FW1A with a XHP50.2 feeler list

“please PM me when its alive!”

Thanks,
Mike

please PM me when its alive!

This is a great idea. Would definitely be in for one.

Since it will be 18650, let me know when it is alive.

How many lumens would we expect from it?

Please pm me when it is alive.

Sorry, you’re right :slight_smile:
Soooo if it has 5000K tint and a copper pill, I’d be interested

First post says 2500+ :slight_smile:

please PM me when its alive!

updated

Martin, are we getting close to release?

Want one!

The Ti/Cu body would be my first choice, but Cu or Al/Cu are interesting too.

Please PM me when it’s live!

Forgive my naivete… how many K…? 4…5…6?

It will be the 3V version so 6000K or 6500K as those are all that really exist right now AFAIK.

Please PM me when it’s live!

There is no significant difference in heat between one XHP70.2 and three XP-L HI. Neither one can provide a sustainable output of 1000 lm in a light this small with a linear driver. The XHP50.2 is simply four XP-G3 dies packed together under one dome.

There are reasons the FW3A did not use XP-G3, XP-L2, XHP35.2, XHP50.2, or XHP70.2. The main reason was because these emitters all make ugly rainbow beams. The warm frequencies and cool frequencies are split apart like a prism so they land in different parts of the beam. This is okay when the beam is fully integrated with a strong diffuser, like in a light bulb, but in a flashlight it makes things look pretty bad.

For example, here was a beam shot Hogo posted, comparing XP-G3 (left) to XP-G2 (right) in the same flashlight. The XP-G3’s integrated tint falls directly on the ANSI white “pure white” line, but when used in a flashlight it is quite clearly not white:

Here is another example, posted by someone on Reddit. It shows a XHP50.2 or XHP70.2, I forget which. They make the same type of beam though:

I have some lights with these emitters too, and the beams are among the ugliest I have ever seen. I have them because there was a time period when Sofirn sent me their returned items for QA purposes… and there was a high rate of returns because the beams were ugly.

These latest-generation Cree LEDs are designed to be used in bulk behind diffusion sheets, not focused with a reflector. Their poor beam quality is the reason why so many flashlights started using SST-20 and LH351D emitters instead of Cree, and why flashlights are still using previous-generation Cree LEDs like XP-L HI and XP-G2.

If a XHP50.2 version of the FW1A is produced, Lumintop and vendors should be prepared for a potentially high return rate. The pictures above do not exaggerate the amount of color shift within the beam… they actually look like that.

Sorry, but I did the testing and the results are eloquent:
Sofirt C8F 3*XP-L(990Lm) – 58°C 20 minutes later. And (650Lm) 50°C 20 minutes later.
Sofirt SP33 1*XHP50.2(1000Lm) – 41°C 20 minutes later and 47°C 60 minutes later.
BLF Q8 4*XP-L(~1100Lm “Group6-Mid2”) – 47°C 20 minutes later.
Air temperature 24°C, without cooling. Testing was carried out repeatedly, with different instances of flashlights.

Those lights are much bigger than the FW3A and FW1A. Of course they can sustain more lumens — they have a lot more thermal mass and surface area.

The FW3A / FW1A can sustain about 300 to 500 lumens without getting hot. This does not change simply by switching from XP-L HI to XHP50.2. The sustainable brightness is determined mostly by the physical size of the host.

The FW1A weighs 56 g. The Sofirn SP33 weighs 144 g. If we apply that ratio to the SP33’s 1000 lumens, the result is about 390 lumens for FW1A. This is only a rough estimate, but I hope it at least shows the general idea that a smaller light has a lower sustainable brightness level.

Additionally, the XHP50.2 has a lower Vf than XP-L HI does. It may be more efficient in terms of lumens per Watt, but the FW1A driver is linear so it cannot convert extra voltage into photons. The extra voltage gets burned off as heat instead. This gives an efficiency advantage to LEDs with higher Vf, because it burns off less voltage as heat. For a linear driver, efficiency is measured in lumens per Amp instead of lumens per Watt, because lower Vf simply makes the driver run hotter.

I think the comparison Sofirt C8F 3*XP-L and Sofirt SP33 1*XHP50.2 explains my position. We see a significant difference, with an advantage in the size of the C8F.
Perhaps because of the compact body, the difference between the FW3A and FW1A XHP50 will be less significant, but I’m sure it will be noticeable.

The XHP50.2 isn’t that bad. I have one in a Zebralight SC600w MkIV Plus, and it’s acceptable. Sure, it ain’t perfect, but it’s better than the XP-L’s and XM-L’s. Not nearly as good as the XHP35 HI and XPL-HI, and certainly not as good as the Nichia 219’s, but it’s decent.

Here is a beam shot. Auto white-balance, auto-exposure. It’s more-or-less what I see with my eyes. If I’m white-wall hunting, the XHP50.2 isn’t a good beam. But in normal use, I can’t see its imperfections.

That comparison doesn’t tell us anything because there are so many variables, least of all the size/shape/design of the lights, not to mention different thermal stepdowns (many Sofirn lights beside the C8F are set to 55C for instance).

3xX-PL HD is 165lm/W vs XHP50.2 at 185 lm/W at roughly 1000 lumen output.