FLIR , IRST in air to air mission

Cockpit, radar, helmet-mounted display, and other avionics
  • Author
  • Message
Offline

hornetfinn

Elite 3K

Elite 3K

  • Posts: 3064
  • Joined: 13 Mar 2013, 08:31
  • Location: Finland

Unread post10 Oct 2018, 08:21

Thank you ricnunes and Spurts! I really like the fact that here matters are discussed and argued in very respectful and professional matter. I like being questioned and challenged and learn something myself in the process. :D
Offline

aasm

Banned

  • Posts: 77
  • Joined: 12 Aug 2018, 13:01

Unread post10 Oct 2018, 13:24

Thank you hornetfinn. About OSF, there was a job offer at Thales about two years ago for ISF IR mentioning both IR channel and India.
Offline

aasm

Banned

  • Posts: 77
  • Joined: 12 Aug 2018, 13:01

Unread post11 Oct 2018, 10:13

Interesto,g diagram about neaer IR and its use vs vegetation (green curve). See inversion in near IR

Image
Offline
User avatar

BDF

Senior member

Senior member

  • Posts: 268
  • Joined: 23 Nov 2006, 13:54

Unread post20 Apr 2020, 17:10

hornetfinn wrote:Latest IRST/FLIR systems like EOTS and podded systems with 3rd gen staring arrays will likely have very impressive detection and tracking ranges. For example these documents have interesting information:

https://www.google.fi/url?sa=t&rct=j&q=&esrc=s&source=web&cd=1&cad=rja&uact=8&ved=0CB4QFjAAahUKEwiwltStp4XHAhUDAXMKHYWdB8o&url=http%3A%2F%2Fwww.dtic.mil%2Fcgi-bin%2FGetTRDoc%3FAD%3DADA481411&ei=lGG7VbD8IYOCzAOFu57QDA&usg=AFQjCNGgptw8MO8yIfdZ58h1lY-4q4URhg&bvm=bv.99261572,d.bGQ

http://www.wat.edu.pl/review/optor/14(1)1.pdf

http://antonirogalski.com/wp-content/uploads/2011/12/Third-generation-infrared-photodetector-arrays.pdf

It seems like 3rd gen MWIR systems usually have about 2-3 times longer ranges compared to 2nd generation FLIR systems. Technologically for example Pirate and targeting pods of 1990s are mostly 2nd generation IR technology. EOTS and DAS use 3rd gen MWIR system. I think many people will be surprised how powerful system EOTS+DAS is in air-to-air combat.


Its been my understanding that IRSTs suffer a substantial range hit when conducting volume search because of the limited FOV; i.e. the "soda straw" analogy. Rule of thumb I've often heard is the actual detections ranges when conducting a volume search is about 1/3rd of "brochure" ranges which are usually cued detection ranges. Would CCD arrays improve upon volume search times?

Sorry to dig up a zombie thread!
When it comes to fighting Raptors, "We die wholesale..."
Offline

hornetfinn

Elite 3K

Elite 3K

  • Posts: 3064
  • Joined: 13 Mar 2013, 08:31
  • Location: Finland

Unread post29 Apr 2020, 12:08

BDF wrote:
hornetfinn wrote:Latest IRST/FLIR systems like EOTS and podded systems with 3rd gen staring arrays will likely have very impressive detection and tracking ranges. For example these documents have interesting information:

https://www.google.fi/url?sa=t&rct=j&q=&esrc=s&source=web&cd=1&cad=rja&uact=8&ved=0CB4QFjAAahUKEwiwltStp4XHAhUDAXMKHYWdB8o&url=http%3A%2F%2Fwww.dtic.mil%2Fcgi-bin%2FGetTRDoc%3FAD%3DADA481411&ei=lGG7VbD8IYOCzAOFu57QDA&usg=AFQjCNGgptw8MO8yIfdZ58h1lY-4q4URhg&bvm=bv.99261572,d.bGQ

http://www.wat.edu.pl/review/optor/14(1)1.pdf

http://antonirogalski.com/wp-content/uploads/2011/12/Third-generation-infrared-photodetector-arrays.pdf

It seems like 3rd gen MWIR systems usually have about 2-3 times longer ranges compared to 2nd generation FLIR systems. Technologically for example Pirate and targeting pods of 1990s are mostly 2nd generation IR technology. EOTS and DAS use 3rd gen MWIR system. I think many people will be surprised how powerful system EOTS+DAS is in air-to-air combat.


