【ツ】Interest check + ideas for a new multicolor flashlight

Interested

[quote=Lux-Perpetua]

Okay, some more thoughts and rough ideas…

Multicolor flashlight with reflector

Let’s just imagine we had a quintuple reflector as in Acebeam’s X65 flashlight. Now, the X65 is a heavy duty flashlight with a large reflector. So, please take this as reference for the reflector design only. Getting there entire reflector smaller the configuration could be:

center emitter: Cree XP-L HD V6 3D (4.885K) or for even more throw Cree XP-L HI V3 3D (4.885K) or - more cost effective - Luminus SST-20 4.000K 95CRI / SST-20 5.000K 70CRI

adjacent emitters: Cree XP-E2 red/green/blue/amber or just three colors and UV (e.g. LG LEUVA33W70RL00 365nm or Seoul Viosys Z5 365nm).

the adjacent reflectors would at least need to have a structured surface (orange peel) to reduce artifacts. However, I’m not sure if the shifted angle (not being in the center of the host) would cause an uneven beam shape.

Personally i like the reflector in the photo, and at first we have to decide the colors :will be red, blue, green,white, UV added may make things more complex,

Thank you for all the information. I am collecting now and will officially started this project process with factory in a near future, we will have a small meaning with factory, and i will also feedback their opinions here :face_with_monocle:

Please note that with a reflector like mentioned above there's a high risk of an uneven beam pattern ("Mickey Mouse" pattern), see here for example...

If you intend to use a reflector, make sure each LED has its own reflector (see Wurkkos WK30 for example). However, the small WK30 reflector geometry wastes some potential we could gain with a deeper and wider reflector design. But then again you need a big flashlight head to take more area for multiple reflectors into account. That would interfere with the idea of a lightweigt and compact design we have in mind and it would also raise costs. The real downside with a reflector design is that color mixing will get almost impossible as the correct alignment of different LED sizes with different die areas and a uniform reflector geometry could possibly end in a nightmare.

Bottom line: If we want to go on with the idea of infinite color mixing, we shoud better focus on TIR optics.

On a side note:

Lumintop has recently changed their supplier for optics with the FW3A. They now appear to use a Chinese optic (without glas lens on top) that is more floody/frosty than Carclo's 10511 they used before. Since Carclo optics are more expensive and hard to source in China, this alternative optic or the supplier behind might be worth looking at. However, it's a triple optic design, so we would not have RGBW. If white is a prerequisite we need to use a quad optic design like Carclo's 1062x (where x=1 is used for clear narrow spot and x=4 for wide frosty spot) or we could go on with the LEDIL ANNA 40mm optic with 7 LEDs in place.

The hardest part will still be the driver and the UI. Maybe Sofirn can adopt the Dr. Jones driver design and develop something comparable on their own. Without knowing the Dr. Jones driver in detail I could imagine a basic user interface like this...

From OFF

single click = ON (white mode)

double click = ON (colorful mode)

triple click = ON (colorful strobe modes...consecutive double clicks switch to party-on-mode, psychedelic-drug-frenzy-mode, gaudy lightning storm mode... whatever :-D )

ten clicks = ATR configuration mode*

* Advanced Thermal Regulation can be set between 45°C - 55°C - 65°C as follows:

After 10 clicks the flashlight shows green color => click to set 45°C or wait for next color >>> color changes to blue => click to set 55°C or wait for next color >>> color changes to red => click to set 65°C or wait for flashlight shutting off without changes to settings.

There will be no time-triggered stepdowns. ATR default configuration from factory is set to 45°C.

