UV Led, yes we have arrived!

[cyberlocc]

Now, this is simply proof of concept!

DO NOT BUY AND USE THIS Bulb!

Fran Baines has tested it to verify, that it does in fact work. This is not a gimmick, this light does in fact produce UVB and UVA, in VERY large amounts.

She said the beam is narrow, and will have a fun write up soon. The tech is here. The design needs to change (were going to need a strip of these not a bulb version) and may need some refinement.

This is proof of concept only.

https://www.amazon.com/gp/aw/d/B07RKSCL2K/ref=ox_sc_act_image_1?smid=A3GYIS37IB3XTP&psc=1

The UVI of this bulb from 10 inches away is 18, DO NOT ATTEMPT TO USE THIS BULB! not because it doesn't produce UV, but because it does, in an extremely large amount!


https://www.amazon.com/gp/aw/d/B07RKSCL2K/ref=ox_sc_act_image_1?smid=A3GYIS37IB3XTP&psc=1



Fran's last thoughts thus far.

A strip light will be needed, but we can make that happen.

Again I want to emphasize, DO NOT BUY AND USE THIS LIGHT! This is simply a glimpse into the future. Not a working product for today!

@Goose502 you asked, I have delivered your answer. Very soon my friend. Very soon.

 

[cyberlocc]

I will see if I can track this down tommorow.

See what Diodes they are using, and hopefully by the time Fran has some data, I will have a link to a strip. Or a way for us to procure one, whether it be a group buy or something else.

Fran didn't think it would work, as it's only 3ws. I knew different (Fran is not a fan of LEDs lol) that means, we are looking at likely a 7-10w fixture for a 2ft cage, with proper setup and diffusion.

This will allow full dimming of UV in tandem with Sunlike LEDs, and we will be closer to the sun than ever before. IMO. These out out very little light that isn't UV as well. No more wasted power!

 

[Goose502]

Keep me posted brother! I’ve been dreaming of using LEDs for UVB some day. I wouldn’t mind building my own fixture, I’d just need some directions...like step by step.

 

[nightanole]

"The UVI of this bulb from 10 inches away is 18 "


So like how big of an area are we talking? UVI 18 over a 6"x6" area, or like the size of a golf ball? With some math it may work out if you pop it 18" above the cage so you get a good spread. Still my main concern is just how ugly they made the bulb look


Also who has the UVI 11 measurements? The manufacture says 102 µW/cm² at 20cm and 11µW/cm² at 1 foot. The UVI to µW/cm² is 1:40.

 

[cyberlocc]

"The UVI of this bulb from 10 inches away is 18 "


So like how big of an area are we talking? UVI 18 over a 6"x6" area, or like the size of a golf ball? With some math it may work out if you pop it 18" above the cage so you get a good spread. Still my main concern is just how ugly they made the bulb look


Also who has the UVI 11 measurements? The manufacture says 102 µW/cm² at 20cm and 11µW/cm² at 1 foot. The UVI to µW/cm² is 1:40.

I'm really not sure, about the spread Fran made that comment about spread, it was the first brought up.

As to the UVI it's a video on Facebook. Pre that video everyone thought the thing was a scam, "no way 3 watts is enough" ect. (They really are not LED fans on that group )

Some loose numbers, I have from the video. I got a load of SS's but don't want to post all of them, so I will write it out. This is going to be strange, and I will explain after.

3" - 3.1 UVI
6" - 23.2 UVI
10" - 17.9 UVI
15" - 7.5 UVI
20" - 3.5 UVI

Okay, so now you are thinking. What the heck, how is 3" 3.1 UVI, the Manufacturer says much lower numbers, and this doesn't make any sense.

It does. The bulb consists of Many LEDs, the very center posses that orangish diode, with a 120 capsule on it. That is a UVB LED. That is the only one.

There is then, 4 UVA LEDs in the non visible spectrum and a few more visible area of UVA diodes as well.

The manufacturers, UVB section might be right. The UVI readings are just that, and that's why I really dislike UVI being touted as the best metric. UVI means UVB and UVA. All light between 280-400nm is read by the solar meter as UVI. 390-400 is the violet light spectrum, the visible area of UVA, used in alot of fish LEDs, Atinic bulbs ect.

What I think is happening here, is we are seeing just the UVB or mostly the UVB at 3", we are so close to the light, not all the UVA LEDs are affecting it yet. At 5 inches, they have are all present. The UVB being produced isn't causing the High UVI, the UVA is.

