Sofirn C8F 21700 Teardown and Review

You’re not going to be able to get the lights working on the switch unless you cut pcb traces.

As far as the two wires to operate the switch, see below.

The black wire goes to the driver ground.

The white wire goes to the driver switch positive.

Thanks for telling me. I would have needed to try all combinations otherwise. Do you know where I would need to cut traces to get the switch leds working? In the mean time I will try to solder the black and white wires.

I show all the circuits a few posts above. You’ll need to replace the double leds with single units, cut a trace then solder a jumper wire - I think. I’ll have to look at it later tonight.

I’m already happy if the switch works. I already killed the green leds. Maybe I could fit a Q8 switch inside. Thanks for the nice helpful circuit drawings.

I managed to complete the TA driver with a working switch. But now there is another problem: Outside of the driver cavity the driver works perfectly when connecting a battery. But as soon as I screw it in with the retaining ring it stops working and shorts the battery! I have no idea where the unwanted contact is.

if your batt + spring is compressing too much or sideways it can touch a 7135 leg on the back side of the driver. I suppose you could test with an 18650 and the spacer to see if the shorter battery helps.

Is the retaining ring touching any of the components on the driver? You may have to increase the diameter of the retaining ring so that it doesn’t touch any components.

I tested the driver with a 18650 battery holder holding one cable to the spring and the other to the outer ring. I didn’t compress the spring. The driver retaining ring is touching the plastic part of the 7135. Shouldn’t be a problem?! So far I didn’t screw the battery tube on with a battery because I fear it would explode then (maybe not but who knows?).

I have a Utorch C8F which seems to have a timed ramp down to medium after being on Turbo for 5 minutes (+ a few seconds). Didn’t see any sign of temperature control during this time.

All the LEDs in the C8F are just wired in parallel, right?

The main leds? Yeah, that’s why they run on a 3.6v battery with a FET driver.

Thank you contactcr and JasonWW for your help! :+1: I managed to complete the C8F successfully now apart from the lighted side switch. I resoldered the wires on the MCPCB and isolated the reflector with kapton tape. Lexel gave me this hint in the TA drivers thread, he said shorting wires are the only possible cause he could think about. It must have been the positive wire contacting the copper of the PCB or the reflector. After screwing everything together again it worked well. :student:

I forgot you had to unsolder the mcpcb wires for a driver swap. I thought yours were still stock. Oh well, glad you got it sorted.

LoL all of this current draw optimization treadmill is pretty crazy. I see the MOSFET used in the new 21700 C8F should provide some extra edge over the 18650 version, but it seems you do not get the expected results because of some “unaccounted” roadblock. At least on paper the newer PSMMN2R0-30YLE MOSFET is better than the AON6418 used in the 18650 C8F, but please note both MOSFETs are outside their optimal RDSon to VGS zone in the graphs, RDSon creeps much higher under ≈3.8V for both according to datasheets. Maybe going back to the AON6418 would “fix” this (= AON6418 “better” in practice)? Or using a MOSFET which actually works well with low VGS values.

I do not really care about flashlight drag racing. To me the actual performance of my flashlights is the one I can always count/rely on nearly irrespective of battery SoC or temperature.

Preheating batteries to attain higher current draw LoL!

Cheers :-)

I wish I understood more about the mosfet specifications, but I barely understand how they work. Lol

I’m not sure why you talk about preheating a battery to get better performance. I don’t think that was mentioned in this thread.

I agree that trying to get maximum amperage is not very practical. It just heats up the head and runs down the battery quicker.

I switched over to using the Sofirn SP33 v2. It’s got a nice regulated output. It doesn’t build up heat very fast and the battery lasts a longer time.

Yes it's been mentioned by… yourself!

Cheers ^:)

Oh yeah! I forgot. Lol

I recently ordered another Sofirn C8F-21700, and noticed that the tailspring is now not factory-bypassed (left), but instead uses a double-spring (right).

I think they now have the same tailspring as the Sofirn C8G (which also uses the double-spring). (I tested that the C8G and C8F-21700 are interchangeable)

Any idea if which one is better? Which can pass higher current?

I notice that Sofirn's factory bypass wires tends to get detached when using high-drain batteries (at least that's what happened on the Sofirn C8A and Sofirn C8F-18650 when I was testing Turbo mode for a bit longer time, and notice the factory spring-bypass wire on both sides got detached probably due to heat -- and I'm not good at soldering small parts)..

Properly done spring bypasses require wire coiling. Without it, chances for the wire to break are substantially higher. So either do a custom spring bypass, or use the double spring is my advice.

:-)

As Barkuti said. Properly done wire bypass gives higher current but double springs far more reliable. Even a properly done wire bypass can break over years.