A method I like to use to determine actual current in setups that I can’t get a good measurement otherwise is a two step approach and does require a power supply and a bit of time, but results are as real as you can get when currents are as high as they are with the 70.2 and triples, quads, etc.
1.) Ideally while installed in the light/host, generate a Vf vs I curve. Do this by soldering on long temporary leads to the LED connected to a PS. For most accuracy record values at multiple times, for example: turn-on, 3sec, 10sec, 30sec, and 120sec. This will result in 5 different tables/graphs.
2.) Reconnect driver and then just record your voltages. You now have a look-up table to determine your current. Remember that LED temp will play into these numbers so make sure you match it to the right table.
Besides the one in the CREE specs? No, not that I'm aware of. Sorry, no time for setting something like that up, besides not knowing WTH I'm doing to accomplish that....
This is why I bought a led lenser flashlight for EDC.
I can choose between spot and throw and not have to deal with inconveniences of both at the same time.
This probably isn’t for you though, it’s far from cheap.
If you want more suggestions, instead of discussing it in this thread just make another topic
@Jason:
*expensive does not neccessarily equal accurate, as I have found many times, and also the opposite I have often found: cheap and simple can be very accurate. Sometimes you can not get around an expensive component though
*with these high currents, a (even well bypassed) tailcap can lead to significantly lower current than a thick copper loop, although I would be surprised if that is the 5 amp difference we are looking for to explain our differences.
Maybe we should measure emitter temperature when we measure amp load and lumen output?
Maybe your emitter is getting much hotter as someone mentioned. Adding a temperature curve to the results might explain the drop in output your getting.
I forgot, did you reflow your 70.2 onto the mcpcb yourself? If not, maybe it wasn’t done well and there is too much solder paste in there with the emitter floating on top? Was this already discussed?
I reflowed it myself, and re-did it for the second set of measurements, I believe I did it well, there was enough solder and not a huge excess, and earlier tests did not find a measurable effect of thicker/thinner layers of solder between led and PCB.
Measuring MCPCB temperature would be a very nice addition to my test set-up, but it would add to the work-load of doing emitter tests. In practice that will lead to less emitter tests because I’m limited in hobby time and also the extra work will raise my treshold for doing tests, which is already higher than when I started doing them
Djozz, you are measuring the current at the led? Right?
The others at the tailcap before a driver?
Maybe that can explain something.
With these high currents weird things can happen.
The best dog walking light will vary by person but there are a lot of options. At the top of my list would be an L6 / S70s if you want something you can knock someone over the head with if needed. The Q8 is also a great option for this.
In smaller packages an X6 or X7 triple makes for a great overall light with some good output.
Just updated my MT07 with a XHP70.2 (same P2 3B), but from KD. I did make resistor mods, but had somewhat disappointing results with updated to a XHP70 P2 1C - now I'm not even sure it is what I thought it was....
Significant gains. The boost in throw is very encouraging. The MT07 has an OP, but it's a big head for sure at 86 mm, and 72.5 mm I.D. Sorry, didn't measure amps.