Now we need to either wait for Fran's readings. Or someone with a Solar Meter, 6.2 and 6.5 needs to play with them, and see what the beam is like, ect.

A 6.5 only is not going to get it. That works fine, for a metric when we are looking at our Flos, that we know the percentage of UVA, and of UVB. It falls flat, when we actually need to solve for percentages of each. A spectrometer capable of UV, down to 280 or a 6.2, a 6.5 and the UVA one (forget the number) would be best.

 

[nightanole]

Yea a 6.5 is for measuring nature light. And a 6.2 is for measuring hazardous light.

So really we need a 3rd more scientific meter that will give a spectrum graph. We certainly dont want a repeat of a decade ago with the first rev cfls...

 

[cyberlocc]

Yea a 6.5 is for measuring nature light. And a 6.2 is for measuring hazardous light.

So really we need a 3rd more scientific meter that will give a spectrum graph. We certainly dont want a repeat of a decade ago with the first rev cfls...

Well, not really.

We know that the bulb does not produce UVC, so thats good, we are fine from that aspect.

We need a 6.2 to Measure only UVB, as that is all a 6.2 does, is measure UVB, 250-320 is all a 6.2 detects, and will give us the µW measurements we need, which we can match back to old way measurements.

The 6.5, will give us the UVI, however that is a result of UVB and UVA, of which there may be too much UVA. 30% UVA, on the 6% and the 10% Arcadia, so 30% UVA equivalent is what we want.

So maybe we want a Solar Meter 4.0 or 4.2, to gauge that aspect, and match it to our Flos.

Then due to Frans question of if the spectrum could create UVB (and IDK why she didn't try this, Maybe she did) we just need a Solar Meter 6.4. A 6.4 will tell us, by its calculation and weighting, how much said UV light will create D3. If the number is similar or higher than our 6% flos, I think largely we would be fine.


I have a feeling, that what Fran is talking about it not being close to the sun, is the spread. There is 1 UVB led. Which tells me, thats going to have a spike in a single area. The highest D3 synthesis is found in the 294nm mark, and right around it. However very little of said light makes it here, so it makes up a very small amount of light that we use in nature for D3 synthesis.

So while 300-320 is not very efficient at creating D3, there is a lot more of it. Which in turn means there is more D3 created. Lets go completely laymens, and rough example for those reading this later. This is by no means an accurate example, just an example to pain the picture.

Say there is 5 watts of UV in the Spectrum of 294-300, and it is 100% efficiency of creating D3. Now there is 300-320, which is 50% efficient at creating D3, however there is 30 watts of it. Even though, the 300-320 is only 50% efficient, there is 6x more of it, this means there is more D3 made by the 300-320.

Im not sure how accurate this sun spectrum is, but its a decent visual of why I think she is saying these LEDs are different.

Never mind the other lights, we are not talking about them. Notice how early 300, and before 300, is almost immeasurable? However look at the slope up to the first line, the first line is our 320nm.

Thats where Fran takes issue I believe. I am assuming, these LEDs are pumping 295, which allows for a little UV light, to be very efficient at creating D3.


If this works, and its going to take a lot of testing to verify, we will have to throw the "Ferguson Zones" that a bunch of folks worked really hard on, out the window. Because the delivery has changed, we are using less UVB to produce the same or more D3, as we are using a more efficient area. We will have to separately and measure UVA, and UVB separately, on design of the LED fixture, and we will have completely different UVI numbers to shoot for.

Once the bulb is designed, correctly, then we can go back to 6.5s and measure basing from UVI, however the numbers may very well be completely different.

 

[chino]

I read this and I feel like I'm reading chinese! But keep on doing what ever you guys are doing! Let us know when we can buy it and use this tech.

 

[nightanole]

I read this and I feel like I'm reading chinese! But keep on doing what ever you guys are doing! Let us know when we can buy it and use this tech.

Basically this is REALLY REALLY artificial light. So meters and optimal output based on natural light go out the window.

If you want to pull the rest of your hair out, there is another thread on IR heat emitters, and how they(IR output type and amount) can vary a lot vs output from the sun

 

[cyberlocc]

So there was an update by Fran today, we are still not getting a full report. However this is better than nothing.


