Hacker News new | past | comments | ask | show | jobs | submit login
Why are submarines demagnetized? (epfl.ch)
341 points by fgeorgy on Jan 27, 2016 | hide | past | favorite | 114 comments



At my previous job, I worked on software and algorithms for degaussing systems, it's an interesting field. A fun fact, more modern systems are moving away from copper wire for deperming and towards high temperature super conductors. It's quite expensive and difficult to wrap modern ships with tons of copper wire, so the plan is the ceramic wire will alleviate that. I left the company, so I'm not sure where that tech left off.


I spent a summer working at an accelerator physics lab, where a constant problem was degaussing large beam steering magnets. We always used a DC current to drive the magnet through smaller Hysteresis loops, until the magnetization is roughly zero. Is there a better way?


4,000 Amps is a ridiculous amount of current. Though when you have a nuclear power plant nearby...

Also, this line made me shudder:

> "...stringing an electric cable the length of the ship and pulsing 2,000 amps through the cable."

EDIT: Okay maybe not that ridiculous, apparently the Luxor hotel has higher-amp lightbulbs http://www.wolframalpha.com/input/?i=4000+amps


The degaussing facility uses shore power. They could mean 4000 ampere-turns, too.

Source: I went through one of these back in the day.


Oh, it's a degausser? So it's not running regularly while at sea?


Correct. The magnetic signatures seen are the permanent magnetic moments embedded in the steel of the ships, which are picked up by passive magnetic-anomaly detectors. These have to be removed by randomization in a normal all-at-once degaussing process.

It is hypothetically possible to have an active-detection process too, but presumably that would take the form of emitting a pulse of electromagnetic radiation and looking for its echoes, in other words it's just another form of vision. The standard problem applies: find a frequency at which water is mostly transparent. Unfortunately the absorption spectrum for water is https://en.wikipedia.org/wiki/Electromagnetic_absorption_by_... , it has a minimum in the visible range.

In addition to being one of the more stunning confirmations of evolution (of course our eyes are only sensitive to the least-absorbed wavelengths of light in water -- our ancestors evolved eyes in the ocean! there weren't many other useful wavelengths down there!) this has the nasty property that it's unlikely that we'd be able to detect a sub this way with any better approach than shining a bright laser or spotlight or what have you, and watching what's reflected back with our own eyes.


The ships undergoing this look kinda crazy. Couldn't find the exact image I was thinking of but here is a sub with wires wrapped around it:

http://systems.polyamp.com/images/stories/DepermingSub.jpg


Huh. It's like naval rope bondage.

(Which is probably somebody's fetish)


I'm a bit unclear on how degaussing the hull prevents the sub from showing up as a magnetic anomaly. All that would do is keep the vessel from affecting nearby compasses. It still distorts the local geomagnetic field, doesn't it, just by virtue of being made from a ferromagnetic material...?


"All that would do is keep the vessel from affecting nearby compasses."

"It still distorts the local geomagnetic field."

These two statements are trivially identical.


Qualitatively, but not quantitatively. A large metal object might not affect a conventional compass reading in a noticeable way, but if the enemy is flying around with cryocooled Josephson-junction magnetometers or whatever, they'll still see a distortion in the local field.

I'd think that the permeability of a submarine hull -- magnetized or not -- would be very different from that of the surrounding seawater.


Oh it doesn't completely erase the signal, but it minimizes it. P8 Orions have a plasma induction magnetometer in the back that is supposed to be able to see the degaussed submarines. The Russians did something even simpler: they took photographs from space and saw where submarine propellers were causing churn in the ocean cyanobacteria.


I think your understanding of evolution is a bit off, since there are other creatures with common ancestor (birds) that can see into long UV.


And magnetic anomalies, apparently, if they're migratory, via polarisation of uv iirc.


another hypothesis is that spin-coupled electrons preferentially and directionally emit photons from a cryptochrome protein. But tbh, we don't really know yet.

http://www.ks.uiuc.edu/Research/cryptochrome/


Thanks for your aclaration, it helped me understand the idea :)


The degausser is a pier-like structure constructed at shore facilities for submarines.

