-
I had been experimenting with my Seek Compact Pro looking at a clear (not white coated/frosted) light bulb, and I noticed something very interesting. I could see the filament. If the Seek only saw LWIR, or even if it saw both LWIR and MWIR, you wouldn't see the filament, because these wavelengths are absorbed by the glass. As the glass got hot, you would see the glass start to glow in the LWIR part of the spectrum, but you wouldn't see the filament inside. However, I was able to see the filament (though only dimly). This suggests that at least some SWIR is getting through the lens of the imager, as SWIR is the only IR wavelength range short enough to pass through the ordinary glass of the lightbulb. It would seem that instead of using germanium (which passes only LWIR), the Seek Thermal company is using a cheaper material that passes a wide range of IR wavelengths (even as far down as SWIR). I wonder how this affects the accuracy of the imager's ability to measure temperature?
One thing is good about this though, is that it means I should be able to use an external filter made out of a specific material, to pass a specific part of the IR band (such as an SWIR-pass or MWIR-pass filter), to turn my imager into one that works specifically at those other wavelengths (at the expense of losing the ability to measure temperature). In doing so, it would mean I could turn my Seek Compact Pro into a cheap SWIR or MWIR imager, instead of having to shell out well over $10,000 for a dedicated InGaAs SWIR imager or InSb MWIR imager. A small window/filter made out of a material that would pass a specific range of wavelengths might cost a couple hundred dollars, instead of $15,000 to $20,000. Such a window/filter could easily be glued to the front of the lens of my Seek Compact Pro (or put on with tape, if I wanted to be able to easily remove it later). -
That's a fascinating discovery.
I think the Seek lenses are generally made of moulded chalcogenide glasses, mainly I believe for cost reasons (a moulded glass lens is way cheaper than a diamond-turned germanium lens, even if the glass itself is fairly pricey). Schott, for example, offer various glasses with transmission between about 0.7µm (just-visible deep red) all the way out to >15µm:
So, yes, it does look as though there's a prospect of some transmission through the optical system (though it's unlikely to operate particularly well outside its design frequency range - the lens components will all be of sub-optimal configuration, leading to intrinsically blurry images). The coatings on the lenses (if any?) will be what really determine their out-of-band properties.
But you're right, a suitable filter should mean you can limit the passband to wavelengths other than the original design, though at a (significant?) performance penalty. How well the FPA actually responds to those other wavelengths is also an open question.
(You may be aware that at the other end of the spectrum some conventional digital cameras have, or can be modified to have, a response from about 1100nm to about 350nm, near-IR to near-UV, and that some supposedly visible light only lenses can perform surprisingly well outside their design bandwidths. Below is a sequence of images in near-IR, visible, and near-UV taken with regular lenses and suitable filters on a Fujifilm IS PRO, a long-discontinued 'full-spectrum' camera).
For comparison, here's a very poor thermal image of a different flower vs its visible light equivalent.
Thermal images of flowers are challenging, unless you can find a warm-blooded plant.
You may have discovered the first example of a comparable MWIR-LWIR imaging setup.
-
I would not be too sure it is actually very sensitive to SWIR (or even MWIR). If you look at the extiction coeffient of fused silica (most likely material of a glass bulb) (i.e. here: https://www.google.de/url?sa=i&rct=j&q=&esrc=s&source=images&cd=&cad=rja&uact=8&ved=2ahUKEwjazZH19IrfAhUFGuwKHVjKBScQjRx6BAgBEAU&url=https%3A%2F%2Fwww.researchgate.net%2Ffigure%2FRoom-temperature-absorption-spectrum-of-fused-silica-reproduced-from-Ref-9-The_fig1_243575847&psig=AOvVaw0KwiRBA0Wj-kdn1TmFbhXc&ust=1544175701318413) you will notice, that MWIR Transmission at 4.5µm for a 0.5 mm glass plate (I think the glass of a bulb is even thinner) is in the Region of 60% and even for the LWIR low end around 8µm you still end up with roughly 1% of Transmission. That is not very much but still enough to see the really hot filament through the glass.
Regarding transmission (or blocking) it is always wise to have look at logarithmic plots - not the linear ones. Very ofter a 'zero' in a linear plot may well be only 0.1% and cause quite a bit of trouble with very intense out-of-band sources...
The only real proof usually is a narrow-band source like a SWIR laser but given enough power the nature of the bolometer detector may well cause an image of the laser too - or at least some heating of irradiated surfaces?!
By the way: an image of the bulb/filament would be interesting! -
By the way: an image of the bulb/filament would be interesting!
I second that. -
So a cheap filter to block out MWIR and LWIR? Just glass? I mean you could probably get a UV glass filter for analog cameras for less then 5$ and they are made from very thing glass.
-
A UV-pass filter is expensive - good ones (Baader U) are ~US$200 and up (I'm leaving aside Wood's Glass filters because they leak lots of red and NIR). UV-cut filters, OTOH, are often little more than plain glass and nearer $1.
It might be interesting to try the Seek with a 950nm low pass filter (~ US$10). Made of glass and opaque to the eye, they will nevertheless pass some (relatively) longer wavelengths, probably down to 2µm or so.
Sapphire transmits down to about 5µm (and largely blocks longer wavelengths, useful in this context). Small sapphire windows aren't as expensive as you might expect - an uncoated 10mm diameter sapphire window is ~ US$15. They're also available with various coatings, though they quickly become expensive - a half-inch diameter window coated for 2-5µm is about GBP60. -
Either way, the transmission to the sensor is tiny in the end and you need high amplification to see anything. By then the microbolometers picks up its own heat and you need to cool it again. There are no options for long integration times on any of the consumer cameras I know of. So good luck.
-
Here's a picture demonstrating it. This is a 29W halogen bulb, intended as a replacement for a 60W standard incandescent bulb, and screws into a regular socket. It has its filament inside a quartz tube, which is inside a normal glass bulb. Not sure what the transmission spectrum of quartz or normal glass is, or even what type of glass they use for the normal glass. What I do know is this, the glass bulb is blurry, while the filament is sharp. This suggests that there is a significant difference in wavelength between the LWIR coming from the warm bulb, and the emission from the filament that is getting through the glass bulb and also detected by what is (supposedly) an LWIR imager. I'm guessing that it's probably not MWIR, as that would be absorbed by the glass outer bulb, and possibly also by the quartz inner tube. It probably isn't NIR (what you can detect with a modified visible-light digital camera) as that would be unlikely to pass through the optics of even a poorly designed thermal imager lens. My guess is that what I'm seeing is SWIR coming from the bulb, at around 2000nm (2um), which is too long of a wavelength to be seen by a silicon based detector, but not too short of a wavelength to be seen by a microbolometer (which does not operate on photoelectric effect, but rather by heat, which means ANY wavelength that can heat up the microbolometer's sensing pixels, will produce an image, even if only a weak image).
This means that with a simple short-pass filter (like glass or plastic, which blocks MWIR and LWIR, but passes SWIR) I should be able to create a camera that can see SWIR wavelengths maybe down to about 1500nm (not sure what the actual pass band of the Seek lens is), and allow me to explore SWIR phenomenon for a much lower price than buying an imager with an InGaAs focal plane array.
-
That's weird, when I uploaded the JPEG as an attachment, it butchered the JPEG's orientation metadata. It shows it as landscape (wider than tall), when it should be portrait (taller than wide). But that's just the preview. If you click on the thumbnail to blow it up, and then right click and go to View Image, it will load the image into the browser and show it correctly.
-
This is indeed strange. It's not a reflection, although it might look like it it's a very powerful source behind there. The sun usually emits across a really broad wavelength, yet there it not even a hint of the sun if I look through any of my glass towards the sun. I gotta try my collection of photographic filters as well.
-
A while ago I demonstrated how a simple photographic UV Skylight filter can act as a very inexpensive attenuator for use with LW cameras when viewing high energy sources. This lightbulb is a high energy emitter as Halogen lamp filaments run very hot indeed. My first thoughts on this are that you are simply seeing the high energy radiating filament through an attenuator that is the halogen quartz capsule and thin lamp glass envelope. Thin glass acts as an attenuator at LW frequencies and no such crude attenuator completely strops the transmission of energy. This is little different to the use of Silicon lenses and windows on the Lepton core. The thickness of the Silicon dictates whether is is useable and a certain amount of transmission loss will be expected.
I am sorry to say, this lamp experiment does not strike me as a very useful method of detecting the SW sensitivity of a LW camera. The correct methodology is to place a known SW only passband filter of known transmission figure between the camera and a suitable source if SW energy. Good luck finding such a filter at an affordable price for such an experiment. I do have some specialist filters but nothing at SW I am sorry to say. That being said, I do have a SW 'flame filter' from an Inframetrics PM280 SW camera. I will need to see whether it has a specification that helps us determine its usefulness.
Fraser -
Here is my post on using a glass UV filter as an attenuator. Note that the photographic glass is single layer coated for UV and approximately 1mm thick. The Thermal energy source in my experiment is so powerful that some energy gets through the glass and is seen by the LW camera.
https://www.eevblog.com/forum/thermal-imaging/range-extender-for-thermal-cameras-cheap-option-from-fraser/msg941515/#msg941515
Apply this idea to a light bulb and you will see that the envelope glass is just replacing the UV filter in the flight path of the thermal energy from the filament to the camera. To prove a camera has SW sensitivity, you need to remove the other possibilities from the equation, namely LW and MW energy. A tight skirted SW filter would be capable of that.
Fraser -
Well it looks like my Inframetrics flame filter is exactly what is needed. It is a tight skirted SW filter that is used to see through flames in order to image what is beyond. Flames radiate a lot of energy at some wavelengths and very little at others. The filter makes good use of this fact. Take a look at the graph in the included document. The Flame filter produces a narrow 'window' at around 3.9um for the camera to see through the flames energy spectrum. These filters were very expensive, now you know why. I will test the filter with one if my cameras tomorrow but I do not own a Seek anymore so cannot test that.
https://www.flir.com/discover/instruments/furnace-boiler/petroval/
Fraser
-
Here's another reason I think it's SWIR. I believe I mentioned this in my first post, but it seems you didn't read that part of my post, or else I planned to post it but forgot. One HUGE reason I think it is NOT the LWIR coming from inside the bulb, is the fact that I was using the camera close to the bulb, but I had the lens's focus on my Seek set for distant objects. The result is that the edge of the lightbulb (which is emitting LWIR) is blurred so appears like a wide gradient. Meanwhile the image of the filament appears very tight and in focus, almost no blurring). This suggests a much shorter wavelength than LWIR. I have created an annotated image showing this. The narrow image of the filament, but the wide (due to out-of-focus blur) image of the side of the bulb. I've attached this annotated image to this post.
-
Ben321,
Enjoy your experiments and I hope you find what you are searching for.
Fraser
-
I'll just park this here and get on with more important work.....
https://www.ebay.com/itm/FJW-Optical-Systems-Find-R-Scope-85400A-Infrared-Viewing-IR-Camera-400nm-1800nm/264062712861?_trkparms=aid%3D111001%26algo%3DREC.SEED%26ao%3D1%26asc%3D20160908105057%26meid%3Dc483317b382b4ab7a760cdf741490547%26pid%3D100675%26rk%3D1%26rkt%3D15%26sd%3D264062712861%26itm%3D264062712861&_trksid=p2481888.c100675.m4236&_trkparms=pageci%3Aecebf528-fa04-11e8-8ab1-74dbd1805b08%7Cparentrq%3A880f652a1670aca48f738579fff39622%7Ciid%3A1 -
About 40 years ago I bought a 175W UV bulb that consisted of a small mercury discharge tube ballasted by a filament and enclosed in a Wood's Glass envelope. You could barely see the UV when you looked straight at it (though it was actually very bright; the eyes simply don't respond to those wavelengths). You could, however, see the filament as a deep, deep red as its light crept out below the lower (deep red) cutoff of the Wood's Glass bulb. It looked something like the image below, which I borrowed from YouTube video Sylvania - mercury vapour lamp UV - HSW125 & HSBW160W by Mr26lumix. The difference in real life is that the red filament is a much, much deeper ruby colour than you see here.
I wonder if we're seeing something broadly similar?
I must see if I can find an incandescent bulb somewhere and point some LWIR cameras at it. I don't expect to see much because they mostly have germanium lenses - not noted for their visible light transmission - though I may be able to dig out an uncoated ZnSe lens that I can form an image with. I'll see what I can do over the (busy) weekend and report back if I have any success.
-
Interesting picture Ultraviolet
My 'new' Proxxon micro mill has just arrived so I will be spending time with that this weekend.
I do have a scientific SWIR camera hiding somewhere in the lab/attic/garage or shed though. It has a load of specialist bandpass filters with it. From memory it looks like the Find-R-Scope and uses some specialist Vidicon tube, like the EEV LWIR Pevicon's, to operate at SWIR. I got it many yeas ago but never had a use for it really. Time to dig it out and see if it still works maybe. I will be honest though.... I am not feeling that motivated to do so at the moment. The OP wants to prove his Seek is providing SWIR coverage. I know for a fact that the microbolometer has a 7um high pass filter coating on its microbolometer window so anything I do with a real SWIR camera is not really helping his cause. I wish him well with his experiments though. He may wish to do some more reading on microbolometer design and Sun safe sensors though. Even if some SWIR leakage occurs through the optical path filtration, sensitivity at that wavelength is miniscule to the point of being pretty much useless for general imaging at ambient temperatures. A filtered high quality CCD would likely work as well despite it being well outside its design limits.
Time to go play with my micro mill now
Fraser -
This article was found in 10 seconds of searching using Google. It may be of interest to the OP ?
Note the very steep cutoff at 7um. As stated, there is a very good reason for that and the SEEK microbolometer is no different.
https://www.photonics.com/Articles/Measuring_the_Spectral_Response_of_IR_Cameras/a22473
Fraser -
A useful,SWIR test for the OP instead of the light bulb with its crazy hot source
This article discusses SWIR imaging and shows a simple test for camera SWIR response using a wine glass and a 100C thermal source.. Generating 100C is simple (just think boiling water) and wine glasses are common For finer glass, use a Champagne flute ! This test is more reliable than the light bulb test the OP tried.
https://www.photonics.com/Articles/SWIR_Imaging_An_Industrial_Processing_Tool/a25134
Now I really must go play with my new toy
Fraser -
I had a brief search for the SWIR Vidicon camera and filters but they are likely buried in one of many many storage boxes in the garage so seem too much effort to find at the moment. During the search I did did another path into SWIR imaging though. For many years there have been wavelength converters to convert various optical wavelengths into the visible light domain. The wavelengths converted are dictated by the target coating used and they are capable of SWIR operation with the right coating. They are similar in principle to image intensifiers but normally have little or no gain as they are converters and not amplifiers. I found a really nice NIR+SWIR converter that I bought some time ago and it even has the potted drive module on it
An interesting little bit of optical kit Such units are coupled to the users eye using a lens and eyecup or to a camera using a suitable coupling adapter and lens.
Another project I never took forward !
Fraser
-
A more common wavelength converter that I know the OP is already aware of.
The FJW Find-R-Scope units that are available in both hand held and static mount. My example is the latter and has an external HT generator to drive the tube. These units are common and often used in science to observe NIR and wavelengths entering SWIR. Lasers beam shape monitoring is one use that I am aware of. My unit is definitely NIR and may even be the wider spectral response model tube but I would need to check. The best that these particular FJW units can offer is 1.3um though so not the whole SWIR spectrum.
Fraser
-
Sadly none of my recent posts help the OP in his quest to determine the SWIR response of his SEEK Camera. I posted only for the interest of the readership on the topic of SWIR imaging. For that reason I will withdraw from the discussion rather than decrease the signal to noise ratio I hope the OP manages to measure his cameras SWIR response in the real world scenarios in which he would find it useful.
As a final side note ..... In the professional world SWIR imaging is considered cheaper and simpler than LWIR imaging as it does not require very expensive Germanium optics or sophisticated microbolometers ! it is supposed to be simpler to build an imager for these wavelengths and normal photographic lenses may be used for most, if not all, of the SWIR spectrum
Fraser -
Sadly none of my recent posts help the OP in his quest to determine the SWIR response of his SEEK Camera. I posted only for the interest of the readership on the topic of SWIR imaging. For that reason I will withdraw from the discussion rather than decrease the signal to noise ratio
I hope the OP manages to measure his cameras SWIR response in the real world scenarios in which he would find it useful.
As a final side note ..... In the professional world SWIR imaging is considered cheaper and simpler than LWIR imaging as it does not require very expensive Germanium optics or sophisticated microbolometers ! it is supposed to be simpler to build an imager for these wavelengths and normal photographic lenses may be used for most, if not all, of the SWIR spectrum
Fraser
In the professional world, it's far more expensive to get an SWIR camera than an LWIR camera, most likely due to the difficult manufacturing process (the fact that it needs a readout chip below the InGaAs sensor chip, and then the 2 chips have to be precisely welded together without damaging them). Keep in mind, that a 640x480 VOx LWIR camera (at its cheapest) sells for about $3000 new. Meanwhile, the cheapest 640x480 InGaAs SWIR camera I have found, costs about $10000, and that's a used one on eBay (new ones go for well over $10000, which I know, because I have emailed several companies that sell them, and I had to email them asking for the price because they don't post the price on their websites, probably because these things are so expensive that if the price were listed it would scare off any potential customers).
The SWIR cameras you say that are cheaper than LWIR cameras, are probably actually NIR cameras (with a maximum wavelength of about 1200nm, above which the sensitivity is well below 1% of its maximum sensitivity). This maximum wavelength that it is sensitive to being 1200nm is due to the electro-optical characteristics of the element silicon (which is used in NIR sensors). VOx microbolometers don't depend on the photo-electric emission of electrons (electrons generated from a photon striking the sensor), but rather work by sensing heat (basically an array of microscopic electronic thermometers, and heat can be generated by any wavelength, because any wavelength is capable of heating a target that is illuminated by the light source). -
I have an Electrophysics 7290A Vidicon based NIR-SWIR that does penetrate into what is traditionally called the SWIR band. It uses a lead oxysulphide target coating and being a scanned analogue target rather than pixel based, resolution is decent IIRC its coverage extends up to around 1.9um. The enhanced 7290A-06 version operates up to 2.2um. The 7290A-06 requires optics designed for use at 2.2um in order to achieve best performance.
Thermal imaging occurs at around 250C observed object temperature. I have seen mention of thermal imaging as low as 150C being possible though. Interestingly, lowering the Vidicon tubes target to 77K provides significant improvement in the cameras capabilities. Now that would be an interesting looking cryo-cooled camera
An FAQ about my camera is to be found here.....
http://site.cascadelaser.com/sof7290A-faq.pdf
Datasheet is here.......
http://www.lambdaphoto.co.uk/pdfs/EC-MicronViewer7290A-rev07_Lambda.pdf
The 7290A cost me less than £100 and is easy to repair (except for the specialist Vidicon tube of course!)
Science still uses Vidicon tube based cameras in some specialist roles. The technology remains valid even in the 21st Century
Fraser