UV LED's, drive voltage and current questions

Hi all.

I am building a simple UV LED lamp and 100 of these have just arrived http://www.ebay.co.uk/itm/230930659493 and I am testing them on my bench supply, but either I'm being totally stupid or the data on the ebay listing is not correct.

If I drive them with constant voltage (no current limit) at 3.1V one LED is drawing almost 40mA, then if I drive in constant current mode at 20mA I get no light output, I have to increase the limit to around 35mA+ to get any decent light output.

I'm confused and basically trying to select resistor values but not sure what I should be driving these at?

What is the foward voltage you measure with a meter:
Try 5V and 100R

400nm is close to the limit of a humans eyesight so it is expected that these will not look very bright when driven at the correct current.

The required current is stated at 20mA so I would be using that.

I don't know it really doesn't seem to be outputting anything at 20mA and if none of the light is visable then I'm wasting my time :slight_smile:

I did have one running for a few hours at 30mA so maybe that is safe?

I want to run them from 12v in strings of 3.

dtokez:
I did have one running for a few hours at 30mA so maybe that is safe?

As far as I can tell, there is no datasheet available. Which means the seller is the only one who can answer that question.

Bear in mind that 400nm is at the threshold for UVA radiation which, apparently, can cause eye damage.

dtokez:
I'm confused and basically trying to select resistor values but not sure what I should be driving these at?

You should be driving them at 20mA.

(which can be a voltage anywhere between 3.0 and 3.2V, according to the listing)

dtokez:
I have to increase the limit to around 35mA+ to get any decent light output.

How are you measuring the output?

If I drive them with constant voltage (no current limit) at 3.1V one LED is drawing almost 40 mA, then if I drive in constant current mode at 20 mA I get no light output, I have to increase the limit to around 35 mA+ to get any decent light output.

more current, more bandwidth. A typical human eye will respond to wavelengths from about 390 to 700 nm. An other factor is perception of the visible spectrum, differences among gender and age. The UV photo diode will help to measure the output.

dtokez:
I am building a simple UV LED lamp and 100 of these have just arrived http://www.ebay.co.uk/itm/230930659493 and I am testing them on my bench supply, but either I'm being totally stupid or the data on the ebay listing is not correct.

Features:
Size: 5mm
wavelength(nm): 390-400
Forward Voltage (V):DC3.0~3.2
Forward Current (mA):20
View Angle: About 25 degree
Luminous Intensity: 800-1000mcd
Lens Color:Water Clear
Emitting Colour:UV

"Luminous Intensity: 800-1000mcd" could be misleading, "100-200mcd" is more closed. The drive current is 20 mA, practically 15-18 mA.

dtokez:
I am building a simple UV LED lamp and 100 of these have just arrived http://www.ebay.co.uk/itm/230930659493 and I am testing them on my bench supply, but either I'm being totally stupid or the data on the ebay listing is not correct.

if I were you, I will build simple UV LED lamp with few 1W UV LEDs, but not 100 of T1 3/4, 5mm UV LEDs. This could save a lot of headache.

agree 100%, here since you can not see, but only by feel. but once you feel it, might be a way too late.

100 of these have just arrived

100 of them at 30mA - NOT in my house.

thanks for all the responses guys - all noted. Safety (eyes) is not a major concern since I'm planning to make a lamp to cure UV active nail varnish for my other half so you won't be able to look directly at the diodes and I will also have a countdown timer for a minute or so. I went with the idea of using lots of 5mm LED's because I need a large coverage area - enough to shine evenly over a hand.

I'm still confused, If I drive with 20mA the forward voltage is only 2.84V, and with a constant voltage of 3.1v, the LED draws 35mA. So.. should I still only drive at 20mA? I really can't see any output whatsoever at 20mA I know its UV and I can't see the whole spectrum but looking at it reflected off white paper I don't see anything. Another test I have done is to use a camera to pick up the spectrum and I cant see anything until I hit >25mA.
I will have to test with the nail varnish stuff and see if it will cure it.

dtokez:
I will have to test with the nail varnish stuff and see if it will cure it.

Don't forget to pick a colour that matches your clothes :slight_smile: (That's what my wife says all the time anyway so it must be right ...)

dtokez:
I'm still confused, If I drive with 20mA the forward voltage is only 2.84V

It is what it is.

dtokez:
So.. should I still only drive at 20mA?

Yes.

dtokez:
I will have to test with the nail varnish stuff and see if it will cure it.

That's the only test that makes any sense. Everything else you've done? Wrong.

Before you start to test, the best bet is drop a line to gel manufacturer to ask LED-curable gels specification ( how many watts, wavelength, and time to cure).

The wavelength on LED gel lights is much narrower than that of the UV/compact fluorescent lights. This ­narrow wavelength emits the right amount of the specific UV-A wavelength that’s needed to cure LED-curable gels, which is why LED-curable gels cure faster in LED lights than in UV/CFL gel lights. As most UV gels require the use of 350 nm. to cure, most UV lamps emit a wavelength ranging from roughly 320 to 400 nm. But an LED manufacturer that uses an LED gel that is rated for 375 nm. will use bulbs that emit from only 370 to 380 nm. “LED” stands for “light emitting diode,” which refers to the type of bulb an LED gel lamp uses.

for my understanding UV/compact fluorescent lights use at this application is 40 watts, your 0.02*3=0.06W single UV diode ( even 100 of them, still only 6 W) might take too long to cure.

Nails Magazine

A decent LED is expected to last at least 50,000 hours. If you left it on all day everyday it should last for a little over 2000 days, or something like 5 and a half years.

UV Lamp Life Normal lamp life is 1000 to 3000 hours and under optimal conditions can be up to 4000 hours.

The LED is the way to go!

sonnyyu:
Before you start to test, the best bet is drop a line to gel manufacturer to ask LED-curable gels specification ( how many watts, wavelength, and time to cure).

The wavelength on LED gel lights is much narrower than that of the UV/compact fluorescent lights. This ­narrow wavelength emits the right amount of the specific UV-A wavelength that’s needed to cure LED-curable gels, which is why LED-curable gels cure faster in LED lights than in UV/CFL gel lights. As most UV gels require the use of 350 nm. to cure, most UV lamps emit a wavelength ranging from roughly 320 to 400 nm. But an LED manufacturer that uses an LED gel that is rated for 375 nm. will use bulbs that emit from only 370 to 380 nm. “LED” stands for “light emitting diode,” which refers to the type of bulb an LED gel lamp uses.

for my understanding UV/compact fluorescent lights use at this application is 40 watts, your 0.02*3=0.06W single UV diode ( even 100 of them, still only 6 W) might take too long to cure.

Nails Magazine

Thanks, that is some useful info :slight_smile: I will have a test with a few LED's later an see if it seems to cure it