"I'm planning to write this up properly but right now I'm swamped by other work. This is the spectrum. The visible light is very weak with a very strong, undesirable blue content. The beam is very narrow and the animals would need to be very close for any benefit, but the gradient is steep so the suitable UVB zone is very small. So I would certainly not recommend this as a vivarium lamp. " - Dr. Frances Baines

Now lets compare to an Arcadia Lamp.

I dont know that I agree with that being a bad spectrum. China also makes our UVB Flos, the peaks are fairly similar.


I'm not sure the beam angle, or why she says very close to it, seeing the other readings. Were going to have to wait for a full write up, that is more in depth, to get the full picture. The flos are nothing like sunlight either, they are less like sunlight than ALOT of LEDs. that spikeyness of Flos is super unnatural.

I will keep the thread updated, and if anyone with a 6.2 wants to pick up the bulbs and play with them, let us know your findings!

 

[skoram]

The 6.5, will give us the UVI, however that is a result of UVB and UVA, of which there may be too much UVA. 30% UVA, on the 6% and the 10% Arcadia, so 30% UVA equivalent is what we want.

Can you share why you think the 6.5's UVI index includes UVA and may therefore be inaccurate for measuring UVB levels for D3 synthesis? I own a 6.5 meter so this is of particular concern to me.

Solarmeter's own 6.5 product page states that, although it covers 280-400nm, the response is heavily weighted towards the D3 synthesis spectrum, moreso than the 6.2 according to these graphs from their site:



From the graphs you can see that there is almost no response from either meter above 320nm and the 6.5's response peaks between 295 and 300nm, as opposed to 300+ for the 6.2.

Just curious what your thoughts are on this and if you have any reasons or evidence to doubt these claims. I am a natural skeptic and always question what companies tell us through their marketing.

 

[nightanole]

UVA

315–400nm

UVB

280–315nm

Im with skoram, Both units are peak sensitive at peak D3 production. the 6.2 is 3 times more sensitive at 310nm vs the 6.5. But both are measuring zilch by 320nm

 

[cyberlocc]

Can you share why you think the 6.5's UVI index includes UVA and may therefore be inaccurate for measuring UVB levels for D3 synthesis? I own a 6.5 meter so this is of particular concern to me.

Solarmeter's own 6.5 product page states that, although it covers 280-400nm, the response is heavily weighted towards the D3 synthesis spectrum, moreso than the 6.2 according to these graphs from their site:

View attachment 259419View attachment 259420

From the graphs you can see that there is almost no response from either meter above 320nm and the 6.5's response peaks between 295 and 300nm, as opposed to 300+ for the 6.2.

Just curious what your thoughts are on this and if you have any reasons or evidence to doubt these claims. I am a natural skeptic and always question what companies tell us through their marketing.

That graph is kind of decieving. It's weighted, as they said, not void of UVA readings as your graph makes it seem. See this one, with the calculation from the non R.

It still reads UVA, just the weighting of UVA is low. However if you have extreme, amounts of UVA like this LED does, it will throw off the readings.

Like Night said, this is extremely unnatural light in this LED.

They are made for the sun, and to balance sunburns in human beings, that is what UVI is for. Dump on unnatural levels of UVA, and the reading is off.


It tells you that right here, again. From their datasheets.

6.2: "Response 280-322 nm UVB"

6.5: "280-400 nm Diffey Erythemal Action Spectrum"

Not measuring Zilch at 320, on the 6.5.


Also the EAS, they are basing off of, isnt about D3 (It does seem to sort of apply to us, and fit our needs pretty well) but EAS is about Humans and Sunburns. Which is what the 6.5 was designed for, to measure sunlight to see how long you can sit outside until you get a burn, or tan, depending on skin type.


Again not knocking the 6.5, its great with Ferguson zones ect, its the best we have. What its not good at, is ensuring a Lamps design, they even say that. to use a 6.2 to ensure Lamp Design parameters. With this LED, thats what we are trying to do. So A 6.2, and UVA meter, would be better, or better yet as fran has done, a spectrometer with UV abilities.

 

[skoram]

That graph is kind of decieving. It's weighted, as they said, not void of UVA readings as your graph makes it seem. See this one, with the calculation from the non R.
View attachment 259438

It still reads UVA, just the weighting of UVA is low. However if you have extreme, amounts of UVA like this LED does, it will throw off the readings.

Like Night said, this is extremely unnatural light in this LED.

