Chinese OTF lumens??
Sorry for doubting your numbers, but that sounded awfully high..
Nope it just sticks in high.
I checked it three times and as my other numbers are coming in right on it seems pretty close to me. My X8 with a single MT-G2 is showing 2,000 and my SRK is at 2500 and it’s the original Fandy Fire one. Fandy Fire 5x is 2,000 C-12 XM-L2 from lightmall is 1,000.
Ok, thanks for clarifying. Your numbers may be correct though.. Maybe Ive just read too many conservative numbers lately, or... Who knows.. I don't even know how that driver behaves in that configuration.
If you one day manage to get some decent amp readings, it would be interesting to see what your driver outputs to the emitters.
I might have been off on the one I took after the walk. It seems pretty consistent now at 3500.
There's two of them, so sounds about right.
Well a DRY driver works fine to drive two MT-G2’s in parallel with two king kongs. One MT-G2 with a DRY and very heavy wires is putting out close to 2,000 lumens and two with a DRY is close to 3,500. Pretty nice actually. Now I’m tempted to go for three again but it would drain the batteries pretty fast I would think. I wish they had a better 32650 because then I would try to put them in the J-19 and run it with two batteries to three emitters.
Also I would like to report that I did go for it and popped three 18650’s in with the two MT-G2’s and it didn’t kill the driver or the emitters but it would have if I left it on.
Also I’m working the bugs out on my lumen tester so take them with a grain of salt but I think they are pretty close. I get 1,000 lumens with an XM-L2 C-12 and 2500 with my Fandy Fire SRK, it’s one of the first ones.
Is that on two King Kongs as well?
-Garry
No, it’s two sanyo 3400’s protected.
You’ve measured that output, what about the amperage that’s delivering those numbers? I’d like to see how my K3 compares. djozz’s chart shows I’m around 2330 lumens at 4.05A on 2 18650 PDs.
My Multi meter stinks. But for what it’s worth I get just over 3A on it and about 2.7 on the XM-L2 C12 that is supposed to be 4a from lightmalls. So it consistently measures low.
There are a couple people around that are running the same basic setup. It’s an X8 direct drive with 20 gauge wire. I’m sure one of them at least can get you a better measurement.
Ahh, ok…I had to resort to using 2” long pieces of 12ga solid wire from Romex to take my amp readings, my leads gave ultra low readings as well. These short Romex leads give readings that pretty much match the numbers from folks with better meters.
Direct drive on 2 Panasonic PDs showed about 6.7A, on 2 Samsung 20Rs it was up well over 8A! The 4.05A I’m getting now is from the modified Q-Lite driver with 4 chips stacked, the resistor swap and a Zener stacked. I love the hi output but also having a pretty low lo and a respectable (some 700 lumen) med are great benefits! Over discharge protection is also a great benefit from this driver, but I don’t know if it’ll still have it with this mod. I believe I’ve been told that feature will be lost.
I always take numbers with a grain of salt. My own numbers included. :p
Most seems to get 2100-2200 lumen on a original SRK. Some who are more conservative, (like Texaspyro) seem to generally be a few hundred below that.
Unless you know yours are pushing way more amps into the emitters than the others, I would try to adjust the numbers so that they are more consistent with others. The less guesswork you do the better... And if you do guesswork, I would say its better to try and stay conservative. I hope you one day are able to get some decent amp readings on top of your lumen numbers.
Good luck on working out the bugs on your lumen tester.
I hear you but I’m pretty sure they are close. My fandy fire SRK is one of the first ones and it turns on at 2500, after that the reading goes quickly down. The MT-G2 (X8) stays at 2,000 and that is also consistent with what others have gotten from the same set up.
Well anyway, this will probably be the last time I post numbers. I stayed away from it for a long time for just this reason. It’s also why I doubt I will bother posting amp readings, it becomes the same argument again. How do you take the reading is the spring compressed more or less, should you take it while wired or at the emitter or while on etc etc etc.
As far as I can tell everyone is making their best guess.
I guess we could try looking at the emitter on high, then timing how long it takes for our vision to return. 
That’s worth a try, :bigsmile:
The reason you lose low voltage warning is the two resistors that the mcu uses to “read” the voltage take their input from the power + pin of the mcu. The Zener fixes that voltage so that it is a constant that doesn’t drop until the combined. Voltage of the two cells dips below the Zener voltage. The resistor values are wrong for two cells and I think you would also have to use a 6V Zener to get close to reading the cells at even 3V each but I could be way off on this.
The over-discharge protection isn’t lost, just effectively disabled by the Zener. I haven’t had too much trouble taking cells down too far in the past, but it is always a good idea to keep a mental check on runtime and amperage level used as a safety margin to help prevent fatal error. Fatal to the cells is the most likely scenario, but potentially fatal in other ways as well.
Pulled cells from my K3 and checked em when I finished my last “adjustments” and found them at around 3.8V low for me, lol! The more lights I get the harder it is to keep up with the cells. My memory is horrible! The worst part of that is that I might remember specific things but give those attributes to the wrong light. And then would swear by it, bet on it, get totally blown away that something isn’t so, when I just swapped “memories”
Sucks getting old! Should’ve never quit drinking…
“I Picked a bad day to quit ( ).”
Just watch the output and be watchful for dimming. Low mode is the most dangerous to cell with this mod and why I never actually used it in a light. I wanted to make a separate battery guage pcb on an 18mm board but didn’t get enough eagle under my belt to pull it off. I figured out a board layout but not the eagle part. I won’t need that for the Elf since it has it’s own monitoring system.
Voltage protection still 'works' after the zener/resistor, in that the MCU is still looking for it, but it's being fed an invalid voltage reading. I guess since the other resistors it uses for that aren't altered it would still trip at 3 volts, but that would be with each cell down to 1.5v (once input voltage falls below the zener's rating actual battery voltage is fed to the MCU). Best to use protected cells, pretty much for the same reason protected cells are best for any series-multi-cell light. Not always convenient to do that, though.
Taking the cells down to ~2.75v each is I guess technically 'safe', but not best for battery life. After all most cells have their advertised capacity measured when taken down to 2.75v, I think some are even rated down to 2.5v. And at 2.75v each the light output will be noticeably low, it's pretty easy to spot.