

5 Answers5

23

the Titanic communicated with Canada, 400 miles away, with relatively low-powered equipment

Quote from this website: -

The Titanic's "wireless" equipment was the most powerful in use at the time. The main transmitter was a rotary spark design, powered by a 5 kW motor alternator, fed from the ship's lighting circuit.

The equipment operated into a 4 wire antenna suspended between the ship's 2 masts, some 250 feet above the sea. There was also a battery powered emergency transmitter.

The main transmitter was housed in a special room, known as the "Silent Room". This room was located next door to the operating room, and specially insulated to reduce interference to the main receiver.

The equipment's guaranteed working range was 250 miles, but communications could be maintained for up to 400 miles during daylight and up to 2000 miles at night.

enter image description here

So, if you class 5 kW as low power then that's OK but things have moved on since then. For instance, as tubes/valves were developed radio receivers became more sensitive and this means that transmit powers could reduce considerably.

You have to realize that these transmissions are actual electromagnetic waves and they attenuate only very gradually with distance. For instance, comparing against a contactless battery charger, its magnetic field reduces with distance cubed beyond about the diameter of the coils whereas, the H field in a proper EM transmission reduces linearly with distance.

Just consider the Voyager 1 probe and its transmissions from beyond Pluto. The transmitter power is only 20 watts but the biggest thing on it was the parabolic dish: -

enter image description here

And doesn't this mean that there couldn't have been very many people using the systems, since operators within hundreds of miles would all be jamming the airwaves? It seems this would produce a lot of cross-talk.

This was indeed a big problem and there was a famous transmission from RMS Titanic that suggested that SS Californian should "shut-up" because it was blocking a transmission from Cape race on the Canada coast: -

Titanic's on-duty wireless operator, Jack Phillips, was busy clearing a backlog of passengers' messages with the wireless station at Cape Race, Newfoundland, 800 miles (1,300 km) away, at the time. Evans' message that SS Californian was stopped and surrounded by ice, due to the relative proximity of the two ships, drowned out a separate message Phillips had been in the process of receiving from Cape Race, and he rebuked Evans: "Shut up, shut up! I am busy; I am working Cape Race!" Evans listened for a little while longer, and at 23:35 he turned off the wireless and went to bed. Five minutes later, Titanic hit an iceberg. Twenty-five minutes after that, she transmitted her first distress call.

Quote taken from here, the Wiki page for the steamship Californian.

