What's the downside of 12a linear+fet vs 9a linear+fet

LED Flashlights – General Info
21

This is an old diagram, but change a few of the labels and the same idea still applies…

On one of the early FET+1 style lights, the ramp was calculated to exactly match up with the bottom channel’s ~130 lm regulated output plus the top channel’s ~1500 lm direct-drive output. The result ends up looking visually linear.

Take that same driver though, and give it 4X as many LEDs, and the relative brightnesses are suddenly 130 lm and 4000 lm. The ratio changed. So with no changes to the driver or firmware, suddenly the ramp is the wrong shape. Instead of being linear, it has an “elbow”.

Or replace the LED with something much weaker, and the ratio changes in the opposite direction… like 100 lm and 300 lm. So it gets an inverted elbow.

For Hank’s linear+DDFET lights, it’s kind of a similar situation. But in this case, the ramp is calculated to be linear with 5A on the bottom and maybe 15A on top. When you change it to be 12A bottom and 15A top, the ramp shape then looks like the picture on the right. It ramps up faster until it hits the channel boundary, and then suddenly the rest of the ramp increases slowly instead.

This can be fixed with firmware modifications, of course. If the 5A limit hits at ramp step 120/150, perhaps the 12A limit should hit at 130/150 or even 140/150. The exact spot is generally determined by measuring output of each channel and applying a curve to see where things land, then usually trying to adjust it until things line up with a round number.

But that type of fine-tuning ideally would be done for each emitter type on each driver type on each host type… and that’s not feasible. So the firmware is built as more of a “one size fits all” or at least “one size fits many”.

Anyway, that’s only one aspect of all this. There are other factors too… like the driver details tend to affect the moon level in ways firmware can’t fix. And it changes the thermal characteristics and regulation pattern, since it moves a few Watts of heat to a different part of the light.

The impact is likely not huge, but also likely not a thing you’ll be able to find hard data about… because there are so many different hardware combinations, and so few measurements and reviews. So the exact consequences of the hardware change are probably unknown, and the best anyone can do is make educated guesses.

5 Thanks
22

The only part scaring me (at the moment) is removing the press fit switch bezel. I’ve got another metal switch bezel coming with the switch as I can see the easiest way to remove it will be cutting the bezel. I don’t want to mark the flashlight body- of course.

I don’t know- I’m only saying what I was told regarding my D4SV2. (Note: If it’s an RGB backlit switch it can be hooked up in parallel with the RGB aux lights on the 12A driver.)

Somehow he misread my Email request. :slightly_frowning_face:

His response was (to me questioning retrofitting a backlit switch myself):

I’m afraid it’s not possible since both the hardware and firmware are diffeent.

23

Ahh thanks thats perfect. That clears things up, appreciate it TK.

Strange coincidence but ive also been emailing Hank about a lighted switch d4sv2 single channel w2 recently lmao. So your experience is concerning. It wouldnt make it or break it for me but I thought I’d double check just in case.

He wrote back an hour ago “the one with the lighted switch is with the 12a+fet driver”.

I did ask if RBG made any difference in an earlier email. He just calls them all lighted switch and doesn’t differentiate. I assumed that meant they would all be the same.
So ya idk

24

He was adamant no lighted switch could be fitted initially- until I questioned the RGB run in parallel with the aux.

I was disappointed when it arrived with no backlit switch after requesting it. He said I didn’t mention any light colour for the switch so he thought I wanted the black switch. At that stage I didn’t know you could choose a colour and figured it would just turn up with RGB (when I later found the Noctigon site with the option listed.) I wrote "I was wondering if it’s possible to get the backlit on/off switch fitted in place of the standard black switch please? " He replied “Yes, I’ll add it to the order”, or similar.

It’s costing US$13 in parts (delivered) and it’s not going to be an easy task to change it. But it’s bugging me as the lighted switch would be so much better with how I use it.

25

Hm now I’m debating whether its worth even asking for a backlit switch on the chance there’s some issue. Do the new backlight ones have that firmware issue still?

Maybe Ill just get the tint ramping one, comes with the switch and right firmware off the bat. But I think I’d still just want 4 W2s… Seems pointless… Unless, do I get two 9a drivers that way? Thats something

26

Sounds like what you’re looking for is a regular D4Sv2 w/ W2 LEDs, a linear + DDFET driver, and a lighted switch. The switch, if I understand correctly, is a single color… so RGB is used only for the aux LEDs up front.

The difference between 9A and 12A on the linear channel … probably doesn’t matter. It might be possible to tell the difference with 2 otherwise-identical models side by side being used simultaneously, but I doubt anyone would notice the difference otherwise.

I’d pick one of the brighter colors for the switch, like cool white or cyan or green. I find it very effective as an extra-low moon mode with several years of runtime. My D4K uses just 0.09 mA in standby, but is bright enough for me to see things up to 3 meters away. I’m pretty photosensitive though, so YMMV.

2 Thanks
27

The 4 x W2’s are impressive. Even with the narrow (stock) optic you get some useful spill. I do need to get another 26650 as I’m using a protected Klarus one at the moment (came with my G20L.) It works fine even on turbo, but an unprotected would probably be even better.

I just found this thread on Reddit https://www.reddit.com/r/Hanklights/comments/107vgtk/anyway_to_get_a_lighted_switch_for_the_d4sv2/

The d4sv2 w2 option is my best seller my stock gas the new 12A driver and lighted switch the D4k if you don’t see something in stock or available let me know I can always swap the emitters, I stock all color temps

(“gas” = “has”?)

