Author Topic: Why don't a multi-spectrum IR camera exist?  (Read 5656 times)

0 Members and 1 Guest are viewing this topic.

Offline sam1275Topic starter

  • Frequent Contributor
  • **
  • Posts: 401
  • Country: us
Why don't a multi-spectrum IR camera exist?
« on: October 10, 2016, 07:30:25 am »
I've read that MCT array can detect 1-14um, all over the SWIR-MWIR-LWIR range, that would be amazing for night vision. But I haven't seen a product that claim archiving this in a single camera, some multi-spectrum system use several separate cameras to do this, and all camera with the MCT array just waste it's power by limit it's range using a band-pass filter. Why? Is that the lens cannot pass all those IR, or other reason?
« Last Edit: October 10, 2016, 08:52:28 am by sam1275 »
 

Offline razberik

  • Frequent Contributor
  • **
  • Posts: 265
  • Country: cz
 

Offline helius

  • Super Contributor
  • ***
  • Posts: 3628
  • Country: us
Re: Why don't a multi-spectrum camera exist?
« Reply #2 on: October 10, 2016, 07:59:49 am »
Those cameras are visible/NIR only. They are definitely not thermal cameras.
Also the one that has both visible and NIR bands is a line camera, meaning it only captures a line, not an image. These types of cameras are only useful in industrial inspection and perhaps astronomy applications.
 

Offline sam1275Topic starter

  • Frequent Contributor
  • **
  • Posts: 401
  • Country: us
Re: Why don't a multi-spectrum camera exist?
« Reply #3 on: October 10, 2016, 08:37:26 am »
Thank you razberik, but I think Helius is right, they are far less than an ideal multi-spectrum camera. Besides, I think there are already some products which are sensitive to visible-NIR-SWIR, such as Indigo's VisGaAs which respond to 400-1700nm.
I'm not going to actually buy one because that must be expensive, but just want to know why they don't use all the ability with an already existing technology such as MCT, to make a king of night vision.
« Last Edit: October 10, 2016, 08:53:49 am by sam1275 »
 

Online Fraser

  • Super Contributor
  • ***
  • Posts: 13125
  • Country: gb
Re: Why don't a multi-spectrum IR camera exist?
« Reply #4 on: October 10, 2016, 01:46:05 pm »
Multi Thermal Spectrum cameras do exist but they use two cooled sensors sharing a single optical block. Germanium optics can be coated to cover SWIR through to LWIR.

A limitation with MCT is the need to 'tune' the band-gap for the spectrum to which it will respond, via its chemistry. MCT cannot cover SWIR, MWIR and LWIR with a single chemical mix. It also requires cooling to very low temperatures, which is normally a complex and expensive requirement in thermal cameras.

Take a look at the MCT Wikipedia page. It describes the situation quite well.

https://en.m.wikipedia.org/wiki/Mercury_cadmium_telluride

There are many materials that will work well as thermal camera sensors. Sadly they are not 'wide and' and many require crypto cooling to reduce their internal noise levels.

Fraser
« Last Edit: October 10, 2016, 02:08:32 pm by Fraser »
If I have helped you please consider a donation : https://gofund.me/c86b0a2c
 

Online Fraser

  • Super Contributor
  • ***
  • Posts: 13125
  • Country: gb
Re: Why don't a multi-spectrum IR camera exist?
« Reply #5 on: October 10, 2016, 02:06:44 pm »
An affordable wide thermal spectrum coverage camera is commonly found on ebay. It is the Inframetrics 610 model. This is the more cabable brother to the Inframetrics 525 'single band' camera that was very popular in Industry during the early days of thermography. There are a lot of these languishing in corners, unused and unloved in the steel, oil and mining industiesy etc.

The camera is Liquid Nitrogen cooled and contains two separate Dewar's for the two detectors. It is a scanning type thermal camera so each detector is a single pixel with servo driven mirrors creating the image raster. If you have a source of Liquid Nitrogen these are very affordable cameras that still work fine. I attach pictures of just the camera but the system comprises the camera head and a controller that has independent controls for each detector. It is effectively two thermal cameras in one case, sharing a single wide-band Germanium optical block.

