Slashdot Mirror
Plastic Fiber Could Make Optical Networking a DIY Project
An anonymous reader writes "A new European project using plastic fiber and off-the-shelf components could make optical networking so cheap and simple that installation could be a DIY job for even a non-technical person. The object of EU-funded POF-ALL project is to find a technical solution to the rising cost of taking optical fiber right into the home." A mere "few hundred metres" of 100mbps (since plastic is thus far dimmer than glass) would suffice to wire any home I'm likely to occupy.
What's the benefit of 100mbps plasti-fiber over gigabit cat-6?
Honestly, unless it becomes a lot cheaper than Cat-5 UTP, I think it's going to be a non-starter. Now, if it was 100 GBPS, that would be a different story...
Logic is the beginning of reason, not the end of it.
Surely one thing we've learned in the field of computer technology- Whatever seems like more than enough is just too little. Not that this stuff doesn't look cool, it's just that in a year or two something 10 times better will come along, and cool new applications for many Gbps will probably arrive shortly after that... Those holo-hookers won't animate themselves.
"Be light, stinging, insolent and melancholy"
Slightly off topic, but I've been looking for a decent (inexpensive) source for fiber optic cables for doing a small star-ceiling.
I was actually thinking of using these guys - , but I would be interested if anybody could come up with alternative recommendations. I poked around a little and I can't seem to find any consumer sources for plastic fiber. (you know, other than the bait and tackle shop)
Does TOSLINK optical audio not count as a DIY network? I didn't pay anyone to hook up my AV stuff.
Because my favorite cable is a TOS-LINK cable with a clear sheath, over the fiber optics.
(Yes, I am a nerd with a favorite kind of cable.)
Granted there it is a step up to go from a 6-foot cable to 100 feet, but it isn't that big of a deal. Bi-directional communication is another thing that would be needed to make a real network.
Amazon.com has a bunch of 100-foot fiber optic cables, so I don't think that fiber itself is the issue, getting the network cards cheap enough is more of an issue, I think.
If I have nothing to hide, don't search me
The article is Slashdotted and the summary doesn't address it, so... what are the advantages of 100Mbps plastic fiber over wired and wireless Ethernet?
Where do you think plastic comes from? Yes start making fiber from plastic and watch the price of petrol skyrocket! Stupid idea.
is in the trenching and pulling cables. That's not going to change by going to plastic etc.
Engineering is the art of compromise.
[quote]100mbps would suffice to wire any home I'm likely to occupy.[/quote]
Yes, and I think there is a world market for maybe five computers.
Also, 10 megabytes should be enough for anyone.
I picked up some surplus loose buffer tube multi-mode fiber. I'd like to try my hand at putting the terminations on the fibers? I got the fan-out tubes but I'm stuck on how to put ends on. Suggestions please.
Plastic fibre slashes optical network costs
Wed, 01/09/2008 - 19:49 - Wire Services
A new European project using plastic fiber and off-the-shelf components could make optical networking so cheap and simple that installation could be a DIY job for even a non-technical person.
The object of EU-funded POF-ALL project is to find a technical solution to the rising cost of taking optical fiber right into the home.
The project partners decided to focus on the cabling inside buildings, which would typically account for 30% of the cost of laying an optical fibre from the exchange into the home. This last hundred metres or so is known as the 'edge' network.
"We realised that we could lower the cost of this edge installation by using a simpler technology," Alessandro Nocivelli, the founder and CEO of Luceat SpA, one of the partners in the project, said. "If we could employ a technology which is so simple to use that anyone can install it, that would relieve telecom companies of 30% of the cost of the access network, which means up to several billion euro if you consider the European Union as a whole."
Plastic fibres use harmless green or red light that is easily visible to the eye, as opposed to glass fibres which use infrared laser light that could potentially cause eye damage.
"I have a two-year-old child," says Nocivelli, "and I would never install a glass optical fibre in my own home, even though I have been working with glass optical fibers for many years."
Plastic fibres are also much thicker than glass fibres, a millimetre or more, and can be handled without special tools or techniques.
"You don't need to be trained to handle and install it. You just cut it with scissors, plug it in and it works. It's as easy as that," Nocivelli adds.
On the downside, plastic fibres absorb light more than glass, which limits their useful length to a few hundred metres.
