Type 346 radar speculation

ricebunny

ACCESS: Restricted
Joined
21 December 2020
Messages
12
Reaction score
14
Based on what I have gathered, the first version of the radar is a S-band AESA, with an octagonal antenna of about 4m diameter. The AESA is built out of quad TRMs, with each quad capable of 100W peak RF power. Total number of quads is estimated to be 1250, for a total of 5000 T/R elements and 125kW peak RF power per antenna. First version of the radar employed air cooling.

The S-band antenna is sandwiched between two C-band arrays. According to one rumor, the function of the C-band arrays is to provide uplink/downlink communication with the principle weapon of the the Type 052C destroyer: HQ-9. The missile itself uses active radar homing to target. Another theory is that the C-band arrays are illuminators and the missiles SARH or TVM. However their rectangular shape does not seem optimal for that role.

Only known photo of the S-band/C-band antenna, taken from a type 052C fitting out
1620239879002.png

What do you guys think? Was the Chinese state of the art really at this level in the early 2000s? The USN put to sea its first S-band AESA (SPY-4) radar a full decade later and was also late to the party in introducing a long range ARH SAM.
 
Given that it originally had a folding transmitter I think at least one of its methods of operation was as a space fed offset feed reflectarray. I remain very skeptical at the essays on Chinese equipment that appear on wikipedia, but do agree with assessments that the radome covers multiple antennas for search and track/illumination functions. Note that the notion that the whole thing was a massive AESA was, in the west at least, largely based on a single assessment by, IIRC, Friedman, who noted the large ducts to the array face installation on the test ship and concluded it must be AESA because the array face needed a lot of cooling. Image of the array with the early curved radome and what I suspect to be the feed "stick" unfolded is attached.
 

Attachments

  • reflectarray.JPG
    reflectarray.JPG
    57.9 KB · Views: 63
Interesting theory. I always thought that the retractable rod was something like a near field calibration device.

I have Friedman’s last World Naval Weapon Systems book (2006). Therein he speculates about it bring derived from the Ukranian Kvant bureau radar, but I think there was no evidence since to support that thesis.

Furthermore, he speculated that the radar operates in C-band, because the state-of-the-art at the time was not at a level where sufficient power could be packaged in S-band modules small enough to populate the array. It’s interesting to point out that the land version of HQ-9 used a C-band search and illuminate radar. However, wouldn’t such a large array be prohibitively expensive if it is densely populated with C-band T/Rs?
 
"I always thought that the retractable rod was something like a near field calibration device." It's possible, it was censored in some early pictures which was what made me suspicious. :D
 
I think that the original radar that appeared on the Type 052C may have been a C-band radar as this is most consistent with the ground based HQ-9. The actual C-band array should be quite a bit smaller than the radome, because it contained an IFF interrogator, missile uplink and downlink antennas for TVM and possibly side lobe cancellation antennas. Also, the array does not need to be as dense in the periphery which would help keep the manufacture cost manageable.
 
The story of Type-346 from the designer apparently.


C-band was considered but S-band ended up being selected, with C-band uplink/downlink for missile guidance. The elements are quad brick design with length of 45 cm, width of 20 cm and height of 5 cm. The peak power of those quads are 100 W which indicates 25 Watt/ elements. Assuming 20% duty cycle it will be 4 Watt Average. It curiously claimed 4768 TRM's instead of 5000, there probably some modules left passive as means to control sidelobe of the array.

Further proof of S-band instead of C, can be gleaned from the mentioned TRM dimension. 20 cm, which means each channel have width of 5 cm, this is half wavelength of 10 cm (3 GHz) or S-band, there might be cold plates in between the transmit channel but it's unlikely to be so thick in separating the module that it can be C-Band. I lean to 3.4 GHz tho or even higher (4 GHz).

I also made some simple estimates based on Almaz-Antey paper on radar design which includes the weight data and a Chinese paper on TRM weight reduction. The result was, for this 4 Channel modules and 3.4 GHz. Each quad will weigh about 7.4 Kg. For comparison, US SPY-6 AMDR 6-channel TRIIM weigh about 5.3 Kg.

7.4 Kg for each quad makes each face of Type 346 described in the link weighs about 9.2 Tonne. With total "top deck" weight of 36.8 tonne. The under-deck weight should be lighter as the lack of waveguides and signal generator assemblies.


