Type 052C/052D Class Destroyers


AndrewS

Colonel
Registered Member
Procurement cost. And to be exact - powerplant still constitutes a significant part of the price(double digits).
But when we go to operation costs(and logistics load) - powerplant becomes of huge importance - the more active ship is, the costlier it gets.

That's assuming 052D can't do the same networking(or there is no 055 in TG).
Given that 052Ds are very recent ships, and their radar/cooperative engagement/battle management system development schedule proceeded at least partially in parallel with 055 - why such an assumption?

As per an old Arleigh Burke cost analysis I've read, fuel accounts for approximately 10% of the total lifetime cost.

It just doesn't seem that significant.
 

Gloire_bb

Senior Member
Registered Member
As per an old Arleigh Burke cost analysis I've read, fuel accounts for approximately 10% of the total lifetime cost.

It just doesn't seem that significant.
Total lifetime cost includes crew expenses and modernization(s), which may include overhaul/replacement of powerplant.
This masks things quite a bit.

In day-to-day operations, fuel is still a pain. +in wartime it is simply a limited consumable.
 

AndrewS

Colonel
Registered Member
Total lifetime cost includes crew expenses and modernization(s), which may include overhaul/replacement of powerplant.
This masks things quite a bit.

In day-to-day operations, fuel is still a pain. +in wartime it is simply a limited consumable.

What you have is just an opinion.

Where are the numbers???
 

AndrewS

Colonel
Registered Member
Arleigh Burke IIA - total cost of lifetime ownership
R&D: $72M (2%)
Procurement: $1484M (49%)
Personnel: $897M (29%)
Fuel: $331M (11%)
Other O&S (maintenance) : $258M (8%)

So you can see that procurement and personnel account for 78% of all costs.
Ongoing powerplant maintenance should come under Other O&S.

Life cycle costs of selected warships
Please, Log in or Register to view URLs content!
 

Tam

Brigadier
Registered Member
If they can communicate with satellites, high flying UAV or aircraft - they can almost certainly communicate with other ships.
That is the only logical conclusion



Given that the Gallium AESAs are amongst the most expensive part of the Type-052D, and will still be useful for the rest of their lifetimes - I doubt we will see them upgraded. It would be better to build new ships with Gallium Nitride AESAs instead

There are only 39 modern AEGIS destroyers so far, and I expect the Chinese Navy to be aiming for a fleet of 90 AEGIS destroyers in total.

They probably will be upgraded to a new generation of AESA module. Even for the reasons of logistics and maintenance, it needs to be done. Even if you think the ship doesn't need to, and the old radar's performance remains adequate, it still needs to be done.

Think of it. After some time, production of the Gallium Arsenide modules would have stopped, with the remaining stock going into warehouses as spares. With the introduction of new ships, the new ships will be using Gallium Nitride modules, and all production will be shifted to using Gallium Nitride.

Over time, as the 052Cs and 052Ds wear out their arrays and burn out older the modules, the remaining stock will eventually be depleted. Electronics burn out, and they will eventually expire, much sooner than the rest of the ship. All this kit is highly modularized. They are meant to be taken out and replaced quickly. An AESA array has similarities to a server farm. Like any computing architecture, everything is designed in consideration for things to come, including upgrades.

So what you are going to do, is to refit the radars to use the new module standard that is current and available at the time of the refitting event. They get new modules, the cooling system is overhauled, and the backend is updated to account for the new modules.

It will not affect the production of new ships. Instead, the more modules you make, both for new ships and for refitting older ones, the lower the cost of each module, driven purely by volume amortization. It also makes things a lot easier. For example, low production units would require FPGAs. FPGAs are temperature hot and expensive, not to mention China ain't exactly the leader in this field. You can forego FPGAs and use ASICs instead. ASIC runs cooler, and is cheaper, but requires a large production run. You want to order the ASICs specific to these radars by the many thousands, and the only way you want to justify such a large volume to plan the total number of ships you want to produce over time that would use the same module, plus the number of existing ships you plan to upgrade with. The 40+ 052C/D lifetime total would represent an enormous market for such chips to add to the demand for the new builds, that it would make a serious dent in the price of such chips when compared if these chips are only going for new builds.

Think of the Sovremenny refits. The radar refits are mostly sidegrades. Fregat MAE to Type 382, Mineral ME to Type 366, MR184 to Type 344, MR123 to Type 349, MR90 to the Chinese illuminators. At best there is only a bit of a slopegrade here. The main reason why the Russian ones are being replaced is for maintenance reasons. The Chinese cannot maintain the Russian ones indefinitely as parts and warranty supplies run out. Much easier to replace with the Chinese radars where they have control of the parts supply.


