.
How Long Should a Lighting System Last?
.
The lighting industry is taking the Circular Economy seriously. It has been one of the most frequent topics of conversation in the past couple of years. Unlike many industries we even have metrics in the shape of TM66 that we can use to measure and specify the “Circularity” of the lighting equipment we specify.
For this to be meaningful we need to understand the life of the fittings and controls but, more importantly, the life of the lighting system or installation. This may be determined by the life of the equipment we specify but it will also depend on various external factors, including refurbishment cycles for retail or hospitality and lease lengths for commercial properties. This is certainly something I try to determine at the beginning of a project, though raising the question with clients often elicits unexpected responses.
Such a recent discussion with a commercial landlord had them bemoaning the demise of the lamp. Apparently, they include a full re-lamping in the dilapidations charges at the end of leases. They also do not consider extended warranties beyond 5 years worthwhile as they can argue any subsequent failures are fair ‘wear and tear’ and therefore dealt with at the tenants’ cost not theirs.
Such arguments also mitigate against any kind of leasing of equipment or paying for lighting by the delivered lumen. It is more economically advantageous to spend the least amount on a system if you have the opportunity to pass resultant costs onto leaseholders! This tends to reflect what we have seen previously, with building owners unwilling to spend capital on energy savings when tenants and operators bear the running costs of energy.
How should we calculate system life? Typically, we see lighting equipment life expressed in running hours. Building or fit-out life and leases are measured in years. As designers we are left with the challenge of fitting one of these metrics into the other. Frequently not knowing the kind of operation that will use a speculative building we are taking quite broad guesses. We can say, for example, we would expect an office to be used around 2,500 hours a year and High Street retail will be similar; but a retail outlet in a mall may be as much as 4,500 hours. This of course does not account for any dimming, presence detection, daylight harvesting or other method of control.
Ultimately, with the increasing sophistication of lighting controls and even control gear that keep their own record of operating hours we will have more records of actual use. But it will likely be many years before we have data from these sources in sufficient quantities to create effective estimates of usage. Meanwhile, we need to be making good estimates ourselves.
So working backwards from the our estimated usage we can take the magic 50,000 hours that the lighting industry has settled on as the” life” of LEDs and, in many cases, the drivers. But not all of these “50,000 hours” are the same. In LEDs this is usually the Lnn figure, as the time to a certain level of lumen depreciation. But that can be anywhere between 50% and 90% of initial lumens and, bear in mind, this is a statistical probability based on relatively short-term testing.
When we get to electronic gear the question as to what 50,000 hours means is even more abstract. Is it the hours connected to power? Is it the hours running at full output? The latter is OK, but use the former and you will be out of luck claiming on your 5-year warranty when the gear fails at 5 years and 8 months, being 50,000 hours after initial switch on, regardless of actual light usage!
Given the cheapness of lighting equipment it is not really practical to expect the level of life prediction on components to be comparable to other industries. In aviation, for example, Mean Time Between Failures is calculated and devices are changed before they meet that number to ensure reliability of the system. We may see this coming to cars when we get true self-driving vehicles, and for the same critical safety reasons. We may not see the same life-critical status with most lighting, but this does not preclude manufacturers looking more closely at product and component life to help us determine better determination of lighting system life.
So, despite these challenges, what can, and should we do about system life? Every now and then a project will surface where this becomes a significant issue. The Scottish Parliament Debating Chamber was one such that came to us. When we started on this project the building was a mere 14 years old and the existing lighting was beginning to exhibit reliability problems. Added to this, the equipment was obsolete, and the manufacturer was not able, nor that interested, in providing spares, suggesting the wholesale replacement of all the fittings.
So that they did not get caught out again the client was keen to determine a realistic useful life for the replacement lighting scheme and we worked out together what was realistic. As a public building with a very fixed pattern of operation we could calculate quite accurately the operating hours. Dividing this into the magic 50,000 hours produced an expected life of 25 years. When we started asking manufacturers, particularly for the electronic gear, if they could expect their gear to work continuously for 25 years, we found that this was seen as near impossible. Particular components, specifically the electrolytic capacitors, tend to have limited life even at the design operating temperatures. Of course, the cheaper they are the shorter their life.
