Armytek Wizard Pro mod with Virence E21A 2000K

LED Flashlights – General Info 18650 Flashlights
1

So this is my very first flashlight mod and first time soldering and doing any manipulation with electronics. I was initially interested in getting a flashlight with white, but as little blue light as possible, so it doesn’t block melatonin production at night so I can sleep better if i use the flashlgiht. I decided on getting something in the color temperature of around 2000K. I have 5 more wizard pros I broke open and am awaiting the E21As from Clemence.

Thanks to clemence (Virence) and bob_mcbob for answering all of my questions regarding the mod.

Finished lights: https://imgur.com/jK3r1iO

The left bezel is the original metal bezel. The right two flashlights have a 3d printed bezel in nylon that i outsourced. I’m not really sure why it has white residue all over it, but the color is supposed to be black when I cut through one.

Progress: https://imgur.com/NghmHIv

So I started off by clamping down the flashlight with trigger clamps and a U-shaped metal bracket and towels so i wouldn’t scratch or bend anything. Then I broke the glass with a small screw driver and popped off the bezel with a large screw driver. Took out the 2 o-rings and the optic. Unscrewed the mcpcb and led. Unsoldered the switch wires. Used a heatgun to 300F and got the battery tube to unscrew, some harder to release than others. Took out the driver and cleaned up the silicone potting compound. Cleaned up the thermal paste. Unsoldered all the wires and replaced the black/red with 24 AWG and switch wires with 28 AWG. Soldered everything back to the new mcpcb with the E21A in 2000K and put it all back together with new thermal paste and thermal silicone potting material. I did get some flux on the E21A LEDs but some warm IPA and a brush took care of that. As i mentioned above, I outsourced a 3d print file that I made for the nylon bezel and just hand pressed those in; these are only for ease of removal in the future. The metal bezel, I used a parallel clamp with folded paper on top/bottom and that got the bezel in nicely.

I think the 3 biggest issues I encountered:

1. removal of the metal bezel. but once i figured it out, all the others were super easy.

2. removal of the battery tube. none of them had loctite which is nice, but there were a couple of flashlights from the 8 i broke open that had extra glue or whatever was used. I ended up increasing the temp to about 450F and using gardening gloves that had really good grip on it.

3. soldering the wires on the virence mcpcb. it seemed like copper pulled away heat really really quickly, so it was just a bit tricky to solder. I used flux for the first flashlight and it got all over the LEDs, but for subsequent flashlights, the solder wire with flux in it was sufficient. i used AMTECH SAC305 Lead-free Solder Wire 2.2% Rosin Core (NC-61) - 0.031”

I’m super satisfied with the build and the light coming from it… and that’s why i’m building 5 more lol. As for things I’m going to change for the next 5, I’m going to use 22 AWG for the black/red wires and use less viscous thermally conductive potting compound.

2

Nice mod. My 2000K E21A turned out to be closer to 2300K, about the color of an acetylene flame. Those headlamps are an interesting parallel to the acetylene headlamps miners and cavers use. One of their reasons for sticking with with older gas driven lamps being the light color, which is matched in quality and warmth with these emitters. That being said I use the 2000K for the same use, a late at night low-blue white light. These emitters are a very rich color.

3

thanks! it was quite the process and fun too. whoa 2300K eh? how did you end up measuring the CCT yourself?

I hardly run into any posts talking about late night usage and low blue light, glad you’re on a similar page as me.

4

My Tiara Pro with 4xE21A 2000k measured:
CCT: 2010K
DUV: of –0.0002
CRI: 95.3
R9: 91.3
Rg: 101

Blue light output is on par or less than the 2200k Good Night light bulb that I returned. The Good Night light bulb has a yellow shade that makes everything look yellowish. The Duv was about 0.005. CRI was 90. The E21A looked so much better. I wish someone can make E21A light bulbs. It would beat anything on the market including the SORAA bulbs that I have.

5

I ended up running a third party app on my phone to measure CCT. I guess I just got a 2200K sample, but still a gorgeous light. Red wood look so rich under under the light. I was just blown away with how rich it made my wood cutting board and furniture look. I wish they would use LEDs like these in bulbs. I would like to see LED street lights/architectural area lights with these emitters, what wonderful experience it would be to be under this rich light in the city.