As for datalinks, there are quite a few forms of datalink antennas. The ones I see with PLAN ships often take the form of poles, either long or short, or small cylinders. These are likely to be monopoles or dipoles. They can communicate with another ship easy, or with a plane or helicopter low in height and closer to the horizon. But as the plane goes up and over the ship, the signal gets weaker and weaker until it reaches a null area. For this, you need a datalink that is shaped like a dome, with a dish shaped antenna on a gymbal. The antenna would follow the airplane or satellite as it goes across the sky. You can also see a few of them in PLAN ships, and one of them is called "Lightbulb", which the Mineral ME Dr link or its Chinese derivative the Type 366-2. This is used to obtain targeting data against ships for over the horizon. If you look at the Type 022, it only has the dipole kind of datalinks, but if you look at the 056, you will see both.
 
Last edited:

AndrewS

Colonel
Registered Member
They probably will be upgraded to a new generation of AESA module. Even for the reasons of logistics and maintenance, it needs to be done. Even if you think the ship doesn't need to, and the old radar's performance remains adequate, it still needs to be done.

Think of it. After some time, production of the Gallium Arsenide modules would have stopped, with the remaining stock going into warehouses as spares. With the introduction of new ships, the new ships will be using Gallium Nitride modules, and all production will be shifted to using Gallium Nitride.

Over time, as the 052Cs and 052Ds wear out their arrays and burn out older the modules, the remaining stock will eventually be depleted. Electronics burn out, and they will eventually expire, much sooner than the rest of the ship. All this kit is highly modularized. They are meant to be taken out and replaced quickly. An AESA array has similarities to a server farm. Like any computing architecture, everything is designed in consideration for things to come, including upgrades.

So what you are going to do, is to refit the radars to use the new module standard that is current and available at the time of the refitting event. They get new modules, the cooling system is overhauled, and the backend is updated to account for the new modules.

It will not affect the production of new ships. Instead, the more modules you make, both for new ships and for refitting older ones, the lower the cost of each module, driven purely by volume amortization. It also makes things a lot easier. For example, low production units would require FPGAs. FPGAs are temperature hot and expensive, not to mention China ain't exactly the leader in this field. You can forego FPGAs and use ASICs instead. ASIC runs cooler, and is cheaper, but requires a large production run. You want to order the ASICs specific to these radars by the many thousands, and the only way you want to justify such a large volume to plan the total number of ships you want to produce over time that would use the same module, plus the number of existing ships you plan to upgrade with. The 40+ 052C/D lifetime total would represent an enormous market for such chips to add to the demand for the new builds, that it would make a serious dent in the price of such chips when compared if these chips are only going for new builds.

Think of the Sovremenny refits. The radar refits are mostly sidegrades. Fregat MAE to Type 382, Mineral ME to Type 366, MR184 to Type 344, MR123 to Type 349, MR90 to the Chinese illuminators. At best there is only a bit of a slopegrade here. The main reason why the Russian ones are being replaced is for maintenance reasons. The Chinese cannot maintain the Russian ones indefinitely as parts and warranty supplies run out. Much easier to replace with the Chinese radars where they have control of the parts supply.


As for datalinks, there are quite a few forms of datalink antennas. The ones I see with PLAN ships often take the form of poles, either long or short, or small cylinders. These are likely to be monopoles or dipoles. They can communicate with another ship easy, or with a plane or helicopter low in height and closer to the horizon. But as the plane goes up and over the ship, the signal gets weaker and weaker until it reaches a null area. For this, you need a datalink that is shaped like a dome, with a dish shaped antenna on a gymbal. The antenna would follow the airplane or satellite as it goes across the sky. You can also see a few of them in PLAN ships, and one of them is called "Lightbulb", which the Mineral ME Dr link or its Chinese derivative the Type 366-2. This is used to obtain targeting data against ships for over the horizon. If you look at the Type 022, it only has the dipole kind of datalinks, but if you look at the 056, you will see both.

Couple of points

1. You would expect them to produce enough spare AESA modules for the remaining lifetime of the ships. Unit production cost for semiconductors is negligible compared to the huge upfront design cost, like we see in the civilian world

2. Upgrading and running new cooling systems is a major job.

3. AESA modules for 40+ AEGIS ships is not a major production run. It's actually a tiny production run compared to fab capacity

4. If it really were worth replacing GaA with GaN AESA modules, then why haven't the US done this with the Arleigh Burkes? It's exactly the same logic.

5. With the Sovremenny, there was no doubt that the Air Defence system was obsolete. But given all the Russian systems are tied together and that China almost certainly didn't have access to the source codes, it would be easier to replace everything. In comparison, the GaA AESA modules on the Type-052D will still be useful in 30 years time. GaN AESA modules on a Type-052D would face the same power availability and cooling limitations
 

MarKoz81

New Member
Registered Member
If they can communicate with satellites, high flying UAV or aircraft - they can almost certainly communicate with other ships.

