Driver Quiz... eSwitch question inside

Hey guys!

Which driver is that ?
and how can i “help it” a little :wink: ?

Manual says it takes 4A out of the 4 18650s and feeds the 1S3P wired XM-L U2 with it

LED+ = Bat+ = Charger+ :open_mouth:
Modes: H/M/L (doubleclick = Disco)

I think both of those big resistors are the sense resistors, but I do not know for sure someone will have to second that.

Yup, those are the sense resistors in parallel. You could try doubling the current and replacing the wires with slightly thicker ones. (Given that the light has adequate heatsinking of course. :laughing:

As others have mentioned, try the sense resistors. If you still need more juice try a new FET

is the fet on my driver weaker?

I thought that these two guys are the sense resistors…
nice that I was not wrong…

Check out the mod/increasing output section of the review. ;)



i have the 3 -LED Version… :wink:

if i find a way to improve heatsinking i will write some words…

Usually it’s not a problem with 3 leds, but I saw a difference with my 6 xml2 srk. At that much power little things start becoming more noticeable.


i soldered a tiny bridge… :wink:

Tailcap meassured: (using full NCR18650B)
parasitic drain 0,012A
before OFF - 3,3A - 1,6A - 0,5A - OFF
after: OFF - 6,6A - 6,6A - 6,6A - OFF 8)

BUT: is it common that a bridged driver goes off regulation in all Modes
or did i break something?

Edith had an idea:
i just have to use a pushbutton which takes 7-10A and use that instead of the bridge as an TURBO :wink:
just put that thing into the empty charging hole…

fun fact:
i have a Battery test station :wink:
1 NCR18650B delivers 5 A
a unprotected CGR18650DA got me 6,5 A :smiley:
some other protecteds between 4,5 an 5 …

If you read the link/section I gave you it said it did not work to bridge. I used two R100 resistors.

i did read your Link - but i had to try :wink:

i think using a switch for momentary bridging will be where i go with this Light
-> 4 Modes incomming!

works nice - but the switch i had on hand has a too high resistance…

has anyone a describtion on building a fet-based switch?

A N-channel FET is just like a relay but with no moving parts. Put ~5v (or battery voltage, in this case) onto the gate pin, and it enables current flow thru from the drain (the big pad on the base) to the source pin (ground). No 'circuit' needed really since it doesn't need a latching function, only momentary. Hold the momentary switch, FET turns on and bypasses the resistors.

edit: Oops, forgot to add that a 10k-20k resistor will be needed between the gate pin & source pin, to bleed off any residual voltage at the gate when the switch is released. Otherwise the FET could remain partially on.

And yes, the original DTU40N06 is garbage. The datasheet only lists the on-resistance at 10v instead of the more applicable 4.5v, but at 10v it's 0.013 ohms. The Vishay SUD70N02 at the same 10v is only 0.0026 (yes, there's supposed to be an extra zero in there, not a typo). Numbers at 4.5v are usually a little less than double, the 70N02 is 4.2 milliohms while the 40N06 would be something like 22-24 milliohms.