6

yeah i would love it if someone made E21A bulbs. I tried out the soraa healthy a19 bulbs and i hated them. they made me kinda dizzy and had a greenish tint on it. returned immediately. i should figure out how to measure the cct on mine.

7

yeah that would be amazing. they changed the street lamps years ago on my street to horrid blue leds. i would love it if they were 2000K.

8

The SORAA bulbs uses violet emitters instead of blue emitters. Violet has shorter wavelength than blue so it is higher energy, which might be even worst than blue. I’m guessing that’s what might be causing your dizziness. I did measure it to have higher amount of violet wavelength than other light bulbs of the same color temperature.

9

Maybe not a perfect 2000K, but Chicago is beginning to test the 2200K Lumican LED street light. They had show on it someone posted to Facebook. The color I found is hard to get right on camera. Pittsburgh as well should be trailing the 2200K Lumican. They are testing all lights 2700K or less. The Lumican can be had in color temperatures as low as 1750K with around a 70 CRI, nearly the same level of warmth as LPS/SOX, but with much higher color rendering. It’s a trippy experience.

10

I don’t think Healthy is the product line you want if the priority is CRI and tint. Their product page indicates that uses a violet emitter and has a big gap in the spectrum to minimize output in the range with the most potential for sleep disruption. Radiant should be the line most people want. I think Radiant also uses violet emitters to maximize color rendering, but without skipping the blue tones.

However, yes, it would be awesome to be able to buy household lights using some of the best emitters now available, like E21a or Optisolis.

The Yuji A19 bulbs should be great, but they aren’t quite omnidirectional, and they’re even more expensive than the Soraa bulbs.

11

To be honest, I don’t think I’d want that warm for general lighting, even in residential areas.

But definitely not the harsh cool whites they’re using in most places. I think I’m fortunate in that most of the streetlights around my area look like they’re in the 4500-5000K range.

I’m fine with that on the major highways, but in residential areas, I’d much prefer they stay below 3500K.

12

Unfortunately, 3500K, 4500K, and 5000K, are all way beyond AMA guidance for outdoor lighting. Those color temperatures of light suppress melatonin and leptin leading to multiple health implications. The sleep aspect regarding blue rich LEDs has been addressed .

Regarding those LED street lights listed above 3000K.

“High-intensity LED lighting designs emit a large amount of blue light that appears white to the naked eye and create worse nighttime glare than conventional lighting. Discomfort and disability from intense, blue-rich LED lighting can decrease visual acuity and safety, resulting in concerns and creating a road hazard.”

From Harvard regarding the implications with obesity …

“A Harvard study shed a little bit of light on the possible connection to diabetes and possibly obesity. The researchers put 10 people on a schedule that gradually shifted the timing of their circadian rhythms. Their blood sugar levels increased, throwing them into a prediabetic state, and levels of leptin, a hormone that leaves people feeling full after a meal, went down.”

There are further peer reviewed studies that are coming to similar conclusions. Light above 3000K doesn’t belong anywhere lighting our streets at night. Further, studies conducted in Davis, CA showed that people were nearly 20-1 against the original 4000K fixtures. Their early implementation of warm white LED street lights indicated that even 3000K was too high, and that eventually LED them to a custom order of 2700K Leotek Green Cobra Jr.s’. 2700K should really be the maximum permitted in residential areas with 3000K/2700K reserved more for the larger streets.

The EU, as well, their, outdoor lighting guidance for roadways calls for 3000K max with warmer color temperatures having relaxed efficiency standards thus passively encouraging better light.

Also, I would be really careful about using voilet pump emitters. These Soraa lights were tested by Dr. Travis Longcore as having substantially greater negative impacts on wildlife then most warm white light sources that included warm white blue-pump LED, metal halide, and CFL, and even higher CCT cool white LED street lights.

13

Luminous intensity is even more important than color temperature, which the AMA guidance acknowledges, but unfortunately did not discuss. Frankly, the references cited in the AMA list were a bit disappointing in how high the ratio of mainstream media articles was compared to actual research papers. It’s not that the AMA guidance is wrong. It’s just really limited in its utility due to the limited depth and breadth of research it relied on.

