Tag Archives: Sofirn

Sofirn SC33

The SC33 is a high-output, general-purpose flashlight running on a single included rechargeable 21700 battery. It features a well-regulated and efficient boost circuit for maximum performance.

  1. Introduction
  2. Manufacturer Specifications
  3. Package Details
  4. Build
  5. User Interface
  6. Circuit Measures
  7. Emitter Measures
  8. Beamshots
  9. Testing Results
  10. Runtimes
  11. Pros and Cons
  12. Overall Rating
  13. Preliminary Conclusions
  14. Acknowledgement

Introduction

I’m still catching up on my backlog – this SC33 sample was received at the end of last summer.

Following the release of the popular Wurkkos TS22 earlier last year, there was much interest in the enthusiast community to see a similar model with auto-lockout disabled. Wurkkos’ sister company Sofirn didn’t take long to come out with the SC33 to meet this need.

The SC33 features the Cree XHP70.3 HI emitter (which the TS22 was eventually upgraded to as well), available in both the neutral-white 5000K and cool-white 6500K tints. You can also get it as a kit with a Sofirn-branded 21700 included.

Let’s see how it performs in my testing.

Manufacturer Specifications

Note: As always, these are simply what the manufacturer provides – scroll down to see my actual runtimes.

FeatureSpecs
MakerSofirn
ModelSC33
EmitterXHP70.3 HI
Tint5000 K
Max Output (Lumens)5,200
Min Output (Lumens)10
Max Runtime300 hours
Max Beam Intensity (cd)26,750 cd
Max Beam Distance (m)327 m
Constant Levels6
FlashingStrobe, SOS, Beacon
Battery1x21700
Weight (w/o battery)110 g
Weight (with battery)-
Length131 mm
Head Diameter32 mm
Body Diameter-
WaterproofIP68

Package Details





The SC33 came in the modern “cellphone box” style packaging common for higher-end Sofirn and Wurkkos models. Included:

  • Sofirn SC33 flashlight
  • Sofirn-branded 5000mAh 21700 battery
  • Pocket clip (attached by three hex screws)
  • Wrist lanyard
  • USB-C charging cable
  • 2 Spare O-rings
  • Manual

It’s a decent package for a “budget” build, but I would like to see a holster included.

Build


From left to right: Wurkkos 21700 (5000mAh), Vapcell 21700 F56 (5600mAh), Emisar D4K, Wurkkos WK15, Armytek Wizard C2 Max, Wurkkos TS22, Sofrin SC29, Speras E21, Wuben X1 Falcon, Sofirn SC33, Sofirn SP35T, Cyansky P50R, Cyansky P25.









The SC33 is considerably heftier than the TS22 – the handfeel is quite noticeably different. From the pics alone, you might think this is a compact light – it is actually one of the most beefy electronic-switch controlled 1×21700 lights I’ve tested. This is a key discriminating factor to keep in mind, if you looking at the choosing between models.

While it is not as long as the Sofirn SP35T (with its physical tail clicky), the SC33 is heavier. Like the earlier Sofirn light, thie SC33 features a dual spring design. This makes the light very suitable for tactical purposes, but it does mean you have to accept greater length.

The SC33 is a distinctive build – most notable for the charging port in the tail and attached pocket clip. I’m not sure why they went this route for charging, since it also necessitates a dual-body wall design for the current path. This explains the greater width of the SC33.The charging port has a built-in rubber dust cover. This should provide decent splash-resistance, but I wouldn’t recommend dunking the light in water.

The pocket clip is affixed by 3 small hex screws. The clip is thus removable, but not something to swap on or off easily. Note it only positions the light in a single direction (i.e., bezel down). So it wouldn’t be suitable to mount on a ball cap, for example.

The electronic tailcap switch has a metal cover with circular ridge detail. I found the switch traverse and performance consistent with most modern electronic switches – a bit soft, but decent enough. It is a slightly audible click. Given the switch cover’s slightly recessed nature, it is able to tailstand stably (although is still near enough the surface that accidental activation is possible).

Body threads at the head are square-cut but not anodized – again, another aspect required by the tail charging feature. A physical lockout is not possible on this light, you will need rely on the an electronic lockout.

The SC33 uses a concentric ring design on the body, with some additional cut-outs, to help with grip. With the included clip installed, I would say grip is excellent. Also thanks to included clip, the light will not roll when layed on its side. Anodizing looks to be good quality, with no damage on my sample. I would describe the finish as matte.

Inside, the light comes with a Sofirn-branded standard-sized 5000mAh 21700 battery, with a slightly raised flat-top. There is a good size spring in the head, ensuring good contact.

There is a small LED at the base of the head, that shows the charging or battery charge status (scroll down for details).

This is a very solid light, with good grip and handfeel. But it is more substantial than most in this class, which also makes it less “pocketable”.



The SC33 comes with a Cree XHP70.3 HI emitter, which is my preferred high-output choice (due to the reduced colour aberrations compared to the HD emitters). The reflector is fairly deep and heavily textured (heavy orange peel, HOP). There doesn’t seem to be any kind of anti-reflective coating on the lens.

The bezel is crenelated aluminum. Scalloping is not too aggressive, but would do some damage if you struck someone with it. Light can headstand stably.

User Interface

The SC33 uses a straightforward user interface, and one that is very similar to the Wurkkos TS22.

To start, you have a choice between Stepped Ramping mode (default), or continuously-variable Smooth Ramping mode. To switch between these modes, click 4 times when the light is On.

Stepped Ramping mode (default) levels: Moonlight, Eco, Low, Mid, High, Turbo, Strobe, SOS, and Beacon.

From OFF:

  • Press-and-hold: Turns on in Monlight mode.
  • Single-click: Turns On in last memorized mode.
  • Double-click: Turns On in Turbo (or, if auto-lockout engaged, turns On in the last memorized mode).
  • Triple-click: Turns On in Strobe.
  • 4 clicks: Activates Lockout mode. Double-click to unlock and return to memorized mode.

From ON:

  • Press-and-hold: Advances through main output modes.
  • Single-click: Turns Off.
  • Double-click: Jumps to Turbo. Single click returns to last memorized mode.
  • Triple-click: Jumps to Strobe.
  • 4 clicks: Switch between Stepped Ramping mode and Smooth Ramping mode.

From Lockout:

  • Press-and-hold: Momentary Moonlight mode, until you release
  • Single-click: Nothing (does a double flash to show lockout status)
  • Double-click: Turn on in last memorized mode.
  • 4 clicks: Disable Auto-Lockout.

Mode memory:

Yes, for constant output modes.

Shortcuts:

  • Moonlight mode: Press and hold the switch.
  • Turbo mode: Double-click the switch from On or Off.

Battery indicator:

When first activating the light, the indicator on the side switch shows the battery capacity remaining (lasts for ~5 secs upon activation):

  • Solid green: ~75-100%
  • Flashing green: ~50-70%
  • Solid red: ~25-50%
  • Flashing red: 0-25%

Low voltage warning:

Yes, the main light will step down as the battery is running low. It will then turn Off at ~2.95V

Lockout mode:

No physically, only electronic lockout. Enabled by default, auto-lockout can be disengaged.

Reviewer Comments:

I find this interface fairly intuitive, with a click to turn On in constant output modes, and press-and-hold for Moonlight (from Off). Otherwise, press-and-hold advances modes, single click turns Off/On, and shortcuts/blinky modes are multiple clicks. This is the most common arrangement, and works well. And you have the choice of smooth ramping or stepped, which is appreciated.

Like most enthusiasts, I’m not a fan of auto-lockouts, so it’s nice to have the option to disable it here. But I should note that auto-lockout is engaged by default.

Circuit Measures

No Pulse-Width Modulation (PWM):

Moon:
Moon

Eco:
Eco

Lo:
Lo

Med:
Med

High:
Hi

Turbo:
Turbo

There is no sign of PWM, the circuit appears to be fully current-controlled. There is no circuit noise on any level. This is similar to the TS22, except that model showed some (non-visible) high-frequency circuit noise on Turbo.

Note that circuit noise is not a concern, as it common to many lights. But do find that its absence (as in this case) bodes very well in terms of regulation and output/runtime efficiency. Scroll down to see actual results.

Strobes:

Strobe:

Strobe alternates between 8 Hz and 14 Hz every ~2 secs or so. Very disorienting and distracting.

SOS:

A standard SOS mode, relatively slow.

Beacon:

A single flash beacon once every 2 secs (0.5 Hz).

Charging:

There is a small LED in the head which shows solid red when the light is charging. Changes to solid green when the charging is complete.

Resting voltage <3.0V

Resting voltage >3.0V

The SC33 has a two-stage charging feature, as seen on many modern lights (i.e., where there is a lower initial charging rate when the cell is heavily discharged). The initial charging rate here is ~0.3A, which jumps to ~1.5A once the cell is over 3.0V resting voltage. This is a decent charging rate for the class, and will charge a 21700 cell quickly.

Standby / Parasitic Drain:

Due to electronic switch, there is bound to be a parasitic standby drain when the light is not in use. But because of physical build, I wasn’t able to measure.

Emitter Measures

In this section, I directly measure key emitter characteristics in terms of colour temperature, tint, and colour rendition. Please see my Emitter Measures page to learn more about what these terms mean, and how I am measuring them. As tint in particular can shift across levels, I typically stick with the highest stably regulated level for all my reported measures.

As explained on that page, since I am using an inexpensive uncalibrated device, you can only make relative comparisons across my reviews (i.e., don’t take these numbers as absolutely accurate values, but as relatively consistent across lights in my testing).

