I hope it’s a buck driver but I have a feeling it’s linear.
Waiting for a reply.
Yes, we can do the tint mixing.
CH2 is also on the flood side.
We are open to more options as soon as the LED is avaiable.
It’s linear driver.
Regards,
Hank
Thanks for the update. Too bad it’s linear.
What leds did you ask about? I want an sst40 5000k
Ch2 is flood? Are center intended for throw?
Sst40 is a bigger led. Would need a different board I guess
no i think its because ch2 offers leds like osram w1, so people might expect it to throw, but it would just be a bit throwier than ch1, but not actually throw
CH1 is both linear and FET with PWM. Anduril could probably be modified so that CH1 is FET-only with PWM, if that was better for your use case.
Buck drivers are better, but linear drivers are still reasonably efficient. It is inaccurate to say that 1/3 of battery capacity will turn to heat. Even if you had a 100% efficient buck driver, most of the energy in your battery would still turn to heat.
Wall-plug efficiency of almost all leds is still less than 50. For example, state-of-the-art SST-20 4000K CRI95 led has 43 WPE at 350 mA. If you have a Noctigon K9.3, that is a total of 3.15A for 9 leds. WPE is even lower if you drive them with more current.
That number comes from this presentation:
A buck driver would generate less heat in the driver than a linear driver, but even in the worst case scenario, which is a 100% charged battery, the majority of heat would still come from the leds. You would likely be able to notice that a buck driver is more efficient, because the flashlight would get hot less quickly and be able to sustain slightly more lumens, but it would still get hot.
If you have a 100% charged battery, it is reasonable to estimate that 20–35 of energy will turn to heat in a linear driver. However, a buck driver is not 100% efficient either. It is reasonable to estimate that 5–15 of energy will turn to heat in a buck driver.
Once you drain the battery to 50%, the difference in the amount of heat generated by a linear driver compared to a buck driver becomes much smaller.
To really compare the overall efficiency of a buck driver compared to a linear driver, you would need to measure it starting with a fully charged battery until it was empty. It would vary depending on the exact type of led, the number of leds, how much current they were using, the battery, and whether you charged the battery to 4.1V or 4.2V.
You can get slightly better efficiency and runtime with a linear driver if you use a battery like Samsung 50E and if you only charge it to 4.1V.
This datasheet has an interesting comparison of driver efficiency between Lume1-FW3X buck-boost driver and stock FW3A driver.
That Lume1 vs linear driver efficiency comparison was done using a single emitter. The efficiency wouldn’t be as good for the linear at 3A if it was done with 9 LEDs as they’d be running at a lower voltage.
Right. Also, the stock FW3A driver is linear but not constant current, it uses PWM. It is quite different from the linear constant current driver used in Noctigon K9.3.
Xpl-hi
I was thinking that in worse case scenario using red SST20. If we use 9 led at 9A for main light, each led is 1A.
Using datasheet https://download.luminus.com/datasheets/PDS-003040-Rev-04-SST-20-DR.pdf
Forward voltage for red led at 1A (because all in parallel) is about 0.25V+2.05V or 2.3V. For full battery at 4.2V, at 9A maybe the voltage drop to 4V, so wasted energy in driver is 15.3W while leds only use 20.7W, so efficiency is 57% to led and rest as heat in driver. Of course using higher voltage white led will be much better, so maybe red led not best choice for efficiency. At low brightness with red led voltage of about 1.8V, the result is actually more power burning in driver than leds, so i hope driver is heatsink well to flashlight body.
Wow Light, my mind is blown… I don’t have any monochromatic lights but I’d love to get some. I did some research on the XP-EL and found that it comes in Red, Red-Orange, Amber, Amber, Green, Blue and Royal-Blue
I’m thinking of getting some D4VS2’s in different colors and definitely at least one in 2000K. With the 26650 I can just leave them on for hours or even a 26350
But now I’m completely undecided on what 9.3’s to get, especially with the variable of being able to tiny mix thrown in. I’m thinking of getting switch backlighting that will allow me to quickly know which light is which. Red for amber etc
Do you have a list of all the Hank lights you own?
Thanks
I love Hanks lights, what is the main importance of this one?
I have an M43, two D4, K1, D1, and a D18. I’m not sure this one would fill any role that is not covered by those. But what am I missing here?
21700
USB C charging
2 channels (seems fun)
And Anduril 2 loaded on already.
I’m grabbing one for sure.
I’ll do xpl-hi 5000k with deep red on CH2.
Things I want to ask Hank, but do not want to bother him because, ya know, he is busy.
“Thank you Hank for another great creation. I know you are busy, I just have two questions for you before I purchase the K9.3
I was wondering if you plan to offer Xhp35hi or sst40 in the inner channel?
I also would like to know the lumen output on channel 2 for the yellow osram and the sst20 2700k?
Thank you Hank, I appreciate everything you do for the Flashlight community.”
Can anyone answer this for me or should I wait a week, let stuff settle down, then try to email Hank.
Thank you all
XHP35 HI needs 12V, while the K9.3 is using a 3V linear driver, so I believe that’s not likely to happen. SST-40 may require a new MCPCB (other than that 3XP one used right now), so that’s not likely to happen either.
9A for 3 outer emitters = 3A per outer emitter. Assume a J3 bin SST-20 2700K or MA bin W1 Yellow is used. From their datasheets, the LED output is roughly ~500/~400 lumens. The OTF ratio is usually close to 80~85%, so the OTF lumen of a single emitter is ~410/330. Thus total OTF output is (roughly) ~1200 (SST-20 2700K)/~1000 (W1 Yellow) lumens.
Just fyi, I agree with toobadorz on all the points made -- nice job in answering/helping!
Oh man, I appreciate the rapid response! I think this makes my mind up for this guy then! I’ll sit on it for one more day or so, and then place the order and wait
Hello! First post to this forum, got into flashlights earlier this year when I discovered the /flashlight subreddit. Big fan of the Noctigon Emisar lights that Hank puts out.
Going to pick up one, maybe two K9.3 lights and have been fiddling with configuration options since the page launched. Here’s the three options I have specced so far, trying to decide which ones to pull the trigger on.
1) “Forest fire” [saw a similar build in a reddit thread, liked the idea of super warm K9.3]
Color: Green
Switch: Amber
CH1: E31A 2000K
CH2: W1 yellow [This is where I’m the least sure. The SST20 Red could go with this super warm build, but I do already have a D4V2 with deep red. Have not seen the W1 yellow in person yet.]
2) “Every day use”
Color: Dark Grey
Switch: Green
CH1: SST20 4000K
CH2: W1 Green
3) “Blaster” [inspiration is a star wars storm trooper]
Color: Black
Switch: Red
CH1: XP-L 4000K
CH2: SST20 Deep red
Trouble is, I keep changing my mind of which one to get. For my D4V2’s, I usually like warmer temperature that are more comfortable on the eyes, so #1 is very appealing. It in terms of usability, I think #2 would make a better all around flashlight. Then I start thinking the K9.3 should be a special light and to take full advantage of all those LEDs then the super white XP-L with red will be a very cool build.
Also have not seen the amber switch or W1 yellow in person. Would they be a good match in the #1 build?
It’s too bad the switch isn’t a selectable RGB light, that would really lighten the load for choosing all these options.