The Future Has Arrived

[colinbm]

Girls & Boys.....The Future Has Arrived.

Here is a #1 of what I hope will be a new approach to UV LED Lighting.

This #1 has four different wavelength LEDs on each 20mm star, they are 365, 375, 385 & 395nm & the output is 10 Watts.

I have assembled nine of these LEDs to produce nearly 90watts of mixed UVA.

This is all in development at the moment.

Enjoy the dream....

 

 

[ulf]

Nice build.

I like that you are taking the cooling needs seriously.

Hope you get good Image results from it.

Is it intended for fluorescence photography?

 

I had a similar dream that stalled.

In my dream the current for each led-type could be independently controlled to shift the colour-balance.

 

You are not producing almost 90W UV-A, but still dangerous amounts if you are not protecting yourself.

I think the conversion efficiency to light at best will be 30-35% and 25W UV-light is still impressive.

 

My dream stalled as I went down the path of UV-converted flashes instead.

The advantage with flashes are very short exposure time and portability.

[colinbm]

Thanks ULF, & for your help getting me here....

I intend to use it for both UV & Fluorescent photography & have filters in front.

I have also gone down the path of UV-converted flashes, I will see what the advantages are either way.

Cheers

Col

[dabateman]

The future looks messy like my grad school builds. You just need an optical bench and a floating Aluminum table and your all set.

 

Doesn't look like a hand held light that you could run around with either.

I hope it works for applications.

 

I checked Amazon and no longer see the cheap 385nm LED bulbs I was considering. I wouldn't mind taking a LED bulb apart to see if it could drive something like this. My 365nm ones I think are 9W for the full LED grid.

The 385nm ones were 7W. I still see the 7W 395nm ones though. But not as cheap as I would want.

 

For me the ideal set up might be 345nm, 365nm, 385nm and then maybe a 405nm or not. The 340nm leds are still way too expensive though.

I wonder if Jonathan ever finished building his. It was similar to this 340/365/385 idea.

 

[colinbm]

Thanks Dave, actually I put in a request just like this today, so we'll have to see how it pans out.....

[JMC]

I wonder if Jonathan ever finished building his. It was similar to this 340/365/385 idea.

 

No David, I gave up - my skill level wasn't high enough to get something working which I was happy with. I donated all my LED stuff to Ulf.

[ulf]

Doesn't look like a hand held light that you could run around with either.

 

If the portable generator or cable drum isn't to heavy you can run around at leat a bit.

 

For me the ideal set up might be 345nm, 365nm, 385nm and then maybe a 405nm or not. The 340nm leds are still way too expensive though.

 

Are any of the 340nm or 345nm LEDs reaching a meaningful conversion efficiency like the 365nm LEDs?

 

Most marketing specifications show the maximum input power or current handling ratings, while it is the optical emission power that is the interesting parameter for us.

That too is in stated in the data-sheets, but sometimes overlooked.

Naturally the input power must be handled too, but only with sufficient cooling.

 

I guess that the drive to develop efficient UV-LEDs has mainly been for resin- and glue-curing and that works well with 365nm or longer wavelengths.

 

What mass application would benefit from very many LEDs with shorter wavelengths? They might be there, but I know too little about them.

[Stefano]

Regarding 340 nm LEDs... I have one, I will write about it, wait for it. They are roughly 2-3% efficient.

[ulf]

Regarding 340 nm LEDs... I have one, I will write about it, wait for it. They are roughly 2-3% efficient.

That is a tiny bit lower that the 30% efficiency of the 365nm LEDs

[Stefano]

What happend to the bottom-right chip of the module in the first column, second row? Did you burn it?

[colinbm]

What happend to the bottom-right chip of the module in the first column, second row? Did you burn it?

No, I touched it with the soldiering iron......

[Cadmium]

Impressive.

[colinbm]

Thanks Steve

[ulf]

I intend to use it for both UV & Fluorescent photography & have filters in front.

 

I think the filters will become quite hot after a while, when absorbing all the energy not transmitted through the filter.

As the light is in a wider wavelength-band it will be difficult to find suitable filters and they must be BIG!

 

With those LEDs your filter must pass light close to 400nm, including some violet. A thick UG5, 3 or 4mm might be the best glass, but as this is for illumination ZWB3 might be good enough.

https://www.ebay.com/itm/100-100-3-0mm-ZWB3-UG5-U-330-254nm-UV-Pass-Filter-Visible-Light-Absorbing-Glass/274103207213

The price for proper UG5 will likely be more than 10 times higher.