SC33 on Hi:

The key measures above are the colour temperature of ~5725K, and a negligible positive tint shift (+0.0103 Duv) to a very slight greenish-yellow at this temperature. For CRI (Ra), I measured a combined score of 58.

These values are consistent with the performance of a cool white XHP50.2 emitter, and match my visual experience of this light. Note that there is a tint shift to more yellowish spill, with purplish spillbeam edge, as is common on XHP50.2 HD emitters.

Beamshots

All outdoor beamshots are taken on my Canon PowerShot S5 IS at f/2.7, 0.5 secs exposure, ISO 400, daylight white balance. The bend in the road is approximately 40 meters (~45 yards) from the camera. Learn more about my outdoor beamshots here (scroll down for the floody light position used in this review).

Click on any thumbnail image below to open a full size image in a new window. You can then easily compare beams by switching between tabs.



As you can see above, the SC33 has a similar tint and overall output to the TS22, although with a different profile – the SC33 is more throwy, and the bezel provides for a less sharp demarcation of the spillbeam edge.

Testing Results

My summary tables are generally reported in a manner consistent with the ANSI FL-1 standard for flashlight testing. In addition to the links above, please see my output measures page for more background.

All my output numbers are based on my home-made lightbox setup. As explained on that methodology page, I have devised a method for converting my lightbox relative output values to estimated lumens. Note that my lightbox calibration runs higher than most hobbyists today, but I’ve kept it to remain consistent with my earlier reviews (when the base calibration standard was first established). On average though, I find my lumen estimates are ~20% higher than most other modern reviewers.

My Peak Intensity/Beam Distance are directly measured with a NIST-certified Extech EA31 lightmeter.

SC33 Testing Results

ModeSpec LumensEstimated Lumens @0secEstimated Lumens @30 secsBeam Intensity @0secBeam Intensity @30secsBeam Distance @30secsPWM/Strobe FreqNoise FreqCharging Current <3VCharging Current >3VParasitic DrainWeight w/o BatteryWeight with BatteryCCT (K)DuvCRI
Moonlight108.88.8---NoNo0.32 A1.50 AYes (not measured)12 g189 g---
Eco504747---NoNo0.32 A1.50 AYes (not measured)12 g189 g---
Low300310310---NoNo0.32 A1.50 AYes (not measured)12 g189 g---
Med750710710---NoNo0.32 A1.50 AYes (not measured)12 g189 g---
High1,8001,8001,800---NoNo0.32 A1.50 AYes (not measured)12 g189 g4,6500.005482
Turbo5,2005,2003,90018,000 cd10,000 cd200 mNoNo0.32 A1.50 AYes (not measured)12 g189 g---
Strobe5,200-----8-14 HzNo0.32 A1.50 AYes (not measured)12 g189 g---
SOS-------No0.32 A1.50 AYes (not measured)12 g189 g---
Beacon------0.5 HzNo0.32 A1.50 AYes (not measured)12 g189 g---

The SC33 seems to match fairly closely to the specs on most levels (although keep in mind that my lightbox’s relative calibration is generously high). Turbo is clearly over-stated however, as it only approaches that output on initial activation – and quickly declines by the ANSI FL-1 standard of 30 secs post-activation.

My NIST-calibrated luxmeter is accurately calibrated to an absolute standard, and similarly reports much lower beam intensity on Turbo compared to the specs. But I find these throw measures reasonable for this kind of emitter and output level.

Note that “Moonlight” is a misnomer here, as >8 lumens qualifies as a low in my view.

To view and download full testing results for all modern lights in my testing, check out my Database page.

Runtimes

As always, my runtimes are done under a small cooling fan, for safety and consistency. To learn more about how to interpret runtime graphs, see my runtimes methodology page. Note that on average, my lightbox’s calibration seems to be ~20% higher than most modern reviewers.

Max

Hi

Med

Performance is right where you would expect for a XHP70.3 emitter coupled with a good current-controlled, fully voltage-regulated, high-efficiency circuit. 🙂

It is very much inline with the Wurkkos TS22, but also the various Acebeam and Cyansky lights above. The only real difference is in the actual output levels (both initial and step-down) for each mode. Overall efficiency seems pretty consistent across all these well-regulated models.

Since I know many are interested in how the initial Turbo modes compare (i.e., how quickly and how far the lights step-down), here is a blow-up of the first few mins of runtime on max output:

The SC33 steps down sooner than the Wurkkos and the Acebeam lights, but to a higher stabilized level. As always, it’s your call which pattern you prefer.

Pros and Cons

ProsCons
Light has excellent output/runtime efficiency.Max output is lower than the specs would indicate (although the rest of the output levels seem fairly accurate).
Circuit shows excellent flat-stabilized regulation, with thermally-mediated stepdowns.Light lacks a true moonlight mode.
Solid and sturdy build, with securely attached pocket clipThere are some minor tint shifts across the beam periphery (common on this emitter class).
User interface is fairly sophisticated, and reasonable for the class.
Due to the charging port location and design, physical lockout is not possible, only electronic lockout.
Auto-lockout mode can be disabled.Light is larger and heavier than most in this compact 21700 class.

Overall Rating

Preliminary Conclusions

The SC33 is another strong showing in the high-output 1×21700 class of lights. I would have no issue recommending this light along with the similarly performing Wurkkos TS22, Acebeam E70 and P17, and Cyanski P25. It really comes down to the specifics on the user interface, output level spacing, and build.

There are a number of issues that resulted in losing a full star in my subjective rating system. One of the most significant is the tailcap location of the charging dock, which has required a dual-wall body tube design that significantly raises the size and weight of the light, and prevents physical lockout. While the electronic lockout is serviceable (and the auto-lockout can be easily disengaged), I still prefer a physical lockout on my lights.

The extra bulk and weight is a two-edge sword as well. This may be what allows it to run at higher step-down output level than some of the competition (i.e., extra heat-sinking mass). And some may prefer a heftier light like this. Personally though, I like a more compact light for easier pocketability and carry. I also typically prefer a bi-directional clip (the extra size here reduces the value of a reverse carry option).

Accuracy to published specs is not bad, but not quite as close as some of the competition. I wouldn’t let that hold you back, but I do encourage you to compare the actual performance of lights across my reviews. For example, if maximizing throw is important to you, you may want to look at one of the larger lights with bigger reflectors. More significantly to me, the light lacks a a true moonlight mode, which is disappointing (but may not matter as much to you in a larger light like this).

At the end of the day, I find this to be another quality light from Sofirn. It has a very serviceable user interface, and excellent performance. The price is also very good, especially for the battery-included bundle. As always, it comes down to the specific feature set and build you prefer.

Acknowledgement

The SC33 was supplied by Sofirn for review. As always, all opinions are my own and the light received the same rigourous and objective testing as all other lights that I have reviewed. At the time of review, this light retails for ~$40 USD (~$55 CDN) on sale on the Sofirn website here.

Sofirn SP35T

The SP35T is a tactical-style, general-purpose flashlight running on a single included 21700 battery. It features both tactical and general user interface options.

  1. Introduction
  2. Manufacturer Specifications
  3. Package Details
  4. Build
  5. User Interface
  6. Circuit Measures
  7. Emitter Measures
  8. Beamshots
  9. Testing Results
  10. Runtimes
  11. Pros and Cons
  12. Overall Rating
  13. Preliminary Conclusions
  14. Acknowledgement

Introduction

The SP35T is a popular mid-range model from Sofirn. Equipped with the XHP50.2 emitter, it is rated for relatively high output in the 1×21700 class – similar to many competing models that I’ve recently tested from other makers. What really caught my eye here though was the tactical tailcap switch. That’s not something you see very often any more, outside of the larger “tactical” lights from Wurkkos and Sofirn.

Given the very interconnected (and sometimes interchangeable) nature of parts across Wurkkos and Sofirn lights, I thought the SP35T might be an interesting one to test. Could this be a “Goldilocks” model that strikes just the right balance between output and performance?

Let’s see how it performs in my testing.

Manufacturer Specifications

Note: as always, these are simply what the manufacturer provides – scroll down to see my actual runtimes.

FeatureSpecs
MakerSofirn
ModelSP35T
EmitterXHP50.2
Tint
Max Output (Lumens)3,800
Min Output (Lumens)5
Max Runtime220 hours
Max Beam Intensity (cd)19,625 cd
Max Beam Distance (m)280 m
Constant Levels5
FlashingStrobe, SOS, Beacon
Battery1x21700
Weight (w/o battery)-
Weight (with battery)87 g
Length139.5 mm
Head Diameter28 mm
Body Diameter-
WaterproofIPX8 2m

Package Details




Unlike the modern “cellphone box” style packaging of the newer models from Sofirn and Wurkkos, my SP35T came in the same basic retail packaging as my old IF25A. Oh well, it’s what inside the box that counts I guess. There I found:

  • Sofirn SP35T flashlight
  • Sofirn-branded 5000mAh 21700 battery
  • Pocket clip
  • Wrist lanyard
  • USB-C charging cable
  • 2 Spare O-rings
  • Manual

It’s a decent package for a “budget” build, but I would like to see a holster included. FYI, Wurkkos sells an inexpensive holster (small size for ~$2 USD) that fits this light well.

Build


From left to right: LiitoKala 21700 (5000mAh), Vapcell 21700 F56 (5600mAh), Emisar D4K, Imalent MS03, Convoy S21E, Skilhunt M300, Wurkkos WK15, Wurkkos TS22, Sofirn SP35T, Cyansky P25, Nitecore P20iX, Acebeam E70.










