Dead XM-L- how can I check the driver?-IT"S ALIVE!

[jabe1]

:thinking: Somehow, I killed it.

About 2 weeks ago I finished a build I'd been accumulating parts for. XM-L neutral on a 14mm star, shiningbeam 2.8A driver, in a cutdown Mag C (1x 26650), Mag rebel reflector, with a UCL lense.

It was awesome! Beautiful tint, approx 600-700 lumens OTF, large hotspot with a minimal donut.

The donut effect was bothering me, so I trimmed the PCB to get better focus, which it did.

While using it on a night dog walk, it binked twice and shut down. Fine, I couldn't remember whether or not the driver has low batt. warning, but I used my backup for the balance of our walk.

Getting home, I replaced the battery with a fresh one (Sony IMR18650, until my 26650s get here). clicked the switch, and got the same double flash/shutdown.
I decided to Ohm test all of my wiring... OK.

Next, I checked V at the emitter and it registered 4.0v (?). I then desoldered the emitter, and gave it direct voltage from a protected 16340. Nada.

Everything I know to check is OK, I even checked for continuity from the emitter to the PCB. I cant see any burn marks or any problem areas under the emitter dome either (using a magnifier). The driver board seems clean too, although besides faulty connections or burn marks, much of the electronics is virtually like magic to me.

What gives? Can I trust the driver? How can I test it without risking another XM-L?

I'm determined to rebuild this, as it was a fantastic light.

 

[jabe1]

Re: Dead XM-L

Nobody?

Please read policy #7.
Norm

 

[hanachan]

Re: Dead XM-L

recently my Cree XML from KD died during its lighting test.
dead XML?
if the XML emitted light by direct drive, your XML is still alive.
simply replace the driver with new one.

 

[old4570]

Re: Dead XM-L

Yeah , I usually use a AA x 2 holder [ 3v ] for testing the emitter , if it lights up its alive .

 

[Al Combs]

Re: Dead XM-L- how can I check the driver?

If you made a direct connection with a 16340 after taking the LED out of the circuit and got nothing, it does sound like a dead XM-L. Did you notice it turning blue before it died? Rebel reflectors sit deeper than an H22A or something like it would allow. What sort of heatsink are you using? BTW, like old4570 said, a pair of AA's or a CR123A is a safer test than a 16340. A 16340 has too much voltage.

As far as trusting the driver, was it wired when you got it? If not, how did you wire it up? I remember a case where someone managed to wire a 7135 driver up so it was actually direct drive, bypassing the regulator. Does your meter have a 10 amp scale? If the wiring is OK, turn the light on by using an ammeter at the tailcap. Linear regulators have the same current input as current output. If it reads ≈2.8 amps on high at the tailcap, the regulator is OK.

It does sound like a nice combination for a light.

 

[bshanahan14rulz]

Re: Dead XM-L- how can I check the driver?

Get some black barrel diodes and put a few in series so that their Vf gets close to that of an XM-L at whatever current you suspect the driver is putting out. Then, you end up with a beefy diode with a hefty Vf that doesn't emit light, and is relatively cheap/free

 

[jabe1]

Re: Dead XM-L- how can I check the driver?

Sorry Norm. Overtired and upset about the demise of this light.

Thanks for all the input!

Thanks guys, I did initially try a cr123 cell to test, but decided I would prefer a protection circuit with a LED capable of pulling high amps...I hadn't thought of a few simple alkalines.

Al, it is a H22 flat top with the lip ground off, so it can sit deep into the tube.
Yes, the driver was wired, but I replaced the puny wires provided withe some of a heavier guage. I'm pretty good with my crappy old iron, and I test for continuity after I'm done to ensure no crossed circuits.
This is a three mode driver and was acting properly. I'm afraid it's cooked also, and without a functioning LED for resistance, I'm hesitant to fire it up.

Along those lines, Bshanahan, that's a great idea, and something I have been wondering about as a possible test solution.

I think the bottom line is I'll start over. I wish I could figure out what let go first though.I'd rather not nuke a hard to obtain LED again!

 

[jabe1]

Re: Dead XM-L- how can I check the driver?

It lives again!
I ordered a few XM-L neutrals from Digikey, who ships cheaply and fast!
New driver from Shiningbeam, purchased a cool white (1C) from Cssih, solely for the 12.8mm board.

A couple hours later, some reflow work, and it's working again. The tint of the T5 is not as nice as the last one, but it is a bit brighter. The 12.8mm board is thinner than the previous sanded down 14mm, so i've got a little focus fine tuning to do. I may need to shim it.

These XM-Ls are beasts at 2A or better!