Key.Aero Network
Register Free

Page 2 of 3 FirstFirst 123 LastLast
Results 31 to 60 of 70

Thread: AESA Radar range calculator.

  1. #31
    Join Date
    May 2012
    Posts
    865
    Well have you looked at the equation i posted ? It's not simply like that. The beamwidth is squared.
    I did, but to be honest, I don't reall understand why it can't be calculated my way. Can you explain in layman terms?

  2. #32
    Join Date
    Sep 2015
    Posts
    715
    Well the beamwidth is squared to take account of its cross sectional area. The number of beam positions equation basically states the area of the search sector divided by the cross sectional area of the beamwidth.

    So you can't just divide the horizontal sector with beamwidth and later multiply it with dwell time to get scanning time.

  3. #33
    Join Date
    May 2012
    Posts
    865
    The number of beam positions equation basically states the area of the search sector divided by the cross sectional area of the beamwidth.
    To summary, the number of beam positions equivalent to the area of search sector (in degrees) /area of beamwidth (in degrees). Correct?. It's like filling a rectangle with small squares.
    So if the radar can't scan vertically, wouldn't it make sense to only consider the horizontal length?. Vertical length is irrelevant because radar can't scan up and down.
    Last edited by moon_light; 25th September 2017 at 17:34.

  4. #34
    Join Date
    Sep 2015
    Posts
    715
    The vertical length of the search sector might be irrelevant. But, radar beamwidth still have 2 dimensional nature, thus you can't simply discard it as the target may arrive at the vertical part of the beamwidth. Therefore the squared value

  5. #35
    Join Date
    Dec 2003
    Posts
    126
    Very insightful, thanks.

    I have some general questions:

    - How applicable is the use of the spreadsheet for PESA? Can we take the peakpower, divide it by amount of phaseshifters and put it in a T/R modules? Would the result be representative or are there some effects for AESA that are considered and hence this would deliver false results?

    - How would the parameters of a hypothetical AESA 92N6 S-400 engagement radar look like? Since just a single array is used, we can exclude a continuous wave operation (although many sources for some reason state that the SA-6 with its single engagement antenna is a CW system)? So how would the parameters (PRF, pulsewidth) for a pulsed illumination look like approximately?

    - A general question on AESAs for illumination role I always had is, whether a part of the array could use transmit-only elements with higher peak power per element, and the other part normal T/R modules. I see much higher power levels for transmit-only modules which may could compensate the lower number of receive-capable modules. Even if there would be no benefit for the range performance due to the lower receive modules; in SARH (bi-static) operation, the missile seeker would receive higher RF energy.

  6. #36
    Join Date
    Oct 2013
    Posts
    1,979
    Quote Originally Posted by stealthflanker

    Linear array beamwidth is basically 2/N Thus for array of 24 elements it is 2/24=0.08 Radian or 4.7 degrees
    Does that formula take into account elements spacing?. Elements seem very close together, your example wavelength is 24 cm, 24/2= 12, i doubt that woman wrist can be 12 cm wide. Even 6 cm wide wrist seem very rare
    Last edited by mig-31bm; 27th September 2017 at 14:33.

  7. #37
    Join Date
    Dec 2008
    Location
    Norway
    Posts
    6,369
    Guys.. i hope you don'T mean to debate the L-band IFF transmitter as a magical Russian AESA.

    And likewise ppl whom state that the L-band is ineffective as a radar to detecting Stealth or whatever. Both are dead wrong!

    The Key point is wave lenght. As it is an IFF system, it is not required to constantly hit a Radar Contact with Radio waves. One small return is enough for the N035 complex to catorize the target as Friendly, unknown or Bandit. The onboard computer will memorize the rest of any contacts.
    Thanks

  8. #38
    Join Date
    Jan 2009
    Posts
    4,916
    One small return is enough for the N035 complex
    IIRC, That is not how IFF works.

    It's basically an AESA radio. It's sends out a coded pulse. If the targeted aircraft recognizes the pulse it then sends a coded pulse in reply. It's not a radar that is listening for a reflected return of it's own pulse.
    "The early bird gets the worm but the second mouse gets the cheese."

