newsletter link
mondo arc

Museum Lighting... Now

Issue 62 Aug / Sep 2011

Kevan Shaw of KSLD argues that, despite the LED industry’s greatest efforts, solid state lighting is not always the best solution for museum lighting. Take the Holburne in Bath, England for example.

We recently completed work on the Holburne Museum in Bath, UK. Nearly four years ago when be began work on the project the lighting market was a very different place than it is now. At that time, and while we were still designing, we were watching LED development closely and at the point where the decision had to be made we opted for Tungsten Halogen as the primary display light source. I am confident that we made the correct choice and believe this has been confirmed by the successful results achieved.

If I was making the same decision today it would be next to impossible to choose Tungsten Halogen. In a relatively short time we have seen radical changes. We are seeing huge pressure to reduce energy use, partly to address the sustainability issue but also to address increased costs of operation through rising energy costs. We are also seeing legislation play an increasing role in lighting design. On one hand meeting building energy efficiency targets set by the UK Building Regulations, on the other dealing with the impacts of European Energy Related products legislation that limits what lamps, control gear and, potentially, control system designs may be placed on the market.

Looking back at the Holburne, sustainability was very much a design consideration. We worked hard to achieve usable daylight for the major exhibition spaces. For this to be effective a different philosophy on light exposure needs to be adopted. This is to look at the total exposure over time of objects rather than set arbitrary levels of illumination. Lighting designers can provide the tools for this, however it requires a much greater degree of management to be carried out by curators for this to be effective. It also requires a big catch up by many institutions that lend works to others. Understanding that exposure over time is the real determinant of deterioration due to light is not new, it is quite clearly set out in Thompson’s conservator’s bible, ‘The Museum Environment’ dating back to the 1960s. The approach was heavily reinforced in the ‘Control of damage to museum objects by optical radiation’ (CIE 157:2004) and will feature in the forthcoming revision to CIBSE /SLL LG8 Museums and Galleries. Without this approach using daylight in museums, particularly in a country like the UK with capricious weather, will not be generally satisfactory for lighting exhibits.

We are being led, or rather bulldozed down the road of considering LEDs as the primary choice for exhibition lighting. As it is some thirteen years since I first specified LEDs, I find it difficult to accept this is still new technology. Of course what is available on the market now is a far cry from the products available three years ago, let alone thirteen. Are they suitable for general display use in museums? The answer here is both yes and no. As we have to design schemes with considerably lower power density than we have in the past, LED based fittings must come under detailed consideration. This necessitates a cost - benefit analysis that covers more than the current price of fittings and the potential savings on energy bills over the projected long life of these products.

Quality of light has become recognised as a vital element in museum display. For the most part visual access to museum objects is all that visitors are allowed and therefore lighting to enable you to get the most information and enjoyment is an absolute requirement. Over the last quarter of a century we have adopted the Low Voltage Tungsten Halogen lamp to provide this quality of light. It has near perfect colour rendering, is easily controllable both optically through modest scale reflectors and lenses and by dimming electronically. It has down sides: it operates at high temperature, emits infra red and Ultra Violet unless filtered in the lamp or externally, dimming dramatically reduces the blue end of the spectrum, it fails totally at end of life and this is relatively short (one or two years in a typical museum) necessitating frequent maintenance.

We are told LEDs have none of these vices, however they have a whole bunch of their own. Firstly LED may be a mature technology but it is far from stable. Sadly it is unlikely ever to achieve the product stability that we are used to for other light sources as it is born of the semiconductor electronics industry where constant change is not expected but demanded. In a commercial environment, where there are increasing pressures on efficiency and demands for ever cheaper products, not to mention higher profits for manufacturers, LEDs are following the same trajectory as personal computers and laptops, where, by the time you leave the shop, your purchase is already an out of date model. Unlike computers, where there is no physical limit on memory capacity and none yet apparent on speed of processing, you can always add another core. Converting electricity into light visible to humans has very fixed limits and LED manufacturers are mighty close to those now. We are at a point where there is a very significant trade off between the colour quality of light and energy efficiency.

Coming back to where we are now, we have LVTH IR lamps producing warm white light at a CRI of 99 or better with a smooth curve distribution, at up to 27.5Lm/W. We can get LED systems that are close in colour appearance with a CRI of 95, however these only have an efficiency of 33Lm/W. Of course we are promised better performance than this but not just yet. We also have other things to consider. If we can’t get the best performing LEDs or need to achieve greater energy savings, what are our choices? Unfortunately there are many and there are further compromises to lighting quality, our primary aim in the museum environment.

