Sofirn SP70 Alone $50, PM for AMZ US CODE(LIMITED)

The L6 shelf already looks to be about 6mm thick and can already handle 120 watts. How much thicker does it need to be?

In looking at the L6 cutaway, I can see that if there is a triple emitter version built later on, the shelf is going to be much higher. This would be a great opportunity to use the same basic outer fin design as the single emitter except cut the fins under the shelf much deeper into the head. You will have much more surface area then.

Both single and triple versions would look very similar until you hold them sideways and notice the different fin depths. Does that make sense? I might try and draw a picture tonight.

BTW, the SP70 will use a clear rubber side switch boot like the Q8 to make the switch leds easy to see in the dark and identify the color.

So basically you want Narsil lol

I think he is talking about the very thin walls along the edge of the reflector.

I pointed this out way back when the L6 was still in the prototype stage but it was never addressed. It would significantly improve heat dissipation if we gave the heat a path to use the fins along the top of the light.

As I am sure you know the heat is kind stuck around the shelf area of the light and has a hard time moving higher up the head to use all the fins.

See the Red area below:

Yeah basically, iā€™d prefer narsil as well
And then it could be the Sofirn SP70 BLF edition

:smiley: Yeah, Iā€™m not claiming originality, Iā€™m sure I ripped that off from somewhere.

Also @Sofirn and everybody else, would it still feasible to include some of those little silicone rings to use 18650 cells instead of 26650? I suspect Iā€™m not the only one that has an extensive assortment of 18650 cells but only about one or two 26650s cells.

This seems to be a nobrainer and should definitely be included IMO. :+1:
Thanks for mentioning it Caleb & thanks TA for explaining it. :beer:

Good discussion.

Unless the design and CNC programing has already been finalized, anything we do to increase heat dissipation without significantly increasing cost is a good thing.

Donā€™t we all just want to be able to blast as much light as possible for as long as safely possible before stepping down?

My stock L6 gets too hot to be comfortable to hold in a relatively short time. If weā€™re bumping up the output significantly without making a way for the heat to be released, this will just exacerbate the issue.

This is already NOT a small light. Letā€™s not be afraid to add a few extra ounces of aluminum that could make a difference in performance.

Iā€™ll let the engineers work out the specifics.

If you want I have an stock domed XHP70.2 Astrolux MF02 with TA driver and AndĆŗril. 84mm head and 71mm reflector. I can set andĆŗril ramping max to a specified lumen output and measure intensity and throw. That way we can estimate what we can get with the SP70.

I have an L6 XHP70.2 and it suffer heat transfer to upper fins. The hottest part around the side switch.

For firmware I really want AndĆŗril but Iā€™m ok with Narsil too :smiley:

:+1: larger and if possible deeper.

Trustfire does some extreme finning in some lights, like this one:

Agreeā€¦

These fins are far too close, there wonā€™t be any air movement between them. But in general, I like the idea of making most of the head from fins.

Better heat sinking is one thing we are talking about. On the k70 we moved the fins around to address this and made the shelf 8mm thick.

Okay, Iā€™m going to submit my official list of brightness levels for the SP70 to Sofirn.

I looked at the S70s spacing and L6 spacing.
I then used the Narsil in my L6 and my TA Lumen Tube to record itā€™s 4 level mode, 5 level mode and 6 level mode. I tried them all out seeing how useful they are and where the gap in brightness is too big or too small. I then tested out some custom levels.

I came up with a total of 6 levels which includes a hidden moonlight. Iā€™m not sure how helpful moonlight would be on a light this size, but I figure it should be easy enough to add it as a hidden level. So we end up with 5 normal levels.

Turbo could be anywhere from 5000 lumen on up. For now Iā€™ll say 5000+.

This level of turbo gets you into the issue of extreme heat which really limits run time. On my light I would have it stepping down after 50 to 60 seconds* as itā€™s just too hot to hold. So I think of it as a ā€œnot so practicalā€ burst mode for when you need the max. It also rips through your battery life.

*I imagine a bit more runtime with a bigger head and maybe more mass.

So for a more practical and longer run time ā€œhighā€ mode, I would suggest 2000 to 2500 lumen

1-2 Moonlight
40-50 Low
250-300 Med 1
800-1000 Med 2
2000-2500 High
5000+ Turbo

My recommendation is to cycle through the 4 middle modes and have turbo on a double click.

Yes, 4 main modes is my preferred setup, a hidden moon and turbo is good as well.

I like it Jason. :+1:

Thatā€™s also what I decipheredā€¦ SP models have side switch, while SF are only tail-switch.
Some SP models have both side & tail switch (eg. SP30A, SP31).

Sofirn has an official website, but it doesnā€™t seem to be that well known (a Google search for ā€œSofirnā€ will more likely than not link first to their Sofirn AliExpress Store rather than the ā€œofficialā€ Sofirn website).

http://www.sofirnlight.com/en/product/productlist

There are more series listed in the Sofirn product list (SP, SF, SD, etc). Thorfireā€™s ā€œTKā€ series is also on the list, although no models are listed there yet (many of the other series also donā€™t have any models yet, apart from the well populated SP and SF series).

Jason:
Like the levels :wink:
About turbo:
If unlimited fet driven madness like with a TA driver heat is an issue. But I played with the MF02 XHP70.2 yesterday and from 4000 to 9000 lumen the output bump with naked eye is not much but with 4k it can handle the power without heating up for 10 minutes. But with 9k it gets hot about one and a half minute. Iā€™m trying to say that 6-7k lumens is about enough for turbo because above that most power will be turn in to heat and not light. For real world use it is better longer turbo runs without much heat.

Thanks for the Sofirn website link d_t_a!! I did not know they had oneā€¦ :wink:
I only knew about the Amazon & AliExpress stores.

Here is a rough concept drawing. Itā€™s is somewhat based on the L6 as I like that shape more than the S70.

You can have the same external appearance with both a single emitter and a triple emitter. With the triple emitter you get much more mass under the shelf, plus deeper fins. Itā€™s a win win. :smiley:

Note that I used the L6 75mm head diameter by mistake. The 90mm SP70 will be bigger in relation to the battery tube.

Also note that Iā€™m a skilled photoshop user, but I only had paper and my phone for the above.

I donā€™t know if memory limits the number of modes or not, but if I had to sacrifice a mode Iā€™d say the moonlight would be the least useful for this style of light. If we can keep it, great, but I think it might be more useful to have an additional ā€œHighā€ mode that can still be usable for a while. Something like:

  • 1-2 Moonlight (If possible)
  • 40-50 Low
  • 250-300 Med 1
  • 800-1000 Med 2
  • 2000-2500 High
  • 3500-4000 Highest
  • 5000+ Turbo

Double-click to Turbo sounds good to me too.

(I also would still like to save a little bit of memory space for a hidden strobe and at least one beacon too, as discussed earlier in this thread.)