-Resistor- mod for- olight sr51 driver now running my 2D Magmod

Could anyone help me ID the correct resistors to change/add/bridge to up the current from my SR51's driver. Currently the emitter is getting 3.4A, I'd like to really push it.

What's that metal bar at the bottom? Anyway, down there is where you need to look, the part not shown, around the main FET (the AO4468).

Its just a big ass jumper, there arnt any resistors down there and the only things hiding under the smaller board are two larger SMD ceramic CAP’s and a 8 pin ic (which is identical to the one under the big cap right next to it). I don’t see any main FET’s and none with that marking, only see three baby FET’s on the whole thing.

Some more pic’s:

http://75.65.123.78/AO4468.pdf

The big metal bar IS the sense resistor, looks like. It goes from the source drain pins on the FET to GND, right? And LED- connects to the pins on the other side of the FET?

Ah ok, the drain pins of the FET go to the bar, the other end of the bar is LED~~. Source (3 pins) go direct to batt~~. G (one pin) goes to the right most resistor marked “010” in the biggest bank of resistors down on the bottom right (in that pic)

What options does that give me, either move the LED- direct to the FET drain? Jumper the resistor on the gate?

Thanks for helping so far.

The metal bar sense resistor is probably already very very low, you can try replacing it with fat wire or a ~.040" thick strip of copper, but I doubt there's much left in there to find. That's the same FET used on the 3-toroid SRK drivers, they won't do more than about 4.1-4.2A with the resistors bridged, and that's when powered by four cells in parallel. The FET's on-resistance is just too high. You might could find another SO-8 FET with a Rds(on) down around 3-4mOhms that would work as a drop-in replacement, otherwise I don't think there's much to be done there.

What about a different package FET if I was up for some finanglein’?

Is there a different approach to bumping this driver up, somewhere else to make some improvements?

Here’s where we are now, I measured that bar and it was only reading <1ohm but now that it’s replaced with this solid copper 16 gauge wire it was reading 6+ amps on my test bench however the current emitter I was testing with was an indirect Al mounted XML and it only lasted ~5sec before it died. I don’t see how that bar setup could of possible limited it that much.

Lower resistance FETs in that same package are common, no need to kludge in something different. It's a really straightforward circuit, the only things in the path between LED- and ground is the FET and the sense resistor, and if the sense resistor is already down at something like 0.0015 ohms, the only thing left is the FET.

What's the battery config on this light? I checked the Olight site and can't figure how you get a light that uses either six CR123s or two 18650s...

On 18650’s its 2S, not sure how it runs the 6x123’s.

Got my bench setup swapped out with a proper copper mounted xml2, it’s reading 4.78, I think the 6A+ reading before it died was runaway.

Every solder connection in this light is way over loaded with the solder, I’m going to clean up all the connections and swap the 20AWG wire for 18, if I can get 5A I’ll be happy for now while I look for a lower resistance SOIC FET. So is the 8A rating just a joke or what’s the deal with that?

Edited to add: hmm, the driver no longer has low, still has the hidden strobe tho (double click from on)

The current rating in the spec sheet is for running a load that's not dependent on voltage like a LED is. Since more voltage = more current, resistance in the FET lowers the voltage which also lowers the current. A different kind of load like a DC motor or a straight resistive load doesn't work like that.

Losing modes is pretty common when bridging resistors on a lot of drivers. It'll need at least some small amount of resistance there to go back to working again. A better FET will let you use a reasonable sense resistor and still get more output current. The 4468 is something like 23-25mOhms (for comparison, the Vishay 70N02 is 4.5mOhms).

see here: http://www.digikey.com/product-detail/en/IRF8734TRPBF/IRF8734TRPBFCT-ND/2096660

Thanks for all your help again man.

I’m not to concerned about low, its a dedicated thrower now that its dedomed, wtf do I need low for lol.

The final report-
Stock was 3.4A to the emitter for 58kcd
Modded & dedomed it’s running 4.7A for 123kcd (see post 14) and a much more pleasing tint.

You're lucky it wasn't one that (mis)behaved like this one: https://budgetlightforum.com/t/-/20635 xD

I just received my SR51 yesterday. Really like the beam pattern, but the UI and modes suck. Really high low and not much mode spacing between High and Low. After reading this, I may just going to swap out the whole driver.

Nice to know how to resister mod the stock driver though. Thanks for this thread.

comfychair wrote:

What's the battery config on this light? I checked the Olight site and can't figure how you get a light that uses either six CR123s or two 18650s.

The 18650 holder is 2S. The CR123 holder is 3P2S.

EDIT: Unfortunately, the battery tube can't take 3 18650's and it's too thin walled to bore out. What a shame, there is plenty of vertical room to rig up a 3S carrier in there.

Don’t overlook the stock MCPCB, while it’s really nice looking all copper clad and huge it’s not direct bond, after getting it all back together and measuring lux it didn’t even make it threw the 10sec reading lol. Beam started out nice and neutral and almost instantly there was a major blue shift followed by instant death.

Would it be possible to get any gains from leaving the big bar sense resistor but swapping in a better FET so it would retain factory modes?

Now that I’ve got a new U2 1A dedomed on a XM20 it’s doing 134kcd, I’m happy now and gonna leave it alone but I’m still wondering about the question above.

Yes, that was what I was suggesting earlier. With the high resistance FET and very low sense resistor, it was already running near wide open in stock form. Swapping that around, a low resistance FET and maybe even a higher sense resistor would net you more output while leaving more safety margin.

Dang it C_K… now you make me want to purchase one too :weary:

Man that driver looks complicated. I like the idea of swapping out the FET and also replacing the MCU. Won't be touching the light until after I finish my scratch build though.

If you get a chance, it would be nice to see some beam shots. I think your the first to install dedomed emitter in this light.

So it turns out there is a low still, its just barely dimmer than high lol, its not useful at all and its such a small difference out in the open I couldnt even tell, only when I went into a tunnel was I able to notice.

Beamshots here-

There's what looks like another FET opposite the AO4468, can you get the number off that one, or a pic with the big cap more out of the way?