 

This seller is OK and has good service. I hope that he still is active.

[dabateman]

Yes still active he just recently cut me some 3mm thick 28mm diameter zwb1 and shipped them to me.

Awaiting for them to arrive in the mail now.

[Stefano]

I think the filters will become quite hot after a while, when absorbing all the energy not transmitted through the filter.

As the light is in a wider wavelength-band it will be difficult to find suitable filters and they must be BIG!

I remember that putting a 2 mm thick ZWB2 filter ~1.5 cm from my 10 W 365 nm LED heated it up to ~50°C (I measured it). Does heat degrade ionic glass?

[colinbm]

Thanks for the filter link & ideas....I am looking at the filter problem at the moment....

The actual LED set-up & liquid heat-sink runs cool, but the heat from the front of the actual light is 50 deg C, I am not sure yet if this is too hot for UV short pass filters OR what the rise in temperature will be when I enclose the light for attaching the filter ?

Experiments are continuing....

[colinbm]

Thinking out aloud here...I wonder if I made a dead air pocket with two fused silica filters, if that would block the heat to the chosen filter...?

[Andy Perrin]

Thinking out aloud here...I wonder if I made a dead air pocket with two fused silica filters, if that would block the hear to the chosen filter...?

The filter heats up because it absorbs the out-of-band light, which will happen regardless of any air pockets. The heating mechanism is that it absorbs the light and converts it to heat, not thermal conduction or convection from the LED. If you use a larger filter further from the LED, the temperature won't go as high because the beam will spread out. (Just like in Stefano's experiments, in fact. The idea for those was ok, it's just not a good idea to use your face as a thermometer.)

[Stefano]

Your LEDs have lenses. What is the viewing angle? If it is ~70° like mine, you will have to go further from the LEDs to use a bigger filter.

[Stefano]

 

The filter heats up because it absorbs the out-of-band light, which will happen regardless of any air pockets. The heating mechanism is that it absorbs the light and converts it to heat, not thermal conduction or convection from the LED. If you use a larger filter further from the LED, the temperature won't go as high because the beam will spread out. (Just like in Stefano's experiments, in fact. The idea for those was ok, it's just not a good idea to use your face as a thermometer.)

Does the 15-20% of absorbed UV also account for a significant portion of the heating? Usually UV LEDs don't produce very high amounts of visible light, still enough to heat a filter.

[colinbm]

The filter heats up because it absorbs the out-of-band light, which will happen regardless of any air pockets. The heating mechanism is that it absorbs the light and converts it to heat, not thermal conduction or convection from the LED. If you use a larger filter further from the LED, the temperature won't go as high because the beam will spread out. (Just like in Stefano's experiments, in fact. The idea for those was ok, it's just not a good idea to use your face as a thermometer.)

Definitely no faces for temp sensors....

Stefano, the LED is 60mm square & I have some 72 mm filters that I plan to use....I just don't want to break them unnecessarily...

I could mount the filter at say 300mm from the LEDs, with a tube attached & add a ventilation chimney with a 12v computer fan to pull out the heat....?

Cheers

Col

[Stefano]

 

Definitely no faces for temp sensors....

Stefano, the LED is 60mm square & I have some 72 mm filters that I plan to use....I just don't want to break them unnecessarily...

I could mount the filter at say 300mm from the LEDs, with a tube attached & add a ventilation chimney with a 12v computer fan to pull out the heat....?

Cheers

Col

If your LEDs emit at ~60-70°, you would need a filter ~40 cm wide at that distance to avoid losing as much light as possible.

[Andy Perrin]

Lol 40cm filter!

 

Stefano, yes for sure any absorbed UV will also contribute to the heating. That may actually be the dominant part since you are right that there won’t be much out of band light in this case.

[dabateman]

I never thought twice about the heat off my lights with some filters. But what I have seen is placing a 52mm BG39 filter say 2 cmm away from a white LED bulb causes it to get very hot. Hasn't broken yet. But does oxidize quickly near the center. Maybe due to cooling and condensation forming.

A ZWB1 filter on a 365nm light doesn't get very hot 2 cm away. Also hasn't had any rapid oxidation form.

 

Lee gels work well, don't melt and can isolate some bands for you. I will have to look at my book again but remember a 550 or 505 number gel was great for UV. Has an unusual high UV transmission bump, followed by the yellow curve. So if you just have UV leds would work to cut out blue leakage.