Some ideas

[Cadmium]

That’s a brick basically... I guess it’s heavy too... did you make it by stacking two 4 mm pieces?

 

Yes, I threw my back out getting into the ring!

Yes, it is two 4mm thick pieces.

Andrea has a U-430 4mm thick filter. Can't find the kink to her test of that right now.

But I think U-340 4mm + S8612 1mm is in that graph I posted above.

[JMC]

Steve, I have 2 of your 4mm U-340 and one 2mm too, Not that I'd advise using them all together of course.....

[Cadmium]

I am not advising using them together either.

I am going to try the 8mm stack with UV-Nikkor when the sun comes out.

I think the most that I ever tried before was U-340 4mm + S8612 .75 or 1mm, and that works, but I think its potential is a little limited by the lens.

I could not find your graphs of all those combinations. Do you know where that is?

[Stefano]

If all of you want to try something, I would really like it. I don't have any Hoya filter now to try myself.

[Andrea B.]

Here is a link to the U-340 composite photo which Cadmium referenced above:

https://www.ultravio...lso/#entry14065

 

There are 4 stack combos using the U-340/4.00 mm filter plus some reference filters. At the time I made that composite, I did not have a full thickness range of S8612 filters. I should retest the U-340/4 with some S8612 in 1.25 or 1.50 mm thickness. The 1.0mm S8612 Adding a 1.0 mm U-340 to the 4.0 mm U-340 did not completely supress the vis/IR leak (whichever it was. Not sure.)

 

Edit: Fell into Typo Vat again. Meant to refer to a 1.0 mm U-340 instead of a 1.0 mm S8612. Distracted!

[JMC]

Steve, will check when I have more internet (am in the middle of the Tasmania highlands at the moment with no wifi and little phone signal).

[Cadmium]

Here is a link to the U-340 composite photo which Cadmium referenced above:

https://www.ultravio...lso/#entry14065

 

There are 4 stack combos using the U-340/4.00 mm filter plus some reference filters. At the time I made that composite, I did not have a full thickness range of S8612 filters. I should retest the U-340/4 with some S8612 in 1.25 or 1.50 mm thickness. The 1.0mm S8612 did not completely supress the vis/IR leak (whichever it was. Not sure.)

 

Here is a link to the U-340 composite photo which Cadmium referenced above: https://www.ultravio...lso/#entry14065 There are 4 stack combos using the U-340/4.00 mm filter plus some reference filters. At the time I made that composite, I did not have a full thickness range of S8612 filters. I should retest the U-340/4 with some S8612 in 1.25 or 1.50 mm thickness. The 1.0mm S8612 did not completely supress the vis/IR leak (whichever it was. Not sure.)

 

Andrea, I don't see any stack in that post that is U-340 4mm + S8612 1mm.

You show one that is U-340 4mm + S8612 1.75mm.

You also show one that is U-340 5mm which does show some IR, because you need about 7.5mm to make that work with no IR on it's own (with no S8612 suppression).

The U-340 4mm + S9612 1mm will suppress to OD5, that should be working.

[JMC]

This should be it Steve.

 

https://www.ultravioletphotography.com/content/index.php/topic/2383-good-practice-and-setup-for-spectrometer-measurements/page__view__findpost__p__19186

[Cadmium]

Yes, Jonathan, that is it! Thanks!

So that is a scan. So I think the U-340 4mm + S8612 1mm should be solid for no IR leak.

 

Way back I tried U-340 4mm + S8612 0.75mm, it worked fine too, here it is.

"LUV U 343" = U-340 4mm + S8612 0.75mm, keep in mind this is shot with Kuri 35mm, limited to a maximum of 320nm UV depth reach.

[Andy Perrin]

What are the top and bottom rows Cadmium?

[Andrea B.]

Andrea, I don't see any stack in that post that is U-340 4mm + S8612 1mm.

 

I should not edit Stickies and respond to topics at the same time. Sigh!