They also have a lower data capacity than glass fibres, but that is not an issue for the cable that runs from a conventional glass fibre in the street into a house, or even for laying a network within a block of flats.
The partners have built a system that uses green light to transmit 100 megabits a second over a distance of 300 metres, which is the speed telecom companies hope to offer their customers five to ten years from now, and 50 times as fast as a typical adsl broadband connection.
Their second achievement is to transmit ten times faster still - one gigabit per second - over a 30m fibre, using red light.
By the end of the project in June 2008, they expect to have extended that to 100m.
"Then, of course, we will try to focus on longer distances," says Nocivelli. "We have already demonstrated that plastic fibre would be future-proof not only for the next ten years but for the next 30 years. With that speed in your home you could download a full DVD in thirty seconds."
The POF-ALL members have not had to develop any novel technologies, as they have built their systems using the latest off-the-shelf components and the ingenuity and skill of the ten academic and industrial partners.
Two products are already coming to the market. Luceat is commercialising an optical Ethernet switch (a router) using plastic fiber technology and the Fraunhofer Institute is looking for partners to market an integrated optical transceiver to work at one gigabit a second with plastic fiber.
Home and office networks could be rewired with plastic optical fibre so simply and cheaply it could be a do-it-yourself job.
"It's future-proof," confirms Nocivelli. You run at 100 Mbit/s today, 1 Gbit/s tomorrow and maybe 10 Gbit/s in the future."
A follow-up project, POF-PLUS, is intended to further develop optoelectronic components for plastic fiber and is awaiting a final decision on EU funding.
since about 1985. I designed a system back then, using plastic fiber and off-the-shelf HP transmitter and receiver modules, to pass data between elevator controllers where they are in a coordinated group. Isolation was the main reason, but it's also very convenient. We're still producing the same system today. It's convenient that it uses visible light and termination is very easy since the fibers are relative large. We're using relatively low data rates and the maximum distance I've got to handle is less than 100 feet.
"Do the Right Thing. It will gratify some people and astound the rest." - Mark Twain
1 They say that they can do 100mbits with green light now and hope to do 1Gbits with red light soon... I thought that the higher the frequency the higher the bandwidth?
As to the benefits over Cat5 I can see a few.
One is cost. Copper is getting more and more expensive plastic is cheap. Not only that but it should be lighter to ship and easier to install since it is smaller than CAT5.
Then you have safety. You don't have to worry about shorts and other issues with fiber. Not a huge problem but I can see where some people would like it.
And you have the lack of EM from fiber. No not for safety reasons but for interference with other devices. I have never seen a problem with it but anytime you have a signal running on a wire you have the potental for EM.
See my blog http://ilovecookes.blogspot.com/ for light hearted technical information.
Plastic fibre slashes optical network costs
Wed, 01/09/2008 - 19:49 - Wire Services
A new European project using plastic fiber and off-the-shelf components could make optical networking so cheap and simple that installation could be a DIY job for even a non-technical person.
The object of EU-funded POF-ALL project is to find a technical solution to the rising cost of taking optical fiber right into the home.
The project partners decided to focus on the cabling inside buildings, which would typically account for 30% of the cost of laying an optical fibre from the exchange into the home. This last hundred metres or so is known as the 'edge' network.
"We realised that we could lower the cost of this edge installation by using a simpler technology," Alessandro Nocivelli, the founder and CEO of Luceat SpA, one of the partners in the project, said. "If we could employ a technology which is so simple to use that anyone can install it, that would relieve telecom companies of 30% of the cost of the access network, which means up to several billion euro if you consider the European Union as a whole."
Plastic fibres use harmless green or red light that is easily visible to the eye, as opposed to glass fibres which use infrared laser light that could potentially cause eye damage.
"I have a two-year-old child," says Nocivelli, "and I would never install a glass optical fibre in my own home, even though I have been working with glass optical fibers for many years."
Plastic fibres are also much thicker than glass fibres, a millimetre or more, and can be handled without special tools or techniques.
"You don't need to be trained to handle and install it. You just cut it with scissors, plug it in and it works. It's as easy as that," Nocivelli adds.
On the downside, plastic fibres absorb light more than glass, which limits their useful length to a few hundred metres.
They also have a lower data capacity than glass fibres, but that is not an issue for the cable that runs from a conventional glass fibre in the street into a house, or even for laying a network within a block of flats.
