Sofirn SP35 warning, light got a shelf floating in the air + LED wire came loose

The Miller, I had to watch the video several times until I understood that he used this tool with the Klarus flashlight and not with the Sofirn.

Ah hahaha that makes sense :slight_smile:

Fairly easy to understand the pill is press/friction fit inside the housing it has no way of attaching to the body. It does not screw in place or is not glued down. It is held in place by the force of the reflector and bezel!
Yes there is a lip/shelf around the pill that may touch the light in some areas. But if it fits in so easy and is so easy to pull out how good is it connecting to the body? How much contact area do you have? Like it can rattle around in the light and not make proper contact. Even with that lip its still a fairly small area of contact to the body when it is actually making proper contact? Could be a gap of a hundred of a mm around the pill to the body?

I have similiar built lights with the same design they dont have much of a heat path.

!

!

!

!

This light is a fairly under driven headlamp compared to the XHP50 that is in the Sofirn which is meant to be running at 30w

Ah! I see, the video didn’t show the driver installed so it was unclear the relationship.

Yes, I too wondered about the talk of the heavily indented wood as to how difficult it was to pull the emitter shelf/heat sink. Misleading a bit there, placing the wood on top of this light and explaining it.

Of course it would be easy to foil wrap it and reinstall, but the factory should have a proper fitment not requiring intervention. I’ll ask Tracy about it, see what she says…

The inside of the battery-tube (where the shelf-cup sits) might be slightly cone-shaped, and ideally the shelf-cup too. In this case there could be good thermal contact when pressed in by the reflector (and the shelf-cup would be lose and rattle if not pressed in). Perhaps SOFIRN uses this construction. Can the shelf be moved sideways when pressed down by hand?

I asked Tracy to check into the matter, we’ll see if her engineers give an adequate explanation…

Is it just me? Why do they make a nice metal pad on the driver where the screw holds it down and then NOT use a panhead screw? Wouldn’t a brass pan head make much more contact and carry better ground current? Most lights are this way, how do the engineers miss that important little detail?

You are right with the screws, but the opposite side the whole driver rests with a nice gold plated surface on the blank body
I replaced the driver screws with flat heads from a Laptop

I have stood accused of having beefed up numbers on my builds for a long time. The truth of the matter is that I was willing to spend the time on each and every light to reduce resistance everywhere I could find it, and by going to such great pains to eliminate all waste my lights produced higher outputs than most were accustomed to seeing. There are now a lot more people also doing similar and my numbers are not always higher, those that used to accuse me are silent.

Attention to detail accounts for a lot in most things, I believe that when we see a company paying close attention to every detail, we see a better product and are proud to own it. Perhaps Sofirn will be one of those companies, they are new and are trying, so if they listen to the people like on this forum, they can make slight adjustments and have a far better product for it.

Thank you for the pics and efforts taken on this light to show us what it does well, and where it falls short. Appreciate it. :wink:

I would say it is a manufacturing error. I don’t have this particular light, but I have the SF11 (4 AA), that is similarly constructed. Mine is tightly pressed in.

I was thinking that the P60 mod would work just fine here.

Has anyone heard if Sofirn is associated with another manufacturer? I don’t know if they’re a small company just getting started or a sister to a larger already known company. If they’re just getting started, they’re doing a fine job… missing a few details but they seem to have their ears open as to how to make improvements.

There are a lot of almost light “clones” between Sofirn and Thorfire, but I have no idea if there’s any formal relationship between the two.

I’ve made a few suggestions behind-the-scenes, and apparently some of them were taken, so there seems to be some independent r&d on the Sofirn side. I still say the warm-white SF31 is one of the nicest lights I’ve seen. :smiley: And the SP10A is quite nice, too.

Reminds me, I gotta pull a lot of pix off my phone. Don’t think it’s my phone, but the damned usb cable that ain’t making the connection. And all I got are charging cables… (Insert “duuuh!” here.)

Anyhoo, the C8 with the trip XP-G3s is quite nice, or will be, once the kinks are smoothed out. I don’t think Thorfire has one like that (yet?). Different depths of the reflectors, the position of the integrated shelf, etc., make at least the head a completely different part.

My guess is that the main issue with Sofirn is consistency. Eg, to mount trip XP-G3s, it’s one thing to reflow them onto a single MCPCB. But then the reflector has the 3 holes for the emitters. They either need to match up exactly, or there’ll be stress once the reflector is placed over them, and magnified once it’s torqued down. Enough to shear one or more chips off the MCPCB? Maybe. With 3 separate MCPCBs floating on thermal goop (not adhesive!), they can self-align, within reason.

So it’s growing-pains. You try this, it doesn’t quite work consistently. You try something new, it works a little better. Etc., etc.

“Try again. Fail again. Fail better.”
— Samuel Beckett

I have the Sofirn C8F, and yes there were some issues, but with a 32mm Noctigon sanded down to fit and XP-L V6 3D HD’s on it, it’s a very nice light! Their driver works fine, is limited with the supplied protected cell, but with a Samsung 30Q and the new emitters it makes 3314 lumens. Not a lot required to make that change, and I have informed them of this. Will be doing a review on it hopefully tomorrow.

i was planned to buy this light and after seeing the floating shelf…nope
but damn dusty, that ant on your signature really make me want to wipe my lcd screen :laughing:

Any response meanwhile?

Tracy told me she’d pass on my concerns to the engineers. And thanked me for the feedback. And then asked me for more feedback on future improvements. :slight_smile:

To quote:

“Hello Dale,

Thanks you for your feedback. I have email our engineers about this issues. And we will try to make adjustment in the next batch.

Have a good time

Tracy”

Well… we always need manufacturers for BLF-Editions :innocent:
Sofirn could make their name more famous here if they make a BLF-light
The SP-33 would be a nice start for that I think.
white LED in the switch, different UI (ramping mode like the Emisar?), 21700 sleeve, XP-L HI on DTP and so on

Thank you.
This light looks like Klarus G30 clone (just with different led). Did anyone met same problems with Klarus?
I think we need to open thread and name all lights with same construction (and show difference between bad and good variations). If Im not mistaken all 2x18650 lights with side-to-side cells location before EC4 had similar insert and usually it could “shake” inside.

This light has 3 cells the video are taken without batteries, the LED shelf itself is loose, has only contact to the driver