One of the most common challenges when deploying and maintaining “full fibre” (FTTP) broadband ISP networks is the impact of dust. A single microscopic speck of dust – impossible to see with the human eye – could feasibly lead to a data communication failure and create performance problems for thousands of customers.
All of this becomes a bit easier to understand when you consider that a single fibre strand, carrying pulses of light from a laser, is no thicker than a human hair. In order to see such dust, you’d need a microscope. The presence of dust impacts the transmitted light by causing signal degradation as it absorbs, refracts or reflects some of the light being sent down the fibre.
On top of that, dust can also create permanent defects on the fibre (microscopic scratches or pits), which is another challenge. Plus, any debris that gets inside can then break up and spread wider. Sadly, you can’t just solve all this by wiping the dust away manually because the contact friction, and resultant electrostatic charge, will end up drawing oppositely charged microscopic dust particles onto connector endfaces (like a magnet).
Suffice to say, keeping the endface of such fibres clean is a constant concern for, we hope, every network operator (i.e. dust is to fibre connections, what water ingress is to copper lines). Most modern fibre exchanges are thus built with climate control and dust-filtration, but this won’t catch everything, and ultimately it’s up to the engineers to follow the best possible practices.
CityFibre has just published a useful editorial on this common, albeit rarely discussed, challenge. The article attempts to highlight the benefits of taking a zero-tolerance approach to cleaning practices. Indeed, out of the many tens of thousands of connections made in their Fibre Exchanges (FEXs) last year, they claim not to have suffered a single failure due to dust.
By investing in the right tools and technology and enforcing a zero-tolerance policy on poor fibre inspection, cleaning and testing processes though, we’ve proven that it is possible to prevent dust and debris causing any issues within our FEXs.
Complacency is the enemy of the fibre network though, especially in a FEX. Once a connection is in and the first customer is live you can’t just go back later and disconnect a fibre just to clean it – not unless a fault is reported. It’s therefore essential to ensure steps are followed correctly at the outset. It might take three, five or even ten times longer to do these steps right, but it can be the difference between having to return next week or never!
All the third-party fibre engineers who touch our network (as well as our own people) are required to have the right testing and cleaning equipment. They must then use it to test and clean every fibre connection, every single time with no exceptions. Once ‘scoped’ and cleaned, a pass/fail software check must then be run. If it fails, the clean and test process must be repeated as many times as needed to secure the pass result. Photographic evidence of good connections and successful tests are also required by us.
Today in CityFibre, “Inspect and clean before you connect” is the mantra that every engineer working on our network knows and respects. We acknowledge though that if they didn’t have to inspect and clean fibres meticulously, they could do four to five times more FEX fibre patching jobs each day. This means it’s important to make our quality over quantity priorities crystal clear. It’s not until you see the results of a correctly followed end-to-end process though, can you truly appreciate why this approach is worth it.
At this point we should highlight that we don’t know a huge amount about how different operators in this industry approach cleaning, which makes it hard to construct a useful comparison. Likewise, there’s virtually no solid or accessible data on how many failures, or performance problems, within each operator actually end up being attributed to dust.
Put another way, it’s hard to know whether CityFibre’s approach is the exception to the rule or the norm, but either way we don’t tend to hear much about dust related connection faults. The assumption being the majority of operators probably have a reasonably effective approach to cleaning. We’d welcome feedback from other operators on this.
I’ve seen the damage contamination can do. For important or high power links scoping and checking for contamination is a MUST before lighting the line.
I can see it clearly now, in a coule of yrs time when my (as of yet) none existing FTTP connection starts to experience slowdown, myself contacting tech support/customer services to see what the problem is, the scripted ISP reply will be “due to increased environmental factors out of our control has lead to a buildup in dust, affecting one of our circuits blah, blah blah.. “
Having previously worked for both Openreach and BT. I can say openreach should always clean every ferrule (engineer dependant) but don’t tend to use microscopic probes unless the circuit bit rate is over 1gig or they have a problem circuit and precision test officer is working on it. Every BT electronics engineer (atleast in 2021 when I worked there) were issued with Microscopic probes and they are instructeded to inspect everything. Bare in mind BT engineers connect BTs core PDH, SDH, DWDM. 21cn, gea etc. Most of their connections are 10gig plus.