At just under 140mm, the SP35T is one of the tallest general-purpose 1×21700 lights I’ve handled. This is the side effect of the tactical forward clicky switch (and dual spring design). This makes the light very suitable for tactical purposes, but it does mean you have to accept greater length. As someone with above-average sized hands with long fingers, I find the light comfortable to hold and use in either overhand or underhand grip – but some may find it a bit long.

The SP35T definitely shares a lot close design similarities to recent compact Wurkkos lights I’ve handled. This is not surprising, since these lights come off the same manufacturing production lines (i.e., Sofirn is the OEM manufacturer for Wurkkos).

The tailcap physical forward clicky switch does indeed look and feel identical to the Sofirn C8L that I recently reviewed. It has a pleasantly firm action, with a solid click and predictable firm traverse. It could just be sample variability, but I found the switch on my recent Wurkkos TD01 to be comparatively “soft and squishy” – I like the firmness of this SP35T sample. There are two raised tailcap guards that can serve as the lanyard attachment point. And just like my C8L, it is able to tailstand stably (my TD01 would not).

Tailcap threads are square-cut and anodized, with good feel. Thanks to the anodized tailcap threads, you can easily lock-out this light by a simple twist of the tailcap.

As an aside, the whole tailcap assembly here won’t fit on the TD01 – but it does fit and work on my Wurkkos TS22 perfectly.

There is a raised side-mounted electronic switch on the side of the head, with red and green LEDs underneath to show charge status. Feel and traverse of the electronic switch is very similar to a lot of Wurkkos and Sofirn lights – it’s ok, but could be a bit tighter/firmer (i.e., hard switch covers always have some degree of play).

The side switch shines a bright red when charging the battery through the light’s USB-C charging port (green when fully charged). The port is located on the opposite side from the switch, under a rubber cover. The cover fits pretty well on the SP35T (just like the C8L) – not too too tight, not too loose. I expected waterproofness is reasonable.

There is a reasonable amount of knurling on the light – not super aggressive, but more than most, and certainly enough to help with grip. With the various other design elements and cut-outs, I would say grip is excellent. Thanks to included clip, the light will not roll on its side. Anodizing looks to be good quality for type II, with no damage on my sample. I would describe the finish as matte.

Inside, the light comes with a Sofirn-branded standard-sized 5000mAh 21700 battery, with a slightly raised flat-top. There is a good size spring in the head, ensuring good contact.

This is a solid and well-designed light, with good grip and handfeel. It is a bit longer than most in this class, which is something to keep in mind.



The SP35T comes with a XHP50.2 HD emitter, in cool white tint apparently (I didn’t see any options to select a specific tint). The reflector is fairly shallow and heavily textured (heavy orange peel, HOP). There doesn’t seem to be any kind of anti-reflective coating on the lens.

As expected, there is some tint/colour shifting across the periphery of the beam, with a cool white hotspot surrounded by a yellowish spill except for a purplish shift near the edge of the periphery. This is a well-known issue with HD emitters of the XHP family, especially apparent on the XHP50.2. The heavily textured reflector seems to be help even it out it somewhat – it’s not as pronounced as most that I’ve seen.

The bezel is crenelated black aluminum. Scalloping is not too aggressive, so you can headstand stably.

User Interface

The SP35T has a straightforward user interface, and one that is identical to the Sofirn C8L and Wurkkos TD01. Like many “tactical” lights, you have two sets of possible modes; Mode Group 1 for General use, and Mode Group 2 for Tactical use.

To switch between groups, press-and-hold the side switch for >3 secs when On.

Mode Group 1 (default) available levels: Eco, Low, Medium, High, Turbo, Strobe, SOS, and Beacon.

Mode Group 1, from OFF:

  • Tail switch, partial-press: Momentary On in last memorized mode.
  • Tail switch, single-click: Turns On in last memorized mode.
  • Tail switch, double-press: Turns On in last memorized mode and then jumps to Turbo (click to stay locked-on in Turbo). You have be very rapid on the double-press to jump to Turbo.
  • Side switch, press-and-hold: Nothing – but if you click the tail switch while holding down the side switch, the light will activate in Eco mode.
  • Side switch, single-click: Nothing.

Mode Group 1, from ON:

  • Tail switch, partial-press: Nothing.
  • Tail switch, single-click: Turns Off.
  • Side switch, press-and-hold (3 secs): Switch to Mode Group 2 (see below)
  • Side switch, single-click: Steps up to the next non-Turbo constant output mode (in sequence, Eco > Lo > Med > High).
  • Side switch, double-click: Turbo.
  • Side switch, triple-click: Strobe.
    • Side switch, double-click when in Strobe: Cycle through in sequence Strobe > SOS > Beacon (with no mode memory).

Mode 1, Mode memory:

Yes, for non-Turbo constant output modes.

Mode 1 Shortcuts:

  • Eco mode: Press and hold the side switch while turning on at the tail switch.
  • Turbo mode: Double-click the side switch from On, or double-press the tail switch from Off.

Mode 2 available levels: Medium, Turbo, and Strobe.

Mode 2 functions basically as a stripped-down “tactical” version of Mode 1. The main differences are:

  • Single-click of the side switch from On only selects between Medium and Turbo now.
  • Double-click of the the tail or side switch goes to Strobe instead of Turbo.
  • There is no level memory now.

Otherwise, the two modes function the same way.

Battery indicator:

When first activating the light, the indicator on the side switch shows the battery voltage  (lasts for ~5 secs):

  • Solid green: ~70-100%
  • Flashing green: ~40-70%
  • Solid red: ~10-40%
  • Flashing red: 0-10%

Mode memory:

Yes, in the Mode Group 1 for non-Turbo constant output modes. There is no memory in the Mode Group 2.

Shortcuts:

  • Mode Group 1: Yes, for Eco, Turbo and Strobe (see above).
  • Mode Group 2: Yes, for Eco and Strobe (see above).

Low voltage warning:

Yes, the main light will step down as the battery is running low. It will then turn Off at ~2.95V

Lock-out mode:

Yes, but physically – you lock-out the light by a twist of the tailcap.

Reviewer Comments:

As with the C8L and TD01, I think this is a reasonable dual-mode UI for a tactical light, with a general mode set and a tactical mode set.

The multiple-press functionality of the tactical tailcap switch is a cute feature, if you feel you need a direct shortcut to Turbo from Off. With the firm clicky switch, I found I was able to do this reliably well. But for non-tactical types, I find doing a double-click of the side switch from On even easier.

I prefer General Mode Group, for its general usefulness and versatility. But I suppose “tactical” people will like the lack of mode memory in Tactical Mode Group 2.

Circuit Measures

Pulse-Width Modulation (PWM):

Eco:
Eco

Low:
Lo

Med:
Med

High:
Hi

Turbo:
Turbo

There is low frequency circuit noise on all levels, including Turbo, on the SP35T. It is definitely not PWM (note the simple sine wave below), but is at an an unusually low frequency of 167 Hz according to my soundcard oscilloscope. Shown below on a shorter timescale for the Hi mode.

Turbo

This is certainly different from the C8L, which was completely noise free. Even the TD01, which had circuit noise on all levels except Turbo, was at a more typical high frequency of ~5kHz.

Again, this is not a problem per se, as it was completely undetectable in practice (i.e., I couldn’t even see it when shinning on a fan or running water). But I do find it very unusual, and am not sure what to make of it. Typically, I find this doesn’t bode well for regulation or output/runtime efficiency.

Strobes:

Strobe:


Strobe alternates between 7 Hz and 11 Hz every ~1.5 secs or so. Very distracting.

SOS:

A standard SOS mode, relatively slow.

Beacon:

A single flash beacon once every 2 secs (0.5 Hz).

Charging:

The switch button shows solid red when the light is charging. Changes to solid green when the charging is complete.

Resting voltage <3.0V

Resting voltage >3.0V

The SP35T does not have a two-stage charging feature, as seen on many modern lights (i.e., where there is a lower initial charging rate when the cell is heavily discharged). Mind you, neither does the C8L. The initial charging rate here is ~1.75A, which slowly rose to ~1.85A after a few minutes. I presume it continued to climb from there. This is a pretty high charging rate for the class, and will charge a 21700 cell quickly.

Standby / Parasitic Drain:

None. That is one of the nice things about a physical clicky switch, no standby current. 🙂 And you can always lock-out the light by a twist of the tailcap, to prevent accidental activation.

Emitter Measures

In this section, I directly measure key emitter characteristics in terms of colour temperature, tint, and colour rendition. Please see my Emitter Measures page to learn more about what these terms mean, and how I am measuring them. As tint in particular can shift across levels, I typically stick with the highest stably regulated level for all my reported measures.

As explained on that page, since I am using an inexpensive uncalibrated device, you can only make relative comparisons across my reviews (i.e., don’t take these numbers as absolutely accurate values, but as relatively consistent across lights in my testing).

SP35T on Hi:

The key measures above are the colour temperature of ~5725K, and a negligible positive tint shift (+0.0103 Duv) to a very slight greenish-yellow at this temperature. For CRI (Ra), I measured a combined score of 58.

These values are consistent with the performance of a cool white XHP50.2 emitter, and match my visual experience of this light. Note that there is a tint shift to more yellowish spill, with purplish spillbeam edge, as is common on XHP50.2 HD emitters.