If I could easily obtain Liquid Nitrogen in the UK I would certainly own one of these. As it is, I own a complete Liquid Nitrogen cooled Agema 880 camera system for purely sentimental reasons.

On the pictures of the 610 camera, not the separate Dewar filling hole covers marked L for the LWIR detector and S for the SWIR detector.

As I am sure you know, SWIR and MWIR has advantages over LWIR, especially for looking through thin glass. The atmospheric window properties are also very different to LWIR.

Fraser
« Last Edit: October 10, 2016, 02:32:13 pm by Fraser »
If I have helped you please consider a donation : https://gofund.me/c86b0a2c
 

Online Fraser

  • Super Contributor
  • ***
  • Posts: 13125
  • Country: gb
Re: Why don't a multi-spectrum IR camera exist?
« Reply #6 on: October 10, 2016, 02:24:27 pm »
For anyone interested, the manual for the Inframetrics 610 is still available via the FLIR legacy camera archives.

http://support.flir.com/DocDownload/Assets/dl/06706-000$a.pdf

The manual is excellent and explains the system operation and design.

I extracted the image of the optical path within the camera and attached it here.

For info, the 610 is a dual band MCT detector camera. A modern version uses Sterling cooled detector arrays rather than Liquid Nitrogen cooling and mechanical scene scanning.

Fraser
« Last Edit: October 10, 2016, 02:33:44 pm by Fraser »
If I have helped you please consider a donation : https://gofund.me/c86b0a2c
 

Offline sam1275Topic starter

  • Frequent Contributor
  • **
  • Posts: 401
  • Country: us
Re: Why don't a multi-spectrum IR camera exist?
« Reply #7 on: October 10, 2016, 02:43:38 pm »
Thank you very much Fraser, that Inframetrics camera is very interesting.
I read the wiki page before but didn't read very carefully and missed the part that those wavelength cannot be detected by a single MCT sensor, my fault...
 

Online Fraser

  • Super Contributor
  • ***
  • Posts: 13125
  • Country: gb
Re: Why don't a multi-spectrum IR camera exist?
« Reply #8 on: October 10, 2016, 03:27:40 pm »
The old Inframetrics cameras are interesting and can be very cheap due to sellers not being able to test them (no understanding, or no Liquid Nitrogen). I have bought the telescope lenses for them cheaply with the intention of adapting them to my other thermal cameras. They are very high quality Germanium X3 Keplerian telescopes and cost me less then £100 each  :)

Fraser
« Last Edit: October 10, 2016, 05:27:57 pm by Fraser »
If I have helped you please consider a donation : https://gofund.me/c86b0a2c
 

Offline sam1275Topic starter

  • Frequent Contributor
  • **
  • Posts: 401
  • Country: us
Re: Why don't a multi-spectrum IR camera exist?
« Reply #9 on: October 10, 2016, 05:30:25 pm »
The old Inframetrics cameras are interesting and can be very cheap due to sellers not being able to test them (no understanding, or no Liquid Nitrogen. I have bought the telescpope lenses for them cheaply with the intention of adapting them to my other thermal cameras. They are very high quality Germanium X3 Keplerian telescopes and cost me less then £100 each  :)

Fraser
Thank you for replying.
I posted another thread which have some thoughts related to the cooling:
https://www.eevblog.com/forum/chat/random-thought-when-will-uncooledtec-cooled-wmir-cameras-present/
I mentioned the TEC cooling, will it be able to cool the sensor if in multi-stage configuration? I think maybe not, because I haven't seen a cooled TIC using TEC cooling.
 

Online Fraser

  • Super Contributor
  • ***
  • Posts: 13125
  • Country: gb
Re: Why don't a multi-spectrum IR camera exist?
« Reply #10 on: October 10, 2016, 06:49:04 pm »
Hi Sam,

I have responded in your TEC thread.

In short, it has already been done by AGEMA in the late 1980's. They shelved that approach soon after.