I corrected my error. Meant to refer to U-340/4.0 + U-340/1.0.

I had tested whether adding another millimeter of U-340 thickness would help quell the IR. It didn't.

[Cadmium]

What are the top and bottom rows Cadmium?

 

I think the top row is the same white balance OOC.

The bottom row are white balanced from RAW.

[Cadmium]

Andrea, I don't see any stack in that post that is U-340 4mm + S8612 1mm.

 

I should not edit Stickies and respond to topics at the same time. Sigh!

I corrected my error. Meant to refer to U-340/4.0 + U-340/1.0.

I had tested whether adding another millimeter of U-340 thickness would help quell the IR. It didn't.

 

Yes, U-340 5mm thick, which doesn't suppress IR enough.

[Stefano]

What about Terahertz waves? Call them millimeter waves, submillimeter waves, Tremendously High Frequency band or whatever, but they are beautiful and very hard to detect (THz gap). I perfectly know that I am asking the near-impossible, but I want to see in THz waves with an helicoidal antenna. How can I handle a THz frequency AC low-voltage wave? Diodes are not an option to rectify it, stepping up the voltage with an air-core transformer is also practically impossible (when there is a load on the secondary, the magnetic field generated by it opposes and cancesl out (partially in this case, totally theoretically) the one generated by the primary, in fact lowering its reactance, and so more current can flow through it). Parasitic capacitance (in parallel) is a dead short, and a little series inductance is an open circuit. I was inspired by this video (am I allowed to put the link? I hope so, otherwise I will remove it).

It is a 3 video series, this is the last one.

Can I crank up the frequency?

[dabateman]

What about Terahertz waves? Call them millimeter waves, submillimeter waves, Tremendously High Frequency band or whatever, but they are beautiful and very hard to detect (THz gap). I perfectly know that I am asking the near-impossible, but I want to see in THz waves with an helicoidal antenna. How can I handle a THz frequency AC low-voltage wave? Diodes are not an option to rectify it, stepping up the voltage with an air-core transformer is also practically impossible (when there is a load on the secondary, the magnetic field generated by it opposes and cancesl out (partially in this case, totally theoretically) the one generated by the primary, in fact lowering its reactance, and so more current can flow through it). Parasitic capacitance (in parallel) is a dead short, and a little series inductance is an open circuit. I was inspired by this video (am I allowed to put the link? I hope so, otherwise I will remove it).

It is a 3 video series, this is the last one.

Can I crank up the frequency?

 

See if you can buy a used airport security scanner. Some will just be old enough now to be available. They use terahertz range for metal detection.

 

[Stefano]

Wouldn't buying an airport security scanner be a bit "overkill"? Do I have any chances of building something like in the video, but DIY and with an helical antenna with a diameter of 0.1 mm? I would need a very fine wire to do that. And how can I pick up that signal?

[Andy Perrin]

I would give this a 0% chance of success unless you are an electrical engineer. THz waves are notoriously hard to design electronics for. Unless you buy something off the shelf that does the hard part for you, it’s unlikely to pan out. Also, for waves that small going with a more camera-like geometry and scanning the antenna across the focal plane would be better. Lenses can be made from plastic in THz.

[Stefano]

Even a teeny-tiny voltage signal (possibly rectified, but I am asking a bit too much) would be enough. Can I get something like a microvolt or less?

[Andy Perrin]

Stefano, two things. One, I’m a mechanical engineer, and you really need an EE. Two, at very high frequencies, conductors stop behaving like conductors, and all kinds of parasitic capacitance/inductance/resistance effects are in play. People use waveguides, not wires even. It’s a whole new world. Believe me, this idea has already occurred to me, I got quite into it, and I gave up (at least for now).

 

It’s best to start with parts of the spectrum you know and gradually push at the edges of the envelope than make a jump to a new part of the spectrum where you don’t know what you don’t know even. I started with near IR (it’s the easiest) and then UVA and SWIR. SWIR was a big jump in terms of finding optics that work well, filters and a camera that can handle those wavelengths. For LWIR I bought off-the-shelf stuff so it was easy because FLIR did all the work for me, but there was still the challenge of low resolution.