Beamshots

All outdoor beamshots are taken on my Canon PowerShot S5 IS at f/2.7, 0.5 secs exposure, ISO 400, daylight white balance. The bend in the road is approximately 40 meters (~45 yards) from the camera. Learn more about my outdoor beamshots here (scroll down for the floody light position used in this review).

Click on any thumbnail image below to open a full size image in a new window. You can then easily compare beams by switching between tabs.



As you can see above, the SP35T has a narrower spill than most lights in this class, and somewhat lower output (although to be fair, three of the lights above are XHP70.2 lights). It is a very clean beam though, with little evidence of chromatic/tint aberrations.

Testing Results

My summary tables are generally reported in a manner consistent with the ANSI FL-1 standard for flashlight testing. In addition to the links above, please see my output measures page for more background.

All my output numbers are based on my home-made lightbox setup. As explained on that methodology page, I have devised a method for converting my lightbox relative output values to estimated lumens. Note that my lightbox calibration seems to run higher than most hobbyists today, but I’ve kept it to remain consistent with my earlier reviews (when the calibration standard was first established).

My Peak Intensity/Beam Distance are directly measured with a NIST-certified Extech EA31 lightmeter.

SP35T Testing Results

ModeSpec LumensEstimated Lumens @0secEstimated Lumens @30 secsBeam Intensity @0secBeam Intensity @30secsBeam Distance @30secsPWM/Strobe FreqNoise FreqCharging Current <3VCharging Current >3VParasitic DrainWeight w/o BatteryWeight with BatteryCCT (K)DuvCRI
Eco52.12.1---No164 Hz1.75 A1.85 ANo84 g153 g---
Low1008080---No165 Hz1.75 A1.85 ANo84 g153 g---
Med500385380---No166 Hz1.75 A1.85 ANo84 g153 g---
High1,5001,0501,030---No167 Hz1.75 A1.85 ANo84 g153 g5,7250.010358
Turbo3,8002,9502,85011,200 cd10,400 cd204 mNo168 Hz1.75 A1.85 ANo84 g153 g---
Strobe3,800-----7.0-14.8 Hz168 Hz1.75 A1.85 ANo84 g153 g---
SOS500-----No168 Hz1.75 A1.85 ANo84 g153 g---
Beacon3,800-----2.25 Hz-1.75 A1.85 ANo84 g153 g---

The SP35T clearly has very inflated specs across all its output levels, as measured in my lightbox. It is probably even worse than the numbers above suggest, as I know my lightbox’s relative calibration is generously high for modern high-output lights. This finding is not entirely surprising, since the max output ratings in particular were not realistic for the XHP50.2 emitter.

My NIST-calibrated luxmeter is accurately calibrated to an absolute standard, and similarly reports much lower beam intensity on Turbo (~25-30% less than spec, which is comparable to what my lightbox reports for overall output measures).

To view and download full testing results for all modern lights in my testing, check out my Database page.

Runtimes

As always, my runtimes are done under a small cooling fan, for safety and consistency. To learn more about how to interpret runtime graphs, see my runtimes methodology page.

Med

Hi

Max

According to reports online, the SP35T is supposed to have a buck driver – but I don’t see any evidence of that. Looking at the Med and Hi mode runtimes, it looks like a basic FET driver – and one that performs remarkably similar to the XHP50.2-equipped Wurkkos WK15 that I currently have on hand for testing. I presume these two lights are in fact using the same basic driver. This is a bit disappointing compared to the excellent flat voltage-regulated Wurkkos TS22.

There is one thing that is very different on the SP35T though – the wide oscillations in output on Turbo mode. Presuming this was due to the thermal sensor reacting to my cooling fan, I did a separate test without cooling, as shown in the lighter green above. It is clear that the cooling fan is having a big difference, as the light runs fairly consistently at the much lower level once step-down occurs without cooling. But it is interesting that the oscillations do eventually re-appear later in the run.

To show this more clearly, here are the two Turbo runtimes – with and without cooling – on a longer timeframe:

Max-extended

This is pretty unique in my experience. The step-down from Turbo level is to quite a bit lower level than usual (down to ~450 lumens in my lightbox, just slightly above Med level). Eventually, as the light cools, it starts to step up in output, with widening swings.

By the way, I know the swings under a cooling fan seem a lot more extreme above, but they are not so noticeable in real life. Below is an expanded runtime, to show you that a typical ramp up and back down under cooling actual takes about 7 mins. Here is how it looks in practice:

In the rising stage, you won’t notice the gradual shift over time, it is that slow. But on the ramp down, you are likely to notice the light is dimming fairly quickly.

All that to say, I think this light would benefit from a less sensitive thermal sensor – and a lower step-down level to start with (i.e., ~450 lumens is very low for a modern light).

Pros and Cons

ProsCons
The light has a solid build, with a tactical forward clicky switch in the tail and a side electronic switch.Circuit is not voltage-regulated, producing a slowly decreasing output instead of flat runtimes.
The light has a serviceable dual mode set user interface, identical to the Sofirn C8L and Wurkkos TD01.The circuit is also noticeably less efficient then other current-controlled lights with flat regulation.
Price is reasonably low.The turbo mode steps down to a much lower level than most lights, and oscillates considerably in output (likely due to a poorly calibrated thermal sensor).
XHP50.2 HD emitters produce well known tint shifts across the beam, with a yellowish spill and purplish spillbeam edge.
Output specifications are clearly very inflated.

Overall Rating

Preliminary Conclusions

The SP35T is certainly a solid light, with a very decent physical build and good user interface. The presence of a forward clicky tactical switch here is great, if you are a fan of that design. The user interface is certainly very serviceable, being identical to the C8L.

But as the pros and cons list above demonstrates, the circuit performance is disappointing here. Sure, it produces a reasonable amount of light for a reasonable amount of time – but its performance just doesn’t compare to the well-regulated and efficient C8L or Wurkkos TS22. But the SP35T doesn’t even compare well to other simple FET driver-based lights – due to the unusually low step-down level on Turbo, and the repeated oscillations back up to higher output as it cools.

Moreover, the rated output specs are way off on this light (as in, at least 25-30% below spec, if not more). Its rare nowadays to see such a large mismatch between published specs and actual performance. Since many make their purchasing decisions based on published specs, this is very disappointing.

The XHP50.2 HD emitter is known for a lot of tint/chromatic variation across its beam, and this example is no different. That said, I do find it a bit better than typical, likely due to the heavily textured reflector here. But to put it simply, I think this light would benefit from both an emitter and circuit upgrade.

At the end of the day, I like the physical build (and forward clicky tail switch). The user interface is also quite serviceable. But the circuit performance is sub-standard compared to Sofirn’s other offerings, and to other lights in this class – both in terms of overall output and output/runtime efficiency. And the odd behaviour on Turbo after step-down needs to be corrected. But it still is a reasonable amount of light for a reasonable amount of time, in absolute terms.

Acknowledgement

The SP35T was supplied by Sofirn for review. As always, all opinions are my own and the light received the same rigourous and objective testing as all other lights that I have reviewed. At the time of review, this light retails for ~$30 USD (~$40 CDN) on sale on the Sofirn website here.

Sofirn C8L

The C8L is a budget tactical flashlight featuring fairly high output and very good throw, running on a single included 21700 battery. Also features both tactical and general user interface options.

  1. Introduction
  2. Manufacturer Specifications
  3. Package Details
  4. Build
  5. User Interface
  6. Circuit Measures
  7. Emitter Measures
  8. Beamshots
  9. Testing Results
  10. Runtimes
  11. Pros and Cons
  12. Overall Rating
  13. Preliminary Conclusions
  14. Acknowledgement

Introduction

My first Sofirn review was of their fairly basic 1×21700 IF25A model, which has been around for a couple of years now. While certainly solid and serviceable, it did have a fairly generic build and presentation. In contrast, the newer C8L has been recommended to me as more representative of what Sofirn is producing now.

The C8L is in the style of a “tactical” light, with a larger head (for better throw, and likely higher sustained output due to great thermal mass) and an actual forward clicky tailcap switch. The C8L comes with the XHP50.3 HI emitter, which should provide for excellent throw and decent output.

I thought this would be a good opportunity to expand beyond my initial focus on compact, EDC-style 1×21700 lights. Let’s see how the C8L performs in my testing.

Manufacturer Specifications

Note: as always, these are simply what the manufacturer provides – scroll down to see my actual testing results.

FeatureSpecs
MakerSofirn
ModelC8L
EmitterXHP50.3 HI
Tint6000 K
Max Output (Lumens)3,100
Min Output (Lumens)8
Max Runtime220 hrs
Max Beam Intensity (cd)70,500 cd
Max Beam Distance (m)531 m
Constant Levels5
FlashingStrobe, SOS, Beacon
Battery-
Weight (w/o battery)-
Weight (with battery)151 g
Length156 mm
Head Diameter46.5 mm
Body Diameter-
WaterproofIPX-8 2m

Package Details




The C8L comes in a modern-looking hard cardboard box with printed specs cover and cut-out foam interior. This design offers good protection for the light (e.g., although the outside corner of the box got dinged in the mail, everything inside as unaffected). Inside, I found:

  • Sofirn C8L flashlight
  • Sofirn-branded 5000mAh 21700 battery
  • Spacer to allow the use of 18650 batteries
  • Basic wrist lanyard
  • USB-C Charging cable
  • 2 Spare O-rings
  • Manual

It’s a decent package, consistent with other lights of this class. I would have liked to have seen a holster, since that is my preferred way to carry a light like this, but that is not typically included with budget lights.

