Troubleshooting Fiber Optic.

I know this my be a tad off subject, but i was wondering if any of you guys can help me out, and/or answer a few questions for me.

I work for a company where I pretty much get asked to take care of anything that deal with electronics ( PLC's, Switches Routers, Computers, Nuclear Density Meters, VFD's,...ect). We have several fiber optic cables ran across our property, some are for internet, some are for industrial truck scales, and some is for our PLC system. So, i figured it would be a good idea who learn how to take care of it, and repair when needed.

The cable in question runs between 2 buildings with a fiber to ethernet converters on each end. The fiber run is about 3/4 of a mile. The connection is simply there to provide internet access to a building. On Saturday, I saw that the internet was out, and traced it all the way back to the fiber to ethernet box. I have 2 spares and I replaced the box on each end and still nothing changed. The boxes shows that the copper side is communicating just fine, however, the fiber side shows nothing. So, im thinking it has to the actual fiber optic cable that runs from building.

IM not familiar with fiber at all. The cable comes into both buildings and runs into a Cornning box that has a blue cable that goes into a fiber fan out, that splits into 6 different fiber cables.

I came up with a few questions that Im having a hard time answering.

1.)What is the best way to troubleshoot a cable like this ?

2.)Is the cable i described a multimode, or single mode cable ?

3.) Are there any tools I can buy to make jobs like this easier ?

HoldenC said:

1.)What is the best way to troubleshoot a cable like this ?

Click to expand...

Sometimes the fastest way to tell is swap to unused fibers on both ends and see if communications is restored.
Second method is a quick continuity check. On the receive end you can use your cellphone camera to see the IR transmitter if it is 1310nm (may not work with 1550nm). The other method is someone at the far end with a bright light and just look at the end of the cable and you should see light.

2.)Is the cable i described a multimode, or single mode cable ?

Click to expand...

Blue jacket is almost always single mode.

3.) Are there any tools I can buy to make jobs like this easier ?

Click to expand...

A visible fault locator is a good start. It is a bright source that you can attach directly to the cable, easier than a flashlight, plus if there is a break you can actually see light leaking out of the cable. You should have an optical power meter as well. More expensive but you can see if there is excessive loss in the cable or connectors. Ideally, you would want an OTDR, optical time domain reflectometer. As you can imagine from the name this is a very expensive piece of kit but it you allow you to locate any breaks or damage in the cable even from several km away.
Since you only have a couple of fibers I would just worry about getting a low-end visible fault locator. Also, when you are ordering make sure you get cleaning tools for your fiber.

I want to thank you so much for your reply. It was just what I was looking for.

The cable contains 6 fibers, and I have swapped to all of them and came up with nothing.

I did this other test, that i read about online. Not sure if it worked or not. I have some ethernet to fiber boxes that state you can connect a single cable ( we use ST connectors), from the TX to the RX and see if the activity lights flash. So, i used 3 fiber patch cables to connect fiber 1 to 2, 3 to 4, and 5 to 6. Then I went to the other end and tried a continuity test. I got no activity on any pair. So, im leaning on a broken cable somewhere.

Here is another questions. You stated that those cables are usually a single mode fiber, however the fiber to ethernet boxes state they are for multi-mode, and the patch cables that have always been in a place show 62.5mm on them. Can these cables be on a single mode cable run like this ?

Any construction or landscaping been done? Usually it goes media converters, patch cables, termination, then actual media failure in that order.

Did someone swap connectors around?

A visual fault finder is cheap and can easily be bought. Also when was the last time the fiber connections were cleaned?

OTDR's can be rented, but unless you have some training on it, it will be a super expensive brick.

HoldenC said:

Here is another questions. You stated that those cables are usually a single mode fiber, however the fiber to ethernet boxes state they are for multi-mode, and the patch cables that have always been in a place show 62.5mm on them. Can these cables be on a single mode cable run like this ?

Click to expand...

They shouldn't be. Sometimes with enough power and shorter runs you can get it to work. Single mode is 9um compared with the 62.5 um you have. So on the receive side there isn't a problem as you have a 9 um wide exit going into a 62.5 um entrance. However on the other end you have a 62.5 um exit going into a mere 9 um aperture. Might work if there is enough light in the central core but most likely would fail.

Double check the blue jacket for any markings. Based on the fact that this used to work its quite possible you got a blue jacket that isn't single mode.

BTW when you say blue do you mean dark blue or aqua? Dark blue is supposed to be single mode while aqua is multi-mode 50/125. Sorry I should have picked up on that sooner.

Maxkling said:

Any construction or landscaping been done? Usually it goes media converters, patch cables, termination, then actual media failure in that order.

Did someone swap connectors around?

A visual fault finder is cheap and can easily be bought. Also when was the last time the fiber connections were cleaned?

OTDR's can be rented, but unless you have some training on it, it will be a super expensive brick.

Click to expand...

Thanks for the info on the failures. Ive seemed to rule out media converters and patch cables so far. I see what I can do tomorrow to rule out anything else.

NO construction at all. Our fiber cable actually runs on telephone poles across our property. So, it would be kind of hard for anything to hit it. I did a quick visual check today and couldnt see anything wrong.

