hi,
I’d like to lower the max current of this light and I wonder if I’d just need to change some resistors, here are some pics :
I would assume the current sense resistors to be the 4 biggest (3216/1206) resistors of 40mΩ which looks to be wired in parallel between BAT- and LED- (not 100% sure because without desoldering the top PCB I can’t see well.
Would I just need to increase the resistance ? by how much ? Would +25% mean –25% current ?
thefreeman, I am not necessarily going to answer your question as I know nothing about it, but I would like to ask you, if possible, to take some more photos of the driver, please. If you don’t want to post them here, you can send me PM, please.
I gained a sudden interest with this driver / light and I would like to know if it will “fit” my aim!
Thanks again and I hope someone can step in to help you 
The amount of current going through can be calculated with Ohm's law. Initially you know the sense resistor stack value, and the driving current the driver is stock set at. Considering the driver outputs 8A on turbo (check this out) and the sense resistor stack value is 10mΩ, the sense voltage is Vsense = Idrive × Rsense = 8A × 10mΩ = 80mV.
Now, resolve Rsense2 = Vsense / Idrive2 and you will have the target sense resistor stack value for the desired reduced driving current Idrive2.
Have you looked at the Imgur album I linked ? there are more angles in it, but I can desolder the 2 PCBs to get a better view, I was too lazy to do it yesterday.
Great! thanks for the answer.
I desoldered the 2 PCBs and added more pics to the album : Sofirn IF25 driver - Album on Imgur
better view of the lower PCB :
Hey thefreeman!
Sorry I didn’t see the “album” before, I just saw the photo and didn’t click on it :person_facepalming:
Thank you very much for your time to replying and taking te photos. My main doubt was about the switch being integrated on the driver, but it is not, it is wired so there is a very good possibility that this may be good for what I want 
Thanks once again for your reply on this once again and for the detailed photos :+1:
Still, I presumed “8A” as the stock driving current in the IF25's driver because of info like these tailcap current measurements in Sofirn IF25 USB-C 2500lm Variable Temempeture 2700K-6500K quad led's thread. However considering people reports of its actual output on turbo the actual current is likely lower.
It is very important to know what the actual stock driving current value is to properly determine and set the driver's current, though.
Another way to obtain the sense voltage over the sense resistor stack terminals implies measuring it with a multimeter. 
Yeah 8A in is the max current I measured with my bench power supply, but I’m too lazy to rewire some leds outside the light to measure the actual drive current.
Do you have a schematic?
All 4 resistors are connected to the ground , so yes , these are the culpit . Just take one out ( any of them ) and you'll have 25% less current to the led .
thefreeman didn't say how much drive current he aims for; cera@1967's suggestion is fine if he just aims to somewhat reduce heat output, as it reduces about linearly with power output or current drive. One, or maybe two of the stock R040s. With half of the drive current output would still be above 60% stock but heat would be a lot lower (half peak temperature maximum). 
Yeah I didn’t say 
, it’s just too bright for my use case, mostly as a lantern/ceiling bounce lamp, also as a normal flashlight but I don’t need this high of a turbo, but also for the LEDs, I put sliced LH351D in it and somehow one burned its phosphor layer.
A bit strange at only 4A per LED !? but I noticed afterwards that I sliced them very close to the phosphor layer so maybe that, maybe Vf differences, not dirtiness because they were very clean, anyway lowering the current will be safer.
I’ll try with 6 and 4A with one and two removed.
So I made some measurements, at turbo the drive current increased with the input voltage to above 9A and then it crashed down to ~4A, not sure what was that, I didn’t test again cause I didn’t want to break something.
After removing one resistor the max current Out is now 7.38A , Vout 3.41V (2xLH351D), which is reached at Vin 3.83V, with Iin 7.38A (eff =89%). although the current isn’t very stable, it decrease progressively toward 6.9~7A, but sometimes jumps back up a few 0.1A…
The Vin-Vout of the buck driver seems a bit high no ? that’s why it the Vin needed to be so high to reach the max current originally. the high Vf LH351D didn’t help opposed to the original sst-20.
I suspect the user who measured Iin 7.3A couldn’t get the actual max current due to cell resistance, springs (24mΩ each, uncompressed), connection resistance and ampmetter resistance if he didn’t use a current clamp.
The min drive current didn’t change at all and is still 3.64mA, maybe the lowest level isn’t controlled in the same way ? I didn’t measure in between discrete levels because it is in ramping mode, I had so much trouble switching mode that I didn’t want to get it stuck back in stepped.
So anyway, the max current got reduced, but in practice not really, so I’ll remove another one.
I did some efficiency measurements for curiosity (with the original sense resistance) :
~low :
In 4.197V 0.151A Out 0.184A 2.631V eff= 76.4%
~mid :
In : 3.97V 1.00A Out : 2.80V 1.19A eff = 84.9%
turbo :
In : 3.84V 7.82A Out : 3.46V 7.82A eff = 90.1%
turbo with lower Vin:
In : 3.41V 4.69A Out : 3.18V 4.70A eff= 93.5%
With two resistors removed the maximum Iout current is now ~4.8A, I’ll try the light with this.
Thanks for the help.
With two R040 resistors removed the sense resistor stack value is 20mΩ. If Iout is ≈4.8A, then Vsense = 4.8A × 20mΩ = 96mV.
If this measurement of yours is accurate, this means Sofirn was quite daring concerning the maximum driving current setting for just (down to) 2x SST-20 emitters: 9.6A. I understand that 9.6A driving current is a peak value, attainable only for as long as Vin is high enough and any remaining conditions go well with it (low cell internal resistance, low spring resistances, etc.) to ensure the minimum required energy and voltage is input to the driver for it to deliver the target Vout to the emitters.