Its been my understanding that IRSTs suffer a substantial range hit when conducting volume search because of the limited FOV; i.e. the "soda straw" analogy. Rule of thumb I've often heard is the actual detections ranges when conducting a volume search is about 1/3rd of "brochure" ranges which are usually cued detection ranges. Would CCD arrays improve upon volume search times?

Sorry to dig up a zombie thread!


IMO, it's good to sometimes get back to these old threads.

It's true that IRST systems are not very good for volume search especially compared to modern radars vs. 4th gen jets. IRST systems tend to have 2 or 3 different fields of view (basically magnification) for different tasks. Wide FoV is used for volume search and navigation. Narrow FoV is used for cued (small volume) search, identification and single target tracking. Some targeting pods have more FoVs to select from and can have very narrow FoV for very long range targeting and tracking. Those narrow FoVs can be something like 3-8 degrees in most IRST systems and wide FoV is something like 10-20 degrees. Effective range and FoV is basically inversely proportional to each other. Meaning that 5 degree FoV has 3 times the range compared to 15 degree FoV.

Just take a look at Thales Cathernie MP thermal sight specs: https://www.thalesgroup.com/sites/defau ... asheet.pdf

So I'd say that as a rule-of-thumb, detection range in volume search is about 1/3 of max detection range using narrow FoV. Latest IRST systems probably have almost similar volume search range as earlier generation units had in cued search. Of course modern radar will have far superior volume search ranges against non-VLO targets. That might well change against VLO targets as it's easier to degrade radar signals than lower thermal signature. But IRST systems still have the problem that they are like mechanically scanned radar that they must be slewed relatively slowly over the whole search sector. That means getting usually about 30 frames a second and those frames must have quite a bit of overlap to not have blind spots between frames. Modern AESA radar can do hundreds or even thousands of beams a second being quite equal to one frame in IRST system. So they can scan a huge volume very quickly.

But naturally IRST systems have a lot of good points. They are totally passive systems and they can give very accurate bearing to target. They also have far superior resolution to radar systems and can aid a lot in identifying targets. Sensor fusion is very good to have to maximize the benefits of all the sensors combined. Here F-35 really excels.
Offline

bring_it_on

Forum Veteran

Forum Veteran

  • Posts: 973
  • Joined: 30 Apr 2014, 14:32

Unread post29 Apr 2020, 15:27

hornetfinn wrote:But IRST systems still have the problem that they are like mechanically scanned radar that they must be slewed relatively slowly over the whole search sector. That means getting usually about 30 frames a second and those frames must have quite a bit of overlap to not have blind spots between frames.


Unless someone figures out a way to cooperatively split search sectors in a multi-ship, multi-sensor scenario. :wink:
Offline
User avatar

blindpilot

Elite 1K

Elite 1K

  • Posts: 1271
  • Joined: 01 Mar 2013, 18:21
  • Location: Colorado

Unread post29 Apr 2020, 16:04

BDF wrote:
hornetfinn wrote:Latest IRST/FLIR systems like EOTS and podded systems with 3rd gen staring arrays will likely have very impressive detection and tracking ranges.... It seems like 3rd gen MWIR systems usually have about 2-3 times longer ranges compared to 2nd generation FLIR systems. Technologically for example Pirate and targeting pods of 1990s are mostly 2nd generation IR technology. EOTS and DAS use 3rd gen MWIR system. I think many people will be surprised how powerful system EOTS+DAS is in air-to-air combat.


Its been my understanding that IRSTs suffer a substantial range hit when conducting volume search because of the limited FOV; i.e. the "soda straw" analogy. Rule of thumb I've often heard is the actual detections ranges when conducting a volume search is about 1/3rd of "brochure" ranges which are usually cued detection ranges. Would CCD arrays improve upon volume search times?

Sorry to dig up a zombie thread!


