Roberto Michelena wrote: > Yeah, coincidentally today I met Marc Levine (ex X-Rite) at SGIA, and > he confirmed that it's not a matter of just changing the lamp, it's a > complete science project. I'm a bit surprised to hear that. If an instrument is used purely for reflective measurement, then the critical thing for calibration is the calibration tile, not the lamp. Issues with the lamp may be: Electrical/behavioral compatibility with the drive circuitry - the driver circuit may be setup to give a certain turn on time and help regulate brightness, and this may well depend on how the lamp reacts electrically. Due to their thermal nature, lamps have a negative resistance characteristic. Light output level, directionality and spectrum. These will depend on the filament construction & geometry, the envelope shape, thickness and material. Stability, both short term and long term (aging). Some lamps may not be very consistent. The calibration process typically has sanity check limits, so a replacement lamp has to fall within those limits if the existing firmware/software is going to work. But if one were prepared to source a number of physically similar lamps with the same nominal voltage and wattage, I would have though that it might be possible to find something that is compatible, and $1000 is a fair budget to experiment. Graeme Gill.