Lighting Aging and Degradation

Light emitting diode output decays on a characteristic curve: a comparatively quick fall over the first few hundred hours, then a much slower long term decline. For a new station that means the imaging conditions change fastest in the first months of its life. If the criteria were set tightly during that period, they may fail later as the decline continues; if they were set loosely enough to accommodate it, they may be too loose from the start. Knowing the shape of the curve explains a pattern that otherwise looks mysterious — a station that worked well for two months and then began producing problems.

Three ways it changes

• Overall dimming. The most obvious form: mean grey level falls and contrast with it. In a threshold based algorithm this changes the segmentation directly — a defect that used to separate from its background may stop separating, with nothing else altered.
• Spectral shift. A white source changes colour temperature as it ages, drifting warmer or cooler; a colour source whose channels decay at different rates changes the balance between red, green and blue. A monochrome camera is largely indifferent to this. A colour inspection sees the whole image shift and its colour criteria go out of tolerance.
• Non-uniform decay within the array. In a row of emitters, those near the supply or in the worse thermal position fade faster, so the distribution across the field of view changes rather than the overall level. This is the most insidious of the three, because the centre of the view may look unchanged while the edges have lost a noticeable amount — and the edges are often where the defects are.

What to do about it

The first measure is margin. Since output will fall, a lighting design should not run the source at its rating — both to extend its life and to leave room to compensate later. Drive current can be raised to make up part of the loss, but only if the initial operating point was not already at the ceiling. That margin belongs in the scheme document, rather than being decided by whoever sets up the station.

The second is monitoring, which the reference sample and a uniform card between them provide, as described in the daily check. What deserves emphasis here is the frequency. Because the initial decline is the fastest part of the curve, a new station should be monitored more closely in its first months and less often thereafter. Many sites set one interval for every station and thereby miss exactly the phase where monitoring was most valuable.

Spares, briefly

A light source is a consumable and should have a spare, but how the spare is stored is often overlooked. Buy it with the station, from the same batch, and store it sealed, dry and out of sunlight, with the purchase date recorded. The same batch matters because the source itself ages in storage, and because a model bought three years later is likely to be built with a later process whose characteristics are not necessarily identical. If only a newer batch is available, image it against the source in use under controlled conditions before accepting the swap.