However, we did find one gear manufacturer who was happy to state that there was no reason their gear would not last 25 years if kept within the stated operating temperatures. Finally, we had to find a fitting manufacturer who would work with us to develop the fitting design and be prepared to undertake to maintain it for the 25 years. Moving on to 2022 and the focus on Circular Economy and this issue is being addressed more fully and asking that question today tends to get a less hostile response than I did then.
We also need to consider lighting controls. Power control systems, leading or trailing edge, rely on capacitors though, historically, lighting control systems were built to high standards and continue to function for 20 years or more, sometimes to the chagrin of the manufacturers who like you to replace them sooner! As we are moving away from power control to digital, we are facing new challenges, the primary one being software and firmware obsolescence, and the secondary being the constant updating and short lives of hardware and operating systems for devices used to programme the systems.
Having just dragged a 22-year-old iMac out to read some files for a project that we designed 20-odd years ago, I can confirm that software and operating systems do not last. We regularly see “firmware updating” messages on control systems for the latest wireless systems; how are manufacturers going to deal with these issues over the design life of lighting systems of 20 and more years?
So back to the original question, what should the design life of an installation be? At EFLA/KSLD we have a peculiarly broad-based practice so we can and do simultaneously work on a modern office fit-out and a churchthat’s over 100 years old, retaining its original turn of the 19th century chandeliers fitted with B22 lamp holders and glass shades. In the context of the latter, you could say that lighting should have the same life as the building it is installed in. However, in the modern office the life of the lighting needs to be commensurate with the expected duration of the lease, or a multiple thereof, say 15 or 30 years. In the refurbishment market we regularly see 600 X 600 lay-in fluorescent fittings with ‘Cat 2’ egg-crate louvres that are 30 or 40 years old, still fulfilling their function on their fifth or sixth set of tubes. To achieve that with LED technology we are looking at fittings requiring at least one or two rounds of re-manufacturing with new gear and LEDs.
Alternatively we enter a world of recyclable lighting with minimal integrated resource use. An interesting example of this is the 600 X 600 luminaire that we were recently shown which had a cardboard body, recyclable acrylic light guide and diffuser with control gear that provided constant luminance as the output was programmed to increase over the operating time of the fitting. Personally, I feel a bit uncomfortable with this approach, as it doesn’t really chime with a drive towards reduced resource use.
Of course, we can design for future flexibility. We can select fittings and devise wiring systems and installation details that allow fittings to be easily moved to different locations to suit changes in space planning. We can ensure that lighting control is applied to promote flexibility of use and accommodation of changed requirements.
We are still waiting for transforming luminaires. Although these have been demonstrated by Philips and Osram they have not been brought to market. A luminaire that can change not only its colour temperature but also its beam distribution and direction is a great tool for a very broad range of lighting within a space, varying between a flat wash and defined spotlighting with every combination in between. This would allow a space to transmogrify from a call centre to high-end retail at the swipe of an iPad! Such devices would last much longer than the magic 50,000 hours as the LEDs are not all used at the same time and rarely at full output. They would also add the potential magic of dynamically changing lighting within a space, instantly able to focus on the call centre employee with the highest sales or sweeping across a range of merchandise stopping on the one that matches your desired colour or form. Really endless possibilities!
We do return to data at the end of this. If we know how long an LED fitting has been running, we are able to predict its future useability or reusability. Even if it has reached the magic 50,000 hours we know it is probably still good for more hours at reduced output. So why not re-use that fitting in a less-used area such as a plant room or cupboard? Something with more working life available might be re-applied to a back-of-house corridor or other utility space. This may not even be in the same building but could be re-used elsewhere if a light refurbishment can be accomplished without requiring re-certification and at a suitably low cost.
Ultimately we should now be able to achieve LED fitting life of 30 , 40 or even 50 years by careful design and the development of a refurbishment infrastructure. Electronics life is still a challenge but we may get used to swapping gear in the same way we used to swap lamps in fittings. This seems to be the aim of initiatives such as Zhaga ands might also be the solution to firmware obsolescence though not the best for minimising resource use.
As designers and specifiers it behoves us to ask the challenging questions and set the targets for system life. We know that requirements for the Circular Economy are going to be added to lighting regulations in EcoDesign, both the EU version and our UK divergent BREXIT version. Let’s do the joining-up of the thinking to see how these will assist in ensuring that our lighting systems can stand the test of time.