Slashdot Mirror
Electricity Over Glass
guddan writes "Running a live wire into a passenger jet's fuel tank seems like a bad idea on the face of it. Still, sensors that monitor the fuel tank have to run on electricity, so aircraft makers previously had little choice. But what if power could be delivered over optical fiber instead of copper wire, without fear of short circuits and sparks? In late May, the big laser and optics company JDS Uniphase Corp., in San Jose, Calif., bought a small Silicon Valley firm with the technology to do just that."
"Running a live wire into a passenger jet's fuel tank seems like a bad idea on the face of it. Still, sensors that monitor the fuel tank have to run on electricity, so aircraft makers previously had little choice. But what if power could be delivered over optical fiber instead of copper wire, without fear of short circuits and sparks? In late May, the big laser and optics company JDS Uniphase Corp., in San Jose, Calif., bought a small Silicon Valley firm with the technology to do just that."
What, no one ever heard of vacuum lines? Or maybe pressurized lines? I'm not a rocket scientist, or even a plane scientist, and I could figure that out before I was finished reading the frickin' summary, let alone the frickin' article.
People love to make work for themselves...
Setting that aside, the idea sounds awesome!...what with all the planes we lose every year to short-circuiting wires...BUT, I'll wait to see if this materialized before I get all excited about it.
So, firing laser beams into fuel tanks is a safety feature now?
But what if power could be delivered over optical fiber instead of copper wire, without fear of short circuits and sparks?
You're stilling bringing as much power into the fuel tank. High-power beams of light aren't any safer, a laser can cut inch thick steel.
At least electricity is very well understood, we know how to insulate the wire, we know how much voltage will spark in a given medium, and the low voltage for sensors is very safe.
High energy lightbeams are not at all well understood. Will the fiber heat up? What about light leakage, will that cause an explosion? What if the fragile fiber breaks while the beam is on?
...have been using similar technology for some time.
however, there is a problem with what is called dark current. that is when there is no light hitting the transducer, and there is still a current being developed...
Those who live by the sword, get shot by those who live by the gun...
...They're shooting a laser through fiber into a small solar cell that's inside the fuel tank. explain to me how that is innovative.
There is nothing wrong with running wires into petrol tanks for sensors... Take a good look at how badly made the rheostats in everyone's pertol tanks are made. Most engineers freak out when they see them for the first time.
However the design is what is known as "Intrinsically Safe"... ie, it can't cause an explosion.
Currents, voltages are limited. Components are overrated by a set amount.
I've never heard of any intrinsically safe circuit igniting gasoline.
So what if you use fiber optics to provide the power. It's still electronic circuits in the tank, except now they are a whole lot more complicated and have power generation and regulation circuits, which make it a whole lot more dangerous...
And please don't just say encapsulate the dangerous stuff, because I'm sure that won't explode with a pressure build up if a component dies (as they tend to do in regulated power circuits).
It really scares me how such "great" ideas like this seem sane, when the original technology was probably safer.
GrpA
Enjoy science fiction? "Turing Evolved" - AI, Mecha, Androids and rail-gun battles. What more could you want?
why do they need electricity? if it's to monitor the level of the tanks you could use a purely mechanical device surely?
Why bother with electricity at all. A piece of fiber to an optical encoder would do the job just fine. I can't think of any sensor that couldn't be implemented optically.
Having said the above, the product seems like a solution in search of a problem. I can't recall any incidents where a fire or explosion was caused in an airplane because of faulty wiring in the fuel tank. There are lots of places where an electrical spark could cause an explosion. For instance in a mine, or factory, dust explosions are an ever present danger. To deal with that, we have explosion proof wiring. http://en.wikipedia.org/wiki/Electric_actuator#Explosion_protection In other words, the problem was solved long ago.
...and found that they said "Such transformers are large and necessarily heat up, which can lead to hot spots. To prevent equipment temperatures from rising to dangerous levels and to reduce power leaks, oil and gas are used as insulators. But oil is flammable and can make the transformers explode at high temperatures. The transformers are also expensive to install and maintain."...
Say what?!? Ok...so, yes, I'd much rather have the manufacturer disclaim that they can't be sure that their product won't explode (thusly guaranteeing all hands lost), than use wires that have have never caused a problem in the manner in which the manufacturer of said bomb-like device.
Still...there might be some application for this device, but it certainly WON'T be in a fuel tank.
By the way, millions or even billions of fuel level sending units have been in use in anything with gas gauge for years. How many users of such devices have been killed due to electric failures? I'm guess very very VERY few, if any at all.
I agree with another commenter when they said that they don't want frickin' lasers pointing at their frickin' gas tanks.
Millivolts. Most level sensors are variable resistors, so you only need to exceed the forward min. bias of the resistor (see the spec. sheet) to have accurate results. Above that, it's just a matter of calibration and maintaining a well-regulated power supply.