  9. #39
    Join Date
    Dec 2008
    Location
    Norway
    Posts
    6,369
    My knowhow on Radars are not that Great. But that i knew, spud. Its traditionally called a radar transponder, it contains cryptic ID on your own jet and other jets Allies or own airforce. Remember what happend when the Mig-25 pilot defected. Every military Radar transponder had to be re coded in USSR.

    Anyway my point was you do not have to constantly T/r on the IFF to keep contacts Identified. Idealy Its a one deal ordeal.

    If not you loose a contact over some given time and then re-acq it.. there are limitations.
    Today the Radar transponder is part(integrated and software driven) of the radar complex itself.

    Just don't think of it as a AESA Radar in the traditional way.

    My 2cent is that the Su-35S transpond IFF in X- band from main Array as well(only logical). For N035 it is advs that it is less prone to Jamming. Guess T/r IFF in L-band has its merrits.
    Last edited by haavarla; 27th September 2017 at 17:08.
    Thanks

  10. #40
    Join Date
    Sep 2015
    Posts
    715
    Quote Originally Posted by PeeD
    Very insightful, thanks.

    I have some general questions:

    - How applicable is the use of the spreadsheet for PESA? Can we take the peakpower, divide it by amount of phaseshifters and put it in a T/R modules? Would the result be representative or are there some effects for AESA that are considered and hence this would deliver false results?

    - How would the parameters of a hypothetical AESA 92N6 S-400 engagement radar look like? Since just a single array is used, we can exclude a continuous wave operation (although many sources for some reason state that the SA-6 with its single engagement antenna is a CW system)? So how would the parameters (PRF, pulsewidth) for a pulsed illumination look like approximately?

    - A general question on AESAs for illumination role I always had is, whether a part of the array could use transmit-only elements with higher peak power per element, and the other part normal T/R modules. I see much higher power levels for transmit-only modules which may could compensate the lower number of receive-capable modules. Even if there would be no benefit for the range performance due to the lower receive modules; in SARH (bi-static) operation, the missile seeker would receive higher RF energy.
    1. I will not recommend it for PESA. As it need to be treated differently. Especially for the space feed array. Russian research on Space feed array appears to be ahead of US. To the point where they can realize advanced feed with low sidelobe. As described below :

    https://www.scribd.com/document/2753...-Radar-Systems

    Some technical informations regarding the design of array in S-300 and S-300V

    https://www.scribd.com/document/2890...t-Phased-Array



    2.Hypothetical AESA for 92N6 would be massive. 2.5 sqm array of 92N6 will contain about 12000 TRM assuming it wish to retain same beamwidth as original 30N6/92N6. The maximum attainable peak transmit power with current state of the art technology (1.5 Watt/mm for GaAs) would be 22 Watt multiply it with 12000 it would be 264 Kilowatt for peak transmit power using GaAs. GaN would be even more powerful with 7 Watt/mm power density. X-band module with half wavelength width would attain 105 watt. and the transmit power would be 1.26 MW for 12000 elements. Range figure with GaN would roughly be about 68% higher than GaAs.

    The 5N63 family and her siblings, 30N6 and 92N6 is a pulse doppler design with 100 KHz of PRF. The pulsewidth would be around 1-2 microseconds. There is no info though on how much pulse compression factor involved. Unfortunately so if you desire to calculate, you will have to consult on the table of waveform i provided. In previous page. There also be another factor which i haven't implemented in the sheet which is F4 or Pattern propagation Factor. This would affect ground based radar more than the one airborne. The design duty cycle however could be assumed to be 25% maximum for avoiding/reducing eclipsing issues.

    Regarding Illumination role. Some AESA's do behave similar way as you describe, which is SMART-L where some elements with wide beamwidth will "flood" the area with radar energy while the receive party scans the skies for target return. However it's the transmit module that are less in number than receive as transmit does not need such large beamwidth. Receive module would be more in numbers due to need for narrow beamwidth resolution.

