0

I am not even a novice, but was directed here by several companies when they couldn't help me with finding a tool.

I am setting up an experiment to determine which glow pigments are brightest. Full details of the controls and experiment is on ElementalBreakdown.com

I think what I am testing for is luminance.

The tool I need should be able to be granular down to at least .01 millilumens, if that's even the right unit of measurement.

The object I want to test is a small metal bead with various brands of glow paint applied over an adhesion primer and charged with the same light source.

I have sorted out a lot of details like making sure the actual phosphorescent (not an accurate term since the new paints use strontium aluminate or doped Europium) pigments are all the same size (50 microns).

So now I need to figure out what units to measure in and probably upgrade the cheap lux meter I have.

I don't have experience building electronics so would need a ready made tool or assistance.

Are any current tools sensitive enough to get accurate readings from a 50-80mm sized source at about 3 feet? I'd like to take readings at several intervals like 5, 20, and 60 minutes.

Please help me if you know of such a tool or have advice on what and how to measure. I currently have samples from 5 of the largest companies all claiming to have the brightest product.

ElementalBreakdown
  • 11
  • 1
    Hi EB, welcome to EE.SE. Well, your question isn't a good fit for this electronics-design Q&A site. But I will mention, to measure very low-light levels, special equipment is needed. This is because you'll literally be counting photons. Typically this would be done with a photomultiplier tube, a very specialized piece of gear. These also use a high voltage (thousands of volts) to operate, making them dangerous, even deadly to build and tinker with. I'd suggest looking for a commercial PMT solution. – rdtsc Aug 13 '20 at 17:32
  • 1
    @rdtsc 0.01 millilumens is still trillions of photons per second, whereas PMTs typically saturate at millions to maybe a billion photons per second, so his targets are probably too bright for a PMT. In this case, a CMOS detector is going to work better due to higher dynamic range. – user1850479 Aug 13 '20 at 18:26
  • @user1850479 I see no reason why not to use a PMT for accurate, repeatable, wide-band intensity measurement. Too bright? Increase the distance or reduce the cathode voltage. An imaging sensor is going to return noise for every pixel except those in-focus. Too close = no focus, too far = few pixels. Let me guess - use digital zoom and extrapolate a "usable" post-process value in python? What about spectral linearity? – rdtsc Aug 13 '20 at 21:56
  • Here is a paper on correcting CCD's and another related to this topic. – rdtsc Aug 13 '20 at 21:56
  • @rdtsc PMTs are orders of magnitude more expensive, fragile, have limited dynamic range and will be much more complex to align and read out than a camera. With enough money and attenuation you can make one work, but it would be much better to pick a more appropriate sensor for the power level. An array sensor is probably a better idea as well since he has an extended source he wants to measure. – user1850479 Aug 13 '20 at 23:03

1 Answers1

1

The tool I need should be able to be granular down to at least .01 millilumens, if that's even the right unit of measurement.

Lumens are how bright something looks to the eye, so unless everything you are measuring is exactly the same color, I would probably not use that unit. Power or number of photons per second would be my first choice.

Are any current tools sensitive enough to get accurate readings from a 50-80mm sized source at about 3 feet? I'd like to take readings at several intervals like 5, 20, and 60 minutes.

I would use a camera or smartphone on a tripod. The intensity values in RAW mode are proportional to the number of photons detected. Set your exposure time long enough to see however dim your object is, and be sure that your background is completely black. Note that JPEG is gamma corrected and so pixel value is not linear with intensity.

user1850479
  • 16,883
  • 1
  • 21
  • 46
  • This sounds interesting for a similar side-project. Could you elaborate on how you controlled the phone cam (custom code?), and perhaps the utility to parcel out intensity data from a RAW file? – Chris Knudsen Aug 13 '20 at 19:39
  • I envisioned the "counting photons" as what I am ultimately trying to do. As a side note, I was sure to purchase the 2nd blackest substance on Earth right behind Vantablack which is a pigment made by the artist Stuart Semple called "Black 2.0" which is a "super matte, super flat" acrylic that absorbs 99.(undisclosed) as opposed to 99.6% for Vantablack. – ElementalBreakdown Aug 14 '20 at 18:25
  • @ElementalBreakdown The number of photons per 10 microlumens varies by wavelength, but the minimum value is 41 billion photon/s at 555nm reaching into the trillions at further wavelengths, so you do not want to use photon counting. You don't need Vantablack, just something blackout cloth around the experiment to keep the room lights out. – user1850479 Aug 14 '20 at 18:38
  • @ChrisKnudsen In my professional work I use or build my own sensors. If you want to automate a phone, there are apps for that. For raw files, you can open them in matlab or similar: http://www.rcsumner.net/raw_guide/RAWguide.pdf – user1850479 Aug 14 '20 at 18:41
  • Chris, do you offer your services to build a sensor? Is there a consumer level app you might suggest? Thanks for the reply. – ElementalBreakdown Aug 16 '20 at 16:34
  • I'm not Chris, but I don't see any use for customization here. You're essentially just asking for a camera, and you can get those anywhere. – user1850479 Aug 17 '20 at 01:51
  • @ElementalBreakdown I have a purely hobby-level interest in this as a side project for myself to learn about measuring light intensity filtered by fluids, as well as developing custom code on phones. I'm afraid that in my current state of understanding, I would not be of much help to you. – Chris Knudsen Aug 17 '20 at 12:15
  • Thanks Chris. If it is of any consequence, the samples I am dealing with are fluids of various viscosities, and the pigments are all of a known size (30-50µ). I'm wondering if your code runs from a specific app or console app (that's probably the wrong terminology, I'm thinking about windows and not Mac that I am presently using where you could bring up the console for code) ? – ElementalBreakdown Aug 18 '20 at 13:33