I've already invented a device to tell the level of a fuel tank without using electricity!
Through advanced hydrostatics, I found the level of the fuel tank could be remotely monitored via a capillary tube, from which the fuel level can be calculated from an ocular spectrogram in the VIS range.
Okay, okay.. so it's just looking at the level in a hose connected to the tank... and it's not new.
Power is still power. Whether you're pumping 100 mW of electricity or light into a fuel tank, I don't see a difference.
We already have intrisically safe electrical technology for such things. As long as you limit the power so that there isn't enough to create an ignition source, you're golden.
Personally I'd prefer new sensor technology that allows sensing the desired quantity with either less power or from a safe distance, like ultrasonic level sensors and such.
"I have never let my schooling interfere with my education." - Mark Twain
What's the problem with all the other solutions? Load cells, surface reflectivity, refraction with a modified optical fiber, acoustic ( ever tapped on a gas tank?), a current limiter on old school sensors? I'm glad I got out of aerospace in the 90's. It looks like idiots have taken over.
One of the main applications for this will be when galvanic isolation of the components is required. This has fairly little to do with fuel tanks, but is interesting for various medical applications, applications in humid environments, and so on.
This development would be great for Intrinsically Safe (EEx etc) instrumentation applications.
Current ATEX regs make it awkward to supply anything above about 1Watt at 6V.
Most people resort to pneumatics and/or keeping the computational logic outside the zoned areas.
Disappointingly for IEEE, he article is sparse in terms of technical details, such as the power/size ratio.
How about keeping the empty space in the fuel tank full of an inert gas?
After all, an electric spark can't ignite jet fuel if there's no air to burn it in.
Right, so instead of running electrical cables into the fuel tanks, we'll just shoot lasers into them instead.
It's a lot easier to ensure the power is properly limited. Running a sensor is a low power application (you wouldn't be using a "steel cutting" laser), and the power is limited with the size of the laser diode. There's no other way to get power through the line.
With electric lines, the issue is whether the wire to the sensor is going to short to another wire somewhere else in the wiring harness that will accidently put a lot more power on the line. There are a TON of wires on an aircraft, going every which way, some of which can deliver a lot of power. Short one of those to the sensor line and you can get a spark in the fuel tank.
It's not wasting time, I'm educating myself.
will you please stop spamming that crap!
Tank probes are capacitive and use a very weak signal for excitation. The spec is 25uJ maximum which is WELL under the energy required to ignite fuel. Typical systems use way less energy to make measurements. The problem is more that wiring for OTHER more power consumptive things is routed through the tanks in some designs. Also I agree, optical isn't any better or worse of a method.
I dunno, electricity in glass could lead to some shocking panes.
A small tube is lightly pressurized with a known gas. A mechanical sensor at the far end moves a tiny piston in or out of the tube to measure fuel level or temperature. At the near end, a device emits an acoustic pulse into the gas and measures the return reflection timing. This timing gives the length the piston has moved in the tube. The tube can be made of metal (well grounded to the tank frame at many points) or other reasonably rigid materials.
One tube can even be used for multiple sensors. This would be sorted out by the timings of multiple pulses returned from the various sensors at different distances along the tube. Some of those sensors measure temperature of the gas in the tube itself to maintain calibration. A low angle tube splitter can be used to greatly minimize the impulse reflections between sensors (which would otherwise appears as ghost sensors to the firmware scanning the pulse train, which can easily eliminate them at low levels).
If this has never been done before, this posting hereby constitutes public disclosure of the idea on Monday, 17 December, 2007.
now we need to go OSS in diesel cars
No, probably not, although friction on glass does develop a static charge, and under the exact right bad conditions could conceivably cause a spark. As others have observed in this thread, premise, as presented in the posting, is stupid and promotional.
The safety of stuff in a fuel tank depends on a) how well the risks are understood, and b) how well the engineering to mitigate them is performed.
It's a standard rhetorical ploy to assert that because something is different from an older technology, it is automatically free from the problems of the older technology... and, without saying so in so many words, allowing the listener to infer that it does not have equivalently bad new problems of its own.
The first time I heard groove-skipping on a CD, I laughed out loud. With all the promotion of the digital perfection of the CD, the fact that it suffered from exactly the same problem as a vinyl LP was... delightful.
"How to Do Nothing," kids activities, back in print!
The goggles! Zey do nothing!
Power is still power. Whether you're pumping 100 mW of electricity or light into a fuel tank, I don't see a difference.
There isn't much diffrence between feeding 100 mW opticaly or not.
"where electromagnetic interference is more than just an inconvenience"
Feeding 100 mW sensor and getting a 50 nW signal back with 25 mW of induced ground radar or cell telephone signal on top is the problem. It swamps the signal. In extreme cases such as a close lightning strike, the induced power could be enough to create a spark. The optical is for noise rejection and less for fire safety.