Build


From left to right: Armytek 18650 (3500mAh), Sofirn 21700 (5000mAh), Acebeam 21700 USB-C (5100mAh), Acebeam E70, Acebeam P17, Armytek Doberman Pro, Convoy M21F, Lumintop D3, Nitecore MH12SE, Nitecore P20iX,  Sofirn C8L.









The C8L is a nice and solid build, with very good handfeel. It’s about what I would expect for a tactical-style light in overall dimensions and weight – substantial enough, but still pocket-able.

There is a physical forward clicky switch in the tailcap used for the turning the light on/off. I must say, it’s been awhile since I tested a light with an actual physical clicky switch – it is nice to see them again. Switch feel is good, and you can easily flash/momentary signal with the forward clicky.

There are two raised tailcap guards that allow the light to tailstand, and serve as the wrist lanyard attachment point. I note that some earlier reviews reported the light couldn’t tailstand stably, but my sample does fine. Tailcap threads are square-cut and anodized, with good feel. I always recommend you keep a light stored locked out when not in use. Thanks to the anodized tailcap threads, you can do this easily on the C8L by a simple twist of the tailcap.

There is a raised side-mounted electronic switch on the side of the head, with red and green LEDs underneath to show charge status. Feel and traverse of the electronic switch is decent, but could be a bit tighter/firmer (i.e., the cover has a bit too much play). The switch shines a bright red when charging the battery through the light’s USB-C charging port (green when fully charged). The port is located on the opposite side from the switch, with a thick rubber cover. Like the IF25A, I found the cover to fit rather tightly, making it hard to full depress. But I suppose that should help with waterproofness if you can press it down enough.

There is no actual knurling on the light, but concentric ring reeling around the body tube and a good number of deep cut-outs on the head and tailcap that provide good grip. The head fins have flat areas to help minimize rolling (but it can still roll with enough force). Anodizing looks to be good quality, with no damage on my sample. I would describe the finish as satin. Its a nice package, comfortable and well-balanced in the hand with decent grip – but nothing too sharp to rip through clothing or anything.

Inside, the light comes with a Sofirn-branded standard-sized 5000mAh 21700 battery, with a slightly raised flat-top. A battery sleeve is also included, in case you want to use older 18650 cells. There is a good size spring in the head, ensuring good contact.





The C8L uses the XHP50.3 HI, which is basically a low-profile emitter known for its excellent throw while still maintaining decent high output. Reflector has a light orange peel texture. Together, this should provide for decent throw while minimizing any chromatic aberrations.

The bezel is crenelated black aluminum – not too aggressive, so you can headstand stably. There doesn’t seem to be any kind of anti-reflective coating on the lens (which is surprising nowadays). You can see the reflections of my cell phone camera in the macro shots, for example. Still, at least its not contributing to any chromatic aberrations – the beam is pretty consistently cool white across its full range.

User Interface

Gone is the Anduril user interface – instead, we have a more common dual physical/electronic switch interface.

Personally, I find it great to see the classic forward clicky switch interface again: partially press for momentary on, clicked for locked-on. Easy-peasy. You change output levels by the secondary electronic switch in the head.

But there is some hidden sophistication here, as there are actually two different mode groups you can switch between. A press-and-hold of the electronic switch for 3 secs when the light is on will switch between the two modes. Let me describe them here in detail.

Mode 1 (default) available levels: Eco, Low, Medium, High, Turbo, Strobe, SOS, and Beacon

Mode 1, from OFF:

  • Tail switch, partial-press: Momentary On in last memorized mode
  • Tail switch, single-click: Turns On in last memorized mode
  • Tail switch, double-press: Turns On in last memorized mode and then jumps to Turbo (click to stay locked-on in Turbo). You have be pretty rapid on the double-press to jump to Turbo.
  • Side switch, press-and-hold: Nothing – but if you click the tail switch while holding down the side switch, the light will activate in Eco mode.
  • Side switch, single-click: Nothing

Mode 1, from ON:

  • Tail switch, partial-press: Nothing
  • Tail switch, single-click: Turns Off
  • Side switch, press-and-hold (3 secs): Switch to Mode 2 (see below)
  • Side switch, single-click: Steps up to the next non-Turbo constant output mode (in sequence, Eco > Lo > Med > High)
  • Side switch, double-click: Turbo
  • Side switch, triple-click: Strobe
    • Side switch, double-click when in Strobe: Cycle through in sequence Strobe > SOS > Beacon (with no mode memory)

Mode 1, Mode memory:

Yes, for non-Turbo constant output modes.

Mode 1 Shortcuts:

  • Eco mode: Press and hold the side switch while turning on at the tail switch.
  • Turbo mode: Double-click the side switch from On, or double-press the tail switch from Off.

Mode 2 available levels: Medium, Turbo, and Strobe.

Mode 2 functions basically as a stripped-down version of Mode 1. The main differences are:

  • Single-click of the side switch only selects between Medium and Turbo now.
  • Double-click of the the tail or side switch goes to Strobe instead of Turbo.

Otherwise, the two modes function the same way.

Low voltage warning:

Yes, the main light will step down as the battery is running low. It will then turn Off at ~2.95V

Lock-out mode:

Yes, but physically – you lock-out the light by a twist of the tailcap.

Battery indicator:

When first activating the light, the indicator on the side switch shows the battery voltage  (lasts for ~5 secs):

  • Solid green: ~70-100%
  • Flashing green: ~40-70%
  • Solid red: ~10-40%
  • Flashing red: 0-10%

Reviewer Comments:

The default Mode 1 set is very serviceable, and functions largely as you would expect. I particularly like the shortcut to jump to Eco mode, by holding down the side switch when activating at the tailcap. And shortcuts to Turbo are always appreciated. I also like the very intuitive battery read-out when activating the light.

I have to say though, I really don’t get the point of Mode 2. I could maybe see the value of a scaled down output set that excluded the blinky modes. Or, alternatively, a simple “tactical” interface of just Turbo and Strobe. But I don’t know how many people would want this half-way in-between option.

Circuit Measures

Pulse-Width Modulation (PWM):

Eco:
Eco

Low:
Lo

Turbo:
Turbo

There is no sign of PWM or noise on any level, the circuit appears to be fully current-controlled.

Strobes:

Strobe:


Strobe alternates between 7 Hz and 16 Hz every ~1.75 secs or so. Very distracting.

SOS:
SOS

Beacon:
Beacon

Beacon strobe is a slow 0.5 Hz (i.e., one full power flash every 2 seconds).

Charging:

The switch button shines solid red when the light is charging (switches to solid green when the charging is complete).

Resting voltage <3.0V

Resting voltage >3.0V

The C8L lacks the multi-stage charging feature seen on many lights (i.e., with a lower charging rate for when cells are heavily discharged). Charging rate is reasonably fast for a 21700 cell. Charging terminated at ~4.19V on my sample.

One oddity when charging – if you click the side switch while charging, the light will activate in Eco. I must say I was a bit surprised when I discovered that by accident. But I suppose it could be useful if you need an emergency light/night light while charging.

Standby / Parasitic Drain:

None. That is one of the nice things about a physical clicky switch, no standby current. 🙂 And you can always lock-out the light by a twist of the tailcap, to prevent accidental activation.

 Emitter Measures

This section is a new feature of my reviews, where I directly measure key emitter characteristics in terms of colour temperature, tint, and colour rendition. Please see my Emitter Measures page to learn more about what these terms mean, and how I am measuring them.

As explained on that page, since I am using an inexpensive uncalibrated device, you can only make relative comparisons across my reviews (i.e., don’t take these numbers as absolutely accurate values, but as relatively consistent across lights in my testing).

The key measures above are the colour temperature of ~5970K, and the slight positive tint shift (+0.0101 Duv) to yellow-green at this temperature.

For CRI (Ra), I measured a combined score of 71.

These values are very consistent with the XHP50.3 emitter, and match my visual experience of this light.

Beamshots

All outdoor beamshots are taken on my Canon PowerShot S5 IS at f/2.7, 0.5 secs exposure, ISO 400, daylight white balance. The bend in the road is approximately 40 meters (~45 yards) from the camera. Learn more about my outdoor beamshots here (scroll down for the floody light position used in this review).

Click on any thumbnail image below to open a full size image in a new window. You can then easily compare beams by switching between tabs.



As you can see above, the beam pattern for C8L is very much on the throwy side, with a nice and bright hotspot (as expected for this emitter).

Testing Results

My summary tables are generally reported in a manner consistent with the ANSI FL-1 standard for flashlight testing. In addition to the links above, please see my output measures page for more background.

All my output numbers are based on my home-made lightbox setup. As explained on that methodology page, I have devised a method for converting my lightbox relative output values to estimated lumens. My Peak Intensity/Beam Distance are directly measured with a NIST-certified Extech EA31 lightmeter.

C8L Testing Results

ModeSpec LumensEstimated Lumens @0secEstimated Lumens @30 secsBeam Intensity @0secBeam Intensity @30secsBeam Distance @30secsPWM/Strobe FreqNoise FreqCharging Current <3VCharging Current >3VParasitic DrainWeight w/o BatteryWeight with BatteryCCT (K)DuvCRI
Eco82727---NoNo1.7 A1.8 ANo173 g242 g---
Low100125125---NoNo1.7 A1.8 ANo173 g242 g---
Med500450450---NoNo1.7 A1.8 ANo173 g242 g---
High1,3001,2001,200---NoNo1.7 A1.8 ANo173 g242 g5,9700.010171
Turbo3,1003,1503,10063,600 cd63,200 cd503 mNoNo1.7 A1.8 ANo173 g242 g---
Strobe------7-16 HzNo1.7 A1.8 ANo173 g242 g---
SOS------NoNo1.7 A1.8 ANo173 g242 g---
Beacon------0.5 HzNo1.7 A1.8 ANo173 g242 g---

The Eco mode is not as low in my testing as the specs report (i.e., more of a typical low). The higher outputs in my lightbox seem to correlate pretty well with the specs.

