very interesting……
Interested in buying PCB
Are you willing to list part capacitance and resistance values?
The parts are the same as for all TA 2S drivers
MCU Attiny85 for Narsil
Use 1% resistors all around
R1 : 19.1 k (220 k for e-switch lights, 360k for 2S setups like the L6 driver)
R2 : 4.7 k (47 k for e-switch lights)
R3 : 100 k
R4 : 47 ohm
R5 : 4.7 ohm
R6 : 0 ohm jumper, or can use a solder blob (LDO drivers only)
R7 : 0 ohm jumper, or can use a solder blob (LDO drivers only)
BR : Bleeder resistor for lighted tailcaps, varies between 470-630ohm (optional)
C1 : 10 uF (be sure it is rated for the voltage you plan to use, so for a 4S setup, it needs to be 20v or more)
C2 : 0.1 uF (10uf for LDO drivers, can use the same one as the C1)
OTC : 1 uF (functions as e-switch pads for e-switch lights)
U1 : SOIC-8 footprint Attiny25 for bistro, the 20mm+ versions have pads for the Attiny85, the 25 & 45 also fit on them
U2 : LFPAK56 MOSFET (aka, “FET”). PSMN3R0-30YLDX is a popular cheap option, the SIR800DP is better but costs more.
D1 : Schottky Diode SOD-323 (if using a zener and 2S cells then you will need a 200 ohm resistor here instead)
Z: Zener diode (link is external) SOD-323 (Optional, only needed for 2S+ operation)
LDO : Designed around the MIC5235-5.0YM5 LDO voltage regulator
7135 : Standard 350ma or 380ma 7135’s can be installed on these pads. I prefer 350ma myself. You will need to clip the center pin of the top side 7135 if you plan to use a bleeder resistor. The 7135’s can be purchased from RMM or FastTech
Many thanks, Lexel
This light is also sold as Mateminco MT18. Check out for pictures and additional data. Either the run time data is completely off, or there is going to be a huge thermal step down (although temp is set at 60 C) accompanied with aggressive reduction in output for the fairly high run time of 4 and 3.5 hours for the highest two outputs. Price is $55, but not sure how to interpret Taobao prices.
Edit:
Based on the run time data, I see a possibility that the light steps down aggressively when reaching 60 C to the 1100 lumen mode (for the Nichia version), and stays there until batteries are depleted. So there is no feedback control for output as a function of temperature.
If so it’s a bit unfortunate that there is no 2500~3000 lumen mode where the light can sustain output fairly continuously.
Good catch and it sell for $54.87 too :partying_face:
Can we swap the led with xpl hi v3 ??
This is my concern as well, after five or ten minutes this great big light won’t be much brighter than a single xml2 light.
Why wouldn’t they add a few more modes to it? When they added the 1100 mode they also dropped moon. If it was because doesn’t have enough memory for more modes why not get rid of one of the flashy modes? Why handicap the design that way in the first place?
The board Lexel designed will add a lot of functionality to the light but how much heat will the driver make with all those 7135’s when making 2500 lumens? Is eighteen led’s wired 2S9P on a single channel reliable over the long term? Its hard to say how matched the led’s are and if thermal runaway will happen when running the light hot for a long period of time.
A version of this light that made 7500 lumens max would be so much more useful.
you can modify the sense resistors to do less current
I aggree that a sustained level of like 2500/3500 lumens Nichia/XPG3 should be possible, as you can see the temperatue drops after the light stepped down to 1100 lumens
sorry wrong thread, it was meant for Convoy L6. :person_facepalming:
comment deleted….
I’m thinking about doing that but it would be tripling the value of the sense resistor to get the mode levels where I want. That might be too much of a change and the driver could stop working. Not sure if it will cause the sense resistors to overheat as well.
The other option I’m cosidering is using your driver and limiting the maximum PWM.Then popping the inductors off the old driver and putting them in series with the led’s.
They would smooth out the waveform and do some load balancing as well. If one set of led’s gets too hot and starts drawing too much current the inductor will heat up increasing it’s resistance. They probably have the wrong parameters for this application so there’s a good chance they’ll overheat.
This light looks perfect for task lighting, but the UI is set up so bad.
Is it possible to remove the Carclo optics while keeping the front lens (eg for water proofing).
For example, on my Noctigon Meteor, the Carclo optics are drop in pieces, which can be removed, and the front lens can be retained.
The lens sits directly on top of the rim of the head
so you can replace the optics
on my sample I sanded down the bezel about 0.3mm to get more pressure on the o-ring
do you need measurements how thick the original optic is?
No, you answered my question. I wanted to remove optics to maximize flood. Thanks!
Any idea which bin of Nichia 219C is used?
How should the drive connect to the LEDs, all in parallel?
The initial shipment of this flashlights in China.
There are several flashlights that have this problem.
Original description, problem from MCPCB (DTP)
The LEDs are wired in 2S/3P the Battery carrier is 2S/2P
so a 2S DD driver works
Received, NICHIA
This is not ASTROLUX
Exactly the same product.
How to distinguish it?
ASTROLUX assembly quality is better?