Component No Fault Found

July 1, 2001

Component "No Fault Found"

So then what?

By Thomas Carroll

July 2001

One of the biggest ballyhoos at airline and industry meetings has been the high incidence of "No Fault Found" (NFF) when components are removed from the aircraft. Generally, the assumption is made that the root cause has to be in the line maintenance arena, in that "good" parts are being replaced needlessly. The attention shifts to the line operation, with a plethora of conclusions drawn ahead of time: the training is not adequate; there’s not enough time or desire to troubleshoot; "shotgun" replacement of parts; not using Built-In Test Equipment (BITE); using BITE; the list goes on and on.

Who or what’s at fault with NFF?
Many experts have studied the issue, many investigations have been performed, and many initiatives have been undertaken to reduce NFF incidences over the years. Has it gotten any better? No. Will it get any better? At the rate the experts are going, I guarantee that in 10 years the biggest ballyhoo at airline and industry meetings will be the high incidence of NFF when components are removed from the aircraft!
So then what? Well, let’s go back to the beginning. Granted, there are a number of units that are removed from the aircraft with No Fault Found when they are tested in the shop. Since the focus is already on line maintenance, let’s start there to understand some of the reasons why a component could score NFF.

Perfect and imperfect problems
First and foremost, not all aircraft system problems are neat and pretty or perfect. Oh sure, there are the occasional hard failures when the aircraft is sitting on the ramp (which are few and far between), and these are easy to fix. Most of the time, however, the maintenance technician is wrestling with an "imperfect" problem that only occurs during flight under extreme and dynamic conditions, or a system simply hiccups once or twice.
What should be done when the problem isn’t present? Even the best troubleshooting tree doesn’t say which part to replace or wires to repair when everything is working normally. Built-In Test Equipment isn’t very helpful either when it concludes "System OK." Even the most sophisticated BITE that logs the fault during flight may be unable to show whether the cause was a specific component, an input to that component, or wires running to that component.

Repeat offenders
So then what? When these imperfect faults occur, should the complaint be answered by the maintenance technician as "Unable to duplicate"? How many repeat complaints should be tolerated? Well, before the regulatory agency gets involved because no action is being taken to resolve a repeating problem, the most likely part will inevitably be replaced. There’s a chance that it will not be the right one, but there’s also a chance that it WILL be the right one.
This situation also addresses the assumption that parts are replaced just to move the aircraft. Let’s think about this. Should an inordinate amount of time be taken to check and recheck a system when the fault does not show up? It would be a bad business decision to delay or cancel a flight just to end up replacing the most likely part anyway.
When it comes to those imperfect faults, several NFFs can, and probably will, be generated before the problem is resolved. It’s a fact of life on the line.

"Shotgun" replacements
Here’s a scenario: An aircraft diverts or returns to the field for a system fault, which ends up being one of those imperfect problems.
So then what? Everything is working as advertised, so should the most likely components be replaced one at a time in order to minimize the NFF rate? How many technicians would flirt with experiencing a repeat diversion or return to the field after they performed some maintenance action, just to keep the NFFs down? Obviously, the most prudent course of action would be to replace all the likely parts at the same time, and let the shop sort out the rest.

Beware of rogue units
A number of NFFs are part of the natural order of things, but, there’s a more insidious perpetrator of NFFs prowling around out there — the rogue unit.
A rogue unit is an individual component that repeatedly experiences short service periods, manifests the same fault each time, and whose replacement resolves the system problem. It just doesn’t work on the aircraft. The problem is that it always checks good in the shop. Suffice it to say that a part can be defective and still score NFF. If phantom glitches in the aircraft are considered "gremlins," then rogue units are the devil incarnate!
Since the shop can be generating NFFs also, using the shop bench test as the absolute measure of NFF and the competency of line maintenance is a fallacy. The only true judge and measure of the shop or line maintenance is the big hunk, the aircraft itself.
Should the NFF be ignored? No. But further investigations need to be performed to see if there are line or shop issues, or a combination of both. To accomplish this, some fundamental data sources are required — aircraft maintenance history and component tracking by serial number. If one or the other is missing, then it would be better to ignore the NFF count than to continue focusing solely on line maintenance.
Sifting through the data
Okay, we get the data we need – then what? Post mortem analysis of aircraft maintenance history will enable the sorting out of system design, troubleshooting or BITE issues. Combining that data with component tracking by serial number will identify rogue units, as well as component design, bench test or repair issues.
The world will probably always be focused on component No Fault Found — so then what? First, stop jumping to conclusions. Take a deep breath and approach the issue from a global perspective, combining fundamental aircraft and component data and real-world experience. Only then will the NFF count be meaningful, have a chance of being understood, and quite possibly reduced to an acceptable level.
Or, we’ll talk about it again in 10 years.

Thomas Carroll is the Manager of Reliability Engineering at US Airways at Pittsburgh International Airport, Pittsburgh, PA. During his 28 years in aircraft maintenance, he has worked the line, shop, maintenance control and component reliability engineering disciplines.