Andy aka
  • 434,556
  • 28
  • 351
  • 777
  • So did wireless telegraphs operate at very low frequencies? (Since the lower the frequency, the farther it transmits?) – InterLinked Jan 04 '17 at 13:22
  • 2
    @InterLinked Titanic operated around the 1 MHz area and ionospheric bounce allows radio reception at much greater distance than line of sight would imply. At 250 foot high, the line of sight is only about 20 miles and clearly Titanic could transmit and successfully be received at around 400 miles during day time. Other than the ionosphere, lower frequencies don't actually transmit any further than higher frequencies. – Andy aka Jan 04 '17 at 13:25
  • 2
    Modern ham radio operators communicate around the world with 5mW (yes, milliWatt) transmitted power. – Jon Custer Jan 04 '17 at 15:09
  • Aren't those 5mW raidios handled by repeaters? That's a really low power signal to get several thousand milles without massive noise. – Matthew Whited Jan 04 '17 at 15:48
  • Even that 5mW would have a hard time crossing a house let alone the world. Are you sure you didn't mean .5W or 5W? – Matthew Whited Jan 04 '17 at 15:58
  • 1
    @MatthewWhited YOU need to address your question using the "@" and the name or he might not get a notification to look at these comments. As author of the answer I get notifications and I'l also interested in his response. – Andy aka Jan 04 '17 at 16:01
  • @JonCuster, even bluetooth is 100mW. Could you provide details on that 5mW world wide communication? Intercontinental Shortwave takes over 500kW. Cell phones are around 600mW (I can't find any radio transmitters under 100mW.) – Matthew Whited Jan 04 '17 at 17:23
  • 1
    @Matthew Whited Yes, please do research a bit on the HF propagation. Power levels of 5 mW are actually used for intercontinental contacts. Usually, such low levels are not used for telegraphy. Instead, digital modes with very high level of error correction coding are used. Furthermore, if you look up how digital modulations work, you'll see that many receivers use the "integrate and dump" technique. Received signal strength there depends on the bandwidth and the symbol interval. By using extremely low bandwidths and very long symbol intervals, you can make up for that. – AndrejaKo Jan 04 '17 at 17:24
  • @AndrejaKo - if you consider Morse to be a digital mode! One of my friends has made QRP Morse code contacts (i.e. <5mW radiated power) to every continent from North America, and also from Africa to North and South America as well. It all depends on the propagation in a given band at the time. – Jon Custer Jan 04 '17 at 17:30
  • I can understand how one could construct a transmitter without vacuum tubes, using a spark gap as the amplification element, but I can't think of any way to design a sensitive but workable receiver. The best I can think of would be to have a tuned circuit with high voltage gain attached to a spark cap along with a DC bias-voltage supply that is adjusted to be a smidge smaller than the ionization voltage for the gap, and has enough resistance and capacitance to make an arc unsustainable. I don't know that such a thing would be practical and workable, though. – supercat Jan 04 '17 at 17:30
  • Even looking at that the best I see is 250miles with perfect conditions. – Matthew Whited Jan 04 '17 at 17:31
  • @Matthew Whited Also, the 500 kW that you mention are for stronger international broadcasters that are broadcasting radio using AM for listeners that are using horribly inefficient whip antennas that come portable shortwave receivers. Maritime mobile radios (for example Icom M802) are often 150 W. Land-mobile short-wave radios are around 125 W (for examole Codan Envoy). Amateur radio transceivers are usually, 100 W. In addition to that, instead of traditional AM, a modification called "single side-band", which removed the carrier and the other sideband of AM, is used. – AndrejaKo Jan 04 '17 at 17:38
  • @Matthew Whited If you're so untrusting, why don't you go to say http://websdr.ewi.utwente.nl:8901/ a web radio receiver in Netherlands and see what you can hear there. It can regularly hear ships calling Shanghai radio CHN for example. – AndrejaKo Jan 04 '17 at 17:42
  • 1
    It's all down to antenna gain (both ends) and signal bandwidth and having a good receiver. – Andy aka Jan 04 '17 at 17:44
  • You can have all the gain in the world but if you signal is so weak that all you get is noise it really won’t matter. But if you know your encoding and have ECC you should be able to recover so what eves... just sounds horribly inefficient. if you are doing it just because you can then congrats. but this doesnt seem like off the shelf tech. – Matthew Whited Jan 04 '17 at 17:53
  • 2
    Theoretically, a receiver at room temperature can receive data at 1 kbaud (if properly designed) with an input power level of -124 dBm. At 1MHz, link loss is 32.5 dB + 20log(km). So let's say 10,000 km and therefore link loss is 112.5 dB. With 0 dBm (1 mW), receive power is -112.5 dBm and significantly higher than the power needed by the receiver (on a good day). Throw in some antenna gain and nearly every day is a good day: http://electronics.stackexchange.com/questions/83512/long-range-15-km-low-baud-rate-wireless-communication-in-a-mountain-environme – Andy aka Jan 04 '17 at 18:10
  • @Matthew Whited The "off the shelf tech" comment is quite interesting. This technology isn't and never was "off the shelf tech". The technology has been well-understood, and "common", but it wasn't really off-the-shelf. I'd more compare it to the space technology of current era. Yes, it's quite known that satellites exist and a common person has some sort of idea how they work, but they're still unique and require specialized knowledge to construct, launch and maintain. It's the same thing with that era's radio. cont. – AndrejaKo Jan 04 '17 at 23:30
  • Try looking at for example the [Radio Central project](http://earlyradiohistory.us/1922RCA.htm), or at [Varberg Radio](https://en.wikipedia.org/wiki/Varberg_Radio_Station). Even today, the shortwave long-range coastal stations have relatively large antenna installations in order to have sufficient gain for reception. The receiver antennas are often located at remote sites, in order to have low local noise floor and use directional antennas in order to have gain in the right direction for reception. Take a look at this guide from STO http://aviolinx.com/Documents/STORadio-Pilot-Refresh.pdf – AndrejaKo Jan 04 '17 at 23:38
  • "This room was located next door to the operating room, and specially insulated to reduce interference to the main receiver". What kind of technology was used for the receiver? – user1139880 Jan 05 '17 at 04:04
  • Indeed, with current day receivers the range of the Titanic would now be much more than 400 miles. Maybe even reach around the planet. – vsz Jan 05 '17 at 07:32
  • can you elaborate on _"the H field in a proper EM transmission reduces linearly with distance."_? From what (relatively little) I know of RF this sounds counterintuitive, assuming an isotropic transmission – Alnitak Jan 05 '17 at 07:48
  • @Alnitak power reduces with distance squared therefore the two things that multiply together to make power must each reduce linearly with distance. This is the isotropic scenario. So, do you have issues with power being inversely proportional to distance squared? – Andy aka Jan 05 '17 at 08:24
  • @user1139880 Given that probably 5 mW transmit power could be used today, a power of a million times higher back then speaks volumes about how insensitive receivers were. For sure they used tuned circuits and they had strange apparatus for envelope detecting a received signal based on a magnetizable rotating strip passing by magnets that saturated the strip locally and, then the superimposition of the received signal (again using coils) means that the non-linearities of saturation could act like a diode. The detected signal was picked up by a third coil and fed headphones. – Andy aka Jan 05 '17 at 08:33
  • @Andyaka power inversely proportional to distance squared is what I know and expected. I'm not used to thinking in terms of EM fields instead. – Alnitak Jan 05 '17 at 09:51
6