    Quote Originally Posted by Mig31BM
    Does that formula take into account elements spacing?. Elements seem very close together, your example wavelength is 24 cm, 24/2= 12, i doubt that woman wrist can be 12 cm wide. Even 6 cm wide wrist seem very rare
    Yes, my formulas automatically assume that it's a half wavelength spacing. If there is any less like in the image, the only explanation would be that the array isn't exactly 1250 Mhz but rather lower wavelength. It could operate in 1250 Mhz but with limited scanning angle. Otherwise it might not be in L-band at all but something higher maybe S or C band.

    Quote Originally Posted by Haavarlaa
    Guys.. i hope you don'T mean to debate the L-band IFF transmitter as a magical Russian AESA.
    Well given the controversial nature of the device, one or two debates would occur sooner or later. and naturally i am also interested to see its potential as real Primary radar. The reason is similar as given by Carlo Kopp. Why putting the antenna there ?.

    Traditionally IFF is always in L-band. Implementation however usually embedded together with main nose array Thus the radar can interrogate while also doing scanning. If it placed anywhere else, naturally one would question why. I would love to see other people brewing their own calculations too.

    I really not bought into theory "the beamwidth is too wide" etc as the radar have narrow beamwidth just for 1 side (horizontal) and that's about enough for scanning. The elevation beamwidth is indeed very broad But, treatment with dielectric lens can mitigate it into somewhat acceptable (still wide tho 20-80 degrees) Heightfinding cannot be done in the way usual radar does BUT can use the same way as how E-2 determine altitude via multipath heightfinding.

    E-2 Hawkeye's radar antenna is also a linear array consist of several YAGI elements. The differences is that it scans mechanically while this NIIP array scans electronically. The APY-9 for new E-2D would scan electronically and conceptually similar as one in NIIP.

  11. #41
    Join Date
    Jun 2004
    Location
    Columbia, MD
    Posts
    12,128
    2.Hypothetical AESA for 92N6 would be massive. 2.5 sqm array of 92N6 will contain about 12000 TRM assuming it wish to retain same beamwidth as original 30N6/92N6.
    Do you mean 2.5 sqm or 2.5 m side?
    Old radar types never die; they just phased array

  12. #42
    Join Date
    Sep 2015
    Posts
    715
    Yeah my mistake 2.5 m should be the width of the antenna not area.

  13. #43
    Join Date
    Jun 2004
    Location
    Columbia, MD
    Posts
    12,128
    That's what I thought As far as other X-band radars, you have the TPY-2 @ 9.2 sq. m (25,000 T/R modules), and the MEADS X-Band radar @ 3.5- 4 sq. m (10,000 modules).
    Old radar types never die; they just phased array

  14. #44
    Join Date
    Dec 2008
    Location
    Norway
    Posts
    6,369
    Well given the controversial nature of the device, one or two debates would occur sooner or later. and naturally i am also interested to see its potential as real Primary radar. The reason is similar as given by Carlo Kopp. Why putting the antenna there ?.

    Traditionally IFF is always in L-band. Implementation however usually embedded together with main nose array Thus the radar can interrogate while also doing scanning. If it placed anywhere else, naturally one would question why. I would love to see other people brewing their own calculations too.

    I really not bought into theory "the beamwidth is too wide" etc as the radar have narrow beamwidth just for 1 side (horizontal) and that's about enough for scanning. The elevation beamwidth is indeed very broad But, treatment with dielectric lens can mitigate it into somewhat acceptable (still wide tho 20-80 degrees) Heightfinding cannot be done in the way usual radar does BUT can use the same way as how E-2 determine altitude via multipath heightfinding.

    E-2 Hawkeye's radar antenna is also a linear array consist of several YAGI elements. The differences is that it scans mechanically while this NIIP array scans electronically. The APY-9 for new E-2D would scan electronically and conceptually similar as one in NIIP.
    As radars on Fast flying jets become more powerfull, so does the way of interrogate any contact in the sphere of operation area.

    If you have several Su-35S in the same Op Area, but splitt up in different wings, ideal you want to be able to interrogate the moment any contact pop up on your MFD screen, why the hell not!
    Again its only logical.
    At the very least, you will "KNOW" that those contacts are friendlies, and by that concentrate you vector approach, get enough time to set up an Intercept of any unknowns or Bandits.
    In a way, having very good means to interrogate the airspace(battle space) makes up for good Situational Awareness. This stuff about IFF is important!
    Those LE slate Arrays are a better way of acheiving this.
    The same apply for interrogate towards AWACS and Anti-Air systems.