ON the note about cleaning the connections. Thats something that has never been done. I wasnt even aware that you had to. The connections are in a dusty environment, and im thinking the fiber system was put in about 7 or 8 years ago. Is there a quick way to clean the connections ?

Timbert said:

They shouldn't be. Sometimes with enough power and shorter runs you can get it to work. Single mode is 9um compared with the 62.5 um you have. So on the receive side there isn't a problem as you have a 9 um wide exit going into a 62.5 um entrance. However on the other end you have a 62.5 um exit going into a mere 9 um aperture. Might work if there is enough light in the central core but most likely would fail.

Double check the blue jacket for any markings. Based on the fact that this used to work its quite possible you got a blue jacket that isn't single mode.

BTW when you say blue do you mean dark blue or aqua? Dark blue is supposed to be single mode while aqua is multi-mode 50/125. Sorry I should have picked up on that sooner.

Click to expand...

Ill check in the morning and get back to you on the exact color. With all of the fiber color I cant think. I know the cables that connect from the cabinet to the fiber to ethernet boxes at both locations are 62.5

Is this type of cable ususally referred to as loose buffer, and such requires a fan out kit ? We have orange fiber cable that runs in conduit that has 10 connections and doesnt require fan out kits.

Im new to all this and really appreciate your patience.

HoldenC said:

Ill check in the morning and get back to you on the exact color. With all of the fiber color I cant think. I know the cables that connect from the cabinet to the fiber to ethernet boxes at both locations are 62.5

Is this type of cable ususally referred to as loose buffer, and such requires a fan out kit ? We have orange fiber cable that runs in conduit that has 10 connections and doesnt require fan out kits.

Im new to all this and really appreciate your patience.

Click to expand...

Yes. Sounds like you have multi-mode, loose buffer fiber optic cable. You can clean the connectors with an alcohol (>90%) swab, or you can purchase fiber optic cleaning swabs. You should also use compressed air to clean bulkhead connectors inside the cabinets.

Bit_Bucket_07 said:

You can clean the connectors with an alcohol (>90%) swab

Click to expand...

This reminded me that in certain countries of this world alcohol like that is only possible to purchase with a license from the Government... glad I left that place.

Step one: Get a high quality fiber tester. Expect to spend between $1500 and $5000 for one, don't bother with the $100 **** from eBay. The tester should be able to handle any fiber out there now, FDDI and OM1 through OM4. It should have a built in TDR to determine cable length, or length to fault.

Step two: Test your fiber; DO NOT ALLOW THE TIPS TO TOUCH ANYTHING. If they do, clean them. Test the fiber at both the intended bandwidth, and at least the next higher. Most Fluke testers will actually show all supported bandwidths for a hunk of cable automatically.

That's it. Randomly swapping things around is not sufficient to ensure that the actual communications link is functioning as it should. It is entirely possible (on both fiber AND copper) to have what appears to be a good connection (link lights come on, traffic can pass), but is actually only passing 1 to 5% of the traffic reliably, and wasting most of the bandwidth re-transmitting packets.

Just as a side note... I did a fibre installation course many many moons ago (about 25 years ago) and I remember that the Optical Time Domain Reflectometer (OTDR) used to use a principle called Fresnell Reflection in order to determine where the break/bad bend is in a fibre.

The principle states that a certain percentage of the light (7% or 8% can't remember?) gets reflected when light goes from one medium into another (for instance from glass to air). So, because the speed of light is a constant, the OTDR fires light down the fibre. If it receives a reflection then you can work out extremely accurately *where* the 'leak' is (i.e. half the time from sending the light to receiving it and then do the maths with the speed of light).

Not really much help to the OP of the question but I always found this little titbit of info very interesting and it has stuck with me all these years despite me only working for a year or so in fibre.

rdrast said:

Step one: Get a high quality fiber tester. Expect to spend between $1500 and $5000 for one, don't bother with the $100 **** from eBay. The tester should be able to handle any fiber out there now, FDDI and OM1 through OM4. It should have a built in TDR to determine cable length, or length to fault.

Click to expand...

I wouldn't make that kind of an investment simply for maintaining on-site fiber optic cables. It would make sense for someone who intends to do fiber optic installations on a regular basis. I suppose it would also make sense for maintenance personnel who work for companies that have unlimited budgets.

A simple laser is usually sufficient for basic troubleshooting.

Would anyone have anything to say about this product ?

https://www.amazon.com/gp/product/B00DFTW9SY/ref=ox_sc_act_title_1?smid=A10EAPE4CAYC9P&psc=1

HoldenC said:

Would anyone have anything to say about this product ?

https://www.amazon.com/gp/product/B00DFTW9SY/ref=ox_sc_act_title_1?smid=A10EAPE4CAYC9P&psc=1

Click to expand...

Yes. A simple laser is usually sufficient for basic troubleshooting.

HoldenC said:

Would anyone have anything to say about this product ?

https://www.amazon.com/gp/product/B00DFTW9SY/ref=ox_sc_act_title_1?smid=A10EAPE4CAYC9P&psc=1

Click to expand...

I'm not sure you need something as powerful as 20mW for your application. It would work. It just means you should be careful using it to avoid eye damage. (i.e. don't look directly down the fiber, instead shine it on something or project on a piece of paper.) The 1mW version looks like it is designed for cables less than 5km.