My beam distance measures are slightly lower than the specs, but are within a reasonable range to them (i.e., it is quite a strong thrower).

To see full testing results for all modern lights in my testing, check out my Database page.

Runtimes

As always, my runtimes are done under a small cooling fan, for safety and consistency. To learn more about how to interpret runtime graphs, see my runtimes methodology page.

Max

Hi

Med

The C8L seems to be very efficient, with overall output/runtime performance for the XHP50.3 coming in a little below the XHP70.2/SST-70 lights, but above the SST-40 lights, as you would expect. But max output rivals the higher output emitters, which is very impressive.

The regulation is very stable and flat on the Hi and Med levels, but showed an interesting step pattern on the Turbo run. I assumed this was due to thermal management and the effect of my standard cooling fan, so I did an additional Turbo run without the fan (red trace above). Based on earlier reviews, I expected the C8L to step down to the Hi level and stay there. Instead, without the fan, the C8L stepped down to a lower level than Hi, but still stepped back up to an intermediate level in an apparently thermally-regulated pattern.

I haven’t seen quite this pattern before, so I thought it would be good to compare the runs at a shorter timeframe (by default, all runs are under a cooling fan unless stated otherwise):

Max-extended

Interestingly, the light doesn’t actually step-down, but rather gradually ramps down to lower levels. After some variable period of time (presumably as the light cools), it then ramps back up to a higher level. But doesn’t level off at the defined Hi or Turbo levels, rather at a series of intermediate outputs. This is a fairly distinctive thermal management feature.

In any case, the light is certainly well regulated at every level, with very good efficiency for a XHP50.3 HI emitter.

Pros and Cons

ProsCons
Light has excellent output/runtime efficiency, at all levelsTurbo ramps down to a reduced Hi level eventually, due to heat. However, light ramps back up to intermediate output levels as it cools.
Circuit shows very good regulation overall, with thermally-mediated ramp down/up on Turbo, and step-downs as the battery is almost drained.Lacks a true Moonlight/ultra-low level, but that is not surprising for a thrower.
Uses a dual switch design, with physical tailcap clicky for on/off.Electronic button feel could be improved.
Good build quality and hand feel.

Overall Rating

Preliminary Conclusions

The C8L is a very impressive light. The build quality and hand feel are top notch, on par with with more expensive lights (although I do find the electronic switch cover a bit loose in feel). I would also appreciate a few more package extras (like a belt holster), but this is a very good package for the price.

The beam pattern is what you would expect for the size reflector and emitter – a lot of throw, with decent spill. And there are no obvious chromatic abberations – a consistent (and accurately labelled) 6000K cool white beam.

I like the implementation of the dual switch user interface, with easy shortcuts to min or max output. The UI is very serviceable, although it could use a few tweaks (like a revised second mode set). Output levels are reasonable (note there is no Moonlight mode, but that is not surprising in a larger throw light like this).

Overall output/runtime efficiency seems very good for the emitter type. Regulation pattern is also very flat and stable, but with an interesting thermally-mediated ramp down (and ramp back up) on the Turbo level. It all seems very well thought out.

No surprises, this light works exactly as advertised – and is a great bargain to boot. After testing this model, I am definitely interested in reviewing additional Sofirn lights in the future.

Acknowledgement

The C8L was supplied by Sofirn for review. As always, all opinions are my own and the light received the same rigourous and objective testing as all other lights that I have reviewed. At the time of review, this light retails for ~$40 USD (~$55 CDN).

Sofirn IF25A

The IF25A is a popular compact every-day-carry light, running on an included single 21700 battery. It also features the sophisticated Anduril user interface.

  1. Introduction
  2. Manufacturer Specifications
  3. Package Details
  4. Build
  5. User Interface
  6. Circuit Measures
  7. Emitter Measures
  8. Beamshots
  9. Testing Results
  10. Runtimes
  11. Pros and Cons
  12. Overall Rating
  13. Preliminary Conclusions
  14. Acknowledgement

Introduction

In my absence from reviewing, another popular premium “budget” maker that has come on the scene is Sofirn. Making a series of affordable but powerful lights, I was curious to give them a try. So I purchased their popular compact 1×21700 model, the IF25A.

Note that this light has been around for a while now, so its performance may not be quite as high as some of their newer offerings. And my sample was purchased last fall – it took me awhile to get around to it, given all my requested reviews.

The IF25A is quite a tiny little tank of a light, and features 4x Luminus SST20 emitters. The light comes with a choice of either warm (4000K) or a cool (6500K) white LEDs. This review is of the cool white IF25A, in order to compare with other lights configured for maximum output.

Note this will be the second light I’ve tested that uses the open-source Anduril user interface (although I know it uses an early implementation of Anduril 2). Let’s see how the light performs in my testing.

Manufacturer Specifications

Note: as always, these are simply what the manufacturer provides – scroll down to see my actual testing results.

FeatureSpecs
MakerSofirn
ModelIF25A
Emitter4xSST20
Tint6500K
Max Output (Lumens)4,000
Min Output (Lumens)-
Max Runtime-
Max Beam Intensity (cd)-
Max Beam Distance (m)420 m
Mode Levels5 + Ramp
FlashingStrobe
Battery1x21700
Weight (w/o battery)99 g
Weight (with battery)-
Length106.4 mm
Head Diameter35 mm
Body Diameter35 mm
WaterproofIPX-8

Package Details



The IF25A comes in typical cardboard box, with fairly simply background illustrations and basic light details (I know some of the newer Sofirn models come in fancier packaging). Inside, I found:

  • Sofirn IF25A flashlight
  • Sofirn-branded 4000mAh 21700 battery (note some distributors may include a higher-rated cell)
  • A 18650 battery sleeve
  • Wrist lanyard
  • USB-C Charging cable
  • 2 Spare O-rings
  • Manual

It’s a decent package for a budget build. I would have liked a pocket clip and a higher capacity cell though, to match the competition (again, newer versions may come with a higher capacity cell). I do appreciate the USB-C charging cable – I have plenty of these lying around, but it always nice to see one included in a budget package.

Build


From left to right: LiitoKala 21700 (5000mAh), Fenix ARB-L21-5000U 21700 (5000mAh), Sofirm IF25A, Fenix E35 v3, Convoy S21E, Imalent MS03, Armytek Wizard C2 Pro Max, Acebeam E70, Nitecore P20iX, Nitecore MH12SE, Lumintop D3, Convoy M21F.







The IF25A is shorter than most lights in this class, despite the multi-emitter head. Despite that, it actually feels very robust, with a nice thick body. A surprisingly tiny but substantial build at the same time.

There is a side-mounted electronic switch on the side of the light near the head, with a grippy rubber cover that still allows the red and green LEDs underneath to shine through. Feel and traverse of the electronic switch is good. Note that by default the switch shines a constant low output dual-green when the battery is connected but not in use or charging (brighter green in those conditions). I initially found this constant “locator” feature is a bit annoying – but you can turn it off, or switch it to a brighter green (which actually has an important potential use – scroll down to User Interface for a discussion).

The button also shines a bright red when charging the battery through the light’s integrated USB-C charging port. The port is located on the opposite side from the switch, with a thick rubber cover. I found the cover to fit rather tightly, making it hard to full depress. But I suppose that should help with waterproofness if you can press it down enough.

I always recommend you keep a light stored locked out when not in use. Thanks to the anodized tailcap threads, you can do this easily on the IF25A by a simple twist of the tailcap. Threads are square-cut, and well lubricated. The light can tailstand stably, thanks to the side cut-outs for the wrist lanyard.

There is no actual knurling on the light, just a series of cut-outs at various points. Combined with somewhat glossy anodizing, this makes the light fairly slippery in use – I would have preferred some knurling elements. Note the light can also roll fairly easily, although the slightly flared tailcap helps a little bit with this (as does the rubber port cover). The anodizing seems in good shape on my sample, but I suspect it is only the thinner type II, given the price point.

Note that no pocket clip is included, and it would be a challenge to carry the light that way given the wider head and flared tailcap. Personally, I would have preferred a narrower tailcap and a clip option, since I don’t think thin wrist lanyards are of much use. Not a bad physical build by any means, but it does feel and look somewhat budget. I also find it a bit too short, but I have larger than average hands.

My sample came with a Sofirn-branded standard-sized 4000mAh 21700 battery (higher capacity may be sold by some vendors), with a slightly raised flat-top (also called a wide button by some).

A battery sleeve is also included, in case you want to use older 18650 cells (again, a nice touch in a budget light). I suspect a button-top would also work fine in the light, but it would need to be compact given the small size here (and certainly, none of those 21700 cells with built-in charging ports would fit). Note there is no spring in the head, just a flat contact disc (which helps explain the more compact length).



The IF25A uses 4 non-frosted TIR optics instead of individual reflectors, allowing the head be quite a bit shorter than most. I actually recommend the use of TIR optics for multi-emitters, as it tends to minimize beam artifacts. It also shines a generally “smoother” beam, with less differentiation between hotpot and spill. The beam looks reasonably clean on my sample, with just a few artifacts.

