Author Topic: What's the minimum temperature span of Flir's core?  (Read 1099 times)

0 Members and 1 Guest are viewing this topic.

Offline bibemorTopic starter

  • Newbie
  • Posts: 3
  • Country: us
What's the minimum temperature span of Flir's core?
« on: September 01, 2019, 10:50:55 am »
Hi people.
According to Flir's documents, some of their cores have high-gain/low-gain mode, while some only have low-gain mode.
I know most of their handheld thermal cameras have a minimum temperature span of 4 Kelvin, if the scene is flatter than that, the final picture's contrast will begin to reduce.
How about their cores? What are the equivalent temperature of their high-gain/low-gain mode?
Thank you!
 

Offline Fraser

  • Super Contributor
  • ***
  • Posts: 13388
  • Country: gb
Re: What's the minimum temperature span of Flir's core?
« Reply #1 on: September 01, 2019, 02:33:25 pm »
It would be best to ask FLIR this question as the minimum temperature span is not a common specification on data sheets. The reason is that minimum temperature span is not a very meaningful figure.

1. A Microbolometer can be so noisy or unstable as to make any minimum span below a certain value pointless.

2. A Microbolometer operated without temperature stabilisation is unlikely to provide decent imaging stability at very narrow temperature spans of, say, 2 K. Modern Microbolometer based cameras and cores can come in stabilised or unstabilised types and the cheaper units tend to be unstabilised. The FLIR Ex series being a good example.

3. Industrial and Science grade cameras and cores normally offer the required FPA temperature stability to provide very narrow temperature spans but such performance comes at a cost.

4. The camera cores noise specifications, both NETD and the observation based specifications provide a user with an understanding of the minimum resolvable temperature difference in a scene. Minimum span does not. As you may be aware, the span just sets the ‘window’ size through which the user observes the collected ADU’s. This enables the user display bit depth to be applied to best effect. If there is only noise to be seen at lower temperature differential scenes, then greater bit depth does not help.

High gain and Low gain ...... yes these do help produce better imagery in low thermal contrast scenes but there are limits, as already detailed above. Let us use a Fire fighting camera as a working example......

When a fire fighting camera is used to search in smoke for casualties the room ambient temperature can be close to that of the human body, so producing a low thermal contrast scene. The camera operates in its highest gain mode that can resolve the smallest temperature differential within its hardware limitations. If the camera is temperature stabilised, this high gain mode can be tuned by the manufacturer to provide excellent low Delta T resolving power. Non temperature stabilised cameras, whilst capable of operation in a high gain mode, cannot resolve such low Delta T scenes as well due to the instability in the FPA. The difference in performance of modern fire fighting cameras, stabilised Vs non-stabilised, may not be that great however.

When the fire fighting camera is aimed at a wall on which a fire is present on the other side, the high gain mode soon reaches the limit of its upper temperature boundary. This is often around 120C. At this point the camera switches to a medium gain mode that can cope with a greater temperature range of say 20C to 600C. That range provides less Delta T resolving power in low thermal contrast scenes, but such is not important at the higher temperatures involved with a fire ! The wall may be at 260C and the camera easily copes with the range of temperatures from floor to ceiling as hot gasses collect higher in the room. If direct viewing of the fire or heated metals is required, the camera switches to low gain mode in order to provide adequate scene temperature coverage. The camera may then offer a temperature coverage of 20C to 1000C but at a much reduced Delta T resolving power. Again, this is not important in this usage case.

So from the above you will see that most generic thermal cameras operate in their most sensitive ‘high gain’ mode in daily use but when a higher temperature scene is involved, normally above 120C, a lower gain mode is used to provide the required maximum temperature increase, at the cost of Delta T resolving power.
Some cameras will be set up to provide coverage from -10C up to 260C with just one range. These cameras are a compromise solution and configured for best general operation whilst not excelling at any particular application.

I am not willing to detail how Microbolometers are ‘configured’ for optimum performance in various usage scenarios but that information may be available via internet sources. Fine detail of ‘tuning’ a Microbolometer for best performance is likely a bit sensitive for a public forum.

FLIR will likely be able to fully answer your questions and recommend a camera/core for your needs.

