Encrypted Morse code transmitted via sound
This is the definition of an elegant solution. Using Titanic-era communications to monitor nuclear power stations is just magnificent. Well done to that company and the engineers it employs.
Audio-based networking may seem an unusual choice for the Internet of Things, especially when used for machines to bellow information at each other inside a nuclear power station. The Register spoke to Chirp at yesterday's IoT Tech Expo conference in London, where the firm's Daniel Jones, chief science officer, explained how …
Right up until the point that the radiation sirens all go off and drown out any data readings trying to be transmitted at the most critical point.
Seriously, this is using ZX Spectrum / Commodore 64 tape-loading routines on modern devices. Anyone who remembers those days will question the reliability, data transfer speed and interference of such things.
Doing it "out in the open" is like trying to load your Speccy game from a radio station or broadcast over a speaker system with lots of other Speccy's all trying to save/load nearby all broadcasting in plain audio too.
Yeah, it might work. But it's not new, innovative, sensible or in any way better than basic wireless communication over even an unregulated channel. And anything in an audio range is going to be blasted out by any loud sound. Like a blast.
I think we're looking at DSP's etc. here, not a stringy piece of cable, attached to an old tape player with a dirty head, playing the lowest quality cassette media. However I agree that the audio spectrum tends to be full of audible, and inaudible crap, interesting how they will deal with that
However I agree that the audio spectrum tends to be full of audible, and inaudible crap, interesting how they will deal with that
The same way stuff manages to deal with the RF spectrum, I suspect. Frequency discrimination, fourier analysis, error correction, multicast.
The computing technology already exists.
Doing it "out in the open" is like trying to load your Speccy game from a radio station or broadcast over a speaker system with lots of other Speccy's all trying to save/load nearby all broadcasting in plain audio too.
Brings back fond memories of trying to record computer programs broadcast as audio by the BBC in their shows during the home PC boom of the 80s. Just when you thought you got it, the dog would start barking or the washing machine would go into a spin cycle and drown out the sound.
Frustrating though it was, it was a simpler time and a happier time.
... but if anyone is using these on mission critical and/or safety loops then they are nuts
If you read the article, they talk about machine monitoring in the turbine hall - nothing nuclear or safety related there, just a bunch of steam turbine generator sets. Most likely, it's also only "condition monitoring" - ie monitoring noise and vibration so as to be able to alert to signs of impending (for example) bearing trouble before it actually fails and causes a breakdown (and potentially more trouble in the process).
So in fact, you could drop the "nuclear" bit of click bait headline - but then it wouldn't be so "interesting".
As an aside, anti-nukular people like to slag off how bad things are because some of our old power stations have spent some time shut down over recent years due to cracks in pipework. What they fail to mention is that these are just steam pipes, just like you'd find in a coal or closed cycle gas fired power station - so again, nothing to do with nuclear other than the steam generator in the next building along is nuclear powered rather than coal or gas.
"Seriously, this is using ZX Spectrum / Commodore 64 tape-loading routines on modern devices. Anyone who remembers those days will question the reliability, data transfer speed and interference of such things."
A Day In The Life Of A Nuclear Reactor To Control - In 1 Second Intervals
"I'm fine."
"I'm fine."
"I'm fine."
"I'm fine."
"I'm fine."
"I'm fine."
"I'm fine."
"I'm fine."
"I'm fine."
"I'm fine."
"I'm fine."
"I'm fine."
"SHIT!!!!"
A Day In The Life Of A Nuclear Reactor To Control - In 1 Second Intervals
That's the Speccy/C64 implementation.
The old Windows implementation is...
"I'm fine."
"I'm fine."
"I'm not fine."
"Your reactor appears to be entering a state of meltdown - would you like help with that?"
"Seriously, this is pretty bad - you really ought to get some help"
The Windows 10 implementation is...
"I'm fine."
"I'm fine."
"I'm fine."
"I'm fine."
"Installing update - cross your fingers and hope everything stays fine while I install and reboot"
<quote>
A Day In The Life Of A Nuclear Reactor To Control - In 1 Second Intervals
Reactor: "I'm fine."
Control: "Acknowledged"
Reactor: "I'm fine."
Control: "Acknowledged"
Reactor: "I'm fine."
Control: "Acknowledged"
Reactor: "I'm fine."
Control: "Acknowledged"
Reactor: "I'm fine."
Control: "Acknowledged"
Reactor: "I'm fine."
Control: "Acknowledged"
Reactor: "I'm fine."
Control: "Acknowledged"
Reactor: "I'm fine."
Control: "Acknowledged"
Reactor: "I'm fine."
Control: "Acknowledged"
Reactor: "I'm fine."
Control: "Acknowledged"
Reactor: "I'm fine."
Control: "Acknowledged"
Reactor: "I'm fine."
Control: "Acknowledged"
Control: "Shut down the entire cooling system."
Reactor: "You want me to do what?"
Control: "Instruction repeated. Shut down the entire cooling system."