Here is a degausser for the largest US submarines: https://www.google.com/maps/place/30°46'44.3%22N+81°29'30.6%...


Awesome, thanks for link!


"Oh, it's a degausser"

Ah makes me remember large CRT monitors that went "bonk" when you turned them on (to degauss). I actually still have a Sony 30" TV that does the same thing (and it's quite loud actually).


Going back even older, the first TV's didn't have degaussing coils built in. If you moved the TV, a technician would have to come to your house with a portable coil and do it for you.


Wow what was the reason that moving the TV meant you needed to degauss? Obv. has to do with magnetic field but I always thought it simply had to do with power being turned on (and only because that is when it seemed to degauss).


Simply that moving the TV exposes it to a different localized magnetic field -- just like a compass inside a building. This could also be the result of moving around big loudspeakers near the TV (especially if they aren't magnetically shielded).

Built-in degaussing circuits fire when you turn on the TV presumably because (a) it disrupts the image, so best to do it at turn-on than randomly while you are watching, and (b) because you are unlikely to move the TV while it's on.


It would have been really helpful if the article had made that clear.


It isn't even part of the ship. Its a big coil that the ship pulls into.


Well as far as energy, depends on voltage. That's just 4000 watts if the voltage is just 1 volt.


Whilst that is true, it really all depends on the resistance of the thing that you are passing the current through.

Remember that P = I^2 R (i.e. power goes as the square of the current for a constant resistance)

so for an object of constant resistance then to double the current than you need 4 times the power.


Yes always annoying when amperage is discussed without voltage, it's like when people focus on torque rather than horsepower and conflate torque with power.


Not always. V=IR. More amperage means that resistance has a more significant affect. This is why transmission lines are ran at high voltages. You can get away with smaller wire sizes and significantly less thermal loss.

The first thing I think about when I see 4Ka is that any wires used would be enormous.


V=IR only holds for direct current. The transmission lines can be run at high voltages and relatively small diameters because they carry alternating current, which is a much more efficient way of transmitting electrical energy over long distances. If they were DC then they would be sized more like sewer pipes.


V=IR works perfectly well for AC just as long as you measure the AC in terms of RMS voltage and current. For sinusoidal AC the result is about 0.7 the peak value. There would be no significant difference in wire size.

DC is in fact the better way to transmit energy over long distances. The computer I am typing this on is powered by energy transmitted 900 km in the form of DC.


But as far as dangerousness, only 0.1A can kill a human, whereas voltage is not so important.


Voltage is what drives current so they're directly proportional and their proportion depends on electrical resistance. Given that human body's resistance to electricity is more or less constant, that allows calculating a dangerous voltage just as easily.


The body's resistance is generally not what matters, it's generally the resistance of the connection between the body and the power source. If the connection is made via an air-gap spark you can need thousands of volts to ionize the air to allow the spark to form which allows the current to enter the body (corresponding to tens of thousands of ohms at a killing current, although that resistance is highly non-linear and drops suddenly after the initial spark connection ionizes the air). If the connection is made via EEG or EKG leads attached with conductive paste, the connection resistance can be a fraction of an ohm and the killing voltage can be well under 100 volts. The highly variable nature of the connection resistance is just one reason why the killing threshold is generally described in terms of amps rather than volts. If you're worried about whether a circuit could kill a patient and you must think in terms of voltage, you're not worried about whether you know the body's resistance, you're worried about whether you know the connection resistance.

In practice pretty much everyone who designs safety critical systems where electrocution is a concern focuses on current because it's a much better way to model and analyze the system risks than voltage.


Except, you also need to keep in mind the internal resistance of your power source. If that one is higher enough than the human body resistance, you could have a very high voltage which is actually not very dangerous.


I assume this is the reason I was able to safely use a 5000V power source in school, despite being shocked multiple times (IIRC, the maximum current it would give, even if you shorted the output terminals, was 3mA. Our teacher at the time did initially insist on wearing latex gloves, but they were discarded once we realised the current would just arc through the gloves and leave holes in them).