They are made for the sun, and to balance sunburns in human beings, that is what UVI is for. Dump on unnatural levels of UVA, and the reading is off.


It tells you that right here, again. From their datasheets.

6.2: "Response 280-322 nm UVB"

6.5: "280-400 nm Diffey Erythemal Action Spectrum"

Not measuring Zilch at 320, on the 6.5.


Also the EAS, they are basing off of, isnt about D3 (It does seem to sort of apply to us, and fit our needs pretty well) but EAS is about Humans and Sunburns. Which is what the 6.5 was designed for, to measure sunlight to see how long you can sit outside until you get a burn, or tan, depending on skin type.


Again not knocking the 6.5, its great with Ferguson zones ect, its the best we have. What its not good at, is ensuring a Lamps design, they even say that. to use a 6.2 to ensure Lamp Design parameters. With this LED, thats what we are trying to do. So A 6.2, and UVA meter, would be better, or better yet as fran has done, a spectrometer with UV abilities.

Yes, they don't claim that the 6.5 (or even 6.2 for that matter) don't measure above 320 nm, just that the response is heavily weighted towards the D3 sythesis spectrum. In the graphs I posted you can see that there are dots along the bottom from 320 all the way up to 400nm. One could argue that the logarithmic graph you posted is more deceiving because the unit gaps along the Y axis are not remotely consistent (hence, logarithmic). Each "step" up increases by 0.09, 0.9, 9, and 99. So the "slope" of the graph does not give a correct idea of the relative values as you move down the X axis.

I see your point about the UVA though. Given that the response above 320nm ranges from 1 to 0.1, those LEDs must give off a ridiculously insane amount of UVA if we assume those high UVI readings are mainly attributable to the UVA leds and not the UVB in the center.

It makes me seriously question why they would design the product like this. On a related note, I know that Digi-key sells (or sold) LEDs now that emit light at very specific peak UVB wavelengths. E.g. there were listings for 290, 295, 298nm UVB LEDs, if I recall correctly. I also read an article about a medical research team that used these LEDs to test D3 synthesis in human skin. They confirmed that the peak wavelengths did in fact match the advertised numbers (i.e. "295nm UVB LEDs" did in fact emit at 295nm peak, etc.). I would love to see someone try to make a strip or array of these LEDs and post the results. I would do it myself if the cost per each LED wasn't so darn high.

Edit: I found a link to the article I mentioned:

https://www.nature.com/articles/s41598-017-11362-2

Here is a link to a 295nm UVB LED sold on Digi-key:

https://www.digikey.com/product-det...tronics/MTSM295UV-F1120S/1125-1368-ND/6575164

$54.60 per single LED . Hoping the prices come down soon so I can try some myself.

 

[Graves923]

Geez... Reading all this is giving me flash backs to 10 years ago in the reef aquarium hobby. LEDs were gaining traction while the mainstream used the proven T5 fluorescent bulbs for the reef lighting. Now, almost everyone uses LEDs for some aspect of reef keeping even though T5s give the best colors ?

 

[cyberlocc]

Yes, they don't claim that the 6.5 (or even 6.2 for that matter) don't measure above 320 nm, just that the response is heavily weighted towards the D3 sythesis spectrum. In the graphs I posted you can see that there are dots along the bottom from 320 all the way up to 400nm. One could argue that the logarithmic graph you posted is more deceiving because the unit gaps along the Y axis are not remotely consistent (hence, logarithmic). Each "step" up increases by 0.09, 0.9, 9, and 99. So the "slope" of the graph does not give a correct idea of the relative values as you move down the X axis.

I see your point about the UVA though. Given that the response above 320nm ranges from 1 to 0.1, those LEDs must give off a ridiculously insane amount of UVA if we assume those high UVI readings are mainly attributable to the UVA leds and not the UVB in the center.

It makes me seriously question why they would design the product like this. On a related note, I know that Digi-key sells (or sold) LEDs now that emit light at very specific peak UVB wavelengths. E.g. there were listings for 290, 295, 298nm UVB LEDs, if I recall correctly. I also read an article about a medical research team that used these LEDs to test D3 synthesis in human skin. They confirmed that the peak wavelengths did in fact match the advertised numbers (i.e. "295nm UVB LEDs" did in fact emit at 295nm peak, etc.). I would love to see someone try to make a strip or array of these LEDs and post the results. I would do it myself if the cost per each LED wasn't so darn high.

