Recently I was swamped with almost a dozen proposed updates to Telecommunications Industry Association (TIA) standards to vote on and review. In fiber optic technology, most standards deal with the specification and testing of components. Some testing involves simple geometric measurements, such as the diameter of the fiber core and cladding. Some involve very technical issues, such as bandwidth or dispersion.
Many, perhaps most, of these tests involve the change in attenuation of a component such as fiber, cable, splices or connectors under various environmental conditions. The batch of standards I reviewed were mainly that type.
The first standard was an update of TIA FOTP-78, “Measurement Methods and Test Procedures—Attenuation.” This particular standard may be referred to more than any other, since so many standards involve measuring attenuation changes.
Generally, I review these standards quickly because I attend meetings and take part in discussions about these changes. But I was unfamiliar with FOTP-78, probably because I missed the meetings for both the last revisions and the current proposal, so I took more time carefully reading it.
With any standard, I always start with the terms and definitions section just to make sure I understand them. On this document, I immediately found something very confusing, but basic: the definition of how attenuation is measured and reported.
Now power, attenuation and loss measurements in fiber optics are confusing under any circumstances because they are expressed in decibels (dB). If you have read much on fiber optic testing, you have seen this equation defining dB, which, frankly, almost nobody understands:
I am not going to get technical, but fiber techs are used to reading decibels off a power meter, optical loss test set (OLTS) or optical time-domain reflectometer (OTDR) when making measurements. They know a smaller decibel reading means less loss.
However, many people find it confusing that, with a power meter, decibel loss is a negative number, while, with an OLTS or OTDR, it is a positive number.
The explanation is simple: it’s like profit and loss. In accounting, if your income is greater than your expenses, the difference is a positive number and is called a profit. If your income is less than your expenses, the difference is a negative number but is just called loss without a negative sign.
Going back to FOTP-78, the original version of the standard has been replaced by an international standard, IEC 80793-1-40. Reading standards documents is always a challenge, since the language they are written in (I call it “standardese”) is a cross between technical and legalese. The U.S. standards, which are adapted from international standards, can be even more complicated to understand when they have been translated from another language.
When I got to FOTP-78 Section 3.1, Terms and Definitions, I found the definition for loss measurements had changed. The equation for decibels reversed the ratio of power so that loss would be a positive number, making gain—the opposite of a loss—a negative number.
Now, here is where it gets strange. The definition for decibel-milliwatts (dBm), the measurement term for absolute optical power, remains the same. If you measure optical power, negative dBm means the power is less than 1 milliwatt (mW) and positive means it’s greater than 1 mW. Doesn’t it feel right that something smaller would be represented by a “–“ and something larger by a “+”?
Well, it seems to not be the case for whoever wrote that international standard. Using their definition, if you have attenuation, the signal gets smaller, and you indicate it with a positive number. Confusingly, if you have a gain, the measurement anomaly on an OTDR becomes a negative number.
Why exactly is a gain a negative number? I suspect the writers who changed this definition never thought about what happens if you measure a gain or considered that it might confuse some—or lots—of us.
I am on the case, asking the TIA fiber optic people questions to clarify this.