[Review] Sofirn SP33S (and SP33 V3.0 comparison)

LED Flashlights ā€“ General Info 26650 Flashlights
1

Two months ago Lumeniac already did an amazing review of the Sofirn SP33S, go check it out! (And then please come back and read mine :wink: ) In this review I will focus on the differences to the SP33 V3.0 and if it is worth to upgrade.

About one and a half years ago I got a Sofirn SP33 V3.0 and recently got upgraded to the new Sofirn SP33S ā€“ thank you, Sofirn, for sending me one for free for this review! Still I try to not letting a free light influence my opinion.

If you understand German or want to use a translator, you can also find the review on my website.



What you will get

  • Sofirn 26650 Li-ion battery with 5500 mAh (3.8 V on delivery)
  • 2 replacement o-rings
  • Lanyard
  • USB A-C cable
  • Manual (EN, FR, ES, IT, JP, RU, DE, CN)


Hardware

Here you can see the difference between the SP33 V3.0 (with my switch modification to prevent accidental activation) and the SP33S.



Size and weight are almost identical:
SP33 V3.0 SP33S
Diameter 37.5 mm 36 mm
Length 126 mm 126 mm
Weight (without battery) 140 g 150 g

I really like the new tailcap which is now completely flat and allows a very secure tailstand. A big improvement compared to the SP33 V3.0. The ring for the lanyard is nicely hidden and chamfered! I think this is the best lanyard hole that Iā€™ve seen from Sofirn so far.



No clip, no magnet, no holster. Without a lanyard the SP33S likes to roll away easily.



The battery can be charged internally via USB-C. The charging current was 2.3 A in my case (maybe limited by the power supply?) and stopped at 4.22 V. Also you can use the SP33S as a powerbank. The flashlight works at all time.



The silicone cover was improved compared to the SP33 V3.0 and is now completely flat and doesnā€™t open by accident as it happened with the SP33 V3.0.



Instead of diamond knurling the SP33S got a fine ribbing. Feels nice and doesnā€™t slip.



The battery tube is symmetrical.



Unfortunately Sofirn replaced the spring at the driver side with a brass button. This causes the SP33S to turn off on strong vibrations or when you hit it.


User interface
State Action Function
Off 1C Turn on
Off 1H Eco
Off 2C Turbo
Off 3C Strobe
Off 4C Lockout
Off 5C Configure switch backlight (off/on/blinking)
On 1C Turn off
On 1H Change brightness: Low/Med/High or smooth ramping
On 2C Turbo
On 3C Strobe
On 4C Change ramping mode (steps/smooth)
Turbo 1C Previous brightness (or off)
Turbo 2C Eco
Lockout 1H Momentary Eco
Lockout 4H Exit lockout and return to memorized brightness


A new feature of the SP33S is that you can change the switch backlight while the light is off. Itā€™s either off, on or blinking.



In the first five seconds after turning the light on the LED indicates the battery level (green > 30, red < 30, red blinking = empty).


Illumination

One of the biggest differences between the SP33 V3.0 and the SP33S is the LED:

  • SP33 V3.0 with Cree XHP50.2: Smaller LED, thus smaller spot, slightly more throw, lower brightness.
  • SP33S with Cree XHP70.2: Larger LED, more floody, slightly less throw, higher brightness.

Differences in throw are theoretical because the SP33S has a much better driver, see below.



My version has 6000 ā€“ 6500 K and probably a CRI of around 70. Not great, but lots of light. Thereā€™s also a 5000 K variant. Both lights have an orange peel reflector. While the SP33S still has a tintshift, it is less noticeable than the one of the SP33 V3.0.



Driver and runtime

While the first generation of the SP33 came with a FET driver, the second had a boost driver, then the V3.0 had a FET driver again. And now the SP33S got a boost driver again! This allows the flashlight to keep a more stable output independent of the battery voltage.



On turbo the output still ramps up and down due to the bad thermal regulation. On High it is able to keep the output stable and even higher than turbo of the SP33 V3.0 for most of the time and for a longer time.

The runtime mentioned in the manual (3:40 h for high) is not achieved, but 1:40 h is more realistic and what I expected.

After my runtime tests the battery voltage was around 3.3 V and slowly climbing. I guess the higher current requirement on lower voltage was a bit too much for the batteries.

Here is the relative output for all of the five steps. The spacing is good, but the lowest level could be a bit lower in my opinion. The ramping appears visually mostly linear. There is no PWM at any level, great!



Higher current was measured with a clamp meter directly after turning on with a full battery. Standby current was measured with an oscilloscope via a 100 Ī© shunt resistor.
Mode Brightness (specification) Current
Turbo 5000 lm 12.0 A
High 2000 lm 3.0 A
Med 600 lm 0.7 A
Low 100 lm 0.1 A
Eco 10 lm 0.03 A
Off 150 ĀµA
Off (switch blinking) 2.1 mA
Off (switch on) 3.3 mA


At around 3.0 V the indicator LED starts blinking red when the light is on and at 2.7 V it turns off.


Conclusion

The SP33S is a very good upgrade to the SP33 V3.0. The boost driver allows constant output, it is brighter and the design was improved. Not the nicest beam, but itā€™s bright. With a coupon the SP33S doesnā€™t cost much more than the SP33 V3.0, but is worth every cent you have to pay more!

You can get it directly from Sofirn or via AliExpress.

6 Thanks
2

Nice review/comparison, thank you!

3

What equivalent flashlight design/performance would you select for better thermal regulation?

4

I would say Thrunite TC20...

5

Thermal regulation is mostly a firmware thing.

6

Thanks! :+1:
Althoughā€¦ kind of a bummer that the TC20 is continuous ramping. No menu option for stepped brightness.

7

Someone on TLF asked about the shake resistance of the SP33S. So I made a video:

4 Thanks
8

That is common thing in flashlights that specifically use proprietary battery.

It happened to me with Acebeam E70 when using other than stock cell. Solution is to solder longer BeCu (if possible) spring either on driver or tail and you can go with whatever cell you like and it will not do that thing in video

9

The problem is the brass contact at the positive side, the very spacious inside and the heavy battery. If you want a wide compatibility with various batteries, use springs on both sides.

1 Thank
10

The positive end of the battery is also more easily deformed or dented by "the brass contact at the positive side" than it would be if there was a spring there instead of a brass contact.

I'm fairly careful with my lights, but I have noticed the positive end of some flattop batteries (with weaker construction) eventually end up being pushed into the battery slightly when used in lights that have brass contacts.

2 Thanks
11

Thanks for the review Sammy, the beam pattern pictures are extremely useful. If the SP33S had another spring and user-customizable UI I'd buy it today. 5000lm is a nice feature, but I'd prefer to step it down to something more usable in certain situations. There's something to be said for simplicity, but a lot more to be said for coming simple from the manufacturer and letting the user tailor it to their needs.

(the spring is inexcusable)

12

From my research, the Sofirn SP33S has superior sustained brightness to the Thrunite TC20v2.

13

One SP33S 5000k for sale, in the Classified ā€œWant to Sellā€sectionā€¦ā€¦I bought one too many units as giftsā€¦.great light. Brand New, with factory batteryā€¦.faster delivery than China! :+1:

.

14

I like a lot of things about the SP33S, but the aging XHP70.2 is a dealbreaker at this point. The tint shift is pretty gross, even if the 50.2 looks even worse. The 70.3 is available everywhere, and has been for a while now. Pass. Maybe the next version will have something a little more modern inside.

15

The SP33S is best in the 1x26650 class for sustained brightness, though it seems to time that it can sustain on 40% mode is dependent on ambient temperature. Iā€™ve seen reviews where the sawtooth ATR starts at 15 mins on 40% mode, or this one where is sustains for most of the runtime on 40% (and thus resulting shorter runtime).

2 Thanks
16

The V3.0 has springs on both ends, correct? So with all the improvements on the Sā€¦itā€™s interesting they would go the opposite way here.

Plus, I prefer the little extra throw on the V3.0 to the wider spill of the S. I know some will say just get the SC33 to bridge all the gaps, but the V3.0 is only $27 (including battery) on the Sofirn website right now. Iā€™ll deal with the FET at that priceā€¦especially with this being my first foray intoā€¦not only Sofirn, but flashlights with this many lumens. Speaking of which, in the comparison vids (and even the photos in this review) Iā€™ve seenā€¦ the difference in output between the SC33, and SP33S at 5000 lumensā€”and the V3.0 at 3500, did not appear to be as significant as those numbers might allude (and I would likely only use turbo less than the minute it takes to step down anyway).

Finally, if you study the two photos of the turbo output for the V3.0 and the Sā€¦the somewhat less bright V3.0 appears to be a lot easier on the eyes. Hoping that difference is even more significant in the 5000K version I ordered. I want to see detail, not washing out/reflective glare.

1 Thank
17

Hi SammysHP, thank you for your awesome SP33S review.

Iā€™m really considering to buy SP33S but also kinda waiting for the release of Wurkkos TS26 (Anduril).
If TS26 (non-ā€œSā€) becomes real itā€™ll be on the top of my list, and forget this SP33S :grin::joy: .

Itā€™ll be even more amazing if Wurkkos makes TS26 ever so slightly wider and compatible with 26650 cell :heart_eyes: . TS26, 26650, perfect, right?

What if Sofirn makes SP33A (or whatever the model name)? Anduril, regulated driver, 26650, good emitter.

Right now Iā€™m leaning heavily toward SP33S instead of TS26S because of the configurable button LED. And 26650.

18

This is a very very useful feature.
Why Wurkkos doesnā€™t make TS26Sā€™ swith LED configurable ā€¦

19

Is that a feature or a bug? We need to update the software :grin: LoL

20

Iā€™ve just done similar tests with my TS21 and TS25 and expected similar results but no, they stayed ON, good for them :sunglasses:

Probably due to differences in spring tension / spring ā€œcapacityā€ ?