Gigabit over Plastic Optical Fiber

[thiggins]

I received a pitch for this.
http://www.casacom.ch/components/com_hdflvplayer/videos/5_video.mp4

Note the casacom.ch site is in German...

Thoughts?

 

[azazel1024]

I can't see the linked video, because work. However POF has been around quite a while, but it is only recently the higher quality POF has been around. From what I've seen, the biggest issue is that it is niche for consumer implementations and gigabit is very new, with a number of competing implementations that aren't necessarily compatible.

Glass fiber generally seems to be the way to go. I can't find anything except brief articles on POF, but it looks like the cost is roughly similar to glass based fiber, but POF is massively easier on handling and installation, because it is very durable versus fiber. Downside is bigger cables and much higher attenuation, so the distance you can run various speeds is much shorter than glass fiber whether single mode or multimode.

 

[thiggins]

Ok. But why would you even consider it vs. CAT cable if all you're getting is Gigabit throughput?

 

[azazel1024]

Because there is zero possibility of EMI interference and it won't conduct a charge. So it can be great in industrial environments as well as direct bury outdoors.

Other than that, I can't think of a reason to possibly consider it.

 

[System Error Message]

with plastic that would make it much easier to use at home. I wonder if it makes the SFP modules cheaper or if it is compatible with existing fibre optic modules. I was hoping that plastic fibre optics would be cheaper. I think the plastic ones have been used with audio.

 

[thiggins]

with plastic that would make it much easier to use at home. I wonder if it makes the SFP modules cheaper or if it is compatible with existing fibre optic modules. I was hoping that plastic fibre optics would be cheaper. I think the plastic ones have been used with audio.

But why use it at home vs. CAT cables?

 

[System Error Message]

thinner and more flexible than CAT. Also unlike CAT you dont have to worry about the insulation of cables and you can run them with electrical cables with no interference.

 

[Terry Kennedy]

with plastic that would make it much easier to use at home. I wonder if it makes the SFP modules cheaper or if it is compatible with existing fibre optic modules. I was hoping that plastic fibre optics would be cheaper.

I've never seen a Gigabit Ethernet SFP for plastic. There have been 100Mbit ones, but they don't seem to be very common, and thus have a higher price than regular fiber SFP's. Plus you run into both the vendor lock-in issue and the low level of support for 100Mbit SFPs in general.

There is an IEEE study group for Gigabit Ethernet over POF, but there isn't even a timetable for producing a draft.

I don't see a good usage case for it - based on the status of Gigabit Ethernet over it, faster speeds (be they 2.5, 5, or 10GbE) seem to be very far off (if even possible). It is going to be more expensive than copper, if for no other reason than copper being ubiquitous, while support for any fiber is unusual in low-end switches and even "prosumer" type switches typically only support 1 to 4 SFP ports. Regular glass fiber is not as fragile as people think it is (I've given demonstrations of rolling over "zip cord" type fiber with an occupied office chair, tying it in knots, playing tug-of-war, and so on).

One of the proposed benefits of POF is ease of termination - OptoLock is similar to the back-wire option on electrical receptacles (cut wire and shove into hole) but I'm not sure how will that will work in the field with non-professional termination, particularly at GigE speeds. I used to make my own fiber cables (both single mode and multimode) but cables are now available at such a low cost (the whole cable assembly with connectors costs less than I'd pay for a single connector) that it isn't worth it.

I think the plastic ones have been used with audio.

TOSLINK has used plastic fiber, but in short (15' or less) cables at low bitrates.

 

[htismaqe]

thinner and more flexible than CAT. Also unlike CAT you dont have to worry about the insulation of cables and you can run them with electrical cables with no interference.

With no chance of interference. In reality, people run 1000bT all the time next to electrical cables and interference is still uncommon, if not rare. The biggest home use case for fiber is the one azazel listed - buried cable to outbuildings.

 

[System Error Message]

With no chance of interference. In reality, people run 1000bT all the time next to electrical cables and interference is still uncommon, if not rare. The biggest home use case for fiber is the one azazel listed - buried cable to outbuildings.

Have you read the article on this site relating to influence with cabling performance? I also listed audio which is accurate and also widely used. Although not widely used for LAN cabling because of its price.

 

[htismaqe]

Tim wasn't asking about audio signals. He wanted to know why he would use plastic core fiber vs. copper networking cable.

The simple answer is "you wouldn't".

Again, while interference from electrical wiring is possible with copper cabling, it doesn't really happen in the typical home use, especially when using higher quality cable.

The best use case for fiber in the consumer space is connecting to outbuildings where outdoor burial is required and lengths beyond 100m are necessary. And in that use case, you're almost certainly going to be running the fiber through conduit, which further reduces the need for using plastic over the better-performing glass.

Terry hit the nail on the head when he said "I don't see a good usage case for it".

