I read on reddit that there are some new flashlights from Fireflies coming soon
https://www.reddit.com/r/flashlight/comments/w68nzn/fireflies_from_the_future_the_filth_is_reeeal/
It is not official and it seems like there is 2 LEP lights, and one new 'E12C'. Personally I am a little disappointed that E12C will have a linear driver. But maybe it could be an interesting flashlight with Nicha 519 leds.
This additional info was posted today.
E12C project is going all good
-E12C driver has been upgraded to FET+12.
Possibly i can share some photos next week.
Full ramp 1/255 of single AMC7135 to 100% FET Turbo Moon light to 10 000lumens
-Default smooth ramp = minimum to ~3000 lumens
-Default stepped ramp = 5 steps, ~5 lumens to 3000
lumens
-Simple Ul stepped ramp = 5 steps, ~5 lumens to 2100 lumens (maximum regulated)
Nice Dooski3. Came to do the same thing but I see you’re already on it. Ya love to see it. :+1:
The E07 driver is 7x7135 right ? Adding five more is hardly a driver upgrade, and more risk of a 7135 failure.
If they won’t go boost or buck+FET they could design a custom linear driver like the noctigon one, more reliable, a bit more efficient and with the quantity of 7135 it probably won’t cost more.
I’m not gonna act like I know how drivers work. My understanding is that it can sustain 4.2A and in turn, 2000 Lumen. If that’s the case, that seems like an upgrade over the E07
Can a 519A do 167 Lumen on 350ma? That’s what it would need to do to sustain 2000 Lumen. That’s actually what I was gonna search today. I wanted to see what we can expect from 519A’s if each LED has it’s own 7135 chip.
Also, since I’m not familiar with how drivers tick, I don’t know what Hanks driver composition & cost consisted of and if that’s his permanent solution. But all of these are things Jack has to do because he doesn’t have the drivers he already ultimately spec’d all his lights for. His drivers will return and then he can just plug them right back in. So this IMO is temporary ways to release things people will still be happy using. 2000 Lumen of sustained ultra high CRI light sure feels like a win, and should make me pretty happy. Plus it’s yet another badass looking “Fireflylite”.
I have a whole shelf of awesome Fireflies that people thought they didn’t want (me included) and later wished they bought them when they had the chance. Sounds like this could be a limited release too, so Ima be on it like stink on shit.
I don’t think it can hold 4.2A 2000 lumen for long time before stepping down because it is linear driver and it looks like it is smaller than e12r.
thefreeman, why is custom linear driver more efficienct? If both are linear type, excess power will only be burnt in heat. My guess is fireflies and chinese companies already have a lot of amc7135 in their warehouse so maybe why they decide to build a driver like this since there is a parts shortage.
excuse my ask but is the Nichia 519 led special? What is the benefit of this led over other older Nicha leds?
Using 12x7135 doesn’t mean it’ll sustain 12x350mA, rather it’s the other way around, the best quantity of 7135 to use is around sustained output, since they would be used near 100% duty cycle with no or little duty cycle of direct drive (which is less efficient). But the E12R 219b with a more efficient driver doesn’t sustain that much, about 800lm according to 1lumen, 519A efficiency is similar to 219B at low current (I never swapped them for 519As in my E12R for this reason), so something like 80 LED lumen per LEDs, less than 200mA and so 7x7135 is adequate for this. That said when using 70 CRI XP-L HIs the higher amount of 7135 will be beneficial (slightly).
Convoy has 8×7135 S2+s costing 14.63$ and 5A linear ones costing 16.03$, the price difference of 1.4$ is probably not just the driver, since the LEDs used with the latter usually cost a bit more, and it’s higher end so possibly more margins.
I like firefly lights, actually I prefer then to Emisar-Noctigon, I’m just a bit disappointed that 7135+FET drivers are still used. Regarding the shortage the Texas instrument buck IC is unavailable, but they could have done a small redesign (not the whole board) with other available converters like MP2145 (also 6A), or developed a boost driver (problematic with the 7 LEDs of the E07 though).