Fraser
« Last Edit: October 10, 2016, 08:49:28 pm by Fraser »
If I have helped you please consider a donation : https://gofund.me/c86b0a2c
 

Offline sam1275Topic starter

  • Frequent Contributor
  • **
  • Posts: 401
  • Country: us
Re: Why don't a multi-spectrum IR camera exist?
« Reply #11 on: October 11, 2016, 06:16:22 am »
Hi Sam,

I have responded in your TEC thread.

In short, it has already been done by AGEMA in the late 1980's. They shelved that approach soon after.

Fraser
Thank you Fraser!
 

Offline sam1275Topic starter

  • Frequent Contributor
  • **
  • Posts: 401
  • Country: us
Re: Why don't a multi-spectrum IR camera exist?
« Reply #12 on: October 20, 2016, 06:54:01 pm »
I accidentally found some multi-band thermal camera...
An example: http://www.flir.com/science/display/?id=67896
There are some other models as well.
They use a single cooled INSB detector to sense both SWIR and MWIR.
 

Offline sam1275Topic starter

  • Frequent Contributor
  • **
  • Posts: 401
  • Country: us
Re: Why don't a multi-spectrum IR camera exist?
« Reply #13 on: October 20, 2016, 07:09:40 pm »
I think this is interesting to read:
http://www.photonics.com/EDU/Handbook.aspx?AID=25132
For you guys.
 

Offline sam1275Topic starter

  • Frequent Contributor
  • **
  • Posts: 401
  • Country: us
Re: Why don't a multi-spectrum IR camera exist?
« Reply #14 on: October 21, 2016, 11:58:10 pm »
Multi Thermal Spectrum cameras do exist but they use two cooled sensors sharing a single optical block. Germanium optics can be coated to cover SWIR through to LWIR.

A limitation with MCT is the need to 'tune' the band-gap for the spectrum to which it will respond, via its chemistry. MCT cannot cover SWIR, MWIR and LWIR with a single chemical mix. It also requires cooling to very low temperatures, which is normally a complex and expensive requirement in thermal cameras.

Take a look at the MCT Wikipedia page. It describes the situation quite well.

https://en.m.wikipedia.org/wiki/Mercury_cadmium_telluride

There are many materials that will work well as thermal camera sensors. Sadly they are not 'wide and' and many require crypto cooling to reduce their internal noise levels.

Fraser
Hi Fraser, donno if you can see this, but after some searching, I found that MCT is still a very wide-band sensor, they can be varying by different chemical though, here is a good document example:
http://www.vision-systems.com/articles/print/volume-16/issue-4/features/the-infrared-choice.html
Some cover all over wavelength from short wave to long wave in a single sensor.

Further, MCT can work without deep cooling, aka "hot MCT". Here's an example doc:
http://www.electrophysics.com/dbimages/spiehotmct.pdf
Still need some cooler, but seems better.
 

Online Fraser

  • Super Contributor
  • ***
  • Posts: 13125
  • Country: gb
Re: Why don't a multi-spectrum IR camera exist?
« Reply #15 on: October 22, 2016, 09:45:06 am »
Good research Sam,

Well done on that.

I am out of date with MCT technology. Thermal sensor technology continues to develop as can be seen. The microbolometer is actually one of the poorer performance sensors in terms of noise and a decent low noise replacement that does not need cooling to cryogenic temperatures would be great. The noise in modern microbolometers is managed with software but it is always better to start off with a nice low noise signal with less noise reduction needed.

Sadly we could be looking at another VHS Vs Betamax scenario though. VHS won against the Betamax tape format because it was cheaper and had more support from manufacturers. It was, however, the inferior format in terms of image quality. Microbolometers are now well understood and cheap. They can win against other sensor technologies based purely on cost to manufacture and manufacturer investment in the technology to date. Better wide band sensors may come to market but sadly they may remain expensive due to limited manufacturer uptake. As such they will remain limited to lab environments with their larger budgets etc. In labs, even Liquid Nitrogen cooling is not actually that inconvenient as such cooling is quite common. This means the new sensor technology needs to offer something extra to the scientists. Such as extra wide band performance that you have been researching.

