Knowledge base · Acceptance
How much loss is allowed
No instrument displays the limit, because the limit is calculated. Allowed loss for a section is the fibre plus the joints, and both multipliers come from the project rather than from any program.
Open your file and work it out
This page explains the shape of the answer. You will not find a coefficient, an allowance or a ready figure anywhere on it: those are written into the project or the contract, and only the person holding that document can fill them in.
By Ilya (engineer, 6 years in fibre optics) ·
The formula from the acceptance form
You do not have to derive it: VIAVI firmware ships two ready German acceptance forms, and each carries the formula in its header, directly above the fields where the coefficients are typed: areq = (α − k) · l + as · ns. Length is multiplied and joints are added, and that is all that separates the two terms.
The fibre
α · l is loss per kilometre times length. It grows with every metre, so the length used is the project length rather than the one you end up with in the field: cable in a trench is longer than the distance between manholes.
The joints
as · ns is the allowance for one joint times the number of joints. Length does not enter into it: eight splices over two kilometres cost exactly what eight splices over twenty do.
The correction
k is a correction to the coefficient, also in dB/km. The German form subtracts it from α, which lets a customer tighten the requirement on the fibre without touching the rest of the formula.
The actual value: what gets added to the measurement
The same form carries a second formula, for the actual value: a = am + (α − k) · (l − lm) + as · (nA + nB). What the measurement could not see is added back to what it did.
Two things escape it, both at the ends. The fibre between l and lm: the instrument measures from the first marker to the last, while the cable starts earlier and ends later. And the splices that fall inside the dead zones at either end, nA and nB: they exist, and the event table has no row for them.
So acceptance compares the actual value with the required one, and the actual value gets built up first: one number off the screen is not enough.
On multimode the two methods themselves disagree, and no arithmetic builds that back. The TIA-568 cabling standard requires link attenuation to be measured with a light source and power meter, and adds the OTDR as a second, optional tier of testing. Where acceptance follows such a document, the figure compared with the required one comes from the meter, and an OTDR on the same link reads lower. It works from backscattered light, and the result comes out as though the light filled only the middle of the core, where an offset between the cores at a joint costs less. IEC 61280-4-1 sets the launch condition for multimode fibre, but even an instrument that meets it cannot condition the light on its way back. How large the gap is depends on the instrument, which is why no figure is given here.
Where your file keeps each term
Length. The totals strip holds two of them, and they answer different questions: Optical length is the span between the markers, End of fibre is where the fibre stopped. The first one is lm. The project length l is not in the file at all; it lives in the cable documentation.
Splice count. You read it off the event table, and you read it with your eyes: splices, connectors and bends sit there together, and their allowances differ. Splices hidden by a dead zone never reached the table, so their count comes from the route drawing rather than from the screen.
Fibre attenuation. The table gives the slope of every section between events, in dB/km. That is the measured α, not the required one: comparing it against the project figure makes sense, substituting it into the formula in place of the project figure does not.
A finished verdict. Once the required figure is worked out, type it on the Limits tab into Fibre attenuation, max or Splice / bend loss, max, and the verdict column appears in the table.
Why this page has no numbers
The page answers "how much is allowed" and names no quantity at all. Not out of caution: the coefficient depends on the fibre type, the wavelength, the climate and whatever the contract says, and a figure quoted here would read as a recommendation from us.
We ship no normative values anywhere, on this page, in the application or in the report. An empty limit field means "not checked", not "passed".
The formula is arithmetic rather than a norm: it tells you where to put somebody else's numbers.
Next
The German forms both formulas come from ship inside VIAVI firmware as examples of marking up somebody else's document. Neither carries a normative value: the coefficient fields are blank, and the contractor fills them in.