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

Some comparitive shots:

Initial Measurements:

ON two blue LK 50A's at 4.19V:

  • 11.9 amps (tail, clamp meter)
  • Lumens (Tom E/ANSI, start-30secs): 7850/6930-7240/6390
  • Throw, measured at 5 meters: 194 kcd (881 meters)

For comparison

Clear L6 shown: XHP70 N4 1C, 4.7A, 3790-3700 lumens, 73 kcd

Black L6 shown: XHP70.2 P2 3B, SMO, fully modded w/TA 30 mm driver, NarsilM, 17.9 amps, 9110-8230 lumens, 195 kcd

Your MT07 has an XHP70.2 and you also installed a smooth reflector in it. 296 kcd with the stock reflector.

Just a beast of a light, made extra beastly. Hopefully the SP70 can put out the same big beam at 400 yards and deal with the heat a bit better.

In other news Simon wants to make an XHP35 L6. He wasn’t happy with the results with the current reflectors and is getting another one made.

The “25A” 26650s I ordered arrived. I’m very skeptical of the “25A” rating. IRs seem really high. Mine are between 0.15ohm and 0.25ohm. 0.25ohm seems VERY high for a “high drain” cell.

What brahd are they & where did you get them??

That’s a good result for throw you could get more by slicing the dome. The amperage is down a fair bit to what it should be?

Thank you for those measurements! 194Kcd is decent enough for me.

Sofirn HD 26650 factory rated at 25Amps.

I hope someone is able to take tail Amp measurements with these batteries.

/\ . Thanks cncyana. :beer:

11.9A on the cyan Liitokala is not bad. We know the 70.2 has the potential for more. Maybe the FET or the wires are a bit too high resistance?

194 kcd is more than Sofirn’s measured 160kcd with light OP. If the texture is heavy, then the candela is even more impressive as they only measured 127kcd. Nice.

The texture does look pretty heavy.

You got 195kcd out of your black L6? Hmm, I can’t remember if I got the same, but decided to sacrifice a little in order to get a nicer-looking hotspot. My notes are showing 172kcd. Still close, though. I always go by what looks best as opposed to outright throw numbers.

Just tried a pair of GOLISI 4300's, solder blob'ed one. Got 14.5 amps, think about 205 kcd, same relative bump in lumens. The LED wires look decent, think about 18 AWG, but the double springs might have some loss's compared to bypasses I would use. Got slightly more amps on a pair of 25S charged to 4.21V - about 15 amps, but they would probably drop fast.

I think overall with these results the light lives up to our expectations of what we wanted. Only thing left is how is the UI overall?

Tom what about the tint of SP70 compare to 1C and 3B of your L6?

This took some time - needed to research thru my notes. My notes say the clear L6 has a XHP70 N4 1C, but it's beam tint is well warmer than the black L6 XHP70.2 P2 3B. The clear L6 is about a 4-5 even in tint, I'd say 4000K-4500K. I seem to recall this - think I got a mis-marked LED back then.

The SP70 is warmer than the S70S (~6500K), but cooler than the XHP70.2 3B, so could be a 1C or 2. This is consistent with how the beam looks and how the LED's look under UV light. The old S70S is a strong CW for sure.

Thanks Tom…my SP70 should take longer to arrive after it was return to Sofirn and re-ship :person_facepalming:

I like what I'm seeing for the most part with the UI. A lot in common with our Narsil and Anduril for the basic operations:

  • From OFF, click ON to last mode
  • From OFF, press&hold to moon mode (can click to OFF or dbl click to turbo)
  • When ON, press&hold to cycle thru 4 well spaced modes (moon and turbo are not included in the cycle)
  • Dbl click from OFF or ON goes to max/turbo, click to return to previous mode, or OFF if initially OFF
  • switch LED is ON when the light is OFF, stays ON for a couple secs when the main LED is turned ON, then turns OFF

Ramping works much the same as our standard ramping:

  • to switch between mode set and ramping, 6 clicks
  • direction toggles like NarsilM and Anduril
  • moon mode is not included in ramping, but turbo is - you can ramp up to full max/turbo
  • the ramping is decent, little too long for my taste but appears to logarithmic, not linear, as other makers are doing incorrectly
  • there's no way to force a ramp up from lowest, like NarsilM does (press&hold from OFF), because press&hold from OFF locks you into moon mode

Notes:

  • they implemented moon mode as an independent state, so you can't ramp down to moon (conflicts with what the manual says)
  • I like you can ramp up to full max/turbo (I'm not a fan of an artificial high limit for ramping)
  • in lock out mode (4 clicks), they blink twice for every click - it's a bright blink, too bright
  • on power ON, the light comes ON automatically at the last level used, or mode, not turbo though if accessed via a dbl click, and also not moon mode

@TomE, how close are the springs to one and another?

Since the light has been made with 26650 and 21700 compatibility in mind, the springs weren’t really designed with dual usage in mind to lower resistance.

Here are Sofirn’s specs, and look the inner spring height and thickness.

tail springs:

inner bottom diameter 10
inner top diameter 5.9
inner height 7.6
spring line diameter 0.7mm,

out bottom diameter 16
out top diameter 8
out height 13
spring line diameter 1.0 mm

I don’t even know what type of BeCu they are using.
Could be using C17530 38% IACS, or C17200 22% IACS.

Still, have you tried it with 21700s?
What’s the weight full of cells?
How is the UI?

The UI sounds real good Tom. A big step up from the S70S and L6 UI.

I don’t think it was made to use both. There is just too big a difference in the lengths. I believe they designed it with 18650 and 26650 in mind. This is also why it comes with 18650 adapters and not 21700 adapters.

Got a couple Shockli 4000 21700's charged and solder blob on one (no solder blobs), will try now...

Update:

GOLISI 26650 4300 mAh @4.20B, one cell solder blob'ed: 14.2A

--> Lumens: 8600/7590 - 7820/6900, throw: 203 kcd (901m)

Shockli 21700 4000 mAh @4.19V, as-is stock, no solder blob: 12.7A (springs not compressed much)

--> Lumens: 8640/7620 - 7960/7020, throw: 208 kcd (912m)

So even though the amps measured lower, the 21700's did better. Probably because the spring compression is better. I don't think there's any issues with these 21700's I used - tailcap wasn't hard to tighten up at all. I'd suspect bypasses on the springs would get more amps.

Weight: 671g no cells, 809g w/21700 cells (Shockli).

Not bad.

Try a bypass.

Is there a possibility they might not be using top binned LEDs?

7000 lumens ANSI isn't all that bad. This is close to the best I've seen or gotten - think I got ~9000 Tom E lumens on a single XHP70.2 at best.

Not sure bout the spring spacing - I think I'd have to dis-assemble them to check that.

With bypasses and a pair of 40T's or 2070C's that I got, it should get a nice bump. Only got one 30T and one P42A.