Sofirn SC21 mini

Twice this weekend I’ve put my hand into a very hot pocket ’cause SC21 turned on in my pocket. 4click, people, 4 click.

Or, half twist?

Make a ring of wire and glue it on. Not only stops pocket activation but makes the button easy to find by feel alone. Washers work too.

I’ve never liked side switch pushbutton lights for this reason. If the light needs a software lockout it’s because the switch design is deficient. The d4v2 hasn’t been a problem yet because of the switch guard and fairly stiff pushbutton, but it takes some care to get those details right. Otherwise I’ve generally preferred recessed tail switches.

The Sofirn SP10S definitely has the accidental switch-on problem. Pretty much turns on anytime it goes in my pockets! :rage: That’s why it’s now my bedside flashlight. :+1:

Recess the sideswitch, and guaranteed someone’ll bitch about it being hard to press while wearing gloves, or not being able to find it as easily vs it sticking up proud, etc.

Make it stiff, and someone’ll bitch that it’s hard to activate if you’re arthritic.

Basically, ask 10 people for how to “fix” a light’s design, and you’ll get 24 different and conflicting answers.

Until not that long ago, tail switches were more popular than side switches. I think people generally prefer tail switches. Lights may have transitioned to side switches because the drivers have become more complicated and it’s easier to use an electronic switch that is close to the driver board. I have not yet tried operating my d4v2 with winter gloves and I could see how it might be a problem, but it is not really a cold weather light to begin with.

To some extent yes. But that said, some designs of sideswitches are a lot better than others.

For example: The Zebralight SC51 is a classic light, but Zebralight received a lot of criticism about the switch. It was too easy to press accidentally. It stuck out of the light, was large and had a soft touch.

Zebralight took the criticism to heart and addressed it starting with the original SC600 and then continuing with the SC52 series.

  • They used a smaller switch boot.
  • They deeply recessed the switch
  • They used a different switch that required significantly more pressure to activate.

The result is excellent. Because the switch is deeply recessed, the structure of the light tends to hold clothing and objects in your pocket away from the switch boot. If something does touch the boot, the extra pressure required to activate makes pocket activation less likely.

At the same time, the switch is actually still very easy to locate by feel … much easier than the tiny protruding switch of an SP10. The Zebralight’s deep recess for the switch acts to funnel the thumb directly to the boot.

Result is a light like the SC600 or SC52 can be pocket-carried without lockout and with minimal chance of accidental pocket activation. Yes, there is still a chance it could accidentally activate in the pocket, but that chance is quite low. Much lower than with any other sideswitch light I’ve tried.

Unfortunately, Zebralight changed their switch again and their latest designs don’t quite recess the switch as much. The SC53, SC64 and SC5wII have less secure switches that aren’t quite as well built to resist accidental pocket activation as the SC52.

And then there’s lights like the SP10.
In the SP10, zero design effort was made to resist accidental pocket activation. The button is small, but it is metal, protrudes outside the body of the light, and requires no special pressure to activate. With this light, its either glue a washer around the button or use lockout.

I have a couple SP10s that I modded with Mtn 15mm drivers with Emisar D4 firmware. I also added rings cut from sheet aluminum around the buttons, which turned the protruding buttons into flush buttons. This helps, but still not enough that I would consider pocket carrying one without using lockout.

Of course this problem isn’t unique to sideswitch lights.
Many tailcap lights also have the same problem. Especially so ones with electronic switches that tend to require less travel to activate like the FW3A. This is why every FW3A I have has received the o-ring mod. Some of them have quite strong switches now. My stiffest FW3A switch requires more than 5 lbs of pressure to activate … easily enough that it can be safely pocket-carried without using any form of lockout.

I think the best solution to prevent accidental pocket activation might be something along the following:

  • Insert a sensor or some kind into the light that detects if anything is within an inch of the lens.
  • When you press the button to the turn the light on, the first thing that happens is the light activates the proximity sensor.
  • The light only turns on if the proximity sensor does not detect something next to the lens.

If it worked, you could have a pocket EDC light with a big easy-to-find protruding button, with no need to lock it out and no accidental pocket activation.

How to implement this? … no idea. I’m not an engineer. IR sensor maybe? Or maybe have the main LED turn on at 0.001 lumen moonlight and if the sensor registers a change in brightness when that happens, then the light does not turn on in higher power modes?

This would complicate the hardware, and also be set off by anything in front of the lens on purpose, such as a filter or diffuser. The basic issue is the pushbutton so I hope it is possible to fix the pushbutton.

The best thing is to just turn it a quarter turn to electrically de attach the spring side.

That certainly works.

But one of the nice things about a pocket EDC is how it can be yanked out of the pocket with one hand to instantly give light.

If you use tailcap lockout you probably need 2-hands to unlock it. This slows down the process and removes much of the convenience.

That sounds like it relies on insulation from the anodization in the threads, so a tiny pinhole could defeat it. Doesn’t seem good. In fact one reason I like titanium lights is they’re not anodized, so the threads give an additional conductance path.

Got my Sofirn SC21 today.

My initial impressions:

What I like:

  • Fairly good output for a 16340 light.
  • Small, but not the smallest 16340 light. The Eagletac D25C is smaller. The Wowtac W1 is much smaller (and also has onboard charging).
  • Rubber cap for charging port is well done. It fits securely in its slot and doesn’t feel flimsy or too thin.
  • Feels good in hand.
  • Looks cute
  • Low cost

What I don’t like:

  • Due to protruding button, the risk of accidental pocket activation is high. Using mechanical or electronic lockout is a must when pocket-EDC’ing this light.
  • I do not like the clip. It’s not that the clip is poorly implemented. It’s ok. It stays on the light better than the similar clip on my Thrunite T1. However, I’m not a fan of bi-directional clips since they tend to be harder to use and contain unneeded mass. I also think a black-colored clip would look a bit nicer on this light.
  • The tailcap magnet is not removable. This is surprising, because the tailcap spring is of the type commonly used with removable magnets, but they glued the magnet.
  • The bezel is glued, making modding difficult.

Managed to break the thread locker and unscrew the bezel without damaging the anodizing. Not too difficult to remove. Thankfully they didn’t use Jetbeam levels of threadlocker.

Inside is an 18mm copper star.

How does it hold up performance wise? Quick drop from high or does it seem to be a good EDC in cooler conditions for 20-60 mins?

Is the driver compatible with sc31pro host?

I think there is significant drain while 4click lock. Some readings recently, torch not used apart from checking it’s 4click locked.

Locked
11:11 26-Oct-21
SC21 3.973
09:40 28-Oct-21
SC21 3.808
15:14 30-Oct-21
SC21 3.665

Unlocked
10:24 01-Nov-21
SC21 3.666

Locked
15:06 03-Nov-21
SC21 3.602

Recharging to test again.

After reading your report, I decided to try measuring the idle drain on my Sofirn SC21:
(maybe someone else can also try confirm testing, as my meter and measuring may not be that precise)

while Off (not Locked-out): 0.07mA
while Off and in Lock-out mode: 2.8mA (this seems fairly high parasitic drain)

It seems Lockout mode does consume a fairly high amount of power

(Side comment: I can think of the Skilhunt M150 which also had a similar issue:
while Off, the idle drain is OK, but when in Lock-out mode, it also had high parasitic drain

(Skilhunt M150 also uses around 4mA when in Lockout mode, but when not Locked out, it has very load idle current)

I never use lockout modes myself, but that seems significant and a cause for concern.

Hopefully Barry or someone else from Sofirn can comment and/or look into it?