The partners have built a system that uses green light to transmit 100 megabits a second over a distance of 300 metres, which is the speed telecom companies hope to offer their customers five to ten years from now, and 50 times as fast as a typical adsl broadband connection.
Their second achievement is to transmit ten times faster still - one gigabit per second - over a 30m fibre, using red light.
By the end of the project in June 2008, they expect to have extended that to 100m.
"Then, of course, we will try to focus on longer distances," says Nocivelli. "We have already demonstrated that plastic fibre would be future-proof not only for the next ten years but for the next 30 years. With that speed in your home you could download a full DVD in thirty seconds."
The POF-ALL members have not had to develop any novel technologies, as they have built their systems using the latest off-the-shelf components and the ingenuity and skill of the ten academic and industrial partners.
Two products are already coming to the market. Luceat is commercialising an optical Ethernet switch (a router) using plastic fiber technology and the Fraunhofer Institute is looking for partners to market an integrated optical transceiver to work at one gigabit a second with plastic fiber.
Home and office networks could be rewired with plastic optical fibre so simply and cheaply it could be a do-it-yourself job.
"It's future-proof," confirms Nocivelli. You run at 100 Mbit/s today, 1 Gbit/s tomorrow and maybe 10 Gbit/s in the future."
A follow-up project, POF-PLUS, is intended to further develop optoelectronic components for plastic fiber and is awaiting a final decision on EU funding.
The simple fact is that Copper is about to become VERY expensive. China is buying it as well as working with all countries that have copper mines to aquire full access to the copper. The simple fact is that there is a limited amount of copper and China is about to use 1000x more than what it currently does. In addition, most societies are about to move to electrical cars which will require a lot of copper.
Finally, copper does not go the long distances that Fiber does. I suspect that we will see a lot of uses for these in running from the green box to the home. In fact, I think that the delivery companies will have multiple cables to the home. Basically, dark fiber. It will enable some interesting services.
I prefer the "u" in honour as it seems to be missing these days.
I work for a big telco, the one that runs fiber to the home, we don't put ends on the fiber, at least not directly. what we do is use a fusion splicer, and use a connector with a length of fiber already attached, we then splice that to the end of the fiber..
wanted: one clever sig,apply within
Chums up, let's do this!
Copper is pretty cheap AND so is all it's components.
I have around 12 devices networked in my house.
How much will that Plastic Fiber equipment cost?
Using $5 NIC's, a $80 smart switch, and copper cable seems like a price that will be hard to beat.
With the rising cost of copper products, could plastic be a cheaper?
The best thing my ISP offers at the moment is 10mbs, and they charge me 70 bucks for it. While I enjoy 1MB/s downloads from Bangbros, nearly instant webpage loading, and zero lag in gaming, they would be hard pressed to squeeze any more money out of me for an internet connection. I think it will take several years, if not a decade before we see a full throttle 100mb internet option. We already have cables just fine for dealing with that amount of bandwidth, why re-wire your house when They (Charter in my case) have yet to utilize the capability of the current infrastructure.
"The irony when tending a flock of sheep is the dogs you put in place to protect them are genetically mutated wolves"
You do realize that an optical signal is an "electrical" signal right? So how are you going to make a router out of "optical" components, just a splitter? I don't see how that would work, but maybe I'm just dumb.
I don't see a "Wired" solution being a great option for computer connections at the home anymore. I once envisioned running Cat-5e all over my house. But why??? When wireless N speeds are quite fantastic around 134 Mbit/s (at least that's what it tells me).
I suppose in a network of multiply computers, it might be really nice to have around the house.
It's left blank because I have nothing to say to you punks!
That may be true now but currently plastic is a petrochemical product. As oil prices rise so will the prices of plastic fibers. Copper will rise as well but at least in the US copper can be locally mined thus reducing transportation costs.
FalconShould there be a Law?
back in 96 i worked with a research group trying to develop POF and the day a competing research group demonstrated 1GB over copper, POF was dropped like a hot potato. There are too many issues with POF. From the materials side, first you have to designing a polymer that has a high enough refractive index to be able to transmit signal for more than 6 feet. Then this polymer has to support tubular extrusion at very high rates like > 1000/min, very little variation in I.D/O.D/and wall thickness, be able to be flexible enough to be installed in temperatures ranging from -20F to 110F, be insensitive to moisture, shrinkage, and long term environmental cycling. As for the finished product, it has to cost at least 20% less to install and maintain than copper for an installer to even suggest it to customers and that assumes that a large structured wiring manufacturer is making or partnering with someone to make termination/networking equipment.