This is also one of those things where experience with long range hunting helps. Simple rifle/ammo combinations give a "MOA" spec. (minutes of angle) as in inches variation at 100 yards. I have had iron beed sights that are close to 1 MOA in size, and at 300 yards completely cover the target such that aim is impossible. .. even seeing the deer past the sight is fruitless.

CCD or IIR et al systems usually are range limited by angular resolution. If the target, miles away, only shows up as one pixel, you have no clue what (if anything) you are seeing. With no range data and limited resolution (and even the highest resolution has an angular limit) you do not know if the "pixel" is a fighter or a bomber or a swarm of drones, or nothing at all. It's one pixel. That's the limit of the targeting, even if found. Much like my iron sight 30.30 is pretty much useless at 300 yards, even though I can see that far and the bullet can get there, ... IRST's have limits way before the "detection" range.

Now add to this the soda straw (FOV) limits on volume search and the chance of useful targeting is not great. Scanning millions of pixels to only get a useless "MOA" is waste of effort. Hunting for a deer on the horizon looking down a narrow field of view scope, just to have it masked by the iron sight when you get there will not put venison on the table. Now a 25 yard quick shot with the lever action cowboy rifle ... that'll work just fine.

This discussion might be more productive on a Colorado weekend deer hunt, carrying a lever action and a well sighted .308 for comparison.

MHO,
BP
Offline

sprstdlyscottsmn

Elite 4K

Elite 4K

  • Posts: 4785
  • Joined: 10 Mar 2006, 01:24
  • Location: Phoenix, Az, USA

Unread post29 Apr 2020, 18:31

blindpilot wrote:This is also one of those things where experience with long range hunting helps. Simple rifle/ammo combinations give a "MOA" spec. (minutes of angle) as in inches variation at 100 yards. I have had iron beed sights that are close to 1 MOA in size, and at 300 yards completely cover the target such that aim is impossible. .. even seeing the deer past the sight is fruitless.

I feel this. I have a Mosin Nagant with a 6 MOA front sight post. All but useless even at 100yds.
"Spurts"

-Pilot
-Aerospace Engineer
-Army Medic
-FMS Systems Engineer
Offline

hornetfinn

Elite 3K

Elite 3K

  • Posts: 3064
  • Joined: 13 Mar 2013, 08:31
  • Location: Finland

Unread post30 Apr 2020, 09:07

blindpilot wrote:CCD or IIR et al systems usually are range limited by angular resolution. If the target, miles away, only shows up as one pixel, you have no clue what (if anything) you are seeing. With no range data and limited resolution (and even the highest resolution has an angular limit) you do not know if the "pixel" is a fighter or a bomber or a swarm of drones, or nothing at all. It's one pixel. That's the limit of the targeting, even if found. Much like my iron sight 30.30 is pretty much useless at 300 yards, even though I can see that far and the bullet can get there, ... IRST's have limits way before the "detection" range.


This is true, it takes at least 1.5 pixels x 1.5 pixels to detect something with good probability. Then it takes at least 6x6 pixels for recognition (tank, aircraft, person, ship). Identification requires at least 12x12 pixels. So identification range is usually almost ten times shorter than detection range using the same optics.

Source: https://www.flirmedia.com/MMC/CVS/Tech_ ... 002_EN.pdf

Better angular resolution can be achieved using higher resolution detector with more pixels and/or using better optics with higher magnification (smaller FoV). This is why there are more than one magnification/FoV setting in IRST and FLIR systems. High resolution IRST/FLIR systems can see quite far with suitable optics and can even be used to identify targets dozens of kilometrs away. On the other hand F-35 DAS is very useful short range IRST system and can detect and track a lot of targets that are close to the aircraft or have high thermal signature (like those rockets it has tracked over thousand kilometer away). But it can't see aircraft or cruise missiles beyond what naked eye can do in good conditions. Of course DAS can do it in total darkness and is likely better than naked eye in bad weather conditions. It's also the only reasonable way to make a long range missile launch detector and missile approach warning system.

