Help with setup selection for Photocatalytic laboratory testing via LED

Hi there, please excuse my ignorance. I have no experience with LEDs.

I am building a single top down light source stand that will light a Photocatalytic Concrete sample while submerged in water. The water will be mixed with various different pollutants to test the effectiveness of the self cleaning concrete sample. The concrete sample is small 55mm wide x 25mm high, circular shape. The photocatalytic property activates in two different parts of the light wavelengths, UV 365-380 and the Visible wavelength (5000k) needs to encompass as much of the vis wavelength as possible so a 97CRI and up would be best. The test will show both lighting conditions and the potency of the catalytic reaction under UV and Vis light (5000k). So I would need two LED setups to do both tests individually. This is for a laboratory environment test. The lighting test will run for a period of up 7 days without interruption on each setup. So proper cooling, etc needs to be accounted for. The water pollution degradation process will be visually recorded and then compressed into short time lapse videos for display in youtube and on our website for visual reference.

The chosen LED’s for my project are the:
1537-LZ4-V4UV0R-0000-ND For the UV wavelength test the OSRAM SYLVANIA LuxiGen™ LZ4
‎897-1422-ND For the Visual wavelength test the Seoul Semiconductor SUNLIKE COB SAWS1566A

What I need assistance with is the best lense, driver, holder, clamp, reflector, heatsink, etc selection choice for my application. The lense and reflector setup would focus the light over the 65mm area. And needs to keep the light as pure as possible to keep the test data accurate.

Thank you in advance for any advice or assistance in this regard.

This being for experimental purposes rather than a commercial product, I’d suggest that expediency and sufficiency would be primary concerns. Toward that end, I’d suggest a LCM-40DA driver, LMH020-REFL-0000-0000062 reflectors, and DUALLED-13050 reflectors, and some HSC67-6G thermal grease. A few holes would need to be drilled & tapped for assembly, but one can probably do so in less time that would be required to identify off-the-shelf parts that would be mutually compatible.

Please note that due to differences in the efficacy of the chosen emitters, a proper experiment would require making radiometric measurements of incident light at the sample surface, to allow accounting for the effects of illuminant intensity.

1 Like

Hi Rick,
Thanks for the fast reply.
I should have mentioned that the lighting tests intensity should attempt to mimic the sun.
Which is why I chose 5000k for mid day equivalent light color.
I chose the luxigen because it was the strongest in lumen UV led for the range needed I could find having a 41lm/W being close to the average of the suns UV at mid day being 45lm/W.
Same with the Sunlike cob, chosen because of the high cri and its 105 lm/W being close to the suns 98lm/W.
Although there is a discrepancy between the values. The degradation rate of pollutants in the water over recorded time will give a roughly accurate idea of the time in which the pollutants will be 100% degraded by sunlite.

I hope this explains the reason for the test more clearly.

Also I require the off the shelf compatible parts for each setup. Driver, holder, clamp, reflector, lense, heatsink, etc because I need to spend R2000 on my first order to get the shipping free. The shipping cost of less than added brings my total to closer on R3000. So expediency and efficiency are not my primary concerns for this particular order. I live in South Africa and the shipping costs are high.

The time trials of the photocatalytic reaction will test two different concrete photocatalytic additives. One being UV only the other Vis light. From this test a good comparison can be made between the two and by data justify the price difference.

1804140103 or something like it might be an added convenience for dealing with the visible emitter. Other than that, I don’t know that my suggestions would change.

By “expediency and sufficiency” I mean something that can be obtained quickly and which is capable of meeting the end need, as contrasted with a solution that would pursue optimizations common in commercial products such as cost, efficiency, size, etc.

Availability of complete off-the-shelf product systems allowing assembly of a luminaire with no additional machining is limited in the visible space, and likely nonexistent for your chosen UV emitter due to its uncommon form factor/electrical interface. The combination of potential hole patterns, heat sink shapes/form factors, emitter sizes, etc. is nearly infinite, and the commercial market for such solutions is limited. Especially so for UV applications, due to the much smaller total market compared to visible lighting applications.

Two each of the reflectors, heat sinks, and drivers above, plus some thermal grease and a visible LED mount should put you above the threshold for discounted shipping. If for some reason it doesn’t, we carry a lot of generally-useful items (batteries, for example) that a person can pad an order with.

The emitters in question would appear, to a first approximation, to be in the general ballpark for solar similarity assuming near-field use. Surface irradiance as a function of wavelength is neither a stable nor uniform quantity, so some consideration should be given to the actual light levels needed.

Finally, the reference to lm/W figures in context of a UV LED seems odd; since lumens are a unit of measure used for measuring human visual response, the lack of human visual sensitivity to UV should put that figure at zero…

1 Like

I am going to take your advice and purchase the parts you recommended.
I see the current output max on the LCM-40DA driver is 1.05A and the current max on the Sunlike SAWS1566A is 1.55A. Do I need a driver with higher max output?

I did a conversion from the suns 45W/m2 for that same radiation integrated over the UVA range between 315nm and 400nm to a lumen value. I know its not accurate in the sense of being visible but used it to compare against the LED’s. I got the rough values from here.

Would this driver be good to use ‎‎IDLC-65-1750‎, it has a max current of 1.75A?

I would not recommend it; its listed output voltage range is not well matched to either LED, and you’d be likely to damage the emitter. The test current value indicates the drive level at which most device characteristics are measured, and is generally chosen to be a suitable operating value; “maximum” values communicate a point beyond which damage is likely, and generally should not be understood as an operating limit.

1 Like

Thank you for the shared knowledge and help Rick.
This experience has been educational. Kudo’s to you my friend.

Purchase made.

Respect and regards,