"Already, a Photonic Power device is replacing instrument transformers used in the power grid to measure high currents."
The optical in power substations doesn't have insulation breakdown failures in a lightning strike.
The truth shall set you free!
If they can scale this technology up to usable levels, would the power loss of the conversion outweigh the power loss to heat/resistance in High Transmission Lines? Obviously not over short distances, but imagine how it would play out over the thousands-millions of miles in the electrical grid.
For an idea of the scale of loss versus cost of power: some power companies are currently willing to take the hit in lost power by using aluminium lines instead of copper, because they can engineer the towers holding the lines up to use less steel. (ie: This is possibly an argument against doing this). The cost savings in the tower construction outweighs the power lost in the lines.
Speaking as a former USAF Avionics Specialist, who worked on C-5's, C-141's, and C-130's, and who personally saw a parked C-141 burst into flames on the ramp because of a fuel probe maintenance accident, let me explain things simply.
Design considerations:
JP4, the fuel that makes most jets run, is difficult to ignite. It needs a heat source. You could run a bare wire into a full tank and not have a problem. However, heat that wire up, and get the fuel/air mixture just right, and you have a problem. Big Boomba Problem, to quote JJB.
The big problem is the mostly empty tank and exposed heat sources. The C-5 has a nitrogen purging system. Basically, as fuel empties from a tank, it is replaced by nitrogen. The only way that wing is going to explode is if something other than a bare wire acts on it. Then, you've got bigger problems.
The big problem comes when you open the tank for maintenance. So, there are massive safety considerations. The C-141 that exploded in the mid-90's at Travis AFB in California blew because a jackass tech did not follow lockout/tag out procedures. The 141 doesn't have the nitrogen purge, but the tanks were open anyway. Two senior specialists were standing on top of the aircraft when the wing blew. Several others were in the cargo box. Luckily, aside from bumped elbows and bruised body parts, everyone got out o.k. We towed nearby aircraft to safer distances. There was precious little left of the burnt aircraft that identified it as such.
Most amatuers could make a good guess at a practical design for fuel sensors, but most of the solutions developed as such will end up being to costly, too heavy, will introduce other problems such as high maint., or simply won't work in 3-d, or extreme temperatures.
Politics is the art of looking for trouble, finding it everywhere, diagnosing it incorrectly and applying the wrong fix.
Unless those techniques are patented?
Je fume. Tu fumes. Nous fûmes!
OLD: The deal was finished on May 26, 2005 . The article referenced by the Slashdot story is from October 2005.
NOT NEW TECHNOLOGY: They are merely piping light using fiber optics, and then using the light with photocells to create small amounts of power for use with measuring devices. The measurements are communicated back through the fiber optics, using a different wavelength.
PATENTS? The article says, "Photonic Power owns key patents..." Can the generation of power using light be patented again? Can sending information using fiber optics be patented again? Maybe the company has patents, considering that the U.S. government has become corrupt, but it is difficult to believe that any patents could be valid.
IGNORANT: See this quote from the article referenced in the Slashdot story: "... the company's fastest growing sector is currently electric power transmission. One important application is eliminating the transformers used to step down high currents and voltages to measurable levels."
The article should have said, "... the company's fastest growing sector is currently powering and connecting the measuring devices used in electric power transmission."
The writer does not understand that the idea does not change the measuring system, only the method of transmitting the data. If step down transformers are part of the method of measurement, they will still be required. The "senior research analyst" who was quoted, Vincent Lui, doesn't understand that, either, apparently.
REALITY RULES: If you play video games too much, your brain will become partly useless for other things, and, if then try to be a Slashdot editor, you won't be able to do a good job. (This is a theory that seems to fit the facts.)
This is a useful idea for computer professionals in some cases where voltage isolation is needed, but the Slashdot story was mishandled, as often happens.
Existing sensors I've worked with typically require ~5V and low low amperage and when properly insulated the chances of sparking are very minimal.
With this solution, there still is electronics inside the fuel tank, so I do not see how this solves the problem. The light is converted into electricity inside the tank, so potentially even more problems with the conversion circuitry. However this tech is pretty cool and may have other applications. The 50% efficiency is bothersome as well, that's crap given todays electrical standards.
What interests me for fuel tanks is why they couldn't use some sort of a camera or sonar sensor from the outside of the tank to measure it. Perhaps having a tonight sealed porthole to view into, or maybe even sonar through the metal could determine the amount of fuel.
http://en.wikipedia.org/wiki/Category:Fuel_exhaustion_on_commercial_airliners
People always complain about battery monitors being inaccurate (spending 90% of their time at full charge and 10% of their time on the way down). I always wonder why people can't build a reasonable fuel gauge, as they seem to suffer the same problem. It would seem like these measurement biases could be calibrated out, but I guess it isn't that easy.
