This will be my most powerful build yet. (12) XM-L2@3A

Wow guys…….

5520 lux at 5 meters.


Un-f-n believeable.

Im completely shocked it did that well. The head gets hot real quick. After about 10 seconds I start to get scared. Lol. think I just built a 250.00 dollar 10 second flashlight. What a waste! :~


When is it too hot to hold?

I havent got it too hot to hold yet. I want to do some temp and battery current tests before I do extended run times.


Nice Ken. Love your mod. Need some night shots now. :wink:


My thoughts Exactly!

I'm thinking for a conservative 700 lumens per emitter, you would have 8,400 lumens. More realistic may be 800 lumens, so 9,600 lumens total, but you so need to run it for 30 secs to get a proper reading .

Haha! I had to google the ERMAHGERD to see what it ment!

I will try and get some beamshots this eve. I still want to do the temp and battery current tests too.

I would like to put a HBflex for different modes but the max input voltage is equal to only 5 lithiums. I have 6.


Here are a few pics of it finished up. I did some calculations and it kinda conserns me. I calculated driver efficiency at only 76 percent. Here is how I did it, please correct me if I made a mistake.

The actual battery voltage was 24.18v. My tailcap current was 6.1A. My emitter current is 2.8xA.

I did 24.1x6.1 and got 147.01 that times .76 gave me 111.7 then I divide by vf which is 39.6v. It gave my emitter current that it is set at 2.8x.

Im probably doing it the hard way. Let me know if I did anything wrong.

I think your calculations are mathematically correct as I got the same numbers although I did it a little different.

xml2 vf is 3.4V at 2.8 A according to the tests done by match.

So: 3.4v*12*2.8A=114.24 Watts

You measured 24.18 volts and 6.1 amp at the tail

So 24.18*6.1=147.5W coming from the batteries.

Now 114.2/147.5=.774 or 77% Efficiency same as what you got.

Where I think your problem is with the 24.18 Volts. Is that measured at rest or under load? Under load the voltage would sag and be a lot lower, and when combined with the 6.1A under load the watts into the driver would actually be lower. Raising the efficiency numbers.

I could be wrong though. It’s happened once before.

The batteries were at rest. Good to know you got the same results! Haha! I might have to turn down the current to the emitters if I want to beable to use this light reasonably. Or add more batteries… :smiley:

I like it J) and since you’ve not got it too hot to hold yet, I’d say your still well on the safeside.

The 7*t6 I tested a while back levelled off at 60ºc at the fin parallel with the emitter shelf, that was uncomfortable to touch, but gives you a ball park figure, I’ve just started to order the bits to mod that one into what I would consider a usefull configuration.

I’d say get some temp tests done with an accurate thermometer and keep an eye on cell temperatures considering your configuration, and if you wanted to run the h6flex, just assemble a low resistance dummy cell with a little copper tube and two end caps. it’d certainly make testing the driver simple as you wouldn’t be doing anything irreversible, just fitting the new driver.

Also as you say, it would give more usefull modes - low/high/omfg springs to mind :bigsmile: 8)

I don’t think you need to turn down the current just yet. I’d measure the voltage going into the driver with the light turned on. Then redo your calculations for efficiency I bet it will be much higher.

Okay will do.

Im using the Panasonic NCR18650B batteries. I would rather use a good protected battery that will do the 6-7A I need. Any suggestions which batteries to get? With a button top. I’m having to use neridium magnets to connect my batteries. Im sure that dont help with the added resistance.


Panasonic ncr18650pd cells are supposed to be pretty good but the 7a draw is a killer, most protection circuits wont go that high. enerpower cells are the only ones I’m aware of and they wont be cheap to ship to you

What about a li-mn 18650? Is there a protected li-mn?

I thought the PD’s were good for a 10a draw?

I’ve not pushed mine that far to test, there are though, precious few 10a protection circuits, regardless of what the cells will do, I also dont think hkj was impressed at the performance at 10a.

li nm cells are supposedly safer as in dont get so excited, so dont tend to be protected in the first place, that maybe the way forwards, but li nm does not automatically equate to high draw, check hkj’s comparitor for whats what, the Samsung/sanyo (cant remember which) 20r’s have the lowest internal resistance, but similarly have a low capacity.

As we say in northern England, you dont get owt f’ nowt

Panasonic PD cells are good for 10A draw.

PCB triggers at 11A:

PCB triggers at 9A:

Those are identical to the Keeppower ones:

NCR18650B is not made for such high currents. (I said that in post #21 already..)

Some Samsung 20R’s are likely your best bet. Great for high currents. Won’t get hot at 7 amps.

Available on Ebay.