[Stefano]

Ok, got it. I think that I will probably at least try this, with wires far spaced apart to reduce parasitic capacitance, and minimizing parasitic inductance. Since I don't think that the cutoff is sharp, it should simply get less efficient to use an antenna at higher and higher frequencies. Gradually it becomes basically impossible. Wires are lossy, and waveguides are too small at THz frequencies. All I need is a really small, really weak signal, even if it is noisy etc. I think that at higher frequencies it doesn't become impossible, but simply very difficult, and gradually impossible. Certainly THz waves are the limit, going to higher frequencies is really impossible. We can do this at centimeter waves and millimeter waves. Somewhere someone posted an image from an article that shows "the world" seen at 90 GHz (3.3 mm) waves, with a microwave radiometer. You could see trees, objects and the ground (I think it was made with tiles), which appeared mirror-like. I simply want to push this at the limits.

 

Edit: I didn't read your edit in your post, you said what I meant here. SWIR is a big jump (strange that normal lenses do not work there, shouldn't they be transparent?) and LWIR uses a completely different sensor technology (microbolometers instead of photodiodes). Mercury cadmium telluride should work up to 14 micrometers (according to Wikipedia as always)

[Andy Perrin]

Ok, got it. I think that I will probably at least try this, with wires far spaced apart to reduce parasitic capacitance, and minimizing parasitic inductance. Since I don't think that the cutoff is sharp, it should simply get less efficient to use an antenna at higher and higher frequencies. Gradually it becomes basically impossible. Wires are lossy, and waveguides are too small at THz frequencies. All I need is a really small, really weak signal, even if it is noisy etc. I think that at higher frequencies it doesn't become impossible, but simply very difficult, and gradually impossible. Certainly THz waves are the limit, going to higher frequencies is really impossible. We can do this at centimeter waves and millimeter waves. Somewhere someone posted an image from an article that shows "the world" seen at 90 GHz (3.3 mm) waves, with a microwave radiometer. You could see trees, objects and the ground (I think it was made with tiles), which appeared mirror-like. I simply want to push this at the limits.

Yes, that was ME, in fact, on here! I posted it, and I dug it out of a paper, because, yes, I was quite interested in building one.

 

As I said, this is not the first time I've been down this route. I swear, I am not making this up, I've thought about the problem quite a bit, and I own half a dozen texts on microwave engineering. It is not a particularly easy problem to solve. That doesn't mean I'm completely abandoning it, but as I told you, I think the right way to make progress is to start with the regions of the spectrum that you know and branch out gradually, so that you can feel your way along. My plan when and if I go back to working on it is to start at lower frequencies and try to build something where I can verify the results. It's all very well to make a device, but what makes it science is being able to verify that what you think you are seeing is real.

 

Edit: I didn't read your edit in your post, you said what I meant here. SWIR is a big jump (strange that normal lenses do not work there, shouldn't they be transparent?) and LWIR uses a completely different sensor technology (microbolometers instead of photodiodes). Mercury cadmium telluride should work up to 14 micrometers (according to Wikipedia as always)

The deal with ordinary lenses in SWIR is that (1) like with UV, many of them have coatings that block SWIR, so it's common to lose half your transmission or more, and more importantly, (2) the optical formula is not really correct anymore so far into the infrared, so you get major degradation in the lens quality, both in focus and in aberrations. Better to buy a SWIR lens, which are not all that expensive compared to ordinary lenses.

[dabateman]

To echo, some of what Andy has been saying I have been having a tough time imaging even at 185nm.

The KSS system I have now figured out uses a Hamamatsu 500U or 500S mulialkali photocathode, as the imager manual (page 3):

https://www.google.com/url?sa=t&source=web&rct=j&url=https://usermanual.wiki/Document/03ruvis.475838047.pdf&ved=2ahUKEwjYqLmR9__mAhVJuVkKHRAzBUoQFjAPegQIARAB&usg=AOvVaw1MPrexr1eVr7RFNC1flMpP&cshid=1578896492575

 

claims S-20 with quartz window. And if you screw off the eye magnifier it says Hamamatsu.