You do start to see some odd effects at those voltages though - I recall holding one of the output plates for ten minutes or so, without realising the supply was on. Didn't really think much of it, until I realised I could hold my hand over a piece of tin foil (roughly 1cmx1cm) and it would be attracted to my hand from 2-3cm away (that persisted for five minutes or so).


Through your body, yes. With 1V you can't drive much of a current through ones body.


Even 48 volts won't drive any appreciable current through dry skin, but you can string four car batteries together and weld with them. It's frightening that people are so ignorant about electricity.


Eh, as far as the list of frightening things that people don't know about goes, electricity is pretty far down the list for me. Most people just don't come into contact with it often enough, so simply saying things like "don't get near a downed powerline" is probably good enough.

Stuff like radon or CO are way higher on that list, in my personal opinion. Hell, even people's general ignorance of their countries' political process and what their (elected or not) officials are doing is way more frightening to me than their ignorance about electricity.

Sorry to be so damn pedantic.


I think it's fair to say "invisible" things are generally all pretty high on the list of frightening things people don't know about.

Including but not limited to: software, nuclear radiation, CO and chemistry in general, electricity, microbiology, politics in practice, macro-economics and banking, ... (it's a pretty long list)


Thoughts. Thoughts are the worst.


You're right, ignorance of those things is more dangerous. But people do interact with electricity every day, so I would have thought they'd know the basics a little better.


Not through dry skin, no. But if the thought of it makes you sweat...


Yes but it's a reasonable assumption that if they're pushing 4000 amps through the cables to generate a magentic field big enough to disguise a submarine, then the voltage is not in the single or double digits.


I assume that part of the reason for the "thick copper cables" is to bring the resistance as low as possible, like, under an ohm for a few thousand feet of 2 gauge wire.

Which will still be thousands of volts, but it sounds like it's pulsed thing as well, and they would certainly disable it if they're not in a dive or otherwise trying to be stealthy.


It might still be a little ridiculous - that light is one of the most powerful in the entire world


I'm rather thinking about solar light testing facilities for space crafts here, such as ESA's Large Space Simulator.

The specs say: 'Intensity level with 19 lamps at 20 kW: 2000 W/m' (6m solar beam diameter) [2]

(Also, 'On request, the sun simulator can be equipped with 32 kW lamps'.)

[1] http://www.esa.int/Our_Activities/Space_Engineering_Technolo...

[2] http://emits.sso.esa.int/emits-doc/ESTEC/AD1-AO-5580.pdf


One of the lab techs where I went to university had a poster up in the workshop with a series of nails and the current required to melt them. At the end of the series was a thick bolt, marked "1100 Amp Slow-Blow Fuse"...


Some competition car audio set-ups are using tens of thousands of watts, usually at 14.4 or 18 volts, so the amperage can get into the thousands. The wiring used for that is usually multiple runs of 0-gauge or better. 4/0 is a bit less than half an inch and is rated for 300-400 amps.


I'm unclear on that point, is that current applied after construction, or is it an operational mechanism? If the latter, it seems like that would be a beacon of sorts as well (for some other type of detector)


the luxor hotel also isn't submerged in a giant body of electrically conductive liquid!!


A lot of military submarines have titanium hulls because it's non-magnetic. Interestingly, during the cold war the main source of titanium for the US military was the USSR.

Edit - It looks like the principal reasons for using Titanium was likely to reduce weight and dive depth rather than it's magnetic properties. Thanks to respondents for clarifying that. Much appreciated.


Aluminum, magnesium and austenitic stainless steel are also non-magnetic. But they also have worse strenght/weight ratio than titanium.

All of the above are difficult to weld. This increasest the cost, which means fewer subs. So current submarines are made out of low alloy steel.


The USSR was the only country to build submarines with titanium hulls, a practice which ended with the Cold War.


And only one class the Alfa fast attack submarine the titanium hull was used more to reduce weight (Half the displacement of a comparable Los Angeles class US attack submarine) to allow it to be the fastest submarine ever constructed (double the reported speed of US attack submarines and most likely at least 25% faster than their actual speed) rather than because titanium is non-magnetic. All Alfa subs were decommissioned in 1990 and sold as scrap.