I think that if a fighter pilot has to choose between having modern IRST system or radar system, radar system would win every time. However if he has to choose between having either one or both, he would always choose both, especially when combined with advanced sensor fusion. IRST systems (especially modern ones) do have some very nice features to complement radar and other sensors. Volume search is not their forte though, although they can be very effective in it at shorter ranges. Especially so with spherical system like DAS. F-35 overall IR/optical system is especially powerful with so many IRST systems (DAS), combined FLIR and IRST system (EOTS) and networked sensor fusion.
Offline
User avatar

pmi

Active Member

Active Member

  • Posts: 116
  • Joined: 06 Oct 2011, 09:12

Unread post30 Apr 2020, 15:07

blindpilot wrote:This is also one of those things where experience with long range hunting helps. Simple rifle/ammo combinations give a "MOA" spec. (minutes of angle) as in inches variation at 100 yards. I have had iron beed sights that are close to 1 MOA in size, and at 300 yards completely cover the target such that aim is impossible. .. even seeing the deer past the sight is fruitless.


sprstdlyscottsmn wrote:I feel this. I have a Mosin Nagant with a 6 MOA front sight post. All but useless even at 100yds.


Blindpilot, you might want to check your math. A 1 MOA front sight post is unlikely.

(Sight Radius x 2) / Front Sight Width / 21,600

The front sight tip on an M16a2 works out to 10 (ish) MOA with the sight radius of the average shooter. The 1903a3 front blade is 6 MOA. Most battle rifles are in the area of 8 MOA.

All are easily capable of engaging targets at 100 & 300 yards, and beyond.
Offline
User avatar

blindpilot

Elite 1K

Elite 1K

  • Posts: 1271
  • Joined: 01 Mar 2013, 18:21
  • Location: Colorado

Unread post30 Apr 2020, 15:25

pmi wrote:
blindpilot wrote:This is also one of those things where experience with long range hunting helps. Simple rifle/ammo combinations give a "MOA" spec. (minutes of angle) as in inches variation at 100 yards. I have had iron beed sights that are close to 1 MOA in size, and at 300 yards completely cover the target such that aim is impossible. .. even seeing the deer past the sight is fruitless.


sprstdlyscottsmn wrote:I feel this. I have a Mosin Nagant with a 6 MOA front sight post. All but useless even at 100yds.


Blindpilot, you might want to check your math. A 1 MOA front sight post is unlikely.

(Sight Radius x 2) / Front Sight Width / 21,600

The front sight tip on an M16a2 works out to 10 (ish) MOA with the sight radius of the average shooter. The 1903a3 front blade is 6 MOA. Most battle rifles are in the area of 8 MOA.

All are easily capable of engaging targets at 100 & 300 yards, and beyond.


Yeah it ain't that tiny, my bad, probably an order of magnitude off. Still the analogy of a thumb in front of your aim sight is good for IRST discussions. Thanks for the catch.

BP
Offline

wrightwing

Elite 3K

Elite 3K

  • Posts: 3536
  • Joined: 23 Oct 2008, 15:22

Unread post01 May 2020, 03:17

hornetfinn wrote:


Better angular resolution can be achieved using higher resolution detector with more pixels and/or using better optics with higher magnification (smaller FoV). This is why there are more than one magnification/FoV setting in IRST and FLIR systems. High resolution IRST/FLIR systems can see quite far with suitable optics and can even be used to identify targets dozens of kilometrs away. On the other hand F-35 DAS is very useful short range IRST system and can detect and track a lot of targets that are close to the aircraft or have high thermal signature (like those rockets it has tracked over thousand kilometer away). But it can't see aircraft or cruise missiles beyond what naked eye can do in good conditions. Of course DAS can do it in total darkness and is likely better than naked eye in bad weather conditions. It's also the only reasonable way to make a long range missile launch detector and missile approach warning system.



While the exact range of DAS isn't publicized, it's considerably better than equivalent to Mk1 eyeball range. It wouldnt be much use for situational awareness, were that the case.
Offline
User avatar

BDF

Senior member

Senior member

  • Posts: 268
  • Joined: 23 Nov 2006, 13:54

Unread post12 May 2020, 15:24

hornetfinn wrote:IMO, it's good to sometimes get back to these old threads.