Cold day in hell when this will be in main stream. I'm sure this is a nice grad student project/patent generator...
One is cost. Copper is getting more and more expensive plastic is cheap. Not only that but it should be lighter to ship and easier to install since it is smaller than CAT5.
It's very possible for plastic get more expensive than copper. First most plastic in the US today is a petrochemical product. As oil is depleted the cost of plastic will rise. Secondly because of the depletion of oil transporting it become more expensive as well. However copper can be and is mined in the US.
FalconShould there be a Law?
Spencer's Gifts has been selling these for cool bachelor pads since the 1970's, man. What next, are we going to use lava lamps and blacklights for data transmission, man?
Slashdot "libertarians": Small government for me, big government for those I disagree with. -1, I disagree with you
For some reason, people have it stuck in their heads that plastic fiber is new. It's not. Also, it can carry 10Gb/s just fine. All the 100-300m links are class 1. In fact, I'm looking at a 50Km rated SFP that is Class 1. According to all the safety ratings, you can stare at its laser as long as you like. And wavelength has nothing to do with power. The 50Km SFP that I just mentioned is infrared.
It looks like they're solving problems, badly, that have already been solved. MS Windows and their broken "shortcuts" if anyone remembers? If I didn't think it was just plain ignorance, I would claim this was a well-disguised FUD piece.
our best place to obtain copper may be in the dumps as well as overhead lines that were put in 40 years ago.
I didn't want to say anything about recycling, but you're right, dumps and all the copper cables already laid down may be a better source.
We do have SHITLOADS of coal that can be changed into feedstock for plastics.
A better source of fee stock for plastic may be hemp with bioplastics being renewable. My question then is would bioplastics be good for fiber optics.
FalconShould there be a Law?
A major advantage of optic fiber on CAT6 is distance. CAT6 has a maximum distance of ~75 metres, but with optic fiber you can run connections upwards of 1km.
However as TFA says plastic fiber does not have the distance range glass fiber has. Plastic fiber only can go a few hundred meters. That's more than Cat6 but it's not really that far.
FalconShould there be a Law?
"I have a two-year-old child," says Nocivelli, "and I would never install a glass optical fibre in my own home, even though I have been working with glass optical fibers for many years."
Go ahead and take out the electrical wiring too then. It's far more dangerous.
Other than this text, there is no discernible information contained in this sig.
Aluminum isn't as good of a conductor for power transmission as copper. The reason that it is being used for power lines is that it is a lot cheaper.
Thanks, I didn't know, or recall, that aluminum was used for power transmission.
FalconShould there be a Law?
Nail-on-the-head! The cost of the medium is NOTHING compared to the current cost of the interfacing.
You plug in a usb or firewire or CAT-x or RS-xxx with no problems at all but there AREN'T any optical interfaces built into any computers yet that _I_ know of. The only consumer-level optics I've seen so far are in audio (and higher-end stuff at that) and if you look at the optical/CAT-x converters you'll have quite a heart attack at the prices.
"It's time to take life by the cans." ~ Bender ("Bendin' in the Wind", ep. 3-13)
Off topic, but since you started.. :)
Yes they are ignorant (MS), but not that ignorant, methinks. The main reason for not using symbolic (or hard, for that matter) links was that this requires a fundamental change in the filesystem - and was not nearly as easy to do with FAT (even the new VFAT, later renamed FAT16) filesystem.
Since then, of course, the ".lnk" files have taken a role similar to the old ".pif" files (Program Information File, used to store meta-information about DOS programs in Windows) - in that they can contain various attributes beyond just the path to the referenced file. (As a matter of fact, the extension is not important, so long as it is "executable". You can rename any executable file, be it a batch file, a command file, a DOS/Windows executable, etc, and rename it with a ".lnk" extension, and it will still run fine. This is a bug that is frequently exploited by trojans/viruses).
While you are on that topic, the Macintosh "Aliases" carried forward from Mac OS to Mac OS X (since both use the HFS+ filesystem), and are still widely used on that platform. When viewed in the "Finder" (the counterpart to the Windows Explorer), they appear the same way as symbolic links - but there are some differences. First, if the original file is moved or renamed, the Alias remains valid - while a symbolic link (since it is no more than a pointer by name) breaks. OTOH, it is not supported within the filesystem itself - so you cannot for instance go to the command prompt and say "cd diralias", where "diralias" is an alias to the directory.