Yep, the Alfa was a strange design, Titanium hull, reduced to way below safe minimum shielding to save the weight, noisy as hell (like insanely noisy).

Basically designed as the submarine equivalent of an interceptor, get somewhere fast and open fire.

IIRC the top speed of the alfa was above the speed of the NATO torpedoes in use at the time.


Reported top speed. And it was seen as shocking enough to spur the investment into new torpedo technology.

https://en.wikipedia.org/wiki/Alfa-class_submarine#Impact

Like most fears over high-performance Soviet hardware though, the threat turned out to be overblown compared to the actual numbers build / operational state.


Yep, the Alfa also had a tendency to slowly kill it's crew from radiation exposure.


well thats what you get for putting a fast nuclear reactor with minimal shielding in a relatively tiny sub.


It was also used in the K-278 nuclear attack subs, but only in the inner hull so wouldn't have had so much effect on reducing magnetic signature. It seems I was somewhat miss-informed. Thanks for helping clear that up.


The K-278 was part of project 685/705 of which 2 subs were made the K-222 (PAPA) and K-278 (MIKE) both were prototypes PAPA used a full titanium hull and resulted in the Alfa class subs and the K-278/MIKE was lost with all hands with it being the only sub of it's class.


It was also used on the Sierra class (Project 945) of which at least two have been refurbished and remain in service.


Titanium allows an increase in dive depth but has no effect on speed.


The ship's total mass must be close to the mass of the water it displaces. Less mass spent on the hull means more mass spent on the reactor and steam plant.


The displacement is based on the volume of the sub not the mass. Basically the Alfa was a small and light sub with probably the "biggest" reactor in terms of power output for it's size. What you got is a sub with half the displacement of an LA class submarine with a reactor with equal power output which meant that you could strap on some really powerful engines. So half the displacement with about the same power capabilities means pretty much that you can double your speed.


mass = volume * density. Water has a (nearly) fixed density. So for any given displacement, less mass allocated to the hull means more mass allocated elsewhere.


Yes but the mass doesn't effect the volume of water displaced if you put in a hollow 10 cubic meter steel cube into water or a solid 10 cubic meter steel cube they'll both still displace the same amount of water which is 10 cubic meters.

BTW the density of the sub is also for the most part pretty much equal to the density of water.

The only times where it's going to be either greater or lesser than water is when it's doing a static dive or fast ascent.

When the sub does it's initial dive and takes on water into it's buoyancy tanks it becomes about the average density of salt sea water once that is achieved it can use it's diving planes (foreplanes) to adjust it's depth.


https://www.math.nyu.edu/~crorres/Archimedes/Crown/CrownIntr... was useful for me in correcting some myths about mass and density I'd learned.


Interesting. I knew the US used it in airframes and assumed they just used it a lot less in submarines but you appear to be right, thanks.


The CIA had to setup shell companies to buy titanium for the A-12/SR-71, making submarines out of Titanium didn't made sense for the US as it would be a waste of a "strategic material".

The soviets also had to use titanium in many applications that would require high end steel as they both had plenty of it (almost half the world supply at the time) and their steel industry was decades behind the west and they just could not make the required quality of steel alloys. The soviets ended up using titanium as a replacement and became quite good at creating titanium alloys and working with titanium in general the US would most likely not be able to produce a hull out of titanium (at that time) without investing way too much effort into building that expertise which would be pointless given the titanium resources they had and their ability to make steels which were as good or better for most applications. The soviets had to resort into using titanium for every thing from infantry armor plating to razor blades.


There are places in Crimea where the streetlights are made from titanium.


What I find particularly amusing is that Marvel Comics in 1965 introduced a Soviet enemy of Iron Man: Titanium Man (https://en.m.wikipedia.org/wiki/Titanium_Man) ... Funny how they were "accurate" in mimicking the preference for using Titanium vs.Steel in the USA/USSR arms race.