It's true that IRST systems are not very good for volume search especially compared to modern radars vs. 4th gen jets. IRST systems tend to have 2 or 3 different fields of view (basically magnification) for different tasks. Wide FoV is used for volume search and navigation. Narrow FoV is used for cued (small volume) search, identification and single target tracking. Some targeting pods have more FoVs to select from and can have very narrow FoV for very long range targeting and tracking. Those narrow FoVs can be something like 3-8 degrees in most IRST systems and wide FoV is something like 10-20 degrees. Effective range and FoV is basically inversely proportional to each other. Meaning that 5 degree FoV has 3 times the range compared to 15 degree FoV.

Just take a look at Thales Cathernie MP thermal sight specs: https://www.thalesgroup.com/sites/defau ... asheet.pdf

So I'd say that as a rule-of-thumb, detection range in volume search is about 1/3 of max detection range using narrow FoV. Latest IRST systems probably have almost similar volume search range as earlier generation units had in cued search. Of course modern radar will have far superior volume search ranges against non-VLO targets. That might well change against VLO targets as it's easier to degrade radar signals than lower thermal signature. But IRST systems still have the problem that they are like mechanically scanned radar that they must be slewed relatively slowly over the whole search sector. That means getting usually about 30 frames a second and those frames must have quite a bit of overlap to not have blind spots between frames. Modern AESA radar can do hundreds or even thousands of beams a second being quite equal to one frame in IRST system. So they can scan a huge volume very quickly.

But naturally IRST systems have a lot of good points. They are totally passive systems and they can give very accurate bearing to target. They also have far superior resolution to radar systems and can aid a lot in identifying targets. Sensor fusion is very good to have to maximize the benefits of all the sensors combined. Here F-35 really excels.


A belated thank you for this. This basically confirms what I've been told in the past. I've come to view IRST as a "network enabled sensor." Which raises some interesting questions on how best to enable its capabilities with the "Combat Cloud" ConOps against LO and VLO threats. As useful and incredibly capable networked APG-77s & -81s will be, I suspect they'll be of very limited use for countering J-20 and J-31s (but maybe more so against Felons) to cue IRST sensors. Then there's also the question of multi-target tracking which I believe is another limitation of IRSTs.
When it comes to fighting Raptors, "We die wholesale..."
Offline
User avatar

BDF

Senior member

Senior member

  • Posts: 268
  • Joined: 23 Nov 2006, 13:54

Unread post12 May 2020, 15:33

hornetfinn wrote:
This is true, it takes at least 1.5 pixels x 1.5 pixels to detect something with good probability. Then it takes at least 6x6 pixels for recognition (tank, aircraft, person, ship). Identification requires at least 12x12 pixels. So identification range is usually almost ten times shorter than detection range using the same optics.


This is another interesting point, i.e. there's a substantial difference between so called "blob detection" and a weapons quality track. This 10:1 estimation from detection to positive ID is more than would have thought was necessary but makes sense. This would suggest that countering LO/VLO threats is going to end up with some sporty, close in engagements.
When it comes to fighting Raptors, "We die wholesale..."
Offline
User avatar

ricnunes

Elite 2K

Elite 2K

  • Posts: 2657
  • Joined: 02 Mar 2017, 14:29

Unread post12 May 2020, 16:05

BDF wrote:As useful and incredibly capable networked APG-77s & -81s will be, I suspect they'll be of very limited use for countering J-20 and J-31s (but maybe more so against Felons) to cue IRST sensors. Then there's also the question of multi-target tracking which I believe is another limitation of IRSTs.


I don't think so. The matter being that the J-20 and J-31 have larger RCS compared to the F-35 and F-22.

For instance the J-20 should have a RCS of 0.05 square meters (and this is already in a good case scenario). And even taken a somehow conservative scenario that the APG-81 can detect a 1 square meter target/aircraft at a range of 260 km this means that the same APG-81 would detect a J-20 at a range of 122 km or almost 66 Nautical Miles and this in a wide search.

And even if the J-20 or other aircraft (like the J-31) have a RCS of 0.01 square meters this would translate in a detection range of 82 km or around 44 nautical miles with the APG-81 radar.

Any of the values above would be much higher than possible with the IRST's (specially during wide search).
“Active stealth” is what the ignorant nay sayers call ECM and pretend like it’s new.
PreviousNext

Return to F-35 Avionics

Who is online

Users browsing this forum: No registered users and 5 guests