Re: Destructive Incan Bulb Tests - Updated 3/11/08 (Info Added)
On bulb re-rating, there are different factors at work.
In terms of plain power consumption, it goes up with the square of the voltage. So if you increase the voltage by 1.1 (or 10%), you will increase the power by 1.21 (21%).
Playing against that slightly is the increased filament resistance from the higher temperature. But at bulb working temperatures I think this effect is small and can be ignored.
Another thing the higher filament temperature does is shift the radiation spectrum to shorter wavelengths, out of the infra-red (heat) region and into the visible light region. So this gives relatively more light compared to the increase in power, as you get more visible lumens and fewer infra-red lumens.
The shifting of the emission spectrum is probably the main thing that increases the power law from ^2 to around ^3.
Working against the increase in filament temperature is heat loss by conduction. In small high pressure xenon bulbs, the xenon gas conducts heat away from the filament to the glass bulb and the surroundings. The heat conducted away can't be radiated and is energy lost. This causes a reduction in bulb efficiency and will reduce the power law a bit compared to larger bulbs with lower pressure gas fills.
I am guessing there is a myriad of additional factors that are involved...off the top of my head:
- Accuracy of default specifications that we use as foundation in overdrive tests. (Huge factor in some cases)
- Fluctuation from one company's engineering & manufacturing standards & QC to another brand
- Within a specific company, the variations in assembly, component QC, manufacturing, by location, age of facility & equipment
- Personnel training and methods used in all phases of a bulb's design & manufacturing
- Design changes over time for a given model within the same company
- Purity, type, and concentrations of gas in bulb envelope
- Purity & quality control of filament, glass, ceramic components,
- Component shapes, thickness, coatings
- The age/stress (degree of overdrive & run time) before performance is evaluated
- Ambient temperature, barometric pressure
- Heat transfer efficiency (i.e. my open air destructive tests vs. enclosed reflector-glass lens space), etc.
- Care, transport, packaging, storage, age of bulbs after leaving factory.
I mention some of these factors because I have three different boxes of a particular Osram bulb 64430 model. One box style has a percentage of glass stem fractures/visible cracks around bipins while still in sealed plastic wrapper. Another Osram brand 64430 box style has dramatically better Lux performance and overdrive voltage tolerance, and slight variations in measured amp at a given voltage. The 3rd box outputs more Lux at specific voltages, but flashes at lower voltage. I tested about a dozen of these bulbs before determining that these were consistent within Osram brand of 64430 box style.
Then there are Tungsram version of 64430, and
a Chinese generic version that is 2-3 times as expensive, sold by Top Bulb that had a preposterous default 650L rating that I estimate should have been more like 1/3 of their value --down to a more realistic 230L at 6V.
Remember, people are taking the given default rated values to plug into the Hotrater...and some of those ratings are obviously pulled out of thin air....and trying to figure the corret Hotrater exponent from them.
Without any hesitation, and by far the and most impressive design, manufacturing, and consistently high quality bulbs are the Osram IRC 35, 50, 65W bulbs.
Next, I would put the Philips 5761 & most of the WA bulbs....but I am hesitant about the WA only because they are not in any kind of sealed or labeled container/box. Almost all of us have obtained our WA bulbs from Litho123, and it's hard to be sure of exactly what you are getting & specific details when only provided with hand labeled plastic envelopes. I'm not saying anything negative about Litho, rather it is just not the same "standard" as getting an individual bulb in a sealed envelope/box.