I wouldn't be surprised if the Soviets experimented with basalt fiber composite hulls at some point. I think they used it in some ballistic missile cowlings, anyway.

Basalt fiber is similar to e-glass fiber, but has greater stiffness.

I'm guessing they made a few small fiber-wound spheres and did a crush depth test on them, but the results discouraged further experimentation.


Are you sure they make them out of titanium to save weight? I have two anecdotal pieces of evidence: 1. I have a friend who does welding for the navy and was called in to replace part of the hull of a nuclear sub due to a very hilarious design flaw. It was >6" thick and took days to fill in... with steel. 2. I've read "Concepts of Submarine Design" which goes into great detail on the principles and systems of military sub design. The authors make the case that getting more weight is often a problem. This makes sense because every foot or air space needs to be displaced by more weight in the steel of the sub. So if they had a problem with subs weighing too little, they could just make them more spacious. And they're not very spacious.


Third anicdote; titanium is pretty expensive stuff. Online metals is selling a 3"x12"x12" piece for $1.7k. If your hull was 200' long, 15' in diameter you'd need $486M just for the raw titanium on the hull. That seems high just for the hull, even for a nuclear sub. Google is saying the whole thing is probably about $2B and I'd expect all the other stuff in there to cost a lot.


A naval yard is not paying small-lot retail prices for their materials.


The supply has something to do with the price and the USSR had no shortage of titanium, so the accounting in your example would have been valid only for one side of the iron curtain.


Making a sub bigger increases drag, particularly if you are making the frontal profile bigger.


What was the design flaw?


During WW1, the British Royal Navy became concerned that U-boats could enter its harbor at Scapa Flow. In addition to installing nets, mines and hydrophones, large loops of wire were installed on the channels that could be opened for navigation, encircling the entire channel, in order to detect vessels by induction.

There were no successful attacks in WW1, but perhaps they were less careful in WW2 (or submarine degaussing rendered the technique ineffective), as a U-boat torpedoed and sank the battleship HMS Royal Oak in the harbor, with great loss of life.

Update: It seems that Gunther Prien made his attack before the defenses were fully restored.


Not limited to subs, the same principle was also used on ships via degaussing coils: https://en.wikipedia.org/wiki/Degaussing


Related, interesting HN discussion a month ago:

https://news.ycombinator.com/item?id=10810390

Submarines can alter their magnetic fields, but gravitational fields are a good bit more tricky, and can apparently also be exploited for detection.


I was reading this https://earth.esa.int/documents/973910/1006684/RR3.pdf (Satellite Gravimetry) earlier today after seeing this article as I remembered nuclear submarines used gravimeters.

Is there any information on them being used to track other subs, as I was wondering if that was possible, I know they're apparently used to map the sea floor though.

Edit:

This is interesting:

"He showed a figure of gravity increasing by about 2 microgal over a 4-h period as men shoveled snow from the roof of the SG station" -

http://ethesis.helsinki.fi/julkaisut/mat/fysik/vk/virtanen/s...


The mass of a bucket of water, and an equally full bucket with a sub in it, is equal. Otherwise the sub would sink like an anchor or float like a boat.


True. The gravity gradients from the sub may be different, though.


Do you have any evidence a submarine conducting normal operations has been detected with a gravimeter?



Yup. My dad was a RAD/MAD operator on a P3 during the Vietnam War. One of my indelible memories from childhood is him explaining how it worked and a story about seeing a Soviet sub on the surface. Of course as a kid having watched Das Boot I'm expecting there to be guys scrambling to the conning tower, dive dive dive, depth charges, etc. He's like "nah, the skipper just circled the thing and we waved at each other." Seeing this story on HN makes me wish I could hear the story again but he passed of cancer a few years back.


Vietnam was a proxy war and neither the US or Russia wanted it to escalate into World War III. US policy was to stringently avoid anything that might lead to escalation.


Tom Scott talked a bit about magnetic warships and the magnetic mines in one of his "Things you might not know" videos: https://www.youtube.com/watch?v=LG-X23aVud0&list=PL96C35uN7x...