In any case, every bureaucracy (whether it be the EU, the US, MS, or Apple) tend to have a "not invented here" attitude to outside inputs, and so keep reinventing and making incompatible standards. It's called "progress"..
Five to ten years from now? I think they mean last month. http://abonnez-vous.orange.fr/fibre/fibreres/avoirlafibre/Default.aspx?md=10
Watch this Heartland Institute video
Unfortunately the content of the article doesn't really address the fundamental challenge they are trying to address. The primary costs in fibre to the home are not in the home cabling but in the fibre construction work to connect up to every home (or out in front to a pit). The cost of glass fibre is also not that significant in comparison to the in home construction costs, and hence the whole thrust of this initiative does not seem to materially help their stated goal. The primary costs are installation related, not cabling type.
...and I would never install a glass optical fibre in my own home" is a self justifying statement to rationalize their plastic fibre work. A sharp piece of copper wire, a kitchen knife, electricity, laundry cleaners and drugstore items are going to be far bigger threats to children than a terminated optical fibre cable that would most like terminate in a box on a wall, where ethernet via catX cable (or optical patchcords) would connect to home networking gear. For termination of fibre to the home CPE, think of something located near the home electrical switchboard, which may be internally in a wiring closet, or external. In addition optical levels in access networks are also practically not at the levels to cause harm (one would have to work very hard e.g. have a microscope and focus on the end of a terminated fibre - in additiona with an equipment at the exchange end without any optical safety).
:-). Let the Bit Torrents run free.....
I suspect to, their whole statement about "..I have a two-year-old child,"
I suspect also home data distribution will be ethernet based and this will also undermine their stated research purpose. So if I have a PC/Mac with Gige, DVR with 100mEth, a home network with an CatX interconnections, guess what technology people will look to connect with? Inside the house, CatX is likely to be most common. Gigabit Ethernet switches are already affordable and work well over wired connections and have sufficient range to work through most homes. The interconnect speed to a fibre network right now is in a few 10s of Mbits, with hopefully up to 100M being more common within 3- 5 years. Hence inside the home, even 100M looks fine today, with GigE being fine for a house 'backbone' for the next 5 years. And I hear we expect to have more economic 10G Eth wired connections in early 2012, so there is a path to an economic house 10G backbone next decade...
We do the same. The downside is you end up with several joins and you need an expensive fusing machine to do it.
I've heard (but not seen in action) that there are now fibre equivalents of crimping kits. You clean the end of the fibre and clamp it to the ends. It's supposedly much faster and only creates one join so less loss. This would probably be a better option to just terminate a few cables.
Good article here, although it reckons using crimps wouldn't suit your situation as it takes some practice.
There's a lot invested in the CAT5/6 market: switches, routers, motherboard NICs, know-how, etc.
To displace copper any time soon, fiber will have to be a lot bigger value.
This means either a lot cheaper, a lot faster, a lot better-in-some-other-way, or some combination that adds up to "a lot."
The plusses vs. CAT5e or 6:
* Small cost savings for the wires in a typical home, bigger cost savings for larger facilities.
* Electrical isolation and immune from radio noise.
* Can be coiled without becoming an antenna.
The minuses vs. CAT5e:
* No power-over-Ethernet.
* not compatible with most motherboards and NICs, meaning increased costs during early-adopter period.
* not compatible with most switches and routers, meaning increased costs during early-adopter period.
The uncertainties:
* How many years before plastic degrades?
* How many years before it's signal-carrying capacity is too slow?
* What happens if it becomes a market flop?
* What is the future cost of plastic?
* What is the future cost of copper?
* What is the future cost of glass?
* What is the future carrying capacity of copper?
* What is the future carrying capacity of plastic?
Knowledge is how to play a game, intelligence is how to win, wisdom is knowing what game to play.
... in the same discussion. Impressive.
Good link; It's packed with info. Thanks!
With more and more devices becoming data-aware, wouldn't it make sense to revise the household electrical outlets to have some kind of standardized data jack? This way, I can plug in my 60" HD-Youtube-compatible TV into the wall with ONE cord and have it work!
No, I will not work for your startup