For directional antennas installed at heights of 20m to 40m the maximum distance between two such antennas is within approximately 25km to 45km respectively. Antennas at greater distances are obscured by radar horizon.

Directional antennas are useful for secure communication within a task force where ships are spread 25-30km apart and those distances are also relevant for effective ranges of active sonar and SHORAD coverage. You can bounce the signal if your system has sufficient capacity which multiplies this range by number of ships. Beyond that you need relays in the air.

There are only 39 modern AEGIS destroyers so far, and I expect the Chinese Navy to be aiming for a fleet of 90 AEGIS destroyers in total.

PLAN doesn't have AEGIS destroyers and won't have any unless it captures one. The only AEGIS destroyers in service are in US, Japanese, Korean and Australian navies.

AEGIS is a specific proprietary combat management system produced by Lockheed Martin not a type of architecture like Link or MIL-STD.

USN has so many AEGIS ships because in the 90s E-2s had weaker radars and even weaker computers. AEGIS is a product of 1970s. Right now the processing power is sufficient that instead of huge centralized sensors it's better to develop integrated networks of distributed sensors which are more efficient and more survivable.

I recommend you watch the presentation on YT given by the project manager of Ford CVN project. He said that the one major change that he would introduce having learned from the project would be the lack of expensive radars - AN/SPY-3 and AN/SPY-6. He stated that they are far inferior to aerial assets combined with the escort and that the only radar that a carrier needs right now is for handling traffic. The radars on Ford took too much money, space and time.

PLAN procurement choices so far are extremely practical. With an upgrade to the C4 systems and a humble Type 054A or Type 056A can have situational awareness of a Flight III Burke. Americans are moving in the same direction but they are limited by their past procurement choices which is why their fleet is what it is at the moment.

CVNs require minimum 2-3 DDGs and 1 CG for escort per CBG. LHDs require 1 DDG and 1 CG for escort per ESG. 11 CVNs mandated by law and 9-11 LHDs give us between 51 and 66 AEGIS ships just for escort duties. Add BMD requirements and lack of alternatives for cheaper ships because of LCS concept and you have your answer for why USN has so many AEGIS ships.
 
Last edited:

AndrewS

Colonel
Registered Member
For directional antennas installed at heights of 20m to 40m the maximum distance between two such antennas is within approximately 25km to 45km respectively. Antennas at greater distances are obscured by radar horizon.

Directional antennas are useful for secure communication within a task force where ships are spread 25-30km apart and those distances are also relevant for effective ranges of active sonar and SHORAD coverage. You can bounce the signal if your system has sufficient capacity which multiplies this range by number of ships. Beyond that you need relays in the air.



AEGIS is a specific proprietary combat management system produced by Lockheed Martin not a type of architecture like Link or MIL-STD.

The only AEGIS destroyers in service are in US, Japanese, Korean and Australian navies. PLAN doesn't have AEGIS destroyers and won't have any unless it captures one.

USN has so many AEGIS ships because in the 90s E-2s had weaker radars and even weaker computers. Right now the processing power is sufficient that instead of huge centralized sensors it's better to develop integrated networks of distributed sensors which are more efficient and more survivable.

I recommend you watch the presentation on YT given by the project managed of Ford CVN project. The one major change that he said he would introduce having learned from the project would be the lack of expensive radars. He stated that they are far inferior to aerial assets combined with the escort and that the only radar that a carrier needs right now is for handling traffic. The radars on Ford took too much money, space and time.

With an upgrade to the C4 systems and a humble Type 054A or Type 056A can have situational awareness of a Flight III Burke. Americans are moving in the same direction but they are limited by their past procurement choices which is why their fleet is what it is at the moment.

We've seen Chinese destroyers with similar AEGIS layouts in terms of 4 AESA panels + CEC datalinks + combat management system. Hence I find it easier to call them AEGIS destroyers versus the older types.

I agree that airborne radars are superior to ship-based radars, but there are 2 points here

1. The Chinese Navy doesn't have carrier-launched AWACs aircraft yet

2. In recent years, it is the USA which has focused on expensive and very high-performance AESA radars due to incoming supersonic/hypersonic missiles. In comparison, Chinese destroyers do not face large numbers of supersonic/hypersonic missiles. So I suspect that contributes to the much lower cost of Chinese destroyers.

On the Ford, yes, I think the SPY-6 radar is a mistake as the carrier design should be focused on its airwing. The Ford only has point air defence missiles and will always be accompanied by escorts. So a simpler radar and CEC links should be sufficient. So that's a problem with US Navy requirements rather than the radar itself. Although, if you can conduct last-ditch EW attacks on an incoming missile, that may be an acceptable justification for the SPY-6.

Anyway, back on topic.
 

