I had two Q5s and one regular CE at home before Christmas.
One of the Q5s was subjectively somewhat weaker than the CE whereas the other was roughly equal subjectively in output. (The Q5s were warmer in tint.) When put in a lightbox, the weaker one showed just lower output than the CE, and the other some 15 % higher output, where the bin differences should give a 30% improvement on average. All of the above was with L2D body and fresh and tested eneloops.
I didn't know quite what to do with it - the sample size was too small. Maybe I had a really good P4? But I suspect that Wade is on the right track, that we need to consider the Vf of the individual LED.
I proceeded to measure the current draw for the two Q5 and the P4 heads. The weaker Q5 head drew less current from the batteries, and when subjected to a runtime test, this was confirmed by correspondingly longer runtimes.
Could it be that the driver circuit of the L1/P2/L2 strives to output a given current, and is designed to achieve this within a certain window of Vf voltages? And, when confronted with a LED with high Vf, output is limited by the achievable voltage, rather than target current?
Cree doesn't bin by Vf or efficiency, and if the above situation is common, it really creates problems for manufacturers and retailers alike - the output you get is luck-of-the-draw. Fenix can't know without testing what the Vf will be of a given LED, thus they can't say whether the output of their "premium" light, which both cost more to produce and buy, will actually yield higher output than their vanilla variation. And of course this problem, unless dealt with, is dumped in the laps of the retailers who have the direct customer contact.
I asked the Fenix-store, where I had bought my lights, if my conjecture above was correct. Of course, they didn't answer.
(This is still assuming that everyone is honest in their dealing - that the lower performing parts isn't because someone in the long Cree-distributor-manufacturer-retailer chain decides to make an extra buck by fudging the bins. I've seen a lot of accusations fly around here regarding reputable vs non-reputable, chinese vs. american, and so on, because people get less than they expected and proceed to point the finger at someone.)
I had planned to buy a couple of L2D-Rebel100s to use as bike lights, due to their generally lower Vf and thus AA friendly nature. Now, with the recent recall, I just don't know what to do.
