Re: Destructive Incan Bulb Tests
DMC, AWR reminded me that Osram didn't specify the amps at default voltage, we just divided Watts/Default Voltage and assumed...
here is a typical link from them. So we could never have been accurate without knowing the measured default voltage amps. :thumbsup:
The "predicted" values of 100W/12V=8.3A, but you can see the 64625 has a measured of 9.1A and 64623 of 8.8A. Eventually a new Hotrater spreadsheet will include actual default values and more accurate formulas to reflect real overdrive results.
Plasmaman, without a soft starting driver your cold start reality is accurate. However, trying to reproduce a realistic cold start seems very hard to control variables. You have issues of the test platform/switch resistance, battery factors (type/combination/manufacturer quality/age of cell/internal resistance/hookup resistance/level of charge), behaviour of bench power supply vs. batteries on startup...it just got too hard to make sense out of.
I was also thinking of putting a couple bulbs in AW's D starting direct driver just to see how the lux measures at the same voltages, and on his low/med/high power settings. But, remember AW's (& AWR's) are soft starting, so these flash points should still be realistic. The more important issue is the tested run time...to see how that holds up.
I think measuring 2 representative WA bulbs against their predicted life points would likely extrapolate well for all other WA's, since their numbers were in better correlation with their website rating predictor. Then I would likely just do two run times each of Osrams: 64430, 64275, & a couple of the 12V'ers. Will likely test predicted run times of 5761, C-43, and Tungsram 64430 (whenever I get them from Slovenia)!
JJ, that is how I tested, turning dial from zero, and switching to fine tune dial when I was getting close to target voltage. Then I stepped up from tested voltage to higher levels in chart without going back to 0V each time.
Any variation from readings you got with 5761 could have been from resistance differences between setups, and I tried to keep mine as simple as possible...14AWG wire soldered to test lead plugs, other end soldered to KIU bipin leads, & shrinkwrapped solder connections.
I did verify the reported Amps on default and several other points for each bulb with my Fluke + 100A=100mV shunt, and Fluke tested voltage on a few bulbs with wires on bipins to make sure it correlated with the display I was reading on Power Supply...which it did.
If there is a real difference (assuming both our test methods and measuring were the same and calibrated), it could always be a manufacturer QA issue from batches to batches. To prove that theory, we would have to exchange new bulbs and see if we each got different readings with other batches.