    Again, ppl talking about these L-band array being an agumenting ways to the main radar array in order to increase radar power and resolution is not correct. At least not directly. They are for IFF purpose.
    Last edited by haavarla; 29th September 2017 at 01:17.
    Thanks

  15. #45
    Join Date
    Oct 2013
    Posts
    1,979
    Quote Originally Posted by stealthflanker
    Yes, my formulas automatically assume that it's a half wavelength spacing. If there is any less like in the image, the only explanation would be that the array isn't exactly 1250 Mhz but rather lower wavelength. It could operate in 1250 Mhz but with limited scanning angle. Otherwise it might not be in L-band at all but something higher maybe S or C band.
    can use the same way as how E-2 determine altitude via multipath heightfinding
    Judge by the elements spacing, i think the radar probably operate at 2.5Ghz. Btw, can you elaborate multipath heightfinding?

  16. #46
    Join Date
    Sep 2015
    Posts
    715
    @Mig31bm.

    Well if it's 2.5 GHz then using it as IFF is kinda suspect as both Russia and US use L-band for IFF purpose, and having S-band, despite having better gain doesn't seem to have its merit in detecting low RCS target. Regarding Multipath height finding, you can read it here

    http://www.radartutorial.eu/01.basics/rb63.en.html
    ------------

    Regarding the AESA calculator, im still workin on it... hmm finding better and more relevant/correct method to calculate. Current plan is to add ECM. The ECM being added is Noise jammer and Cross Eye jamming.

    The noise jammer will provide calculations of Burn through range, while the cross eye will calculate possible error generated by the jamming. Counter-Countermeasure for cross eye however im afraid can't be really taken into factor because :
    1.Limitations of excel, can't really do complex integrals.
    2.There are no real literature published regarding cross eye counter-countermeasure. According to Introduction to Airborne Radar by Stimson. The method relevant to counter cross eye is currently classified. However let's see what i could find.

  17. #47
    Join Date
    Dec 2003
    Posts
    126
    I get some values that look too high with the calculator. You already said that it is not fully applicable to ground based radars but I wonder what the penalty would look like?

    Here is a example T/R module for a S-band radar such as the Ground Master: http://micro.apitech.co.uk/pdf/aesa/...module_TRM.pdf

    With 100µs pulsewidth and a duty cycle of 20%, we get a resulting, possible PRF of 2 khz. Now anything between 3000-5000 modules at 100w, the resulting range is in the range of 3000km (RCS 2m˛, dwell time 0,1s).

    I'm sure operation at max. module performance level is not the usual operation regime. But even at half the performance parameters, the range is still 1700km. These performance levels would result in very high range performances, especially against VLO assets.

  18. #48
    Join Date
    Sep 2015
    Posts
    715
    Radar Horizon. You will not get 1000km range unless the aircraft is flying outerspace or in extremely high altitude.

    The calculator has not taking factor of path propagation (The F^4) in the calculation. Which could be ideal for airborne situation but not ground. There are also STC (Sensitivity Time Control) and possible MTI employment which will limit the detection range (esp minimum) even further. This is not implemented yet as it basically making a whole new calculator as MTI and STC is range dependent and calculation must be performed iteratively.

  19. #49
    Join Date
    Dec 2003
    Posts
    126
    Okay, so is it the maturity of advanced S-band silicon T/R modules that has created such high range performances?

    The effects you mentioned should create negligible range performance penalty, am I right?

    As a example: Such a S-band AESA in form of a ground based system like the Ground Master or an airborne AESA AWACS would have following performances with ~4000 modules:

    50% Pd volume search: 300km against a -40dbsm target and that in 10cm S-band where RAM and RAS should be much lower performing. Means if we are generous and give a -10dbms RAM/RAS performance in S-band, its range performance would be in fact 300km against an effectively -30dbsm target (0,001m˛ RCS, shape only). That would be a typical long range EW radar range performance against almost any known stealth fighter target.