Hope this helps

Fraser
« Last Edit: September 01, 2019, 03:00:54 pm by Fraser »
If I have helped you please consider a donation : https://gofund.me/c86b0a2c
 

Offline Fraser

  • Super Contributor
  • ***
  • Posts: 13388
  • Country: gb
Re: What's the minimum temperature span of Flir's core?
« Reply #2 on: September 01, 2019, 02:43:54 pm »
I own some Keysight thermal cameras and they helpfully detail the different thermal sensitivity offered in each range of the camera. Take a look at the difference in thermal sensitivity on the highest spec 3 range camera.......

https://literature.cdn.keysight.com/litweb/pdf/5991-4297EN.pdf?id=2467122

In précis, here it is.....


U5857A. (-20C to 1200C)

Range 1. -20C to 120C   Sensitivity = 0.07C @ 30C

Range 2      0C to 350C  Sensitivity = 0.1C @ 30C

Range 3.  250C to 1200C Sensitivity = 0.5C @ 250C


Interestingly, the specifications of the lower spec 2 Range version of the camera are inferior in the mid range.....


U5856A (-20C to 650C)

Range 1. -20C to 120C   Sensitivity = 0.07C @ 30C

Range 2    23C to 650C.  Sensitivity = 0.5C @ 30C

Range 3.   Not available


And the lowest spec version that I have provides decent performance up to 350C.......


U5855A (-20C to 350C)

Range 1. -20C to 120C   Sensitivity = 0.07C @ 30C

Range 2     0C to 350C   Sensitivity = 0.1C @ 30C

Range 3.  Not available

It is all about how you set up the Microbolometer  ;)


Fraser
« Last Edit: September 01, 2019, 03:00:18 pm by Fraser »
If I have helped you please consider a donation : https://gofund.me/c86b0a2c
 

Offline Fraser

  • Super Contributor
  • ***
  • Posts: 13388
  • Country: gb
Re: What's the minimum temperature span of Flir's core?
« Reply #3 on: September 01, 2019, 03:14:10 pm »
Oh, almost forgot .... FLIR use the terms high and low gain to describe their version of how the microbolomters are configured. If manufacturing a camera with a temperature range limited to, say, 0C to 100C it is often possible to operate the Microbolometer and its processing stages in a high gain mode for best thermal sensitivity. If the camera has a greater temperature range of, say, -20C to 260C, and only one range is used, the Microbolometer and processing stages will need to be configured in a lower gain mode. Hence why FLIR like to make out that their High gain mode is something special. In truth it is just a standard high gain mode with its associated temperature range limitations. Marketing, Marketing, Marketing  :)

Fraser
If I have helped you please consider a donation : https://gofund.me/c86b0a2c
 

Offline Fraser

  • Super Contributor
  • ***
  • Posts: 13388
  • Country: gb
Re: What's the minimum temperature span of Flir's core?
« Reply #4 on: September 01, 2019, 03:36:21 pm »
As is often the case, it would help forum members to provide more helpful comment if we are told a little more about the intended use and the data that needs to be collected. If you know the required minimum resolvable Delta T in the thermal scene, that would be helpful. The duration of a test can also be pertinent as thermal drift can be an issue at very small temperature spans. There is also the path of thermal analysis on a PC to consider. A PC can plot temperature differentials in a scene that a user might miss. The nature of the temperature differentials can also be characterised and documented easily. It is all about the use scenario and desired data output.

Fraser
If I have helped you please consider a donation : https://gofund.me/c86b0a2c
 

Offline bibemorTopic starter

  • Newbie
  • Posts: 3
  • Country: us
Re: What's the minimum temperature span of Flir's core?
« Reply #5 on: September 01, 2019, 04:23:20 pm »
Wow, think you so much Fraser! I didn't expect anyone can give so much words to a single question...
It's actually one of my friend want to know something about it, and he didn't tell me too much details of what he wants to measure with it...
 

Offline Fraser

  • Super Contributor
  • ***
  • Posts: 13388
  • Country: gb
Re: What's the minimum temperature span of Flir's core?
« Reply #6 on: September 01, 2019, 05:18:54 pm »
Your friend needs to provide more detail.

Fraser
If I have helped you please consider a donation : https://gofund.me/c86b0a2c
 

Offline frogg

  • Regular Contributor
  • *
  • Posts: 131
  • Country: us
Re: What's the minimum temperature span of Flir's core?
« Reply #7 on: September 04, 2019, 09:05:48 am »
"your friend" needs to learn how to read technical literature.
« Last Edit: September 04, 2019, 09:07:41 am by frogg »
 


Share me

Digg  Facebook  SlashDot  Delicious  Technorati  Twitter  Google  Yahoo
Smf