Fraser
« Last Edit: October 22, 2016, 08:37:28 pm by Fraser »
If I have helped you please consider a donation : https://gofund.me/c86b0a2c
 

Offline Bill W

  • Super Contributor
  • ***
  • Posts: 1099
  • Country: gb
    • Fire TICS
Re: Why don't a multi-spectrum IR camera exist?
« Reply #16 on: October 22, 2016, 06:35:06 pm »

Further, MCT can work without deep cooling, aka "hot MCT". Here's an example doc:
http://www.electrophysics.com/dbimages/spiehotmct.pdf
Still need some cooler, but seems better.

HOT MCT quoted there is still running at around 130K (so -150°C), so going from liquid nitrogen up to very good Stirling coolers ?
This also pushes the sensitive band below 10um, so has less energy from an ambient scene to image.

Bill

Offline sam1275Topic starter

  • Frequent Contributor
  • **
  • Posts: 401
  • Country: us
Re: Why don't a multi-spectrum IR camera exist?
« Reply #17 on: October 22, 2016, 09:02:25 pm »
Good research Sam,

Well done on that.

I am out of date with MCT technology. Thermal sensor technology continues to develop as can be seen. The microbolometer is actually one of the poorer performance sensors in terms of noise and a decent low noise replacement that does not need cooling to cryogenic temperatures would be great. The noise in modern microbolometers is managed with software but it is always better to start off with a nice low noise signal with less noise reduction needed.

Sadly we could be looking at another VHS Vs Betamax scenario though. VHS won against the Betamax tape format because it was cheaper and had more support from manufacturers. It was, however, the inferior format in terms of image quality. Microbolometers are now well understood and cheap. They can win against other sensor technologies based purely on cost to manufacture and manufacturer investment in the technology to date. Better wide band sensors may come to market but sadly they may remain expensive due to limited manufacturer uptake. As such they will remain limited to lab environments with their larger budgets etc. In labs, even Liquid Nitrogen cooling is not actually that inconvenient as such cooling is quite common. This means the new sensor technology needs to offer something extra to the scientists. Such as extra wide band performance that you have been researching.

Fraser
Thank you Fraser.
If you look closely at that doc, they even measured microbolometers, and the result is a very flat line with very low sensitivity. This make sense to me because microbolometers are not actually detect certain photons, as you know they work by changing self temperature from received energy, then read the resistance/charge difference, so it's not odd to have such a flat line in the chart.
However nearly all microbolometers have a built-in filter to only allow LWIR pass through, Flir says it will help prevent sun light from influence the image. But even if they don't have the filter, I think we can still barely MWIR/SWIR by a microbolometer, because heat energy are too few at those wavelengths compare with LWIR, Flir once told me MWIR camera need to be 1000 times more sensitive than LWIR ones, similar thing is also shown in the document as a curve.
I haven't found any wide range MCT based camera, and I don't know why.
Another thing to consider is, it's may not always good to see a very wide range all at once, since all information will added together and may messed some details. Maybe in the future they will use a filter wheel or mask to distinguish different wavelength while have a wide response, just like bayer-filter in color cameras... That's just some of my silly thoughts though.
Sam
 

Offline sam1275Topic starter

  • Frequent Contributor
  • **
  • Posts: 401
  • Country: us
Re: Why don't a multi-spectrum IR camera exist?
« Reply #18 on: October 22, 2016, 09:11:54 pm »

Further, MCT can work without deep cooling, aka "hot MCT". Here's an example doc:
http://www.electrophysics.com/dbimages/spiehotmct.pdf
Still need some cooler, but seems better.

HOT MCT quoted there is still running at around 130K (so -150°C), so going from liquid nitrogen up to very good Stirling coolers ?
This also pushes the sensitive band below 10um, so has less energy from an ambient scene to image.

Bill
Thank you Bill.
Yes it still need cooling, but it is better than 77k, it will longer the cooler's life and may reduce power usage and cooling time, I remember some hot MCT product claims 20k hours MTBF, which is 2 times longer than the most 10k hours.
Sam
 


Share me

Digg  Facebook  SlashDot  Delicious  Technorati  Twitter  Google  Yahoo
Smf