What about colour appearance? By their nature, white LEDs start out as blue light with a phosphor that converts this to yellow giving a combined “white “ appearance. This white may look warm but still has a blue edge and is quite lacking in deep reds and cyan. In some super critical applications, such as surgical lighting, red LEDs are added to the system to fill this important hole in the LED spectrum. This is to accurately assess the flesh tones of the patient on the operating table. How about the flesh tones in portraits? The blue peak also has a more insidious characteristic. As you move up the spectrum towards blue, light becomes more damaging. This is why we take every effort to shield museum objects from UV. At an optical level we also have two alternate compromises with LEDs if we are comparing these to the existing range of LVTH reflector lamps or fittings. We can either choose a fitting containing several LEDs with individual optics risking multiple shadows or a remote phosphor device that has a large diameter making it cumbersome and difficult to deliver a tightly focused beam of light from a relatively small size of fitting. This results in a wide spill of light around the area we want lit.

The thing I find most difficult to deal with is the constant change. We know this should be offering benefits over time, however it makes things difficult if you want to add or change lighting. If each time you want to add fittings they will have subtly different performance characteristics, how do you achieve a consistency in display? What is worse is that even if you have an open cheque book and buy as many fittings as you would ever want at the beginning of a project, LEDs change in output and colour appearance throughout their lives so a new fitting from your original stock will look different to fittings bought at the same time but that have been running for a couple of years.

Finally the myth of ‘less maintenance’. Anyone who has re-focused or re-lamped fittings in an exhibition after a couple of years will know what a filthy job it is. Being forced to re-lamp generally results in a cleaning cycle for the lens and reflector if this is not part of the lamp that is being replaced. The longer the re-lamping cycle the worse the dirt problem. Imagine our LEDs after five years or so untouched by requirements of maintenance. Dirty lenses mean less light output and possibly, more worrying, dirty heat sinks with consequent overheating and shortening of the life of LED and electronics. So maybe you will still be doing maintenance on a five year cycle replacing the whole fitting and not just a lamp!

Sadly from the perspective of fitting design we seem to be moving backwards, returning to the early days of LVTH with LED arrays right at the front of the fitting giving poor glare control. This becomes particularly apparent in low light museum exhibitions. We are also still seeing variations in colour appearance between fittings particularly between different manufacturers of LED. Many fitting manufacturers use different LEDs in different fittings and even those that don’t are still hostage to the vagaries of LED production and binning processes. We have also managed to get LVTH fittings specifically designed for the museum market with individual self contained dimming and positive tool operated locking. We are starting that uphill struggle again with the majority of LED fittings being tailored to the retail market where such details are unnecessary and add unwanted cost. As LED fittings remain significantly more costly than an ‘equivalent’ LVTH fitting, fighting for extra museum facilities on these fittings is a steep uphill struggle.

What is absolutely clear is that we are not in a position just to swap one fitting for another and achieve an energy saving with no other consequences. Even more significantly we cannot just swap an LED replacement lamp for an LVTH lamp. These technologies are so different that they are just not interchangeable much as the lighting industry would have you believe. One recently much publicised project has claimed to save lots of energy by changing from LVTH to LED. The claim is that the same number of fittings are used. However the LVTH scheme used two fittings each with a different colour, one filtered blue, to balance the daylight appearance. Now there are two identical LED fittings both running to achieve the original light levels. True, the appearance of the LED is closer to daylight and less energy is consumed. However the magical transformation to warm light in the evenings is lost leaving a rather cool and dry appearance to art that formerly glowed with a richness bestowed by warm incandescent light.

LVTH is not dead yet. Despite strenuous efforts of the Eurocrats to consign it to the same bin as the incandescent lamp, the current draft energy related products legislation for reflector lamps and LED replacement lamps acknowledges that there are still no true equivalent lamp replacement products so LVTHIR will remain with us well beyond 2015. When it comes to new projects however, we will probably not be able to specify LVTHIR exclusively as we still need to meet Building Regulations requirements of 22 lm/W for display lighting and we are significantly limited with power density to satisfy BREEAM, LEED and the demands of building operators to cut the electricity bill.

These are a new set of challenges for the lighting designer. On one hand the flexibility of working with lighting exposure rather than being stuck with illuminance values opens up possibilities to use dynamic lighting more effectively; on the other a move to more challenging light source selection and specification may limit what can be achieved in colour rendering and appearance. From a personal perspective I left the Holburne with some sadness in the belief that we are leaving a golden age of museum lighting both in colour appearance of LVTH and its ability and flexibility to achieve beautifully rendered and modeled lighting.


Kevan Shaw
Related Articles


Follow us on…

Follow Mondo Arc Magazine on Twitter Follow Mondo Arc Magazine on Facebook Follow Mondo Arc Magazine on Linked In

mondo arc india

darc awards DWLF IALD PLDC LRO