How about using empty Biro's?

what are Biro's?

I reported a lensing effect from the curved top of the gel caps, and wanted to answer the question of whether the fluorescence could be improved if the top wasn't curved. I made a small tray for open gel caps and did another olive oil test. The results matched identically to the #4 closed cap:

Results match the closed #4 cap almost exactly.

I then tried to see if it would improve the results with bigger caps, like the 0 size. nope, no measureable fluorescence, even with the lensing effect gone.

Hi, Mat - I've noticed that olive oil fluorescence is much dimmer than crude oil fluorescence, and I shipped you some crude samples this morning. That'll help somewhat. See these two vials below, with the olive oil being the pinkish one:

But the SANM camera seems to be much less sensitive than the SYBA cam for some reason; look at these two scans:

SANM: https://spectralworkbench.org/analyze/spectrum/33804

SYBA: https://spectralworkbench.org/analyze/spectrum/33797

Both uncalibrated, but here's the graph for intensity:

And for the SANM I cut the slit wider, too -- like 1.5mm! It's too bad. I wonder if it's just the SANM cam that's giving us trouble and that we don't necessarily need a MUCH better camera, just a different one.

Hmmm... yeah. The SYBA was a good camera. it has 2x the silicon. I've been searching for 1/3" CCD security cameras-- that would put us back to the size of the SYBA.

that is heartening to see, that is a nice response curve for the oil

Woohoo! Mathew, this is brilliant. Could you include a closeup shot of the gel cap w/out liquid -- or I'll just search online -- are they relatively clear? I wonder what the optical properties are like ... looks like a really neat little mixing container in which to do colorimetry experiments, too -- collect a water sample in a larger container, then pour it out into a bunch of little gel cap 'vials' and use strips (or sub-strips) of pH testing paper, etc ... supercool!

Ah, found a pic:

yes-- I used even smaller ones, the #4 caps. they have a nice "snap" closed that seals them when pushed together.

gelatin passes UV, I believe. they are quite transparent. I don't think you can put water in them though-- they're very water soluble :-(

OK- complication-- gelatin will fluoresce when hit with UV: http://www.sciencedirect.com/science/article/pii/S0315546387710980

Oh a whim, I just searched for 'gelatin reagent' and got this:

"The Thermo Scientific™ Remel 12.5% Gelatin Reagent, Acidified Mercuric Chloride is for detection of gelatinase activity of microorganisms. - See more at: http://www.thermoscientific.com/content/tfs/en/product/remel-12-5-gelatin-reagent.html#sthash.1KgRYJy2.dpuf"

The idea behind gelatin capsules is that they dissolve in water + the gut, yeah? Could we exploit some of these properties as an indication of the presence of a microorganism? First thought was: put a colored dye in the capsule, put capsule in other non-colored liquid (one that doesn't dissolve gelatin? or not quickly?); then expose system to microorganism ... when the color of the entire container turns the color of the liquid, we've reached a threshold ...

I also tried using a reflective metallic surface behind the cuvette to make the fluorescence brighter:

It was hard to see, naked eye or by camera, any difference. In part, this is because the laser travels through sample material and gets dimmer the further along the beam path it goes. So I bracketed the cuvette with reflective tape with two cuvettes without it:

However, you can see in the middle cuvette, there's a big dropoff in brightness where the reflective material ends:

So I cut a narrower strip of reflective material, and used a Canon camera to do a very low exposure test. It does seem to increase brightness quite a bit!

I think next step would be a test in a spectrometer, to see if, say, with the SYBA cam above, how high we can get the brightness vs. an un-modified cuvette. The reflective material could go behind the container, too, BTW -- it prob. wouldn't have to be adhered directly.

Not only excellent research- that last photo has some high-modernist artistic merit!

Gel caps fluoresce and we can't use them.

And styrene is dissolved by oils. and we can't use it either.

I'm looking to test out a new board since it seems like the consensus is that the SANM isn't as good for florescence. I'm trying to compare other SYBA webcams comparable in price but unfortunately, the specs for the former webcam, the SYBA SY-CAM63014 do not include the the same descriptors as the ones I am looking at such as the SYBA CL-CAM50001. Does anyone have an opinion on which factors are making the SNAM camera less sensitive or is it a manufacturing issue of better parts or chips. I know @matthew mentioned that the SYBA has 2x the sensor size, but what is the descriptor that would tell you that? Of course in the end, this might just require some good old fashioned experimentation but if you have any thoughts on what I should be looking for I'd appreciate it.

I'm also contacting SYBA

We had a truly terrible experience working with and ordering from SYBA, and one reason we moved to a different camera was to find a company which was easier to work with. So if you do a webcam search, I'd encourage you to look elsewhere -- and please do share what you find, since we're also sourcing new cameras for the Homebrew Oil Testing Kit! (I'm sure @natalie would be very happy to share notes, too, as she's doing a lot of our sourcing!)

Thanks Jeff, I'll connect with @natalie. I'm doing a bit of research by taking a look at people who are hacking cameras. As I've looked at images of the exposed internals, one thing that is occurring to me is that the size of the sensor is always bigger in stand alone webcams vs. laptop. And that makes a lot of obvious sense. So I'm thinking I might focus on those instead of sourcing new laptop cameras.

The sensor size is not definitive. The Infragram webcam is more sensitive than the SYBA, but has the same sensor size as the SANM camera. Its also a pain in the butt to get the filters out of.

Another thing to look out for in camera sourcing is UVC implementation. I've got a stack of cameras here that work on windows and nothing else.

gotcha. Okay. You're referring to the VidView JDEPC-OV04? That's the model number on the datasheet next to "Datasheet for Infragram Webcam" at the bottom of the infragram page. Has it changed or is that still the one?