Gryloc
Enlightened
Hey all. Thanks for the kind words. I completed those excel files in a hurry because I had to go to class. I was looking back through that file and noticed some huge errors. It was not jtr1962's data, but it was my lousy copy-pasting where formulas were not referenced to the right cells.
So, I spent a little more time and practically re-did the excel worksheet to be correct. The new file has two worksheets. The first is a calculator that allows you to insert your own die size (in mm^2) and also change a multiplier number to make up (or compensate) for possible improvements in technology between the XP-G and the XM besides using a larger die. The second shows a series of tables for different die sizes (3, 4, 5, and 6 mm squared). These charts have a place to change a multiplier. Now you can compare the different die sizes, each with their own multipliers. Notice that in some die sizes, the current points may not be exactly at 350mA or 2000mA (due to scaling the current levels with die size). Well, at the bottom of each chart, I added a chart that shows the extrapolated Vf and lumen performance at 350mA and 2000mA. I used linear extrapolation, so technically, the estimates are not perfect. They should still be useful as a rough estimate. Finally, for the second worksheet, I pre-set the multiplier values differently in hopes of finding which die size fit best, so feel free to set all multipliers back to 1.00 to be able to directly compare each table.
I recommend that all of those that downloaded my last file to discard the file and use this one from now. Refer to this one (unless I update the file again). I will no longer think of that old file from here on out. I used rapidshare again, but I am going to cheat the system a little by posting three links (10 downloads each) for the same file. If one link expires, use the next. If someone still wants the file after the links expire, PM me.
Link 1 - Link 2 - Link 3
So, after playing with some multiplier values, it seems like the 4mm^2 die size simulation fits the press release the most. With a multiplier of 1.08, I was able to calculate the following: 163lm at 350mA with an efficiency of 162lm/W, and 758lm at 2000mA with an efficiency of 116lm/W. A 108% increase in surface brightness between after a while since the release of the XP-G seems possible. When I looked at the other die sizes (3 and 5 mm squared), the numbers seem to scale wrong. For example, for the 3mm^2 simulation, I can use a multiplier of 1.1 to reach an efficiency of around 160lm/W at 350mA, but the lumen output or efficiency at 2000mA would be way off.
This is just me ranting now, but I think that if these dies are at 4mm^2, then these would still be great for focusing in medium to large sized reflectors. Even in small reflectors, these would still produce a nice and clean, but floody beam (okay for most EDC lights). These could be the direct competitors to the SST-50, which could give Luminous a run for their money. A 5mm^2 die would not be too bad in comparison, but I am hoping it is 4mm^2 or smaller. Actually if they are 5mm^2 in die size, would this be a step back in surface brightness from the XP-G? I won't complain though. I am not very familiar with the SST products, but the XM is has less thermal resistance than the SST line? If so, I wonder if these can be driven really hard (I wish for 2A/mm^2). Cree does not have the experience like Luminous has with super high drive current levels, but I hope they really did their homework and got things right! Keep the new stuff coming! ...Now, how about a 8mm^2 "XS" to compete with the SST-90? I am dreaming again...
-Tony
So, I spent a little more time and practically re-did the excel worksheet to be correct. The new file has two worksheets. The first is a calculator that allows you to insert your own die size (in mm^2) and also change a multiplier number to make up (or compensate) for possible improvements in technology between the XP-G and the XM besides using a larger die. The second shows a series of tables for different die sizes (3, 4, 5, and 6 mm squared). These charts have a place to change a multiplier. Now you can compare the different die sizes, each with their own multipliers. Notice that in some die sizes, the current points may not be exactly at 350mA or 2000mA (due to scaling the current levels with die size). Well, at the bottom of each chart, I added a chart that shows the extrapolated Vf and lumen performance at 350mA and 2000mA. I used linear extrapolation, so technically, the estimates are not perfect. They should still be useful as a rough estimate. Finally, for the second worksheet, I pre-set the multiplier values differently in hopes of finding which die size fit best, so feel free to set all multipliers back to 1.00 to be able to directly compare each table.
I recommend that all of those that downloaded my last file to discard the file and use this one from now. Refer to this one (unless I update the file again). I will no longer think of that old file from here on out. I used rapidshare again, but I am going to cheat the system a little by posting three links (10 downloads each) for the same file. If one link expires, use the next. If someone still wants the file after the links expire, PM me.
Link 1 - Link 2 - Link 3
So, after playing with some multiplier values, it seems like the 4mm^2 die size simulation fits the press release the most. With a multiplier of 1.08, I was able to calculate the following: 163lm at 350mA with an efficiency of 162lm/W, and 758lm at 2000mA with an efficiency of 116lm/W. A 108% increase in surface brightness between after a while since the release of the XP-G seems possible. When I looked at the other die sizes (3 and 5 mm squared), the numbers seem to scale wrong. For example, for the 3mm^2 simulation, I can use a multiplier of 1.1 to reach an efficiency of around 160lm/W at 350mA, but the lumen output or efficiency at 2000mA would be way off.
This is just me ranting now, but I think that if these dies are at 4mm^2, then these would still be great for focusing in medium to large sized reflectors. Even in small reflectors, these would still produce a nice and clean, but floody beam (okay for most EDC lights). These could be the direct competitors to the SST-50, which could give Luminous a run for their money. A 5mm^2 die would not be too bad in comparison, but I am hoping it is 4mm^2 or smaller. Actually if they are 5mm^2 in die size, would this be a step back in surface brightness from the XP-G? I won't complain though. I am not very familiar with the SST products, but the XM is has less thermal resistance than the SST line? If so, I wonder if these can be driven really hard (I wish for 2A/mm^2). Cree does not have the experience like Luminous has with super high drive current levels, but I hope they really did their homework and got things right! Keep the new stuff coming! ...Now, how about a 8mm^2 "XS" to compete with the SST-90? I am dreaming again...
-Tony
Last edited:
