[Postponed] Sofirn SP10R - deep red AA/14500 EDC

SOS is so 70’s :disguised_face:

Probably don’t need another AA, but will go for it anyway.
I’m in for a new Sofirn, even though my SP10S is still ‘new’.
ARH

Interested!

Please make it simple to use in complete darkness.

Interested



Fair enough. :THUMBS-UP: Personally, I agree with you. Does anyone here really fancy strobes for SP10R or can we just leave them gone? Or in other words: Has anyone ever used strobes with his/her SP10A, SP10B, SP10S or any other 14500 light?

Some people who bought C01R said they were fine with the low mode but they wished it had an even lower low. With the intention to improve preservation of night vision, SP10R could benefit from an ultra-low firefly mode (< 0.5lm). I am not sure about the right mode spacing for low - medium - high yet. Low could still be quite low, let's say 2-3lm to still maintain a level where your night vision will not be impaired. Medium mode maybe 30lm and High mode 100lm whereas Turbo could push the SST20-DR to its spec's limits or even beyond that. Sofirn however has difficulties measuring monochromatic light with their integrating sphere. Maybe using a certain percentage of current applied would the best way to go, e.g.

firefly = 0.05%

low = 0.5%

medium = 10%

high = 40%

turbo = 100%

Talking about Turbo mode:

I don't know high much lumens can be squeezed out of a single SST20-DR. It seems the specifications from Luminus were rated down from 3A to 2A for continuous operation. Maybe those 3A is a peak current for pulsed operation? Djozz, did you already have a chance to take a closer look at this LED? :-)

I have a Philips Wake-Up Light at home. They’re frequently on offer so snapped one up a while back. Has a nice sunset mode which dims and lowers CCT as timer runs down. The clock on the front is illuminated but super dim and doesn’t disturb my sleep. When I’m not at home though the SP10B comes with me.

For white light emitters, I think Rey nailed it on the Pineapple and Ti LAN: 0.3 lm, 3 lm, 30 lm, 130 lm. Four levels are probably too much for some people though.

No I have not and not likely will in the near future. I slowed down led testing a lot since last year, I have grown a lot less motivated to do loads of them. Still have a few leds in the pipeline and this is not one of them.

But I do have an educated guess for this led. The comparison is not fair with the 660nm XP-E2 but because its larger die size it must be compared to the 660nm XP-G3. I tested that one a year ago but did not do a proper BLF-post of that test. But here are the numbers for reference:

My luxmeter (as hardly any is) is precisely correct at 660nm, so the lumen numbers are correct and do provide a good measure of visual brightness. It maxes out at 4A with 320 lumen.
(That is extremely good output and efficiency, given that the perceived brightness at 660nm (photo-red) is six times less as 625nm (most common red leds) link . So with the same radiometric power a 625nm led would put out 1900 lumen, a green led even 5000 lumen).

So if the 660nm SST-20 has similar updated technology as the XP-G3 it should perform very well.


Thanks a lot Djozz! I really appreciate you doing these tests and this one will for sure be helpful to Sofirn for orientation when maxing out SST20-DR's performance. :BEER: :THUMBS-UP:

So, I translated your numbers into a chart to visualize the lm/A-curve.

Provided that the characteristics of SST20-DR were more or less similar to XP-G3 DR, would you say that a current of 3A can be considered to be the "sweet spot" for a reasonable maximum current in SP10R? What I see is that above 3A there's only little more to gain with a bad ratio to its power consumption (blue-lined area).

If I project the numbers into ultra-low currents - obviously there's no linear projection - my guess is that we should aim for 6-7mA to get an output of 0.2lm (firefly mode) and 25mA to get an output of 2-3lm (low mode).

We are usually not extremely concerned about longlivety of leds, but my impression always is that red leds suffer more from overcurrent than white leds. Also, their output goes down fast with overtemperature (must faster than white/blue leds) so if the heat path inside a flashlight is worse than my testing rig (which I assume is almost always the case) I would be more careful even. For white leds I usually take 80% of max output as a acceptable compromise between output and abuse, and would go even lower for red leds. So my recommendation for a sweetspot would be at most 2A instead of 3.

Yes, that makes total sense and SP10R will most likely have a small margin for thermal transfer away from the LED. So, my recommendations to Sofirn for testing would look like this. I hope this is a reasonable mode spacing.

Firefly: 0.2lm (I=6-7mA)

Low: 2-3lm (I=25mA)

Medium: 30lm (I=250mA)

High: 100lm (I=800mA)

Turbo: 225lm (I=2.000mA)

Yeah, I would’ve considered 2A or even a bit lower as the “sweet spot”.

Rather than looking at the droop in the parabola, compare it with a straight line extended from the origin at the line’s initial slope. Then compare the droop vs “ideal”. That difference is where you’re putting out more heat and less light.

So if at one point (say, 2A to 3A) you’re putting in 50% more power but only getting 30% more light, that missing 20% is just being burned off as excess heat. (Addendum: that’s excess heat, not just heat from “regular” current powering the LED.)

If the LED can handle it, it might be worth it, but if it’s sensitive to heat and/or overcurrent (red, UV, etc.), lower would be better.

I guess the key point was only that it doesn’t spoil the light. I get that you don’t even want it there to stumble upon, but most significantly, this doesn’t sound like a light you need - it’s more of an “if this light gets everything right, I’m interested, but since it’s only loose interest, that one annoying feature is enough to count me out.”

Pretty much. Like I said, I got my WK30s and Artemis if I need red light, even my HS-801s if inclined to dig ’em out.

Rewinding a bit, when lights were easy to mod, it wouldn’t be much to just throw in a different driver to get what I wanted. Now, with lights having pretty much custom drivers, sideswitches (wired, satellite-board, whatever), modding is a no-go, at least for me. So unless I get it off-the-shelf pretty much the way I want it, I’m either going to be stuck with a UI that pisses me off (even only occasionally), or I just pass on it entirely and save myself the ag.

In this case, yeah, a red light is a pretty specific niche item, so unless I know it’s not going to piss me off, I’d just give it a pass. Personally, I don’t see what’s wrong with low-high and nomem, but adding firefly would be something extra, and, well, that’s it. I wouldn’t want anything beyond that, because a red light isn’t going to be anyone’s EDC, so the light shouldn’t try to be that.

Again, that’s just my opinion. Anyone wants candle-mode or whatever, hey, have at it…

Interested!

Interested

Interested in two.

Especially the firefly and low light modes should be reprogrammable. If they got too low / too bright, they would not be usable for close range usage. Mode Memory should be choosable (or no at all and then allways started with lowest mode). Preserving night vision would be the main feature for me. Thank You.

Yours Frederik

Interested in 2.

Why not ramping UI?

Interested!

Interested in one.