I don’t know what Hank was telling me - if that’s the case? :thinking:

28

You’re using a protected battery with that? Ya definitely get a new one.
Too bad there hasn’t been many high drain 26650s for sale lately for whatever reason. I need some too.
Liionwholesale has literally none…18650batterystore has literally none…theres some Efest 4200mahs here and there, I don’t buy Efests, I’ve never gotten a good one. Maybe that’s why theyre available.

Imrbatteriess has a 3500mah ‘MXJO’. Maybe an option, but idk anything about mxjo, and idk if that’s worth paying for over just using a p28a or another high drain 18650 with an adapter. Ilumn has some keeppowers, not really high drain, and that 6200mah vapcell that I’m kinda suspicious of, no way that’s high drain.

So I ended up just ordering some 26800 QBs, that I’m also kinda suspicious of. But I wanna see what the hype is about. Plus the d4s looks better with the larger tube.

Funny I’ve got 2 Hank lights in the mail with a cyan switch and a green switch. My others are RBG. I could get into cool white. It sounds boring, a white backlight, but it looks good in pictures.

29

It’s hopeless here in Australia. The place I normally buy batteries from hasn’t got any 26650’s, decent brands are hard to find anywhere. I’ve just ordered a 26650 Golisi S43 - that at least has got decent reviews in the past Test/review of Golisi IMR26650 4300mAh (Gold) .

30

Ya, what I should do whenever I feel bad about the cost of getting stuff to Canada is remind myself that Australia exists. That’s rough.

Australia is the only country where theyll spend 30 grand to ship a 6-figure classic Bentley/Mercedes/Rolls across the world on the off chance that a gorgeous $200k classic Bentley will distract customs just long enough that they might not notice the drugs hidden in the tires. Because Australians are so accustomed to paying extra for imported goods and it’s so hard to get stuff past customs that the gangs can afford to lose a couple Ferraris full of coke at the border cuz theyll still make more money selling the stuff in that one Bentley that made it thru than they would have just selling it all in the US in ziploc bags.

That’s wild.

31

Essential the Emisar 4 W2’s in question…

IMG_2855
IMG_28551280×720 164 KB

Not quite able to reach the little barn at 610 yards.

This is actually an X6 but it’s utilizing Simon’s quad MCPCB and spot optic, with W2’s.

2 Thanks
32

Hmm. Not the best throw, but looks like a nice flood/throw middle ground. I do like those kind lights.

Oh I got a question for you. If I wanted to put a 26800 in a 26650 light, and an adapter doesn’t exist, would I try to modify the stock tube to use a 26800 extension adapter, or would I just put the stock tube aside and make a whole new one?

If I didn’t care about how it looked and just wanted something functional. If you need specifics I’m thinking of a catapult pro (no tailcap, body tube and ‘tailcap’ are a solid piece of ano aluminum, threads going into the head are anodized, eswitch in the head).

Probably just make a new one eh?

33

Simon puts a brass adapter on the end of the tube to accept the longer cell. That works, either way you have to cut threads both ends. Easier to dress the adapter. I guess knurling could make a new tube work, I’ve never had much luck with the knurling process though.

34

Mateminco does the same thing, puts it at the end of the tube. Everyone puts it at the end.

That’s the only thing I’m concerned about. No tail cap on the catapult pro, so mine would have to be at the other end of the tube up against the driver…but thats fine, right? I can’t think of any reason it wouldn’t be fine. They all put it at the other end, why, just cheaper? They probably fit more models that way. Maybe keep the better heatsink closer to the heat I guess. Still should be fine to put the brass extension tube up top eh?

Can you think of any reason why not?

35

No reason it won’t work at the driver.

36

Hank usually sets you up for success

37

This is a neat conversation. I am not sure why 12A is necessary. Or why not just do 20A and get rid of FET all together if it is just a resister value that is the difference?

12A will speed up the ramp table and you will have less control over brightness. Instead of 9A in 130 steps there is 12A. You will probably hit 9A by step 120 though. The difference is tiny.

I don’t think it will seem hotter because 9A is already too hot. The lights on 9A step 130 will be too hot to hold in 15 seconds. On 12A they are too hot in 12 seconds? 9A can’t be sustained so 12A makes no sense.

I think lights should only be regulated to amount they can sustain and then you get turbo if you want more. Like the Lume1 driver it makes sense.

38

Unless, of course, you’re driving a triple or quad…

39

The regulated drive inherently has higher resistance than the FET, so you’d be losing output. You can’t beat direct drive.

12A isn’t too much IMO, I can hold my 12A driver lights at max regulated for longer than the light can handle the output before stepping down. I’m glad they aren’t regulated only to thermally sustainable level (which also depends on too many factors - battery, ambient temps, airflow, the specific LEDs you have, the specific bin of those LEDs, even the temperature of your hand on the light. If you want a lower ceiling, you can set one.

40

The DD FET lets extra voltage burn off at the LEDs. The LEDs are good at this, and are soldered directly to a big chunk of copper to pull away heat.

The linear regulator burns off extra voltage at the FET instead. It isn’t sitting on a big chunk of copper, and it has a harder time shedding heat. So it can’t handle as much power; it overheats at a lower power level.

Most regulator circuits are limited by the amount of heat the regulator can handle, not by how much power the LEDs could take.

2 Thanks