[Optisolis/E17A/E21A] Jetbeam Jet-µ

You’re right. Although these colored LED aren’t as bright as normal colored LEDs, they have much wider spectrum coverage. I was surprised too how well I can see under the Red, Green, and Amber E17As. Unlike Blue which has very specific uses. Amber E17A renders color better than LPS and one could mistaken it from 1800K warm white if the object shone don’t contain too much colors. I also prefer PC Red for retaining night vision, because I can still somewhat see different colors better than with true single spectrum red color. Green although very relatively harsh to the eyes, brings out details extremely good and has the best color recognition among the other colored E17A. I can’t use the blue for too long, it hurts my eyes. Blue E17A, just like many failed dedomed LEDs in my collection, very useful for illuminating anything that normally glows under UV.

- Clemence

Indeed, they are very surprising.

For each color, the datasheet has spectrum info at the page 10 (I mention 20% to 100% to 20% intensity of spectrum)

Blue (430nm - 447nm - 465nm)

Green (488nm - 521nm - 583nm) (slight deep blue, no near-IR)

Amber (543nm - 608nm - 691nm) (slight deep blue, very slight near-IR)

Red (598nm - 647nm - 735nm) (no deep blue /but figure doesn't contains range 400-450nm, slight near-IR)

Based on this, probably I need to take a red for sure in my next order

Received a jetusolis 6500°k and 5000°k today (ordered on janv 24 (3 weeks ago)). Thank you clemence !
I usually don’t like cool white then each time i turn on the 6.5k, i need time to get used to it.
Once adapted, i have zero problem to consider it as one reference for colors rendition for myself. I will use it amply for inspections in the next days.
I really appreciate, at least, being able to check colors with a beam without a neutral or warm tint.
I can stand cool white…i’m cured !! :partying_face:
Thank you for your work, tests and dedication to high cri.

Is there an E21A and optisolis, both 6.5k beamshot somewhere ?
What to expect of a triple E21A 6.5k in the FW3A ?

Is it too much to hope for a tir lens combined to your Wizard/Tiara E21A module that would fit the FW3A ?

EDIT : blue tones with the jetusolis 6500°k look real good :+1:

If the Wizard/Tiara quad E21A module fit the FW3A, I’m going to buy a bunch of them! :smiley:

I only have Cree XPE2 Green from my Armytek Dobermann to compare with. You can see the difference between Cree XPE2 pure green LED and phosphor converted green E17A. The much wider spectrum coverage makes for better color rendering other than green itself. Depending on the applications, this could be a bad or good thing.

Note: you can clearly see which is Red, Blue, or Yellow under PC Green E17A. It’s impossible to check colors under Green XPE2 (except for the false orange bottle cap which actually pink under white light)

I can’t stand low CRI cool white, it’s too harsh with no real benefit. With very high CRI cool white, the harshness pays off. Truest whites, clean colorful objects. Here’s the beamshot, taken from the product page:

If someone can give me a detailed technical drawing of FW3A MCPCB, I can do that. What I learnt: E21A only benefit from nano-ceramic MCPCB in tightly packed setup. For conventional wide spaced triple or quad, we only gain 300mA extra peak current which already very inefficient. My early test with heavily cooled direct thermal path confirmed that too. Below OEM rated current (1400mA), E21A is very tolerant to any MCPCB types. Looks like it’s time to make another triple and quad E17A/E21A MCPCB!

- Clemence

I remember reading the FW3A will be using the Noctigon 20mm triple mcpcb and will be compatible with Carclo optic. Are you planning to make a traditional quad for the E21A/E17A to fit Carclo optics?

Shhhhhhh……

An FW3A with E21A’s might be very interesting.

However, the FW3A is FET driven for its highest outputs. Isn’t that a concern with the E21A due to their low forward voltage?

Texas Ace tested an E21A quadtrix as peaking around 2.5A per emitter. I assume you want to stay below the peak, but the FW3A is likely capable of providing more than that.

The actual peak current of E21A is 3,3A for both R70 and R9080. But in R9080 the curve is almost flat from 2,5A to 3,2A. Making it’s not worth the heat/risk upping to more than 2,5A.
You’re right about the FET. E21A is better suited for more refined current controlled drivers.

- Clemence

Those above was a single E21A R9080 tested on a copper DTP. Indium solder will provide a teeeeny bit more current. Not worth the hassles …
With “normal” non DTP MCPCB, four E21A packed closer together can only output about 1A max per LED. With nano ceramic MCPCB such as VR16SP4 we gain back the 2,5A max for each LED.

- Clemence

Efficient buck/boost cc drivers that can be readily bought for e-switch lights seem non-existent unless you build it yourself or you’re willing to sacrifice an expensive light which has one. One option might be loneoceans’ boost driver driving triple in series if somebody can make those available with appropriate e-switch firmware.

But for the meantime, I personally wouldn’t mind an FW3A with triple E21A running on just the 1+7 7135 chips with FET disabled or removed. IMHO It would still be totally worth it despite the added heat and loss of output and efficiency. But that’s coming from an E21A fanboy whose willing to sacrifice everything else for these emitters. I blame Clemence for that :stuck_out_tongue: .

You’re welcome and thanks for your support. I also keep 3000K and 6500K Maukka calibrated Jetusolis for my reference lights.
Next project in the making …

- Clemence

Any news about configurable multi-emitter Optisolis floodlight PCBs? :partying_face:

No news for now.

- Clemence

I almost forgot, that Optisolis will be available as COB as well, maybe then there is less need for that kind of solution, but a decent, moderately frosted optic could be useful for that.

COB Optisolis does better job to mix all those weird asymmetric beam from each LEDs in a compact package. But it doesn’t do few things:

- Multi CCT setup

- Low voltage. 36V is harder to implement into portable applications.

  • Seamless daisy chain light panel

You mean with the COB, we wouldn’t be able to mix say 2000k with 3000k?

COB Optisolis has only single CCT options. There’s another interesting new product from Nichia, a matrix of 2700K - 5000K E21A in COB form. But it has the same 36V voltage and currently limited to only R8000.


http://www.nichia.co.jp/en/product/led_search.html?op=cond=whatsnew='tunableled'

- Clemence

Got my 2 Jetusolis’s in 5000k in! One appeared slightly cooler than the other, so I went with the warmer, and gave the cooler to my brother and his wife.

First impressions are that it appears much cooler than 5000k, more akin to the 5700k Nichia 219B R9050 I had on my old Jaxman E2. I wonder if a lot of that is attributed to the full cyan region of the spectrum?— I would guess so. Very pure looking. I can definitely tell it’s slightly green, but it does render reds really well regardless. Tint deviation within beam seems small in general, thanks to its floody nature, subjectively it looks like the outer edges of it are warmer and greener than the center, but that could be because there’s less lux on the outer edges.

I think I’ll try it with a green minus 1/8th filter.