From http://hf.ro/ :

The Titanic's "wireless" equipment was the most powerful in use at the time. The main transmitter was a rotary spark design, powered by a 5 kW motor alternator, fed from the ship's lighting circuit

A spark gap transmitter is the simplest possible form of radio transmitter, modulated with on-off keying (morse code). Even allowing for the inefficiency of spark gap transmission - it sprays RF across a very wide band - a 5kW transmitter is huge.

pjc50
  • 46,540
  • 4
  • 64
  • 126
  • The spark gap itself produces a very wide bandwidth, but the antenna acts as a resonant filter. – WhatRoughBeast Jan 04 '17 at 13:42
  • 1
    According to Wikipedia, a 5KW transmitter is illegal in the United States - even for ham operators... - https://en.wikipedia.org/wiki/Amateur_radio#Privileges – InterLinked Jan 04 '17 at 15:36
  • 6
    At the present day, yes. Back then there weren't really any rules. – pjc50 Jan 04 '17 at 15:51
  • 2
    @InterLinked - 5KW was the input power to the motor-generator, the power delivered to the antenna would be (much?) less. For example, this [1500W Ham amplifier](http://www.ameritron.com/Product.php?productid=AL-1500) is rated to draw 15A @ 240VAC, or about 3000W at full output power. I don't know how efficient a spark-gap transmitter is, but I'm assuming that it's not very efficient. Some countries have higher power limits - Canada allows up to 2.25KW. – Johnny Jan 04 '17 at 21:28
  • For comparison, the [TPz 1A1A5 „Hummel“](https://de.wikipedia.org/wiki/Fuchs_(Panzer)#TPz_1A1A5_.E2.80.9EHummel.E2.80.9C) [(picture)](https://de.wikipedia.org/wiki/Fuchs_(Panzer)#/media/File:IL%C3%9C_der_Bundeswehr_am_24.09.2012_--_ELOKA.jpg) is a military-grade HF **jammer** that operates off a 15kW generator... – DevSolar Jan 05 '17 at 12:41
3

Even as far back as the early 1900s, telegrams transmitted wirelessly could reach hundreds of miles. For instance, the Titanic communicated with Canada, 400 miles away, with relatively low-powered equipment. Given that telegraphs are very simple, how could these pulses travel so far?

Besides the fact, as others have pointed out, that the power really wasn't very low, morse is simply a very low-bandwidth signal. You can get a message across using very small amounts of received power, as long as you don't want to send very much information in a given period of time. WiFi carries a billion bits per second from one room to another. A TV channel sends tens of millions of bits per second over maybe a hundred mile radius. Morse code keyed by hand is equivalent to about ten bits per second, give or take a factor of two, and in bad conditions it could be less.

And would these pulses still travel that far today with the same equipment?

Sure. And if you assume the same transmitter but a modern receiver, you could probably receive the signal over a considerably longer distance, because a good modern receiver has higher sensitivity, cleaner amplification, and the aid of computer algorithms.

And doesn't this mean that there couldn't have been very many people using the systems, since operators within hundreds of miles would all be jamming the airwaves? It seems this would produce a lot of cross-talk. Or were there multiple frequencies available for wireless telegraphy?

Some of both. There were plenty of frequencies available for multiple stations even back in the 1910s, and if you look at modern usage you'll see that Morse code allows for very narrow channel spacing, with potentially hundreds of conversations going on in parallel in the space of a few Megahertz. But the equipment in use at the time had poor frequency stability and very bad wideband noise, and couldn't just change channels at the drop of a hat, so in reality there were few channels in use, and there were issues with interference. Nonetheless there were quite a few ships and shore stations making regular contact as early as 1910.

hobbs
  • 6,719
  • 1
  • 19
  • 31
  • 1
    With a modern system, you could probably [bounce the signal off the Moon](https://en.wikipedia.org/wiki/Earth–Moon–Earth_communication) and still receive it. – Mark Jan 05 '17 at 02:32
  • 1
    @Mark the ionosphere is a lot closer, and you need relatively little power to achieve a decent bandwidth. To even detect the *existence* of a lunar reflection requires a very high ERP, which means either extreme transmit power levels or large arrays of directional antennas. It can be done by a radio amateur with a large back yard, but only at very low bandwidth. – Chris Stratton Jan 07 '17 at 21:06
3

Given that telegraphs are very simple, how could these pulses travel so far?