China itself tho already quite deep in Active Array research, this is example of Chinese L-band TRM from 1990's.

1734272268893.png
 
Translated Chinese paper on Midlife update option for early Type-052C's Type-346.


Some interesting passage.

1736615232118.png

Backfit of Type 346A to Type 052C wasn't straightforward as the she would need considerable rework on the cooling and power generation system. Indicating the considerable differences between Type 346 and 346A power wise. It's perhaps close to the US attempt to backfit AMDR to early Burkes, which, due to the lower power and cooling capacity to the Flight II, the AMDR there has reduced RMA compared to Flt III. Also concern with increased topside weight due to the fact no more heavy waveguides and transmitters below deck to help with CG.

The paper provide some good datapoints on the approximate size of the type-346 radar and its evolution to 346A.

It's been established that the 346 was having 4678 TRM's and the frequency based on available TRM dimension turned to be 2940 MHz. Assuming 4 m diameter antenna. The 346A have 5700 TRM from 20% gain of area from changes to liquid cooling and it no longer need the curved cover. The average power could be uprated by twice to further improve the range. This yield range of roughly 426 km or 42% more from baseline Type 346 (300 km). Improved signal processing can of course add more. For me 346A basically closed the gap at least in detection range with AN/SPY-1 B/D on US warships.

The final puzzle is of course the Type 346B for Type-55. An article suggest 40% larger area compared to 346A which said to be 20% larger than baseline Type 346. With 2.94 GHz frequency The type 346A have about 4.3 m diameter antenna thus 14 sqm of area compared to 12.25 sqm of Type-346. Type 346B if 40% increase in area to be accepted it would have 20.6 sqm area or in terms of diameter 5.12 m. Same 2.94 GHz frequency would indicates TRM count of 7980. The range advantage assuming same TRM power as Type 346A would be 1.83 or 83% increase or almost Twice compared to baseline 346 or 549 km.
 
Is there any speculation how the overall performance of the Type 346A/B radar compares to the new AN/SPY-6(V)? Both seem to be modern GaN powered AESA arrays with similar antenna diameter
 
Is there any speculation how the overall performance of the Type 346A/B radar compares to the new AN/SPY-6(V)? Both seem to be modern GaN powered AESA arrays with similar antenna diameter

No i guess. Rough estimates tho can be made from TRM Numbers. Which are known.

Type-346A : 5700
Type-346B : 7980

SPY-6 Backfit to Flt-II : 24 RMA or 3456 TRM's (as 1 RMA contains 144 TRM channels)
SPY-6 Flt-III : 37 RMA or 5328 TRM's

Discounting other differences like frequency, cooling and power generation capacity (SPY 6 is approximately 3.4-3.7 GHz same as SPY-1 while Type 346 is 2.94 GHz, closer to AN/APY-1/2 for E-3 Sentry) Type 346A Has clear advantage over SPY-6 in terms of TRM. In range that would translate into, using SPY-6 as reference :

Rfactor = ((5700/3456)^3)^(1/4)
Rfactor = 1.45 So Type 346A has 45% advantage in range vs 24 RMA SPY-6 and equal to 37 RMA. While for the 346B it would have 35% range advantage vs 37 RMA SPY-6 and almost twice (87%) range advantage vs 24 RMA.


If people ever wonder why N^3 factor for AESA ? Remember in radar range equation there is Power aperture product which denoted as :

Pav*G^2.

Pav is average power of the transmitter while G is gain, why gain is squared ? it's because antenna reciprocity theory which denotes Gain for transmitting is the same as gain for receiving. AESA antenna contains multiple small active elements with its own gain and transmit power, thus their power aperture product is easily translates into the Cube of the numbers of the TRM they have. Increasing number of TRM is not only increase the Gain of the antenna but also transmit power at the same time.


--Edit

So i found another sources or more like compilation of sources about the Type 346. This one includes story of apparent Ukrainian "Kvant" (not Kran as in the articles) Contribution. Tho i am doubting, mainly because they never really deliver active array radar or even finished the development of Mars passat/Sky Watch.


Also. I make a "quick guide" on 346.

Dragon Eye.png
 
Last edited:

Similar threads

Please donate to support the forum.

Back
Top Bottom