Edit: I found a link to the article I mentioned:

https://www.nature.com/articles/s41598-017-11362-2

Here is a link to a 295nm UVB LED sold on Digi-key:

https://www.digikey.com/product-det...tronics/MTSM295UV-F1120S/1125-1368-ND/6575164

$54.60 per single LED . Hoping the prices come down soon so I can try some myself.

Yes I have seen those and mentioned them around a few times. The price is high, the MA are low, Diffusion would prove very difficult with stars.

This LED for the UV is fine, they need to make a strip and less UVA (there is 8 UVA LEDs in this bulb). I would say, 4 UVA 2 of each spectrum, stacked with the UVB closer, in a strip format, with many is the better option.

We could certainly build that, with these bulbs. Yank the boards out and mount them on a strip board, maybe with a diffuser. I want to see what Fran shows, and thinks about them. Then I will wait a bit and see if I can get someone to make a strip, if not I will when I have time and money to play it. Later this summer or so.

Geez... Reading all this is giving me flash backs to 10 years ago in the reef aquarium hobby. LEDs were gaining traction while the mainstream used the proven T5 fluorescent bulbs for the reef lighting. Now, almost everyone uses LEDs for some aspect of reef keeping even though T5s give the best colors ?

They got some LEDs that give good reef look.

The reason that most feel the T5s give better colors than the LEDs is because T5s are spikey and missing so much spectrum. The LEDs everyone pushes for high CRI, but just like with TVs people don't want to see "Natural" they want oversaturated color pop. There is some LEDs that make that happen though.

It's actual more reverse this time. Where the Flos have a more natural UV, and the LEDs are more spikey.

Also, the over saturate color pop T5 reefs do look good lol .

 

[Graves923]

They got some LEDs that give good reef look.

The reason that most feel the T5s give better colors than the LEDs is because T5s are spikey and missing so much spectrum. The LEDs everyone pushes for high CRI, but just like with TVs people don't want to see "Natural" they want oversaturated color pop. There is some LEDs that make that happen though.

Agreed. Once I switched to LEDs i was disappointed in the lack of color my corals gave but I noticed a significant increase in growth and polyp extension. Now i use a combo fixture with both T5 and LEDs for color and coral health.

This makes me curious to see how the chameleon react to switching to full LEDs for UVB, especially in chameleons that don't get any natural sunlight.

 

[cyberlocc]

Agreed. Once I switched to LEDs i was disappointed in the lack of color my corals gave but I noticed a significant increase in growth and polyp extension. Now i use a combo fixture with both T5 and LEDs for color and coral health.

This makes me curious to see how the chameleon react to switching to full LEDs for UVB, especially in chameleons that don't get any natural sunlight.

Well, these LEDs are not perfect, but in time we can get something better.

The Flos spikes in UVA is really not ideal. Chameleons can see UVA 350-400nm. So using the right UVA LEDs, packed with the UVB one, we could setup a better spread. Of more natural UV light. Combo that into a fixture with Tri R diodes (I have a thread on here somewhere about Tri RS). Or even better through smaller UV diodes into Tri R cobs, we would have the best reptile light ever.

The Tri Rs, come the closest to recreation of sunlight of any light source ever. They are just lacking UV, and Infared.

I loved these LEDs tech and spectrum before seeing this video, but this video defiantly sold me lol. I'm planning on replacing all my cobs with these.

 

[nightanole]

hmmm. might have to bust out the soldering iron and decommision a few UVA leds. It would be nice if it wasnt an entire strip in series so we could just disable a leg. Or maybe the driver is smart enough that we can just replace an LED with a jumper? Im not up on my tech, my last project was just a constant current source driver with leds in series and you could add or remove as needed.

 

[cyberlocc]

hmmm. might have to bust out the soldering iron and decommision a few UVA leds. It would be nice if it wasnt an entire strip in series so we could just disable a leg. Or maybe the driver is smart enough that we can just replace an LED with a jumper? Im not up on my tech, my last project was just a constant current source driver with leds in series and you could add or remove as needed.

The UVA amounts might be fine, it's just more than the Flos I think quite a bit more.

The too much UVA comment was a guess. It could be something else entirely causing the issue we seen.

We have Fran saying it doesn't produce much, and the spot is narrow, and we have this guy showing 3uvi right at the fixture, and 20 at 6 inches, the only thing that makes sense is the UVA LEDs To me, but we need to play with it.