By using sufficient power and containing frequencies that supported a propagation that could go around the earth's curvature that distance.

And would these pulses still travel that far today with the same equipment?

Yes. It is known as HF (high frequency) radio. For over the ocean flights, commercial aircraft require some sort of reporting. If they don't have satellite communication, they need to communicate with HF radio (that also extend into the MF bands). HF radio coms need to be attempted with a list of frequencies (based on the distance, time of day, and propagation reports).

Radio waves propagate via line of sight, ground wave, and sky wave. Newfoundland wasn't any where near line of sight. Ground waves can propagate around the earths curvature. A distance of 400 miles would require a very low frequency (and low data rate). Sky waves can refract off of the ionosphere and back down to earth around the curve. Sometimes reflecting off of the earth, back up the ionosphere and refract again (called "skip").

Over ocean flights have traditionally used the skywave refraction when beyond line of sight. It is not entirely reliable, and position reports are sometimes delayed in order wait for the distance to change.

B. Young
  • 131
  • 3
  • 1
    Finally someone who actually understands the issue! One of the unfortunate problems with EESE is that often we get a lot of engineers with no actual experience of a particular topic or application making wild guesses from first principles that are somewhere between wrong and irrelevant. – Chris Stratton Jan 07 '17 at 21:10
  • I'd also just like to add that back then, HF was relatively new and a lot of communication was on low and medium waves. The 600 m (500 kHz) was for better part of a century (and Titanic's time as well) the "distress wave" and 125 kHz to 150 kHz was also maritime-mobile band, with 143 kHz being the calling frequency for the "long continuous wave" in 1930s at least. Back in Titanic's time, ships had to have radios for 600 m and 300 m, but the 1912 Radio Regulations don't go into details of frequencies used as much as the newer ones do. – AndrejaKo Jan 07 '17 at 21:18
  • A little trivia : First time SOS was used a distress signal. Prior to that it was CQD (general call distress) . SOS doesn't stand for anything however it distinct sound in Morse makes it easy to copy. – Old_Fossil Sep 06 '18 at 07:25
1

Consider the following facts:

  1. Signal detection probability is a function of the received signal to noise ratio (SNR)
  2. SNR can be improved upon by:
    • Increasing signal power
    • Decreasing noise power

One way of decreasing noise power is collecting the signal over a longer time period and averaging out the noise using filters or signal redundancies such as parity bits in digital signals. So there is a trade off between data rate and SNR-- you can reduce your data rate to increase your SNR.

Although the detector of the telegraph signal (the listener's ear) is an analog system, the listener's ear/brain effectively "averages" each dash and dot over the duration of the tone, leading to an increase in SNR. Given that a telegraph operator is likely highly skilled at identifying noisy signals, their detection capability will be quite good.

As well, the redundancy of human languages provides another error-correction mechanism. Think about how effortlessly you automatically correct typos in your brain without requiring confirmation from the message sender. (Example: "This szentence h4s a lt of errors.")

Given that 5 kW is a relatively high transmit power for a mobile transmitter (your cell phone is roughly 1 W), and given the redundancies present in the signal itself, it is certainly plausible that communication took place at these ranges.

Robert L.
  • 207
  • 1
  • 7
  • 1
    Like so many others who have posted here, you miss the fundamental point - the challenge to terrestrial radio communications is not power level, but line of sight. Long range is possible where charged layers of the ionosphere, or other above ground objects, reflect the signal beyond the horizon. – Chris Stratton Jan 07 '17 at 21:03
  • @ChrisStratton These are not mutually exclusive points. All propagation of electromagnetic radiation is subject to 1/R^2 path loss, regardless of the path it takes (line of sight or ionospheric bounce.) – Robert L. Jan 11 '17 at 15:28
  • Those losses are not the relevant ones - thinking they are demonstrates a fundamental misunderstanding of the issue. – Chris Stratton Jan 11 '17 at 15:45
  • @ChrisStratton Unless you can transmit across that distance with a transmitter of any power level, losses always matter. Let me know when you've figured out how to transmit hundreds of miles with a 1 femtowatt transmitter. – Robert L. Jan 11 '17 at 15:49
  • That's exactly the point - the power levels involved are orders of magnitude more than needed for the distance-based loss. The *actual* challenge is that we live on a curved planet. – Chris Stratton Jan 11 '17 at 15:51
  • @ChrisStratton Propagation along curved paths still experiences spreading loss. – Robert L. Jan 11 '17 at 15:52
  • Again, the issue is not the spreading loss! That is tiny compared to the horizon problem. Until you understand this, what you are posting is irrelevant noise. – Chris Stratton Jan 11 '17 at 15:56