    I'm somewhat skeptical about those figures. Maybe those pulse width and max. PRF values from the T/R module data sheet are for some reason not applicable?

  20. #50
    Join Date
    Sep 2015
    Posts
    715
    The high performance are simply because of the high power of the TRM and large number of modules.

    And the effects i mentioned..it is significant especially for Early warning application where STC can actually prevent detection of low RCS target and long pulsewidth means low minimum range (thus the radar cannot detect target that are actually close) and possibility of target lost due to eclipsing.

    I really do not recommend using it for ground based radars due to those variables that i haven't take account in. Unless suitable modifications or at least some clear statements about limitations of the calculations.

    Another issue not related to range could be..Whether it would make a practical system.

    4000 modules of planar array would result in about 4x4 meter antenna. Might be practical for ground based but quite bizzare for AEW might end up making the plane look like Chile Phalcon. There could also be cost issue that made this Radar unsuitable for general air defense. It would be more suitable for ABM.

    The cost of radar hardware could be roughly estimated as follows :

    Cost=Nt*Ct+PaV*CaV*Nt

    Nt=Number of Trm
    Ct=Cost of Trm
    PaV=Average Power
    CaV=Cost of producing that average power (let's assume this to be U$ 1)

    The cost of typical TRM in S-band is about U$ 200.

    Name:  trm_price_by_stealthflanker-dbnseb2.jpg
Views: 222
Size:  39.9 KB

    Thus the radar would cost about :

    Cost=4000*200+80000*1*4000
    Cost=320800000

    320 Million dollar radar. Would make sense for ABM applications.
    Last edited by stealthflanker; 10th October 2017 at 20:30.

  21. #51
    Join Date
    Jun 2004
    Location
    Columbia, MD
    Posts
    12,128
    There are already such radars that operate in the S-Band and utilize GaN based PAs along with DBF. The largest currently is the SPY-6 (AMDR-S) at 5000+ modules, while the more mobile radars such as the Giraffe and TPS-80 are in the 2000 module range. Similar sized GaAs based systems also exist. The largest variant of the ELM-2084 should be comparable to the Giraffe in module count (similar rough dimensions).

    https://forum.keypublishing.com/show...23#post2402523

    And the costing data is not right. AMDR does not cost 300 million per face.
    Last edited by bring_it_on; 10th October 2017 at 20:34.
    Old radar types never die; they just phased array

  22. #52
    Join Date
    Sep 2015
    Posts
    715
    Yeah, i know. I just argue that it won't have such range even with the max specification of the module and if it deliver so. Then be ready to accept high cost.

    Reserve should always be exercised. considering as i mentioned before this is noise limited. no clutter and environment factor is only come from attenuation. and No F^4 factor. Plus one must always consider change of target RCS on wavelength.

  23. #53
    Join Date
    Dec 2003
    Posts
    126
    I would expect half wavelength spacing, hence a 2x2m array @ 4000 elements, which would be sufficiently compact also for airborne AEW purpose.

    For such purposes it would also make sense to use lower power (~50W), non-GaN S-band modules, maybe even tile modules to drive the price down to <100$ levels.
    For total costs, in a efficient design, the TRM cost could make up 50% ($400k) of the total radar production cost.

    With all effects taken into account, an airborne array for an AEW platform would have huge range performance or very high anti-VLO performance. So high that it makes it a little suspicious.

  24. #54
    Join Date
    Sep 2015
    Posts
    715
    No @Bring_it_on.

    The equation is correct. I got it from a RAND paper. bit dated but i think still valid. One variable i haven't play around is the cost of producing the average power of that radar. The original paper gave like U$5 which obviously very high. But today with manufacturing advancement i believe the cost is much lower but how low.. i wonder. If we assume that variable to be 50 cents or U$ 0.5 The array cost would halved. Which i assume closer to your data of AMDR cost.
    @Peed

    Well the calculator already assume half wavelength. If you see at the "Weighting algorithm table" tab you may see antenna width. For planar array.

