Agreed, every little bit of resistance adds up and with this much current, could have a significant impact. With this high of a power draw, running the cells in series instead of parallel would probably give you better performance, but then you give up on safe internal charging and it generally makes things like driver design more complex.
Well I tested the EA01 when it came out with the xhp50.2 6500k and came put lower than other reviewers. I saw max like 3600 or 3700 lumens and others were consistently over 4000 on a 30T. I tried different batteries, even different luxmeters and still couldn’t crack 3700 Lumens. I was still over the 3500 advertised though. Just goes to show that you might get a good FET, solid solder joints, a good bin emitter, properly applied thermal paste, aligned optic/reflector, etc, etc. All those can affect output to varying degrees. At the end of the day these are low-cost lights that don’t benefit from exhaustive factory testing like more expensive brands and you get the best cheapest of whatever Mateminco was able to get off the Chinese marketplace that day into the parts bin for that production run. Sometimes you win!
Indeed.
Let’s make our calculations at 100lm/W efficiency, right?
20000 lumens / 100lm/W = 200W.
200W/3.6V = 55A.
That would equate to 18A per cell.
With that much current, everything becomes a bottleneck:
Springs, contacts, board LED wires, MCPCB trace thickness, etc.
Say ChibiM, could you check if the springs are magnetic, just in case?
I’m curious to know…
That burnt spring suspiciously looks silver with the gold plating burnt…
Also, @Sirstinky, the difference is that you still got over the rated brightness 
This one was wildly under the rating 
Now I understand why our own BLF members were so tight about production with the BLF Q8
I did a spring swap on a Ft02S a while back. They share that tailcap spring with the EA01 and FT03 I think also. Looks similar to other Mateminco products also…the outer, taller spring is not magnetic, copper alloy, the inner, smaller spring is magnetic (steel). I removed the inner spring and used a solder wick. I dislike that design somewhat because the spring pcb is really glued onto the tailcap and makes soldering difficult.
The Lumintop D3 I tested has a copper alloy tailspring
Bruh, they took the risk of not using phosphor bronze/BeCu on every spring lmao.
No wonder: at 18A, only a dual phosphor bronze/BeCu spring would allow for good performance, unless all of the coils were to touch each other.
I just checked with a strong neodymium. No signs of attraction to either the outer or inner springs. I’d say no steel present.
Thank you very much for jumping in gchart. That's really helpful
Bummer on the MF01X. I am glad I did not buy one, but I really wanted to. Hopefully they will fix.
I hope they do!
Ooof… that Speras brand is flying a little too close to the sun with that price.
Speras are expensive lights, and I mentioned the L21B because that light does everything performance-wise (and more so) as the T3R, but at over 1/2 the price. Same buck driver, 21700, nearly identical throw, etc. If you dont need the Speras added features, I see no reason to get it over the Convoy.
Yeah, but let’s not forget they have 5 years of warranty.
That’s a bit expensive on their side.
hi
why 3.6? as far as i know should be 3.7v for standart li-ion
not that it would make any significant difference in overall calculation. but i`m always ready to learn and chance are that mentionin 3.6v you know something that i dont know
--
to TS. like conclustion about Speras. that is brave to compare it with another brand model in overall. usually manufacturers are damn unhappy when reviewer mentions another similar model.
Hi ChibiM
I really enjoyed the Gearlight s1000 review.
Looks like the only thing missing from the beam shot comparison, is the MicroFire Falcon H8 
That MicroFire would have burned a hole in the wall ;)
2022-04-13
Nebo Slyde King
work light
https://1lumen.com/review/nebo-slyde-king-2k/
Nebo 12K
12,000 lumen powerhouse
https://1lumen.com/review/nebo-12k/
Gearlight S1000
best sold Amazon light?
Useful reviews and congrats on reaching 350!
Just had a flick through the ‘best sold Amazon light’ and saw this comment under the PWM section:
” I can’t quantify Hz figures, but using the Opple Lightmaster Pro’s “flicker” function, it gives an idea of the severity of the PWM duty-cycle in Low and Medium modes”.
It’s easy enough to work out the Hz from the images you provided, for med mode there’s roughly 13 pulses in 24 milliseconds which is (13/24)*1000= 541 in a second, so 541Hz.
Low mode is about 6 in 23ms which is about 260Hz.
I’ve had toooo many conversations with people about how amazing this kind of light is, which is fine as i get it’s a lot better than what they’re used to, but also toooo many times i’ve hit a wall in trying to explain they’re not as good as the marketing suggests so the actual measured figures will come in handy. Well, assuming i can get past the ‘but it lights up the side of a mountain’ block… 
Thanks for explaining how we can calculate the Hz from this. That's pretty useful, thanks.
I know what you mean. At least people should listen a bit to any enthusiast to learn a bit more. That's the great thing about talking to an enthusiast.. but some people don't want to listen :D
And if they listen, you can explain the differences with higher-end lights, and how they perform vs the Amazon ones.
I hope our measurements can be of help