Note that I did observe some tint shifting on the lower levels (i.e., the lowest output has a definite greenish hue). This is not uncommon on current-controlled lights (e.g., Fenix), but is particularly noticeable here.

The bezel is flat black aluminum, with no crenelations – so, it can headstand stably, but you may not be able to tell if the light is on. I don’t see any form of AR coating on the lens.

There is a very interesting option with this light, using the switch indicator (green LEDs) when the battery and tailcap are connected. As I will explain below, you can increase the output of the default setting, giving yourself in essence an extra “moonlight” mode. Here is what it looks like, close-up on a white wall in the dark (not really this bright, using my cellphone’s auto-adjust mode):

Locator

The green is very even and surprisingly bright with dark-adapted eyes. Scroll down for more details on how to configure.

User Interface

The IF25A uses the open-source Anduril user interface (UI). Although the site I bought it from describes (and displays) the original interface of the inaugural Anduril release, I can confirm that my sample actually uses Anduril 2 (just like the Lumintop D3 I previously reviewed). I understand from ToyKeeper that it is actually an early implementation of it.

Anduril has two distinct UIs mode sets: Simple and Advanced. The labels are a bit misleading, as both are fairly sophisticated – it is just that the Advanced UI has a lot of extra options not available on the scaled-down Simple UI. Advanced UI also has an option for a discrete Stepped level mode, in addition to the continuously-variable smooth Ramping mode (which is the only mode present on the Simple UI).

To switch from the default Simple UI to Advanced UI, you need to do 10 clicks from Off, with a hold on the 10th click.

It’s easier to show the UIs rather than explain them in words, so here is a common pic:

ui-diagram

You can also download a fully described text manual here.

Like the Lumintop D3, my IF25A only has 5 discrete Stepped modes in the Simple UI (all 7 show up under Advanced UI though). Since most people are likely to leave it in the Simple UI, I’ve gone with 5 discrete levels my tables, and labelled L1-L5 as Moonlight, Lo, Med, Hi and Max/Turbo.

Again, check the image and link above for more info, but here is a simplified description of the UI to get you started.

From OFF:

    • Press-and-hold: Turns On in lowest output, in either Ramping mode or Stepped mode depending on which mode is enabled (and which UI you are in)
    • Single-click: Turns on in last memorized mode used (Ramping or Stepped)
    • Double-click: Turns on to Turbo (aka the Ramping max output)
    • Triple-click: Battery check (voltage read out a single time)
    • Triple-click-and-hold: Special strobe modes, but only when in Advanced UI (remembers last strobe mode used)
    • 4 clicks: Lockout mode. In lockout mode you have different options available:
      • Press-and-hold: Momentary Moonlight
      • Double-click-and-hold: Momentary Low
      • 4 clicks: Turns On in memorized output level
      • 4 clicks and hold: Turns On in the lowest level
      • 5 clicks: Turns On in Turbo
      • 10 clicks and hold: Configure the lock timeout threshold (in Advanced UI only), allowing you to pre-set the timeout time of the lock.
    • 7 clicks: (Advanced UI only) Enters auto light config for the button switch (“AUX/Button LEDs”). There are four modes you can switch between; constant low, blinking low, off, constant hi. Click 7 times again to advance to the next option, in sequence. The light auto-memorizes the last option you select. This allows you to use the switch LEDs as an impromptu “green moonlight” mode, as explained below.

From ON:

  • Press-and-hold: Ramps up (or Steps up, depending on the mode). Ramps/steps down if you do it again.
  • Single-click: Turns Off
  • Double-click: Jumps to Turbo
  • Double-click-and-hold: Ramps down (or Steps down)
  • Triple-click: Switch between Ramping and Stepped modes (in Advanced UI only)
  • 4 clicks: Lockout mode (see above for options)

Mode memory:

Yes, the circuit memorizes the last constant On output level in either Ramping or Stepped modes.

Strobe/Blinking modes:

Yes, quite a few actually. The strobe/blinking modes are accessible from Off with a triple-click-and-hold, but in Advanced UI only. You can switch between strobe/blinking modes with 2 clicks, in the following sequence (see testing results below to see what these look like):

  • Candle mode
  • Bike flasher
  • Party strobe
  • Tactical strobe
  • Lightning

Low voltage warning:

Sort of. In operation, the light drops in brightness in steps, and runs for an extended time at a very low level. Apparently it shuts off when the cell is ~2.8V (although I haven’t run it that long to confirm). I haven’t noticed any change in the switch LEDs to indicate low battery voltage (which seems like a missed opportunity, given they have red and green LEDs under there).

Lock-out mode:

Yes. In either Simple UI or Advanced UI, lockout is accessed by 4 clicks from On or Off (repeat to unlock). The lockout mode is unusual with Anduril, as it actually enables momentary operation in the Moonlight/Lo modes. There are other lockout modes available, as explained above. As always though, I recommend you physically lock out at the light at the tailcap, if you want to guarantee no accidental activation.

Temperature check and thermal calibration mode:

This is a little complicated (and beyond the needs of most users), so I will just refer you to the diagram from the manual above. With default settings, I find it steps down fairly quickly due to heat (unsurprising, give low thermal mass in the head). I have not tried to reconfigure my sample.

Reviewer Comments:

Anduril is a fairly sophisticated setup – a choice of Simple or Advanced UI, generally well thought out. Of course, you will never please everyone, and many may prefer a slightly simpler interface (e.g., the Convoy S21E), or even very basic one (e.g., Acebeam E70). While I can see flashaholics enjoying some of the extra items, some of them are really novelties (e.g. candle mode and lightning storm are particularly well done, but when would you practically ever use them other than as a party trick?). At the end of the day, I expect Simple UI is fine for most users. That said, the Stepped mode and bike strobe are something that many could use, so it’s nice to have them as options in the Advanced UI.

One of the main advantages of Anduril in a build like this is the ability to independently control the green LED emitters under the switch cover. By keeping the constant-on default setting (or increasing to brighter green), you can basically turn this into an additional “green moonlight” mode on the light. Simply use the light as a twisty: tighten the tailcap for moonlight and access to all the main modes, loosen the tailcap for a physical lock out. 🙂

Circuit Measures

Pulse-Width Modulation (PWM):

There is no sign of PWM on any level, the circuit appears to be fully current-controlled.

That being said, my oscilloscope was able to detect high frequency noise at all levels except max output, as depicted below. Note that these are not actually a concern, as they are not visible to naked the eye – the light remains flicker free in actual use. But I’ve noticed upon my return to reviewing that a lot of “budget” circuits are showing detectable signs of circuit noise. Using the Simple UI’s 5 discrete Stepped levels:

Level 1 (Moonlight):
L1

Level 3 (Med):
L3

Level 4 (Hi):
L4

Level 5 (Max):
L5

As you can see, noise begins with L1, at ~9.2 kHz. For L2 through L4, noise frequency remains constant at ~15.4 kHz, but increases in amplitude as output levels rise (which is fairly common, as more light is being emitted) – except for L5, which is noise free, oddly. Noise at these super high frequencies is impossible to detect visually, and so again not a concern. I am just including for completeness, as I like to present all my findings.

Strobes:

Tactical Strobe:
Tactical

Tactical strobe is a fairly typical 10.1 Hz, although the light spends more time in the On phase than the Off phase on each cycle.

Party Strobe:
Party

Party strobe is a super-fast (and annoying) short pulse at 25 Hz. I don’t know what kind of rave parties the Anduril folks like to hold, but I won’t be attending any. 😉

Lightning Strobe:
LightningD
Lightning
LightningA
LightningB
LightningC

I’ve shown five consecutive 10-sec cycles above, so you can a feel for the frequency and intensity of light flashes. Lightning strobe is a fairly realistic lightning simulation, with variable intensity and time between flashes.

Bike Strobe:
Bike

Bike strobe is a nice, slow 1 Hz signalling strobe – but with with 4 rapid flashes on every signal (shown in more detail below).

D3-BikeB

Beacon:
Beacon

Beacon strobe is a slow 0.5 Hz.

Candle:
Candle

Candle strobe is a continuous flicker, of varying intensity (again, accurately simulating a candle).

Charging:

The switch button shines a bright red when the light is charging (switches to bright green when the charging is complete).

Resting voltage <3.0V

Resting voltage >3.0V

The IF25A has a two-stage charging feature (i.e., with a lower charging rate for when cells are heavily discharged). With a heavily depleted cell (<3.0V), the initial charging voltage is a relatively low 0.32A. Once the resting voltage of the cell >3.0V, the charging current jumps to ~1.6V. Over the next few mins, it will continue to slowly rise a bit higher.

Note that the light drops to a very low level when the battery is nearly depleted, but there is no specific low voltage warning that I’ve noticed (e.g., there is no warning flash, and the button LED remains green throughout when the light is on unless you disable that feature). It will apparently shut-off at ~2.8V, but I never ran it that far to find out.

Standby / Parasitic Drain:

I measured the standby current (with default low green switch LEDs active) as 78 uA, which is negligible and not a concern (i.e., it would take many years to fully drain the cell). Regardless, I always recommend you lock the light out when not in use to prevent accidental activation (and cut any standby drain). A single twist of the tailcap will lock out this light, thanks to the anodized screw threads. And special bonus, this means you can use the tailcap twisty as a de facto “green moonlight” mode with the switch LEDs.

Emitter Measures

This section is a new feature of my reviews, where I directly measure key emitter characteristics in terms of colour temperature, tint, and colour rendition. Please see my Emitter Measures page to learn more about what these terms mean, and how I am measuring them.