FairAndUnbiased

Junior Member
Registered Member
As for datalinks, there are quite a few forms of datalink antennas. The ones I see with PLAN ships often take the form of poles, either long or short, or small cylinders. These are likely to be monopoles or dipoles. They can communicate with another ship easy, or with a plane or helicopter low in height and closer to the horizon. But as the plane goes up and over the ship, the signal gets weaker and weaker until it reaches a null area. For this, you need a datalink that is shaped like a dome, with a dish shaped antenna on a gymbal. The antenna would follow the airplane or satellite as it goes across the sky. You can also see a few of them in PLAN ships, and one of them is called "Lightbulb", which the Mineral ME Dr link or its Chinese derivative the Type 366-2. This is used to obtain targeting data against ships for over the horizon. If you look at the Type 022, it only has the dipole kind of datalinks, but if you look at the 056, you will see both.
I believe the comm antennas are being migrated to phased arrays as well.
Please, Log in or Register to view URLs content!


Now researchers from the Shanghai Institute for Advanced Communication and Data Science at Shanghai University in China have developed a 28 Gigahertz (GHz) beam-steering antenna array that can be integrated into the metallic casing of 5G mobile phones.

“The antenna elements and arrays are easily integrated on the metallic frame or casing of a mobile phone, which is more suitable for industry mobile phone design,” said Danny Yu, the lead author of the research in the journal
Please, Log in or Register to view URLs content!
. “Compared to all other existing works for mm-Wave 28 GHz band, this work is totally unique in such a sense that it is very close to what the industry is working towards at the moment.”
 

Tam

Brigadier
Registered Member
Couple of points

1. You would expect them to produce enough spare AESA modules for the remaining lifetime of the ships. Unit production cost for semiconductors is negligible compared to the huge upfront design cost, like we see in the civilian world

2. Upgrading and running new cooling systems is a major job.

3. AESA modules for 40+ AEGIS ships is not a major production run. It's actually a tiny production run compared to fab capacity

4. If it really were worth replacing GaA with GaN AESA modules, then why haven't the US done this with the Arleigh Burkes? It's exactly the same logic.

5. With the Sovremenny, there was no doubt that the Air Defence system was obsolete. But given all the Russian systems are tied together and that China almost certainly didn't have access to the source codes, it would be easier to replace everything. In comparison, the GaA AESA modules on the Type-052D will still be useful in 30 years time. GaN AESA modules on a Type-052D would face the same power availability and cooling limitations

I don't think you should make a huge stock of the original version 1.0 modules. What happens if there is a bug or bugs in them? You are going to have a huge inventory mess. You should not put too many eggs in one basket. A complex system will inevitably have bugs in them, and the higher ups are one day, going to request you improve its performance to match changing conditions and add features on them. This means you need to have room for upgrades and revisions which can sometimes take years. You can't put too much inventory on one batch if it turns out to have design bugs in them, and you need to allow for an inventory for debugged, refined and improved modules.

Similar to server architectures, AESA architectures are designed from the start to last for a long time and in consideration for expansions.
Even if it is not an AESA, the SPY-1 has more than a handful of upgrades on its own. Do note that it is now at Baseline 9. Phase shifters have been changed for example, like from 4 bits to 6 bits, this done long ago.

For example, with the 052D, are we going to say the first and second batch of the ships still have the same modules? The first batch are the ones that have some kind of calibration boom or probe under the panel, while the second batch doesn't have it. There needs to be changes in the array and modules itself to allow for some inborn RF measurement and calibration so that you can remove the boom. So in the second 052D batch, its likely there are improvements and refinements in the modules.

We do not know if new cooling systems are needed. Even so, I am not sure how much of a major job is that. Improvements in technology can make a cooling unit more efficient, yet lighter and consumes less power.

GaN makes a module run cooler. The cooler it gets, the more power you can push into it before it reaches its voltage breakdown point. Thus even you do not change the cooling unit, you can still put more power into the module before you reach to the same temperature point at a lesser power using a GaAs unit. For the same temperature, you get more power. It is not necessary to push the modules to the edge for maximum power, your goal post only needs to be better than the previous modules.

GaN is also more sensitive. Thus, even if you emit the same power, you will get more of the echo back. Without any increase in the power emission, you can still greatly benefit from the increase in receive gain. The catch is the potential increase in noise from the sensitivity, which you have to filter out.

Why haven't the US have done so with the Arleigh Burkes? They are already planning to do this. At least with the Flight IIA, the arrays were designed so that one day, they can be changed to an AESA by using the same modules used for the SPY-6. The 052C/D are already AESAs.

With the Sovremennys I am not sure if the Chinese had not access to the source codes. They might do---the later two Sovremennys had Chinese SATCOMs that were retrofitted into them, and these were done already long after the ships had been delivered. These include circular SATCOMs, the type with flat panel as in MIMO phase arrays. Considering that China had managed to reverse engineer both the Fregat and the Mineral radars, I don't see how breaking the source codes would be a challenge.
 
Last edited:

Top