cajampa, I think it’d be pretty easy to use a couple of extra copper mcpcb’s or a disc if you have those available. A 16mm in the pocket, then stack 20’s on top of that to get the right height…probably 2 without actually measuring it out. Then there’d be more copper under the triple to sponge up the heat.
Realistically, you’re not gonna run a triple very long in this light as it will get hot fast. But the small tube shape makes it very easy to carry and a triple would be a huge flood of light from this host that overall wouldn’t be/isn’t much larger than the cell that powers it. I’ve got one of the Convoy tube lights that ComfyChair modded for Richard and it’s a potent little number to be sure! I like that it slips easily into a camera bag or almost anywhere.
The XP-G2’s will get hot as well as they’ll be more overdriven. Remember that they’re rated at 1.5A. A good triple with XP-G2’s will easily pull much more than 4.5A, typically 9A or more. So the heat is still there from the overdrive thing.
Drive 3 XP-L at 9A and you’re within specs, making less heat and far more lumens.
I guess i will have to wait for more nice tint emitters to show up in XP-L (we snagged up the last few available from Richard but those i have already plans for), and maybe take it easy with the low resistance mods, maybe even do some high resistance mods to keep the amp in check if needed on the triple XP-L.
Ok, a more thorough exam… Krono wanted me to go ahead and bypass the tail pcb for him, so I put a 22 ga wire through the board to help keep it reliable and solid. Same 22 ga on the top end. They’ve already got 22 ga on the driver as it turns out, so I left that alone.
So what I’ve done to Krono’s light is put a copper Noctigon under the factory emitter, and add 2 little short pieces of 22 ga wire for the current to run through (the bypasses). Here’s the results…
0.01A at 00.93 lumens
0.03A at 13.97 lumens
0.15A at 62.45 lumens
0.38A at 191.48 lumens
1.48A at 439.19 lumens
2.96A at 852.15 lumens
5.65A at 1490.4 lumens with a 30 second run showing 1418 lumens.
I started with amp readings, then measured the lower outputs first…. so the cell wasn’t fresh full hot for the max Turbo readings. So these numbers are pretty real world like you’d see in normal use. The 30 second run in the light box had the head pretty warm, almost hot, but still it didn’t drop all that much, what, 72 lumens? So not shabby at all!
With 5 of these samples in hand, I am seeing some different numbers on the mosfets and even one completely different set-up (which we knew about) The oddball one uses 2 smaller mosfets and while it pulls more current it doesn’t make as many lumens…. not as efficient so it won’t be our choice at all. There is some variance between the different samples, but it’s slight and really insignificant. Best as I can tell, with limited actual run time clocked, these look pretty much good to go. Get the copper in there on the top and call it the best group buy ever.
Dang! I’m gonna have to be the official poster boy for Eagle Eye! lol
I LOVE their little lights! With 8 hot rodded X6’s ranging from 1800 lumens (single XM-L2) to 5500 lumens (single XHP-70) and now this little beastie, I have far more Eagle Eye lights than any other brand. Worthy, if you ask me. And I have 2 more X6’s to build to complete my top 10 EDC collection.
[qote=DBCstm]Ok, a more thorough exam... Krono wanted me to go ahead and bypass the tail pcb for him, so I put a 22 ga wire through the board to help keep it reliable and solid. Same 22 ga on the top end. They've already got 22 ga on the driver as it turns out, so I left that alone. So what I've done to Krono's light is put a copper Noctigon under the factory emitter, and add 2 little short pieces of 22 ga wire for the current to run through (the bypasses). Here's the results... 0.01A at 00.93 lumens 0.03A at 13.97 lumens 0.15A at 62.45 lumens 0.38A at 191.48 lumens 1.48A at 439.19 lumens 2.96A at 852.15 lumens 5.65A at 1490.4 lumens with a 30 second run showing 1418 lumens. I started with amp readings, then measured the lower outputs first.... so the cell wasn't fresh full hot for the max Turbo readings. So these numbers are pretty real world like you'd see in normal use. The 30 second run in the light box had the head pretty warm, almost hot, but still it didn't drop all that much, what, 72 lumens? So not shabby at all! :) With 5 of these samples in hand, I am seeing some different numbers on the mosfets and even one completely different set-up (which we knew about) The oddball one uses 2 smaller mosfets and while it pulls more current it doesn't make as many lumens.... not as efficient so it won't be our choice at all. There is some variance between the different samples, but it's slight and really insignificant. Best as I can tell, with limited actual run time clocked, these look pretty much good to go. Get the copper in there on the top and call it the best group buy ever. ;)
[/quote]
from those 7 modes listed above, looks like they are using a 380 mA 7135, not a 350 mA - think that's correct? Actually it's a pretty good idea - I should build my DD+1's that way from now on.
Since you think you got a neutral tint, maybe 3C, do you see the tint shift going from mode 4 (.38A) to mode 5?
since you are hitting close to 6A, suppose they are not using those "new" lower max amp XM-L2's. Every XM-L2 T6 4C I've gotten seems to go to 6.5A with no problem.
I just built a hot setup DD+1 ZY-T11 clone (UCL lens, 20 AWG LED wires, + direct from cell to LED, 20 AWG bypass's, etc.) with RMM's XM-L2 U4 1C and it max's out at about 5.12A, though still delivers about 1,700 lumens @30 secs. Wondering do you think a CW in this A6 will also have the limited high amps? Have you tested a CW? They may still be using U2 1A's, and if so, may still be high amp capable.
Looks like the timing works out so I’ll be able to test and update it as soon as I receive it.
DBCstm, thanks for the levels for each mode. It looks like the middle modes may be running a tad bit off spec, but that’s easy enough to recalibrate. The visual difference (cube-root model) between each of the modes ranges from 1.43 to 1.94 instead of being a constant value for each, so I’ll see what I can do to adjust that.
Only things different is the resistors "01C" labeled, if I have that correct (Dale mentioned the FET is different as well). They appear to be connected to the FET and 7135 inputs - tied to ground. Probably to stabilize the signal transitions. I know with my DD+1 driver builds using the SIR800 FET, I get little blinks on transitioning between the 7135 and FET. It also occurs on the DD+1's when just turning an e-switch light ON where it goes to moon initially - you see an intermittently occurring brighter flash for a sec before moon kicks in. Doesn't happen every time. I tried a 100K resistor on the FET input and it helped a little - probably needs a lower resistor...
Just looked it up - a 01C is a 10K, so 10K probably eliminates the flicker. These Eagle Eye guys seem to know their driver designs!
Great Job One and All!!
Ohh - I'm not in on this GB because I'm not a fan of this tube light styling, and can build my own DD+1's anyway, but very interested in how this converison of one of our own driver/firmware combos gets to manufactured in China -- hope this works out, I'm cheering on the sidelines...
thanks Dale.