 

[azazel1024]

Yeah, there isn't really a good use case for it. On distance over 100 meters, I haven't seen any tables of POF, but I do know even with the newer POF that IS capable of gigabit performance, the attenuation is still greater than glass fiber. Since OM2 is capable of "only" something like 430 meters at gigabit speeds for 1000base-sx (which is multimode), I'd imagine that POF is going to be a much shorter distance. Possibly not better than what Cat 5e/6/6a is capable of for gigabit speeds. Now at 100Mbps speeds, POF might be significantly further than.

Now if SFP module price was comparable and POF was cheaper than glass core, I could certainly see the option for outdoor use being very attractive, even if it wasn't for very long runs. Not needing to (or not needing to worry about) use lighting arrestors/surge protectors on the termination points of outdoor cat cabling is very seductive.

I have a run or two I want to make to an outbuilding that is about 100ft from my house at some point. currently nothing there, at some point I'll likely do a wireless bridge, but longer term I want to bury fiber as well as power to the outbuilding (going to install solar+battery+wireless bridge in the short term). If I can do POF SFP modules and POF direct bury cable for cheaper than glass core fiber, I'll certainly do that. However, I suspect I won't be able to, or at least not and also have at least gigabit speeds to the outbuilding (because I can get a wireless link faster than 100Mbps of actual performance at 100ft).

In a house, I see no reason you wouldn't use glass core or just cat cabling. POF isn't really cheaper than glass core fiber and if it is in wall for most of it, no difference in durability to worry about. Short fiber patch cables aren't terribly expensive even compared to cat cabling.

A bit of a solution looking for a problem. Now if POF ever CAN be made much cheaper than glass fiber with very high performance capability, even if it never has the range performance of glass core fiber, that could be extremely attractive. If you can manufacturer it in bulk such that you could sell the stuff for cheaper than cat cabling and similar performance levels to cat cabling in practice, you absolutely would have a real use case for the stuff.

 

[jdabbs]

But why use it at home vs. CAT cables?

How difficult is it to terminate POF vs Cat 5? The product doesn't offer much if already you have a spool of CAT 5, crimpers, ends, and experience and/or a line tester, but it seems preferable to stashing 15 ft of slack behind your TV stand because the shortest sufficient segment was a 50' CAT 5 cable on the shelf at Walmart.

 

[htismaqe]

I bought 50' of pre-terminated Cat6 cable on Amazon recently for $12. Since I was using it for a direct switch-to-switch run between my office and the central MDF, I "stashed" the slack in the attic where the cable was run. Is it perfect? No. It's hard to beat 12 bucks though.

 

[ociordia]

I can't see the linked video, because work. However POF has been around quite a while, but it is only recently the higher quality POF has been around. From what I've seen, the biggest issue is that it is niche for consumer implementations and gigabit is very new, with a number of competing implementations that aren't necessarily compatible.

Glass fiber generally seems to be the way to go. I can't find anything except brief articles on POF, but it looks like the cost is roughly similar to glass based fiber, but POF is massively easier on handling and installation, because it is very durable versus fiber. Downside is bigger cables and much higher attenuation, so the distance you can run various speeds is much shorter than glass fiber whether single mode or multimode.

You are right. POF has been around since several years with 100 Mbps maximum data rate. It has been only since last year that Gigabit speed is available.
There is only one commercial implementation of Gigabit POF that is currently standardized in Europe under ETSI (TS 105 175-1-2) and in process to write the standard in IEEE 802.3 (1000BASE-RH), expected publication date end 2016.

 

[ociordia]

with plastic that would make it much easier to use at home. I wonder if it makes the SFP modules cheaper or if it is compatible with existing fibre optic modules. I was hoping that plastic fibre optics would be cheaper. I think the plastic ones have been used with audio.

There are existing SFP modules with Gigabit POF "flavour". The very same company from the webpage referred by Tim, that is CASACOM, offers an SFP with POF.
Being an SFP means that you can plug it into any SFP device (typically a switch with SFP slots in it). What you cannot do is to use the Gigabit POF SFP with GOF...

 

[ociordia]

Ok. But why would you even consider it vs. CAT cable if all you're getting is Gigabit throughput?

At the end everything comes to installation. If Cat5e/6 is easy to install then POF doesn´t make sense in homes. But typically, at least here in "old Europe", installing copper in already constructed houses is a pain in the back as walls are made out of brick and there are not empty conduits to use.
On the other hand POF can use the very same mains ducts and the installation is much simpler.
But, again, if either you are a lucky USA citizen with those nice walls with holes inside or you don´t mind (or your wife doesn´t) to see an ugly copper cable routing all your home, POF is out of question here.

 

[ociordia]

I've never seen a Gigabit Ethernet SFP for plastic. There have been 100Mbit ones, but they don't seem to be very common, and thus have a higher price than regular fiber SFP's. Plus you run into both the vendor lock-in issue and the low level of support for 100Mbit SFPs in general.