The US Navy has a few pictures of the setup they use for submarines. They use the term 'deperming' rather than 'demagnatizing'.

http://www.navy.mil/view_image.asp?id=38120


What a strange title for an article. It's like back-to-front thinking.

Surely someone uninformed on the topic would want to read something along the lines of "anti-submarine warfare" or even "submarine detection", during the course of which would probably learn about magnetic anomaly detection and attempts to defeat it.

I mean, who and how would someone know (or want to know) why submarines are demagnetized without already knowing something about ASW?


The full title is, "Monday January 25, 2016 Poll Why are submarines demagnetized?" and the article starts off with the results of Monday's poll. So the article we're ready today is a response to the results of the poll, but the site administrators kept the poll question as part of the title.


Interestingly the new P8 Poseidon doesn't have Magnetic Anomaly Detection (MAD) equipment on the US variant, though the Indian one has had it fitted. You need to fly low for it to work which is less efficient for the jet powered P8 compared to the older P3. So if you're generally flying higher and up against submarines which have been demagnetised there's less need to fit MAD.

Surface warships are also degaussed to help avoid magnetically triggered mines.


> Surface warships are also degaussed to help avoid magnetically triggered mines.

I believe the large pipe seen in this photograph running along the hull of the Queen Elizabeth, as fitted out as a WW2 troop transport, contained wires for degaussing.

http://www.ssmaritime.com/RMS-Queen-Elizabeth-War-Trooper-2....



Also known as "degaussing," which brings me back to my school days when I would sometimes degauss the school's CRT monitors when I got bored just for fun. Looks like this, if you didn't live through those times: https://www.youtube.com/watch?v=Sd9nOtX-CJg


There was a lot more ripple if the refresh rate was set higher but the degausser still ran at 60Hz.


Yeah, it made a very satisfying noise and visual effect.



Quick idea: rotate the ship 360 degrees, integrating the magnetic forces. What remains is the magnetic force of the ship itself. In subsequent measurements, subtract this magnetic force to get the component of the magnetic field of the Earth.


I love that simple and direct writing style.


Maybe if one were to have EXTREMELY accurate baseline magnetic field data for the whole world subs could still be detected?

http://www.gfz-potsdam.de/en/media-communication/press-relea...


Look up the noise spectrum data. You'd need a rather dense sensor field to detect a signal outside of the noise levels.


https://www.google.com/maps/place/Naval+Submarine+Base+Kings...

Here's one of the degaussing facilities on the East coast.


https://www.google.com/maps/search/electric+boat+groton+ct/@...

While we're mapping, there's also where they're built: Electric Boat in Groton, CT.


EB doesn't have an externally visible degaussing facility.


Some more information on the advances in sea mines, and how they were countered:

https://medium.com/war-is-boring/how-britain-beat-germanys-w...


The obvious answer is so that nobody catches a submarine while magnet fishing.


thanks, I haven't laughed out loud while reading HN in a long while.


You learn something every day, but thanks to this site I learnt 100 things today :)

Here is the list: http://qi.epfl.ch/en/sondage/list/


Article suggests -- or outright says -- that relatively crude magnometer tech is still used by militaries and 'coalition' expeditionary search vehicles today.

And thus why the MH370 airliner remains missing? Only a thought.


The same people who shot Kennedy hide the plane. You will be disintegrated in 3 minutes.

Or you could use the grey matter between your ears. Modern airliner hulls and structure are made mostly out of aluminium, which is -suprise- antimagnetic.

And finding a Submarine or anything else is hard. Like really hard. You need to know where to look. Because of that, NATO would have to had stopped Red Fleet at the GIUK-Barrier or it would have gotten ugly. Or, look at the problems people had to find ships like Titanic or Bismarck, where they had much more precise locations to start searching for.


> It took them several months to come up with the answer

Maxwell's equations provides the answer: https://en.wikipedia.org/wiki/Maxwell%27s_equations

But then, I personally think Maxwell to be under-appreciated by the mass population whereas at least Faraday people can name even though they may not remember what he achieved.

I bet they were also inspired by Faraday's experiments across the Thames.


TLDR: WWII




Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: