It seems some MWIR cameras can work without sterling coolers

[Ben321]

This one website https://pembrokeinstruments.com/swir-cameras/ has this one wideband IR camera that is sensitive to NIR, SWIR, and the shortest wavelengths of MWIR. It's called the "Zephir 2.9". The specs say 850nm to 2900nm. Resolution is 320x256. It is a HgCdTe camera, and the the neat thing about it is it doesn't use a sterling cooler. It still requires a low operating temperature (-80degC according to the specs, which according to Google's temperature converter is -112degF), but this low temperature is achieved without using any liquid nitrogen or sterling cooler (which is expensive due to the purity of helium required and other manufacturing costs associated with precision mechanics of the cooler). Instead, it uses a 4 stage thermo-electric cooler.
Here's a PDF file with the full specs of the camera. https://pembrokeinstruments.com/wp-content/uploads/2019/06/ZEPHIR_2.9_2019-pembroke.pdf

Despite it not using a sterling cooloer (which should make it a cheaper camera), no price is listed on the website, and that does NOT bode well for it being a low cost device. I've discovered that when a company doesn't put a price on their website, it usually means it's SO EXPENSIVE that they are afraid to put the price on their site, out of the fear that it would scare away potential customers. Usually they want you to request a quote, so they have a chance to not only email you the price, but also a sales pitch on why you should buy it, and not let the price scare you away from buying it. Such devices (ones that they don't want to list a price on their website) I've discovered (after getting prices emailed to me) are typically at least $10000, and often times SEVERAL tens of thousands of dollars.

Now I've emailed this company asking for a price list for their cameras. I hope to hear back from them. I'm concerned they may see my Gmail email address, and realize I'm not a company with big pockets (and thus would unlikely be able to afford the camera), and end up not even sending me a reply email.

[DaJMasta]

There have been similarly cooled TEC based MWIR cameras, like the old Agema Thermovision 470, around which uses a MCT detector as well.  That camera can be cheap if you can find one, but the optics are still pricey and it's heavy and whirs quite a bit for the 100 pixel x 140 line image!  While I'm not certain as to the chemistry/physics of it, it does seem like InSb detectors seem to be preferred for image quality in the MWIR band.

[Ben321]

There have been similarly cooled TEC based MWIR cameras, like the old Agema Thermovision 470, around which uses a MCT detector as well.  That camera can be cheap if you can find one, but the optics are still pricey and it's heavy and whirs quite a bit for the 100 pixel x 140 line image!  While I'm not certain as to the chemistry/physics of it, it does seem like InSb detectors seem to be preferred for image quality in the MWIR band.

Is HgCdTe usually used for LWIR then? Also what do you mean it whirs? You mean the whirl sound of the sterling cooler's motor? I thought you said the Agema one didn't have a sterling cooler. Or do you mean there's a fan to cool the TEC, and the fan motor is loud?

[DaJMasta]

MCT detectors seem to be SWIR and MWIR mostly (though SWIR has a fair bit of InGaAs as well), then VOx (Vanadium Oxide) seems to be the preference for LWIR.  The whirring is because the Thermovision 470's sensor is only a line sensor element, so it's got spinning mirrors at 20k RPM or so to scan fast enough to give you 20fps output at 140 lines, plus it's got a fan for the TEC.

I'd expect a camera like the one above to be much quieter 

[Ben321]

MCT detectors seem to be SWIR and MWIR mostly (though SWIR has a fair bit of InGaAs as well), then VOx (Vanadium Oxide) seems to be the preference for LWIR.  The whirring is because the Thermovision 470's sensor is only a line sensor element, so it's got spinning mirrors at 20k RPM or so to scan fast enough to give you 20fps output at 140 lines, plus it's got a fan for the TEC.

I'd expect a camera like the one above to be much quieter 

A bit of an issue I notice when reading the specs of the camera is that it does NOT use USB. It uses a rare type of camera connector called cameralink. These tend to be used exclusively with science/industrial cameras. And the only way of getting the data from the camera into a computer is via a special PCI-e card, designed to allow the camera to connect to the computer). This precludes the possibility of connecting it to a laptop, and I don't have a desktop. Plus these cameralink cards are quite expensive ($1000 or more). And unlike other video capture cards, I cannot find any Chinese knockoffs of these on eBay.

[DaJMasta]

There are a number of different formats and I don't think CameraLink has that broad interoperability, but an option could be a CameraLink to Gigabit Ethernet converter module.  A lot of high speed/high data cameras will use CameraLink but have modes that aren't as bandwidth hungry (like low resolution images), or are just old enough to not need excessive bandwidth, but while there is some free software for this kind of thing, hardware and software compatibility can sometimes be specific to particular hardware.

There are also PCI and PXI/PXIe cards with CameraLink connections and you can probably find some PCIe cameralink cards <$500 on ebay without too much issue.  Yeah, still expensive, and a gamble unless you have access to the manuals or software, or even have experience with the platform and workflow... but as you said, if the manufacturer wants you to inquire for a price, the price of the supporting accessories may not actually be all that steep.

I have seen some other SWIR band cameras with regular old USB 3 or gigabit ethernet, and depending on your usage, that may do the job for your application - these ones only seem to be rated to 2.9um, so they've only got their toes in the MWIR band.  I can now confirm from testing that somewhere around 150-200C you can see heat emission in SWIR (<2um), so if you're looking for high temperature analysis, that could be an option instead.

[Ben321]

There are a number of different formats and I don't think CameraLink has that broad interoperability, but an option could be a CameraLink to Gigabit Ethernet converter module.  A lot of high speed/high data cameras will use CameraLink but have modes that aren't as bandwidth hungry (like low resolution images), or are just old enough to not need excessive bandwidth, but while there is some free software for this kind of thing, hardware and software compatibility can sometimes be specific to particular hardware.

There are also PCI and PXI/PXIe cards with CameraLink connections and you can probably find some PCIe cameralink cards <$500 on ebay without too much issue.  Yeah, still expensive, and a gamble unless you have access to the manuals or software, or even have experience with the platform and workflow... but as you said, if the manufacturer wants you to inquire for a price, the price of the supporting accessories may not actually be all that steep.

I have seen some other SWIR band cameras with regular old USB 3 or gigabit ethernet, and depending on your usage, that may do the job for your application - these ones only seem to be rated to 2.9um, so they've only got their toes in the MWIR band.  I can now confirm from testing that somewhere around 150-200C you can see heat emission in SWIR (<2um), so if you're looking for high temperature analysis, that could be an option instead.

I know some science cameras use GigE, but what kind of control software do they use? Do they have a builtin webserver, so you just control the camera (and capture video and images) via a web browser interface? Or do they user a proprietary protocol, that requires specialized GigE control software, written by the same company that made the camera?

[DaJMasta]

There is a GigE vision standard and while they generally don't support browser control (to my knowledge), you basically just take GigE vision compliant software and tell it what IP to connect to on the local network.  Vendors have their own flavors and configuration may be specific to some, but you can probably get and image and some base configuration with generalized software so long as they support the standard.

[Fraser]

I wrote about the AGEMA 510 MWIR camera some time ago. It uses the Peltier cooled Lead Selenide (PbSe) detector technology. These detectors operate at around -70 Celsius.

https://www.eevblog.com/forum/thermal-imaging/agema-thermovision-510-a-look-under-her-clothes-by-fraser/

Fraser

[Fraser]

I carried out a teardown of the AGEMA 470 MWIR camera and documented it here......

https://www.eevblog.com/forum/thermal-imaging/agema-thermovision-thv470-camera-complete-teardown-by-fraser/

It uses a Peltier cooled MCT Sprite detector operating at around -70C. The camera is a scanner type but can produce very good MWIR imaging. It is worth studying the MCT Sprite detector as it is a very clever technology. It looks like a single rectangular pixel but it is actually far more capable than a basic single pixel detector 

Fraser

[Max Planck]

Example of a VOx camera operating in the 3-14um band.

https://scdusa-ir.com/wp-content/uploads/2017/09/VOxI-BB.2018-1.pdf

Max

[Ben321]

Example of a VOx camera operating in the 3-14um band.

https://scdusa-ir.com/wp-content/uploads/2017/09/VOxI-BB.2018-1.pdf

Max

So VOx cameras can also do MWIR? Do they require cooling for this? Or can they operate at room temparature, using a shutter calibration, just like they do when detecting LWIR? What lets it detect MWIR? Is it just a different lens material, that passes MWIR instead of LWIR?

[Bill W]

It is a detector system with a broad band window, and does not come with a lens.  Your other questions are all answered in the brochure.

Bill