Getting the best from DALI
Over the last few years we have seen rapid expansion in the use of DALI (digital addressable lighting interface) protocols for the management of lighting. And, without a doubt, DALI can prove to be a better option than alternative dimming options such as DSI (digital serial interface) and 0-10V analogue dimming.
However, DALI should not be viewed as the universal panacea to lighting management and when designing a DALI-based system it is important to understand the system’s limitations as well as its advantages. Taking account of the full picture in this way enables the system designer to get the most from the system and deliver the best solution to the end client.
For example, the key advantages of DALI include the ability to control individual luminaires or zones of luminaires – separately or simultaneously with minimum risk of interference. In addition, as with DSI, DALI provides logarithmic dimming that matches the eye’s sensitivity, wiring is very simple and the system can also be used for testing emergency luminaires.
However, it’s also important to note that communication speed with DALI is relatively slow – 1.2kbits/s, as opposed to, typically, 50kbits/s and above for modern lighting control systems. This can prove to be a limiting factor for a network with high numbers of ballasts, controllers etc.
For example, in a single room application, where the lights need to be switched in less than 16 groups, and with less than 64 ballasts overall, rooms are separately controlled with different controllers operating in each room and the speed of the DALI system should not be a problem.
There are many situations, though, where a true building-wide lighting management system necessitates using DALI in a way that exploits its key benefits while minimising the effects of any disadvantages.
This is because DALI ballasts may be addressed (individually identified) via their long search address (a randomly generated serial number used when commissioning), their group address (1-16) or their individual short address (1-64). Each group address and individual short address is set during commissioning of the system. Consequently, it is theoretically possible to address each ballast but the data transmission speed means there will be a visually discernible delay between the responses of each luminaire that would be unacceptable to most end users.
For this reason, properly designed lighting control systems actually deploy DALI as a number of sub-systems, each sub-system controlling a limited number of ballasts, with the controllers being linked together via a suitable high speed data system. This overcomes the limited number of ballasts, groups and scenes and the slow communications speed.
In addition, the need to assign an individual address to each ballast in the system needs to be catered for during the planning of the project. In theory, the luminaire manufacturers can programme each ballast in the factory but this would mean that each luminaire has to be installed in a specific location – difficult to manage on site. However, this approach is useful when installing chilled beams, where the location is specific and each beam sub-assembly is created off site, allowing the beam manufacturer to allocate the addresses at this time, in a factory environment.
Consequently, most programming of DALI addresses is carried out on site.
There are also potential limitations in the testing of emergency lighting that I would urge specifiers to make their customers aware of so that expectations are realistic. For instance, DALI emergency testing is limited to providing two types of tests, short functional and full duration. This may not always suit facilities managers who are used to conventional control systems performing mains fail simulation tests for a user defined duration.
It’s also important to watch out for compatibility issues with input devices, as the DALI protocol does not cater for these and manufacturers are often implementing their own proprietary commands that do not comply with the DALI standard.
In highlighting these issues I am not trying to deter people from using DALI, simply trying to ensure that this very useful protocol is used to its full advantage – something that will often require specialist input to ensure all of the factors are taken into account.