From ON (white mode)

single click = OFF (mode memory)

click + hold = ramping brightness of white light

double click = 100% white light

quadruple click = sleep timer (each consecutive single click is an increment of 5min for instance)

From ON (colorful mode)

single click = OFF (mode memory)

click + hold = ramping brightness of colorful light

double click = infinite loop ramping through colors ( >>> red > violet > blue > cyan > green > yellow > orange >>>)

click to maintain selected color

triple click = 100% red light

consecutive double click = 100% blue light

consecutive double click = 100% green light

quadruple click = sleep timer (each consecutive single click is an increment of 5min for instance)

From ON (colorful strobe mode)

single click = OFF (no mode memory!)

click + hold = ramping brightness of colorful light

double click = change strobe modes

triple click = change strobe speed of selected strobe mode

quadruple click = sleep timer (each consecutive single click is an increment of 5min for instance)

Interested in this. I like the 7 LED optic with center white and two optic per color.

Finally, i talked with factory and engineers, TIR optic will be first choice for this mixed color flashlight, and LEDs will be 4 or 5 color(amber added if 5 LEDs), the outside design, i’ll say, it is a totally Innovative design for a flashlight, i am not sure if it is a good design for lamp, more like an artwork. I do not like it very much at first time i saw it, but more and more interested in this design now, exciting…

I did some thinking today, taking Carclo's 1062x optics into account again...

This idea may be a bit unrealistic in terms of the optic but I like to share it with you anyway.

  • If we had a quad optic with both clear narrow spot and frosty/pebble wide spot we would be able to get the best of both worlds: A smooth diffused colorful wide spot for close range applications and a distinct, bright hotspot of white light for mid-range applications. Another aspect on the plus side: If we mixed all three colors (red+green+blue) we would (theoretically) get a nice smooth diffused beam of white light, too.
  • With a clear narrow spot for white light we get multiple options of 3535 emitters to influence the beam pattern even more: For a floody beam I would either recommend using Samsung's LH351D 5.000K 90CRI (which Sofirn has stocked up for their SP36S / BLF Andúril edition) or - to get the maximum efficiency and power with the right driver using a separate channel - Cree's XP-L2 5.000K 70CRI which gives you a much better beam/tint profile underneath an optic than in an SMO reflector. For ultimate throw I would either go for Luminus' SST-20 4.000K 95CRI (cheaper than XP-L HI) or - again for better output - for Cree's XP-L HI V3 3D 4.885K 70CRI.

Lux Perpetua - I agree about the mixed optic being really versatile. I’m pretty sure Carclo doesn’t make one that way. However, the use could frost 3 of the 4 optics themselves, such as by sanding or with DC Fix. I think some people have had decent results also polishing a frosted Carclo.

If mixed at the same power level, the result will be white-ish, but likely with a strange tint. Maukka tested this with the color E17A’s:

In theory, by varying each channel individually, you could get one light to produce white of almost any CCT and DUV you wanted, within some really large limit.

But implementing that could complicated fast.

For now, would it be fair to say that a standard quad TIR is the leading option, with a mixed version being a bonus if such an option is identified?

Thank you for your feedback iamlucky13. Let’s see what Wurkkos and Sofirn can do in terms of drivers, emitters and optics.

I fully agree with you that using a standard optic like Carclo‘s 10624 (frosty wide spot) or Kathod‘s PL115140 30° 25mm wide beam optic (https://www.luxeonstar.com/assets/downloads/pl115140.pdf) would be the best way both in terms of best efficiency and modifiability. However, Sofirn may want to check for locally available Chinese optics (see Lumintop‘s plans for alternative, cheaper optic of the FW3A) to reduce prime costs as much as possible.

Eventually, a fully frosted optic might be a useful compromise even for the white LED? I’m not sure about the additional amber LED Wurkkos mentioned as fifth emitter before. It would also mean to find a whole different optic with a larger diameter.

I expect the biggest challenges for Sofirn in designing/sourcing an appropriate driver that can blend any desired color with RGB and is also capable to use more power for the white LED via separate channel. Once that is addressed, finding the right UI will be the next challenge which should be tested thoroughly by involvement of BLF (e.g. run a poll, build prototypes with different emitters and UI versions for testing).