See, pages 5 and 7 of (33 and 35 of the document):

https://www.google.com/url?sa=t&source=web&rct=j&url=https://www.hamamatsu.com/resources/pdf/etd/PMT_handbook_v3aE-Chapter4.pdf&ved=2ahUKEwismdit7__mAhVhkeAKHSi9A4UQFjAAegQIARAB&usg=AOvVaw3tNV7ilfBERBBqaqLUcaTN

 

The range is just to 185nm in case of 500u, but 160nm if indeed the 500S. Interesting the quantum efficiency drops off hard there.

This then sends electrons to a P22 phosphor which emits 530nm light so that I can see it with a back mount camera.

I have a filter capable of imaging this low now.

I also now know that I can use isopropyl alcohol, water and acetone as controls, to see if they turn dark due to there low absorption of 205nm, 190nm, 280nm respectfully. To verify I am seeing what I think I am seeing.

So now it comes to the lens. The KSS100b 60mm f3.5 lens for some reason specifies a minimum wavelengths range of only 230nm. It maybe my problem, due to some lens coating. I do have a quartz 85mm UAT lens which has a coating rated to at least 220nm as being CA free. Hopefully can see deeper. I will test both if I have trouble after my last issue is solved.

That being light source. My germicidal bulbs appear to be near monochromatic at 254nm. Most seem to be very low pressure and/or use titanium dopping to cut out all wavelengths below 230nm.

 

So I have ordered an ozone light. These are medium pressure, higher temperatures and give you 185nm light. See this for example spectrum:

https://www.aquafineuv.com/uv-lamp-technologies

That should solve my light problem, to then see if I have a lens problem.

 

So its never that easy. Was interesting to find that our photocathode is broad spectrum with QE of 20% mostly accross the range, with dropping off to 2% QE at 700nm.

[Andy Perrin]

The range is just to 185nm in case of 500u, but 160nm if indeed the 500S. Interesting the quantum efficiency drops off hard there.

This then sends electrons to a P22 phosphor which emits 530nm light so that I can see it with a back mount camera.

This is quite interesting information. When you were fiddling with it, did you see a way to remove the phosphor without damaging it? I would actually love to try that imager with my SWIR up-conversion phosphor.

[dabateman]

 

This is quite interesting information. When you were fiddling with it, did you see a way to remove the phosphor without damaging it? I would actually love to try that imager with my SWIR up-conversion phosphor.

 

Its not just the phosphor you need to change but the whole module. The photocathode is the front part and will not see anything above 700nm, really less.

 

Look at the graph for the 500S in my second link its doesn't let any IR in.

 

It is then coupled to the P22 530nm phosphor. But for you you would need to swap out the photocathode to get above 900nm. At that point your better just buying an SWIR one, as you just have a tight tube, to scew the components into.

 

But yes after you unscrew the image magnifier, there is a slot on the back phosphor screen to unscrew it. I didn't do that, as I don't want to damage it. But if my dark spots get bigger I might to see if I can improve it.

For the time being I am hoping to keep it for monochrome UVC imaging, 254nm and hopefully 185nm. If I like just the monochrome, I may get a ZWO 1600MM, which I know will image down well to at least 300nm, as its the same sensor as my Em1.

[Andy Perrin]

But if my dark spots get bigger I might to see if I can improve it.

Which dark spots? I am not sure I know what you mean. If it's just unevenness in the image, you can do image subtraction using an image of a white background (PTFE, say) to compensate that. This is how I fix the graininess of my SWIR output.

 

Like, see all the specks on here:

 

After doing optimal subtraction (this is done by a matlab script that minimizes the noise in the resulting image by subtracting a constant times the noise in a white frame) I get: