WARNING - POSSIBLE FAULT WITH LOW VOLTAGE PROTECTION ON SP-40
Here’s what i have sent to Sofirn contact email:
“I have SOFIRN SP40 for almost 6 months. Recently I used it for several hours continuously. My understanding was that it has low voltage protection which protects the cell from discharging under 2,85V. Here is what has happened:
I use NCR18650B cell from Panasonic. The flashlight was running on high (not turbo - one level below turbo) for several hours (3-5 hours) continuously. And then it stopped sharp, it shut down completely. My understanding was that the 2,85V low voltage protection kicked in. Two minutes after that I have removed the cell from the flashlight and measured it with a UNI-T multimeter. The cell measured 2,46V!!! This is a measurement with no load, therefore when the flashlight was still on there was load on the battery and I estimate the voltage must have dropped even lower, to ~2V.
Is this expected behaviour of the flashlight? Is it faulty? My battery cell is not safe to use anymore and I have to buy a new one because of it.”
I will wait for the reply and let you know. Maybe i will also sacrafise another 18650 cell, to test the low voltage protection and see the exact voltage where it cutts of. It could also be that the heat affects the cut-off voltage. I dont know. Also any comment will be much appreciated.
I have contacted the Sofirn. We exchanged several emails as I have made some more tests and send the results back. They were quite responsive. Here are the conclusions if you don’t want to read the whole message:
I was unable to replicate the problematic scenario, where cell is drained <2,5V.
LVP kicked in between 2,64V and 2,746V. The range is quite big; also seems to be heat and current dependent. In my opinion 2,64V is too low for LVP. Many cells should not go lower than 2,85V.
Sofirn claims that this is ok and LVP works fine based on my testing. However, they said, they will talk to engineer, if there is a need to increase the LVP trigger voltage from 2,8V to 3,0V.
So, I have made four tests:
TEST 1: Room temperature. First test with the cell that came with the flashlight (DLGH INR18650) - The low-voltage protection kicked in at 2.703V. Room temperature
TEST 2: Room temperature 23,7C; NCR18650B cut off at 2,746V.
o Cell recovers voltage after cut off (important for next TEST2):
o +1s: 2.756V (+0.010V)
o +2s: 2.775V (+0.029V)
o +3s: 2.289V (+0.043V)
o +4s: 2.795V (+0.049V)
o +5s: 2.805V (+0.059V)
o +6s: 2.813V (+0.067V)
o Data is taken from video.
TEST 3: I suspected that temperature might have something to do with this. I tried to simulate generated heat by long runtime and having it on a head by putting it into regular oven and pre-heat it to ~40C (including the cell NCR18650B) and keep it at ~40C while it was running. The only problem is, I could not measure cell voltage in the oven, so my measurement had 6 seconds delay (until I take the flashlight out of the oven and remove the cell). I measured 2.73V. Assuming the same voltage regain happened in 6 seconds (0.067V), this means cut off happened at around 2.66V
TEST 4: Fully charge NCR18650B and let the battery run on »High« until empty on room temperature. When it cut-off I took me around 8 seconds to remove the cell and measure it with UNI-T voltmeter. It measured 2,716V. From previous testing experience, the battery regained around 0,07-0,08V in that 8 seconds, this means, that the protection kicked somewhere around 2,64V.
Measurements were made with UNI-T UT136B. I took several additional measurements with DT-830B (a cheap multimeter) to validate the accuracy of UNI-T. DT, was showing 0,02V less typically. But in my experience UNI-T is quite accurate in these voltages. I have two chargers for batteries: a professional CRP PlusarTouch competition and XTAR VC4. When I measure full cells with UNI-T they are always 4,2V(+–0,01V). Therefore, I assume my UNI-T is quite accurate.
I hope this helps someone and also, that they make a new version of SP40 with better LVP.
Me personally, I’d have LVP kick in at 3V or even higher (at load, terminal voltage sags). Any lower doesn’t buy much more runtime, as cell voltage at the tail end drops like a rock anyway.
Oh yeh, most cells have that “corner” as high as 3.4V-3.6V, and really start dropping like a rock beyond that. Check HJK’s charts for common cells’ discharge curves.
I agree. I also wrote that to Sofirn in the last email:
“So, I think you should increase the LVP to over 3,00V, even better to 3,3V. Liion cells hold very little charge under 3,3V. Also, the brightness of LED gets exceptionally low.
Check the results of how much capacity cell holds at different voltages on this web site: https://lygte-info.dk/info/BatteryChargePercent%20UK.html
None of the tested cells had more than 3% usable charge at 3,3V. So, there is no benefit to drain li-ion cells below 3,3V. You are decresing cell’s lifetime and it can be dangerous if it discharged too deep. I suggest that you change this on your flashlight.«
Almost a year ago I got a SP40 (for review). I haven’t used it for a few months. The Sofirn battery was in the light (approx. 3.8 V in March), head loosened almost a full turn (the light didn’t turn on), sitting untouched in a box. Today I grabbed it, tightened the head and it didn’t work. Checked the battery voltage: 1.71 V!
Don’t know how this could happen. With the head loosened the circuit shouldn’t be closed. Can’t find a damage in the anodization either. An internal defect of the battery?
I will keep the battery in a fireproof container for a few months the check if it discharges even more. But you all should check the voltage of your unused Sofirn batteries!
I was thinking I had an issue with my XTAR PB2C (new one, not the rubbery one), that it may’ve not gone to sleep and drained the cells down reeeeally low if kept “active” the whole time (a few weeks). It was a Sofirn cell, but I don’t recall if it’s protected or not.
Had to jumpstart it in dumb-charger, and it charged right up, and worked fine in a light, even on turbo.
I wasn’t “blaming” either the powerbank or the cell, just wanted to find out why that happened. Different cell, the PB’s doing fine, so it may very well be the cell that’s draining on its own.
Had one in a SD light (dive-light, forgot which one) that the cell came out dead. Figured I didn’t tailcap-lockout enough, but it was a good full-turn to snug it up.
Dunno. I’d have to keep ’em in boxes untouched and see if they drain. They’re usually topped-off after doing a capacity-test, so…
Well, this does not mean much. In my opinion the cell was fine. These are Lion cells which must be precisely manufactured as you have thousands of these cells in every car, and they cannot just die like LIDL AA batteries. They are waaay to dangerous. So, if it is not a second-hand cell, I doubt its cells fault. Continuation of voltage loss could simply mean cell was already drained too low and became chemically unstable. Of course all this only implies if you had mostly new Sony, LG, Samsung, Sanyo… cell, not Aliexpress knockoff…someting ending with “fire”, liitokala or similar fake garbage.
The flashlight could have some parasitic current flowing through when it is off. I don’t know. It would be interesting to see the results if you can measure it.
At the end of the day, it does not really matter how cheap or bright it is if it is not safe. I am sad to see that the Chinese are generally so focused on sale, that they often forget about the product.