One study that does look in a very cursory manner at intensity vs. color temperature is here:
https://www.sciencedirect.com/science/article/pii/S0003687016301582?via%3Dihub

Basically, low intensity (0.1 lux) 2700K light was their control, and all higher intensity (200 lux) test cases even down to 1600K exhibited more melatonin suppression. I don’t have a lower limit of significant effect handy, but that factor has been more studied in isolation, and a decent figure probably determined. The impression I get is it is somewhere below 10 lux for incandescents, which actually is quite a bit of light, but I don’t know how far below.

Also, I think it’s well worth discriminating between high traffic highway lights not in close proximity to residences versus and those that are in proximity and illuminate low traffic areas, because we have to keep in mind the reason those lights are there, which is safety. More neutral lighting, compared to warm tones, provides better alertness and acuity, and depending on the colors of objects illuminated, can provide significantly better object discrimination and perception. So I suggest that using neutral white lighting in areas where it has a low potential to effect circadian rhythms is prudent.

I don’t see much reason to use violet-pumped LED’s outdoors. It’s just not an application where I see a need to chase the best CRI possible, so the conclusions you reference of Dr. Longcore seem reasonably easy to abide by. From what I’m seeing, his research seems mostly relevant to critically sensitive areas, like sea turtle hatching grounds, where users are being encouraged to use amber or red lights.

14

Also, I apologize, I just realized we’re getting rather far off topic for a thread about Armytek Wizards.

I’m enjoying seeing these mods, and am hoping to do my own Armytek E21A mod before too long. Mine will likely be 3500K.

I also want to try out a good 2000K light, so I think I’m going to mod one of my Fenix E05’s with an E21A. Clemence indicated his Jetbeam Jet-u kit is also compatible with the E05.

15

that’s good to know. i’ll stay away from soraa lights that are violet shifted. i wonder if anyone has made their own light bulbs in e21a at 2000K… i’d want to make some of those. it’s annoying to change halogens so often, but the light is so nice from halogens.

16

I didn’t like the E21A 2000k the first few times I turned it on because I thought it was too yellow. A few months later, I turned it on late at night and I actually started to really like it. It didn’t make everything yellow and felt very warm and cozy. I realize it just takes time for your eyes to adjust to the color temperature. No I modded a bunch of small lights with the E21A 2000k for night time use by family members. I also bring them to hotels and leave them on throughout the night as night light for bathroom visits.

17

it’s not that far off topic, i enjoyed reading the responses. modding is fun and i need to try out more.

i really want to figure out if i can make an a19 and br30 light bulb with e21a in 2000K

18

yes! i’m modding all of my lights so my family members can use them too. glad you warmed up to the 2000K

19

From the AMA guidance directly. While color temperature is an imperfect metric to measure blue, it is what we have for now. The issue here is blue color content, and that higher color temperature LEDs emit unhealthy levels of blue when used in outdoor lighting at night.

“AMA encourages communities to minimize and control blue-rich environmental lighting by using the lowest emission of blue light possible to reduce glare. ”

And regarding intensity

“The AMA also recommends all LED lighting should be properly shielded to minimize glare and detrimental human health and environmental effects, and consideration should be given to utilize the ability of LED lighting to be dimmed for off-peak time periods.”

Regarding research, from Dr. Mario Motta, one of the co-authors of the AMA guidance.

“In 2012, knowing the research activities of many scientists from around world on the effects of nighttime lighting on human physiology, I invited four prominent researchers to help me write a Council on Science and Public Health (CSAPH) report, “Light Pollution: Adverse Health Effects of Nighttime Lighting.” This 27-page report with 134 peer-reviewed references highlighted the adverse health effects of circadian rhythm disturbance. Suppressing melatonin production through excessive night lighting…”

Quoted from the abstract of your own source…

“Melatonin synthesis was tested with weekly intervals in 8 trials from 10pm to 2am (7 polychromatic illuminations and a dim light reference (<0.1 lx)) in 16 students (9 men, semi-recumbent position). Melatonin suppression was almost negligible for CCT <2000 K but increased with increasing CCT…”

20

And then there is this… Along with Space X and other companies sending 20,000 5G satellites into orbit and millions of 5G poles outside your window. 5G LED street lights are here…culling the population.