I thought capacitance based fuel sensors solved most, if not all, of the problems of sparking inside fuel tanks by keeping the powered components on the *outside* of the fuel tank. Is there some problem with accuracy or reliability that makes them unsuitable for commercial aviation that I'm not aware of or is this a solution searching for a problem?
And for all of the people asking how often sparking inside a fuel tank causes a tank to explode, yes, it *does* happen sometimes. The final NTSB report on the airliner that crashed off New York about a decade ago (you know, the one that the conspiracy theorists said was shot down by a hand-held SAM) was due to sparking inside the fuel tank. I'd link to it, but I can't recall the flight number, and I don't have time to search for it right now...
MCSE? No, sir...I don't do Windows. Yes, I am an idealist. What's your point?
Has to be air, doesn't it? 20% oxygen, from what I hear.
Carrying power through glass fibers was one of the main themes of the eco-fantasy 'YV88' written in 1977 - they also had data transport over glass fibers and multi-user chat program. ASEA was using glass fibers to send trigerring pulses to stacked thyristors in the early 70's.
If that was the cause of the explosion, it wasn't due to faulty wiring within the fuel tank. It was due to faulty wiring external to the fuel tank. So, my statement was technically correct.
You could get the same effect by having a short circuit melt a hole in the fuel tank. In fact, there are lots of ways faulty wiring can bring down a plane. For instance, the SwissAir crash may have been caused by improperly installed audio equipment. http://en.wikipedia.org/wiki/Swissair_Flight_111
Most jets (the largest quantity number of them, civilian commercial and private aircraft including everything from jetliners to small turboprops) burn Jet-A, which is a completely different formulation from the old JP-4. JP-4 had a significant amount of lighter molecular weight hydrocarbons (e.g. more of the constituents of gasoline) blended in.
JP-4 was also phased out of use by the USAF over ten years ago. JP-8 is used now, which is a completely different formulation from JP-4 and has much higher flash point than JP-4. JP-4 was a naptha-based fuel and JP-8 is a kerosene-based fuel. Today's Jet-A and JP-8 have very similar base formulations, but they have very different additive packages blended in. JP-8 has a much higher flash point than Jet-A too, since it is tailored for use in military aircraft that need to handle supersonic operations.
This is exactly like Snakes on a Plane.
Well... except the snakes are replaces with fiberoptics and the poison is
replaced with lasers and they're not really in the passenger area.....
Snakes with frikkin lasers in the fuselage.
But other than that... well, it's really nothing alike.
Feeding 100 mW sensor and getting a 50 nW signal back with 25 mW of induced ground radar or cell telephone signal on top is the problem. It swamps the signal. In extreme cases such as a close lightning strike, the induced power could be enough to create a spark. The optical is for noise rejection and less for fire safety.
This makes a lot more sense. We already have industrial protocols like SERCOS for closed loop motion control that are based on fiber, specifically for high data rate and noise immunity. Having done some work on locomotives years ago where there is no reliable ground to use for shielding and it's a very noisy environment, electrically, we talked back then about using fiber for communications. The problems we ran into were cost and installation complexity (you'd have to retrain all the electricians on how to run fiber). We needed something simple and cost effective.
"I have never let my schooling interfere with my education." - Mark Twain
Spin-meisters supreme!
Engineering is the art of compromise.
Why not have the sensors run from an induced current. If it's only possible for the sensors to be supplied with a low voltage, even if they short. doesn't that mostly eliminate the chance of sparking?
Gimli Glider.
If one were going to run a fibre to the fuel tank to begin with, why not just "visually" look down the fibre in the first place?
So they are going to use a higher energy source to power a very low energy device (great power loss due to light absorption/loss within the optic cable, and the inefficiencies of photovoltaics today) swimming in fuel. This (light) energy will still be converted to electricity in the sensor, but now with less electrical fault detection.
from 09 F9 11 02 9D 74 E3 5B D8 41 56 C5 63 56 88 C0
to 45 2F 6E 40 3C DF 10 71 4E 41 DF AA 25 7D 31 3F
With low enough power, there is no risk of arcing. None at all. With the optical approach, there is reisk of arcing, if they deliver enough power. The only thing gaines is less exposed wiring.
Most ACs are not even worth the keystrokes to insult them. Be generically insulted by this and ignored otherwise.
any relation to "Mostly Harmless" ;)
0.8kg/L ~= 1.76 lbs/L.
Am I confused, or should there have been within 1%, the same amount of fuel with either measurement?
Or is it simply that "all-new metric" means the gauges were either not properly labelled or read?
Most modern autos not only have the electronic fuel sending unit inside the tank but also the electric pump and with multiple tanks the electric diverting valve as well. All the components are simply very well insulated and sealed.
Wabi-Sabi
Matthew