“All the third-party fibre engineers who touch our network (as well as our own people) are required to have the right testing and cleaning equipment.”
That comment made me chuckle with the way it’s worded. So 3rd party fibre engineers are having touchy feels with their own CF people too then? I’ll never see those FEXs in the same way again….!
Certainly reads like potential grounds for a few harassment cases in years to come.
Now I’m thinking of this: https://www.youtube.com/watch?v=C4wBLUBa8YI
I think it means the staff need to shower regularly!
As a fibre connectivity developer with 30 years experience, I can only endorse the inspection and cleaning policy described in this article. My personal estimate is that up to 80% of fibre network failures can be ascribed to contaminated connections.
Is this typical practice? All major operators have standards of practice that include inspection and cleaning. Unfortunately, not all individual installers follow best practices. I have seen an installer in a major data centre spit on a connector and wipe it on his shirt, for example. More education for everyone who participates in installation and maintenance of fibre networks is essential.
Thanks for highlighting the issue.
No mentions of the ferrules / connectors at the ONT at the user premises?
Would have thought these would be ripe for contamination with finger prints, grime and muck from users / builders etc…no problem as long as they stay mated (and are clean to begin with)
If he was splicing the fibre, then it’s a different cleaning operation (to cleaning the ferrules / connector ends) before you cleave the fibre In preparing the splice. Once the fibre is stripped bare, it should be cleaned along the length before it’s cleaved, of any remaining residue using alcohol wipes.
Cleaning the fibre ferrules can be done ‘wet’ or ‘dry’. Most operators now use a click pen which is a spool of floss that is advanced across the ferrule end face as the cleaning pen is clicked over the end of the connector (or inside the optics side)
The oils on the skin are a lot worse than the dust. I’m A telecoms engineer, showed a new starter a clean fibre with my scope, then very gently and briefly brushed the ferrule over the back of my hand, and then scoped again, he couldn’t believe how dirty it got
So, fibre engineers are glorified cleaners? Who knew?
Haha. Yes indeed. A trained fibre splicing operative knows all too well how important it is to keep things clean and safe.
When it comes to ‘administration’ (connecting and disconnecting path leads at equipment or at a cross-connect) good practise is to inspect and clean the fibre faces of the cabling AND optics before you commission and when redoing connections.
Contamination is acknowledged in the industry as the number one cause of failures.
Fair. I clean both the fibre and its connector with pens, and as I’ve a could of flanges on each fibre run they also get a clean.
I was lazy one time and ended up having to reclean everything. I was lazy another time and had to replace the fibre run. Best guess a hard bit of dust scratched a ferrule face causing light refraction. There was light however it was arriving at between -17 dBm and -20 dBm on kit with a sensitivity limit of -20 dBm.
Given this fibre was purchased from the United States so I pay a ton for shipping and import taxes best avoided. The fibre itself is quite specific and not so expensive at less than £1 per meter, terminated, but those more than double the price.
Are there similar issues when damaged undersea cables need resplicing?
Different phenomena- a cable break vs dirt/oils/crud on the end of a fibre face.
Analogy: think of the former as dead batteries in a torch vs the latter, someone sticking some pieces of masking tape over the lens.
Sadly not. While I’m sure a ship’s anchor probably leaves some dirt behind on the fibre after it crushes / splits the armoured cable it’s in the issue is more that the single fibre run is now two runs with some ocean / ocean floor in between.
Your email address will not be published. Required fields are marked *
This site uses Akismet to reduce spam. Learn how your comment data is processed.