There is an IEEE study group for Gigabit Ethernet over POF, but there isn't even a timetable for producing a draft.

I don't see a good usage case for it - based on the status of Gigabit Ethernet over it, faster speeds (be they 2.5, 5, or 10GbE) seem to be very far off (if even possible). It is going to be more expensive than copper, if for no other reason than copper being ubiquitous, while support for any fiber is unusual in low-end switches and even "prosumer" type switches typically only support 1 to 4 SFP ports. Regular glass fiber is not as fragile as people think it is (I've given demonstrations of rolling over "zip cord" type fiber with an occupied office chair, tying it in knots, playing tug-of-war, and so on).

One of the proposed benefits of POF is ease of termination - OptoLock is similar to the back-wire option on electrical receptacles (cut wire and shove into hole) but I'm not sure how will that will work in the field with non-professional termination, particularly at GigE speeds. I used to make my own fiber cables (both single mode and multimode) but cables are now available at such a low cost (the whole cable assembly with connectors costs less than I'd pay for a single connector) that it isn't worth it.


TOSLINK has used plastic fiber, but in short (15' or less) cables at low bitrates.

Hi Terry, the IEEE´s 1000BASE-RH will be ready by the end of 2016. The Task force is moving forward really quick.
Of course POF will always be more expensive than copper as the equipment does not implement POF ports and you will always need converters. The point is in the installation. See my reply to another post of this thread on how this makes sense on houses where Copper is hard to install and POF can be router sharing the mains ducts.
Optolock or Fiberdock which is the same works really well in the field. See Homefibre´s webpage for white papers and study cases of several installations done.

 

[ociordia]

Yeah, there isn't really a good use case for it. On distance over 100 meters, I haven't seen any tables of POF, but I do know even with the newer POF that IS capable of gigabit performance, the attenuation is still greater than glass fiber. Since OM2 is capable of "only" something like 430 meters at gigabit speeds for 1000base-sx (which is multimode), I'd imagine that POF is going to be a much shorter distance. Possibly not better than what Cat 5e/6/6a is capable of for gigabit speeds. Now at 100Mbps speeds, POF might be significantly further than.

Now if SFP module price was comparable and POF was cheaper than glass core, I could certainly see the option for outdoor use being very attractive, even if it wasn't for very long runs. Not needing to (or not needing to worry about) use lighting arrestors/surge protectors on the termination points of outdoor cat cabling is very seductive.

I have a run or two I want to make to an outbuilding that is about 100ft from my house at some point. currently nothing there, at some point I'll likely do a wireless bridge, but longer term I want to bury fiber as well as power to the outbuilding (going to install solar+battery+wireless bridge in the short term). If I can do POF SFP modules and POF direct bury cable for cheaper than glass core fiber, I'll certainly do that. However, I suspect I won't be able to, or at least not and also have at least gigabit speeds to the outbuilding (because I can get a wireless link faster than 100Mbps of actual performance at 100ft).

In a house, I see no reason you wouldn't use glass core or just cat cabling. POF isn't really cheaper than glass core fiber and if it is in wall for most of it, no difference in durability to worry about. Short fiber patch cables aren't terribly expensive even compared to cat cabling.

A bit of a solution looking for a problem. Now if POF ever CAN be made much cheaper than glass fiber with very high performance capability, even if it never has the range performance of glass core fiber, that could be extremely attractive. If you can manufacturer it in bulk such that you could sell the stuff for cheaper than cat cabling and similar performance levels to cat cabling in practice, you absolutely would have a real use case for the stuff.

POF today delivers Gigabit up to 80 meters with RED LEDs. With Green LEDs it delivers 100 meters, although these devices are still not in the market (expect them on 2016)
I disagree on the GOF comparison. Obviously you can not compare GOF with POF. POF is a piece of wood compared to such an ideal comm media like GOF. The point here is that if a user ( and I mean a normal user , that is, one that has to call an installer and can not make the installation by himself) were to instal a GOF backbone in his house it will be much more expensive than doing it in POF. You can not just compare the cost of POF itself (30 cents per meter) with GOF but the termination (no needed in POF...) and the skills of the installer (which in turn translates into $$$ charged to you for each hour spent).
There are all these nice "ready" to install GOF systems but go and try to ask for the price...
And finally and once again, if Copper can be installed, POF is out of question... the point of POF comes when copper is hard or impossible to install.

 

[ociordia]

How difficult is it to terminate POF vs Cat 5? The product doesn't offer much if already you have a spool of CAT 5, crimpers, ends, and experience and/or a line tester, but it seems preferable to stashing 15 ft of slack behind your TV stand because the shortest sufficient segment was a 50' CAT 5 cable on the shelf at Walmart.

POF is just cut with a simple cutter (30 cents tools). No polish, nothing else. 5 seconds. 30 cents per meter.