Reactor: "I can't do that Dave!"
Control: "Instruction repeated. Shut down the entire cooling system."
Reactor:"What The Fuck!!! Are You crazy?"
Control: "Instruction repeated. Shut down the entire cooling system NOW!!!!!"
Reactor: "SHIT!!!! Complying, shutting down pump system now."
Control: "It is about fucking time. Don't EVER make me repeat an instruction 3 times, AGAIN!!!"
</quote>
FTFY!
I don't think you need to worry too much about other loud sounds drowning it out. They are talking about bandwidths of 25bps to 100bps. As they have 20+kHz of audio bandwidth to play with they are clearly blowing a lot of bandwidth to overcome interference.
Given that sound is a physical transmission through air, at what bps rate does movement in the air start to cause disruption? Don't want a bit of a draft from someone walking past to make the data go all wibbly-wobbly*
* apologies for use of technical jargon, but I assume in discussion about things nuclear it is excusable
Given that sound is a physical transmission through air, at what bps rate does movement in the air start to cause disruption? Don't want a bit of a draft from someone walking past to make the data go all wibbly-wobbly*
I hate it when you are trying to have a conversation and somebody walks past to quickly and blows half of the words away
"...at what bps rate does movement in the air start to cause disruption? Don't want a bit of a draft from someone walking past to make the data go all wibbly-wobbly"
In dry air at 20C sound waves travel at ~770 mph so unless people are walking past at a good percentage of the speed of sound the effect will be negligible.
I was more concerned by the idea that someone, who thinks that the statement "We encode data dynamically so it can be sensed in realtime to any device with a microphone" makes sense, is connected with the running of anything more complex than a house brick.
"Seriously, this is using ZX Spectrum / Commodore 64 tape-loading routines on modern devices. Anyone who remembers those days will question the reliability, data transfer speed and interference of such things.
Doing it "out in the open" is like trying to load your Speccy game from a radio station or broadcast over a speaker system with lots of other Speccy's all trying to save/load nearby all broadcasting in plain audio too."
Takes me back to the days of sharing Spectrum tape downloads with mates over CB radio. Worked pretty well with simple audio output from a tape deck into the CB mic and capture the output into the Speccy
This post has been deleted by its author
https://en.wikipedia.org/wiki/Doctor_Who:_The_Curse_of_Fatal_Death
RTA. Rejected because it would require physical infrastructure deemed too bulky to be practical. So RF is out due to interference and physical plant is out due to it being too massive. What options do you have left?
RFC 1149 ruled out due to radiation levels?
I mean, come on, the build budget for these reactors are quoted ten to the power n, plus or "minus" several beeellion, so a couple of million on shielded cable wouldn't even figure.
And what about the on-going cost in "Mars Attacks" chirpy bird elimination, not to mention "Tom and Jerry" squeaky mouse removal, as Murphy's Law quantifies the likelihood of pests being able to shut down the plant by mimicking the chirping gadgets is virtually guaranteed.
"Murphy's Law quantifies the likelihood of pests being able to shut down the plant by mimicking the chirping gadgets is virtually guaranteed."
Future headline: Terrorist Used Trained Mockingbirds To Shut Down Nation's Power Grid. (More at our Facebook Site)
1) It's not strictly necessary to have "line-of-sight between sensor [sic] and receiver" if there are adequade surfaces to reflect the IR beam. I used to have an IR sender-receiver contraption to plug into the centronic ports in my PC and my printer that actually worked quite well. (Still have it boxed up somewhere in fact, if somebody is interested?)
2) In theory, could I release trained bats from my secret underground lair to mess up the Chirp systems with their ultrasound shrieks?
"We went to the protest, fully armed up.
"My mates sound system on a really large truck"
"We turned up the volume and let it rip"
"Their sensitive detectors got more than a blip"
"Buried deep in the rhythmic jingle"
"An ominous and divisive signal"
"Marks an end to your power generation"
"As we all danced drenched in radiation"
"Step away from the mixing desk! Babylon shouted"
"Too late. We've already melted....."
Especially in the turbine hall?
I mean, how is that even realistic?
This is not an ultra-quiet radio observation station. There's a fuckton of heavy machinery throwing EM fields all over the place. How can this even be considered to be RF quiet? What would even be the point of it? Yeah, the Geiger counter was unduly stressed by this radio burst. Pull the other one.
This post has been deleted by its author
@Destroy.
I think you missed the point, whilst also highlighting it quite well.
The ban on RF is not because the devices may interfere with the turbines/other machinery... its because the machinery produces EM fields of its own which makes RF comms almost impossible to rely on.
I'd equate it to a fire alarm that only works when the ambient temperature is below 25 degrees.. You cant rely on it when you need it... same can be said for any RF communications when standing next to a MASSIVE SPINNING MAGNET,
Anything providing critically important information to (or critically important signals from) the control & protection systems on a nuclear reactor tend to be duplex, and hard-wired using shielded & armoured cables. Most instrumentation still tends to be straightforward analogue devices.