As explained on that page, since I am using an inexpensive uncalibrated device, you can only make relative comparisons across my reviews (i.e., don’t take these numbers as absolutely accurate values, but as relatively consistent across lights in my testing).

As with all my reviews, the reported CCT and Duv measures above are for the first stable output mode without stepdown – which, in this case, is L3 (Med) on the Stepped Simple UI. The key measures above are the colour temperature of ~5265K, and the very noticeably positive tint shift (+0.0174 Duv) to green-yellow at this temperature.

For CRI (Ra), I measured a combined score of 59 on this level.

These values are very consistent with Luminus SST emitters, and match my visual experience of this light.

Given the subjectively stronger green tint-shift I noticed on the lower outputs, I decided to actually measure the Duv tint shift of each discrete Stepped level on the Advanced UI, as shown below.

L7: ~5225K, Duv +0.0169
L6: ~5240K, Duv +0.0178
L5: ~5165K, Duv +0.0192
L4: ~5000K, Duv +0.0223
L3: ~5000K, Duv +0.0222
L2: ~5030K, Duv +0.0218
L1: ~5095K, Duv +0.0211

As you can see, the CCT doesn’t change much (i.e., typically stays within ~5000-5250K), but the positive (green) tint shift is greater at the lower levels (L1-L4) compared to the higher levels (L5-L7).

Beamshots

All outdoor beamshots are taken on my Canon PowerShot S5 IS at f/2.7, 0.5 secs exposure, ISO 400, daylight white balance. The bend in the road is approximately 40 meters (~45 yards) from the camera. Learn more about my outdoor beamshots here (scroll down for the floody light position used in this review).

Click on any thumbnail image below to open a full size image in a new window. You can then easily compare beams by switching between tabs.


I did a second set of beamshots at this location more recently, showing two more relevant lights:

As you can see above, the beam pattern for IF25A is more on the floody side, as expected give the multi-emitters and shallow TIR optics.

Testing Results

My summary tables are generally reported in a manner consistent with the ANSI FL-1 standard for flashlight testing. In addition to the links above, please see my output measures page for more background.

All my output numbers are based on my home-made lightbox setup. As explained on that methodology page, I have devised a method for converting my lightbox relative output values to estimated lumens. My Peak Intensity/Beam Distance are directly measured with a NIST-certified Extech EA31 lightmeter.

IF25A Testing Results

ModeSpec LumensEstimated Lumens @0secEstimated Lumens @30 secsBeam Intensity @0secBeam Intensity @30secsBeam Distance @30secsPWM/Strobe FreqNoise FreqCharging Current <3VCharging Current >3VParasitic DrainWeight w/o BatteryWeight with BatteryCCT (K)DuvCRI
L1 Moonlight-1.41.4---No9.2 KHz0.32 A1.6 A78 uA98 g163 g---
L2 Lo-7474---No15.4 kHz0.32 A1.6 A78 uA98 g163 g---
L3 Med-450440---No15.4 kHz0.32 A1.6 A78 uA98 g163 g5,2650.0174-
L4 Hi-1,6501,550---No15.4 kHz0.32 A1.6 A78 uA98 g163 g5,2800.016959
L5 Max4,0003,4003,05030,500 cd24,800 cd315 mNoNo0.32 A1.6 A78 uA98 g163 g---
Candle-------No0.32 A1.6 A78 uA98 g163 g---
Bike Strobe------1.0 HzNo0.32 A1.6 A78 uA98 g163 g---
Party Strobe------24 HzNo0.32 A1.6 A78 uA98 g163 g---
Tactical Strobe------10.1 HzNo0.32 A1.6 A78 uA98 g163 g---
Lightning-------No0.32 A1.6 A78 uA98 g163 g---
Beacon------0.50 HzNo0.32 A1.6 A78 uA98 g163 g---

My measured max output is considerably below the rated spec, as are my beam distance measures. The rated specs definitely seem inflated on this sample. That said, I do like the inclusion of the relatively low min output mode here (i.e., ~1.4 lumens). Not quite as low as I would like for Moonlight, but I’ll take what I can get given how rare actual Moonlight modes seem to be nowadays.

To see full testing results for all modern lights in my testing, check out my Database page.

Runtimes

As always, my runtimes are done under a small cooling fan, for safety and consistency. To learn more about how to interpret runtime graphs, see my runtimes methodology page.

Max

Hi

Med

For thermal reasons, both the (L5) Max and (L4) Hi runtimes quickly stepped down from their initial values to ~750 lumens or so, and stayed within a ~100 lumen range of that point for most of their extended runs (scroll down to see this with better resolution). You may be able to adjust this somewhat through the Anduril settings, but ultimately I suspect the low thermal mass of the light is the culprit here.

At the (L3) Med level, you get a semi-regulated runtime pattern (i.e., much closer to direct-drive over most of its run).

Something else you’ll notice in the full runtimes above is that the IF25A regulation pattern is quite “noisy”, demonstrating significant fluctuations in output over time (even at the “regulated” lower level). In practice, these fluctuations are not noticeable (i.e., they are actually fairly minor and slow over human perceptual timescales, and thus unnoticeable).

Here is a blow-up of the (L5) Max output runs, for both Ramping and Discrete Stepped, on a shorter time scale, to show you both the step-down and relative stability:

Max-extended

As you can see, there is no real difference between the discrete Stepped mode set and the Ramping (unlike some earlier reviews out there). It seems like Sofirn fixed that earlier issue.

To better show the effect of the “noisy” pattern above, here is a blow-up in even more detail, over a 2-min portion of the run when the changes in output were the most extreme (on the discrete Max output run):

IF25A-Max-extended

Again, these are not a concern as you will not be able to notice these kinds of subtle outputs changes in real life. Nevertheless, I would prefer to see more stable and consistent regulation across all modes, as the pattern above is a sign of a very basic budget circuit.

Which brings me around to the key point – overall efficiency of the IF25A appears lower than others in this class. Keep in mind my sample IF25A came with a 4000mAh battery, which is 20% lower capacity than most of the competition above. Still, the runtimes are  disappointing compared to lights with fully-regulated current-controlled circuits (e.g., the more expensive Nitecore P20iX, but even the budget Convoy M21F).

Part of this could be due to the relative efficiency of the multiple SST20 emitters (especially in a small head where heat management is an issue). But I doubt that is the main issue, as they are not being driven very hard at these lower “regulated” levels. It seems far more likely that the issue is with the driver. Which is a real shame, since Sofirn opted for the sophisticated Anduril UI – it would have been nice to have seen it paired with an efficient current-controlled circuit here.

I realize this IF25A model is getting a little long in the tooth now, but I still recommend Sofirn look into the improving both the regulation and the efficiency of the circuit on this model while it is still available.

Pros and Cons

ProsCons
Comes with the sophisticated Anduril UI, which has both simple and advanced options.Light steps down quickly on Max/Hi levels to a relatively low ~750 lumens for the remainder of the runtime (for thermal reasons)
Small and sturdy physical build.Driver output efficiency is below average for the class, consistent with entry-level budget lights.
Relatively floody beam, without too many artifacts from the multiple emitters.Regulation pattern is not impressive, both in terms of its relative spiky "noise" pattern, and semi-regulated appearance.
Thanks to the configurable UI, you can use the green LED switch indicator as an effective Moonlight mode.Rated specs for max output and beam distance are over-inflated in my testing.

Overall Rating

Preliminary Conclusions

The Sofirn IF25A is a serviceable build, with a decent (and sophisticated) user interface with a floody beam profile. It is a particularly compact yet sturdy light, which should appeal to many. That said, I would like to have seen a bit more refinement in the physical build (e.g., actual knurling, less-flared tailcap with pocket clip, etc.).

I realize this is an older model that has been around for a while, so it may not be entirely fair to compare it the newer competition. But the overall impression I have is one where they consistently skimped a bit on many aspects of the light. While each one may not seem like a big deal, the overall effect adds up over time. The most significant one is the circuit – both regulation and efficiency were sub-par, consistent with the most basic budget builds (yet many do better, as shown above). I find increasingly that this is something that differentiates the most entry-level budget lights from the intermediate or premium ones. At the end of the day, I find circuit performance really matters to me.

This is also an example where the rated max output and beam distance specs are clearly inflated. Since my return to reviewing, I’ve generally been pleasantly surprised to see much better concordance of my testing results to the rated specs (compared to the Wild West of my early days of reviewing). So when I see a clear mismatch now, as in the case of this light, it leaves a poor impression.

All scoring is relative, and bound to be a bit idiosyncratic, but I initially gave this light 3 stars overall. I’ve since upgraded that to 3.5 stars, as I find the green switch LEDs (especially set to high) to act as a very serviceable “green moonlight” mode. This is one of the plusses of having the sophisticated Anduril 2 UI, combined with a reasonable initial minimum and max output levels on the main emitter. But again, Max/Hi levels step down quickly to ~750 lumens for thermal reasons – which is a bit lower than the competition. And the main beam tint on all levels below Max/Hi is decidedly green on my sample.

The build quality seems relatively decent for a budget brand, and with a few noticeable refinements, this light could easily move into a higher tier. The quality and features inspire enough hope that I am looking forward to reviewing additional Sofirn lights.

Acknowledgement

The IF25A was purchased through an online vendor from China (aliexpress). As always, all opinions are my own and the light received the same rigourous and objective testing as all other lights that I have reviewed. At the time of review, this light retails for ~$35 USD (~$45 CDN).