Personally, I‘d like to see a powerful and efficient XP-L2 in this flashlight but HighCRI is highly anticipated nowadays and maybe a more wanted option.

If a quad optic will be chosen, the best is to find a well frosted one so that at least the colour blending is done well. It leaves white also very floody but people could choose to polish the part over the white cup. Most versatile would be a Carclo optic, so that people can swap it for another type of frosting if they like.

For a proper tint mixing the beams (spot and spill) of white and color LEDs should have the same shape and size - at any distance! This is difficult to achieve with XP-E color LEDs since they are smaller than a powerful white LED.
My RGBW D4 does a proper tint mixing at a distance of a couple of meters but it gets worse if you come closer.
I’ve just tested DC-Fix with this light. It helps a lot but the resulting beam is too floody for my taste.

Is there any variation of shape and size between the XP-E2 red, green and blue? I don’t think so but it would be nice to be absolutely sure. I only know that the output varies as blue and green can emit more lumens per Watt compared to red. Do you think peak temperatures will need to be taken into account when driving the XP-E2s with reasonable currents (< 1.5A)?

Do we need the white LED for color blending (RGBW) or will RGB only suffice?

If white is used separately only, do we still need to have a matching shape and size?

If this is the case, we should better stick to SST20 or LH351B as these are rather small in terms of dome and die size.

The shape and size are the same, and that causes a different beam because the diffraction of different colours on the surface of the led dome is a bit different.

The Luxeon C Colour Line is especially designed for colour mixing but the output is less than the XP-E.

Extra mixing option with the white led would be fun, but also complicates the user interface. Unless you go for a 4 button operation (separate ramping for each led).

Okay, so this is where we are right now, correct?

A quad emitter optic with frosty wide spot to diffuse the colorful beam as much as possible which hopefully helps blending infinite colors by mixing RGB.

So, a Carclo 10624 might be the right optic to start with in terms of prototyping.

@ Flashy Mike: Did you use the 10624 with your Emisar D4 or another one (10621/10622...)?

If that is a 10624 optic I wonder how much influence the diffraction of different colors have in accordance with the opacity of each single color LED, i.e. will spot and spill have the same size, intensity and appearance on a white wall, so they will look congruent when red, green and blue get overlaid?

If not, things might become even more complicated. Maybe there's also need for adjusting the current for each color LED individually to achieve a matching beam pattern? If blue for instance needs 0.5A for 100lm while red needs 0.75A for 100lm it could become challenging. Another aspect is that the reception of colors for the human eye is also dependent on the wavelength, i.e. green light will be noticed with much more intensity than red light at the same output level.

Maybe a stupid question…
What would happen if we use 4x RGBW led? So each color would be emited by 4 emiters?

Something like this...?

I'm afraid it would not create a uniform beam pattern as each color die is off-centered i.a.w. the optic's dome center position. OTOH, it would also increase costs dramatically as the Osram S2WP is quite expensive and so is the old XP-E Color, too.

You would also need modified led pcb’s (or a new one) to suit the RGB leds. In either case you could turn each led 90° to create a more uniform beam.

An optic for each colour gives the most uniform beam

I also did a colour zoomie once. Was much simpler UI in that one though.

With naked eye they appear very similar.

Depends on the body. My RGBW D4 takes 3.5 Amps with a pretty old 30Q and all 3 color LEDs at max (white LED off). The head ist too hot to hold after 2 minutes.

Don’t know - but you will see little lumens with RGB only.

The spot of my white LED (XP-G2) is larger than the color LEDs (XP-E2) spots. It mixes pretty nice with the color LEDs.

It is the stock D4 optics. With all LEDs on max the spot looks pretty nice but there are color artifacts around.

I wonder whether tint mixing is really required. I’d rather prefer a high CRI white LED, offered in diffferent flavours.

If it has a good to great user interface with options I’m in.