No PWM so no pulsing (higher RMS) current, I think ToyKeeper made an output comparison but I don’t remember where. The difference isn’t as big as with a buck or boost driver though.
The 519A is more efficient and has better current handling than 219B and 219C :
I definitely want more too. But since I don’t know what it would take driver-wise to make it happen, I’ve stuck to hearing others take on this. And I think I always hear someone say something like what you said. Something about Jacks setup makes it harder, like the 7 LED’s for instance. Also heard someone else have a great buck driver option, but it was lacking AUX capabilities. So my understanding is that there hasn’t been a straight match that can do everything too, it’s always something.
So it’s not realistic to divide 4.2A by 12? Jack is testing the light and playing with it still, so he must be seeing what’s going on and how it’s working. Hopefully he’s got it, or will get it well figured out. I’m a huge fan of his finished products too, also more than Emisar & Noctigon. So I honestly wanted this regardless, but if he can make it sustain pretty well too, I’ll probably be pretty satisfied until the other big homies return.
Hi again Freeman,
When you get a chance, would you helping me understand why the 7 LED’s complicate a boost driver, but not necessarily a buck driver? O r maybe it does too, I dont know. Would the PL09Q be an easier plugin? How does this idea apply to 12 LED? Is it the odd number of LED’s that complicate what the current does or something like that? Would the PL09Q and E12C be an easier plug-in for a better driver just off of even emitters? I don’t know what makes drivers tick. That’s likely why I don’t get why 7 leds is more complicated.
I know linear isn’t as good as boost, and boost isn’t as good as buck. But then you said that the liner driver can still produce 5a and without 7135 chips, but a linear driver can still use tons of 7135 chips, and not even crack 4a. As you can see, I don’t know shit about drivers. I want to get what Jack could have done in the meantime too. Are there reasons he wouldn’t have done it the way you suggest beside cost? Could he have to eliminate a feature or two like USB charging or auxiliaries?
I know I’m asking a lot, so I would me more than grateful for whatever you can, whenever you can. No worries at all.
Yes it’s the odd number, when using a boost driver the 3V LEDs are put in series or a combination of series/ parallel, for example 4 series (4S = 12V) or 2 series 2 parallel (2S2P = 6V).
Odd numbers work too like 3S = 9V, but higher is limited by the maximum output voltage of the boost converter used, in the case of the comonly used MP3431 it’s 16V, 5S with 519As would actually work, since they have a low Vf of ~3.1V at 2.5A so 5x3.1= 15.5V, but other LEDs with higher Vf wouldn’t, and 7S would not as well, additionally, higher output voltages are less efficient, see The measurements I’ve done of my 6V and 12V drivers for example (2 last tables).
Putting different amount of LEDs in parallel like 2S+2S+3S or 3S+4S would result in pretty bad current disbalance between the LED strings.
It would be much better to simply remove the center LED and do 2S3P.
For single input cell flashlight boost is better than buck since it can have full output regardless of the cell’s voltage voltage whereas with a buck, once the cell’s voltage is close to the LED’s Vf the output starts to drop (same with Linear but can be closer because lower driver resistance), the efficiency and output of available boost converter is also higher. The advantage of buck is that like it is done in the Fireflies buck+FET driver it can be combined with a direct drive FET channel for more output (than boost) when the cell is full.
I was talking about low~high mode and sustained (thermally limited) current where using 7135 chips are more efficient than direct drive. Both custum linear driver and 7135 linear driver can be high current.
Thank you Freeman for the detailed breakdown. Now I will go and learn what it all means exactly little by little.
I missed the release of the e12c! It is now release for preorder. https://www.reddit.com/r/flashlight/comments/xbhpve/fireflylite_e12c_groupbuy/
The most interesting thing is that it has now a new 3A charger and powerbank with usb c! It also has custom aux leds that you can customise yourself with your own color for additional cost. I think I will like to have usb c charging that do not require unscrewing the flashlight, but most likely this way is more waterproof.