  25. #55
    Join Date
    Jun 2004
    Location
    Columbia, MD
    Posts
    12,128
    The equation is correct. I got it from a RAND paper.
    The four year old Engalco report (where your graph comes from) is general and likely based on reported commercial industry costs and may not reflect facilities that have achieved certain MRL status at the back of DOD investments and are sandboxed from commercial supplies since the non-reccuring costs have been funded by the DOD maturation process. Specifically to the AMDR, compared to your formula of $300 Million for a 4000 Module Radar, 2 AMDRs with a combined TR count of 10,656 cost approximately $230 Million in 2017 as per DOD cost data.

    I would also be a little cautious on a dated report since there has been a significant ramp in industrial capacity for S-Band AESA supplies on the GaN side given recent R&D, production and acquisition activity. Some of the largest AESA radar projects in the world and US are in the S-Band GaN space such as Lockheed's LRDR and Space Fence, Raytheon's AMDR and EASR, and Northrop Grumman's TPS-80. The 2015-2020 production ramp at both defense exclusive foundries and commercial and DOD suppliers would have been dramatic. I believe Lockheed's supplier has delivered all TR Modules for Space Fence already and they are being shipped out. Deliveries for LRDR assemblies have probably also begun. The cost reduction on account of all that would probably be more dramatic than was anticipated before all these were awarded.
    Last edited by bring_it_on; 11th October 2017 at 12:23.
    Old radar types never die; they just phased array

  26. #56
    Join Date
    Jun 2005
    Location
    Cemetery Junction
    Posts
    13,549
    Peed:
    I would expect half wavelength spacing, hence a 2x2m array @ 4000 elements, which would be sufficiently compact also for airborne AEW purpose.

    For such purposes it would also make sense to use lower power (~50W), non-GaN S-band modules
    Erieye-ER (one customer so far) has S-band GaN modules. I think the Erieye antenna is 8 x 0.6 metres, so about the same as your 2x2 metres. Dunno how many modules, though, or what power.

    The GaN modules are said to give a massive increase in detection range of VLO objects. Maximum range against big, high RCS things such as ships is still the horizon.
    Juris praecepta sunt haec: honeste vivere, alterum non laedere, suum cuique tribuere.
    Justinian

  27. #57
    Join Date
    Sep 2015
    Posts
    715
    Unless there is real up-to date report. I will stand by those i have. Not that i realize the dramatic price decrease or technological advancement but.. What value should i use ? That is the most important. If there is no real value naturally i will seek the closest i could find. Cautious as i am but.. a calculation cannot be made without a starting value.

  28. #58
    Join Date
    Jun 2004
    Location
    Columbia, MD
    Posts
    12,128
    Right, but there is little value in trying to calculate cost based on numbers that are this off. When you can buy 2, larger (TR count) radars and still save $50 million over the calculated cost of 1 x 4000 module radar then those calculations aren't of much value.
    Old radar types never die; they just phased array

  29. #59
    Join Date
    Sep 2015
    Posts
    715
    Then what should i use as input ? What adjustment need to be made ?.

    I would call it conservative instead of "off"

  30. #60
    Join Date
    Jun 2004
    Location
    Columbia, MD
    Posts
    12,128
    I can't give you he exact information since it isn't easily available and requires access to information that I do not currently have. All I can tell you is that if a 4000 module radar cost calculation comes in at a cost that is $90 Million above the cost of 2 larger radars (10656 modules combined) then it is likely WAY WAY OFF . I would call a margin of 20-30% as being conservative and intentionally leaving some padding for inflation and other escalatory pressures. 2 TPS-80s even @ LRIP will run you just under $100 Million and that would be roughly 4000 combined TR Modules or less than a 1/3 of the cost predicted for a similar sized set up.
    Old radar types never die; they just phased array

Thread Information

Users Browsing this Thread

There are currently 1 users browsing this thread. (0 members and 1 guests)

Posting Permissions

  • You may not post new threads
  • You may not post replies
  • You may not post attachments
  • You may not edit your posts
  •  

 

- Part of the    Network -

KEY AERO AVIATION NEWS

MAGAZINES

AVIATION FORUM

SHOP

 

WEBSITES