There's already plans for Naval PAKFA aren't there? Seems unlikely, Russians seem to have an ego block against buying anything Chinese
Future PLA combat aircraft composition
- Thread starter Totoro
- Start date
When I say "export grade J-XYs" I was referring to land based J-XYs.
My current vision for major J-XY variants is:
- J-XY baseline/carrier variant airframe, for carrier operations for the PLA
- J-XY land based variant airframe, for land based operations for the PLA (derived from above, common subsystems and upgrade path but with deletion of carrier specific provisions like folding wings, reinforced landing gear etc)
- export variant J-XY for general export purposes, based on the J-XY land based variant airframe (equipped with export cleared sensors and subsystems)
For what it's worth, I do not see Russia buying J-XYs even if China was okay with selling it to them.
As for whether there will be a navalized Su-57, I won't hold my breath for it yet.
Last edited:
But you did say earlier China could buy the Su-57, even if not likely it's a possibility. Why do you believe this is the case as I would assume there are no missing capabilities that the Su-57 could fill in (that the Chinese couldn't design and produce themselves if they really needed to anyway).
Because I deem the probability of it is technically "not zero".
That doesn't mean I think it is likely.
plawolf
Lieutenant General
Not going to go point by point or we will be here all day, but a few things to consider.
The USN and marines are looking to procure ~690 F35s between them, with also massive inventories of legacy fighters and future UCAVs on the cards. As such your 300 carrier J21s seem far too conservative. That’s barely enough to fully equip all 10-12 carriers by your own force conpsotiob You need to also factor in spares for maintenance and attrition, as well as reserves and training squadrons.
It is precisely because of the large logistics and infrastructure burden and footprint of 5th gens that I don’t want the PLAAF to operate two kinds long term. Because then you will end up needing two separate sets of spares, support equipment and mechanics who can’t easily cross operate between the two types. There are good reasons why previous hi-lo fighter combos shared the same engine.
On production bottlenecks, I’m sorry but everything is about money at the end of the day unless we are talking about a literal finite and limited amount of raw materials that cannot be substituted, which only really apply to nuclear weapons and even then look at how many the USA and USSR ended up building.
Production bottlenecks are a function of projected production numbers over production years. The more production infrastructure you invest in, the quicker you get through your projected production run and the higher your annualised depreciation costs.
Yes, it takes time to ramp up production, but it is not linear relationship.
It takes a set number of years to set up a new production line principally because of the training time new workers need. But that training time doesn’t double if you train two apprentices instead of one. It might take 3-4 years to get a new production line up and running, but you can potentially get 2-3 new lines set up at the same time if you train all the new workers at the same time, with corresponding additional teaching resources.
All of that cascades down the components chain and can all be expanded to meet the desired prosecution run if there is enough money, just look at F35 annual production numbers as proof.
A 30+ annual production rate in comparison is pretty damn modest, especially when you factor in how competitive the Chinese have been in pretty much every other field of manufacturing.
Once a tech level had been achieved, China rarely have difficulty scaling up production to world leading levels. Again, it all boils down to money and opportunity cost.
But in today’s world, we also need to remember not everything is about cost any more. The grace period that China enjoyed where the risk of a major war is over, and it would be a mistake to not take that into consideration when making cornerstone national defence procurements that will fundamentally determine the combat effectiveness of the Chinese navy and air forces for the next 25-40 years, which will probably be the most important and dangerous period for China since before the founding of the PRC.
Operating costs was a key concern during peacetime when you were just expecting to fly your fighters in training. But with a serious likelihood you will be using them in war, and in a war against a numerically and technically superior foe to boot; you just cannot afford to penny pinch.
I think it is precisely for this same reason that the US is now having serious second thoughts about its commitment to the F35.
The F35 was supposed to be the only stealth game in town. It was mean to kill legacy Russian and Chinese fighters in BVR without them ever having any chance to shoot back. That’s why it’s WVR combat capabilities are very much an afterthought, with no room for IRAAMs and not even a gun on some models, not to mention less than impressive agility, which at best only match up to legacy 3rd gen standards.
Those same issues and limitations also apply to the J21. It might be able to achieve better raw kinetic performance due to not being beefed up to carry heavy bomb loads, but it’s lack of short range missiles is going to be a critical shortcoming compared to something like the J20, and when fighting against enemy 5th gens since BVR is most nerfed by stealth, meaning the likelihood of a stealth v stealth fighter going into WVR is going to be considerably higher than even legacy on legacy fights.
That’s why I was never really keen on it even for the navy, but practically it is the only realistic choice given the likely difficulties, delays and compromises needed to make a J20 carrier capable. And since they will principally be going up against F35s, at least they won’t be at a disadvantage.
At the end of the day, China is never going to match US and US applied F35 production numbers. That means quality becomes that much more critical. Which means J20s being much more worthwhile.
Why? It is a bigger economy by GDP PPP? It could be easily done if 3% of GDP was committed to the defense. Though with that kind of spending I bet the PLA would introduce more 5th gen fighters.
So, a few of my own responses, addressed by main points as I see them.
Re: US shipborne F-35 numbers
The USN and USMC will be procuring some 340 F-35Cs combined, while the USMC will be procuring some 350 F-35Bs.
I do not see why the PLA needs to field an equivalent number of carrierborne J-XYs equal to US total F-35C and F-35Bs. The PLA's procurement of carrierborne J-XY fleetsize should be guided by their own requirements.
Even in the scope of carrier vs carrier combat, seeking to compete directly in terms of number of manned 5th generation aircraft is not something the PLA has to engage in.
US procurement of new F-35Cs and F-35Bs looks formidable, but they are also replacing legacy aircraft and they also have a finite number of carriers and flight deck space is a premium -- for every F-35C they choose to place aboard a Nimitz or a Ford will be a UCAV that they cannot carry.
As for carrierborne J-XY numbers -- this whole time I've been writing "300+" aircraft -- i.e.: over 300, not 300 as a maximal figure. In my last post I clarified a fleet size of 350, and thinking about it again, I am indeed okay with a maximal number of some 350 aircraft tops being produced, which would satisfy their manned fighter requirements up to the early 2040s (assuming they achieve 10 CATOBAR carriers in service by then).
By the mid 2040s I expect them to build a 6th gen carrierborne fighter which will supplement and expand their manned fighter fleet size because I expect them to aim for a "steady state" carrier fleet of 12 CATOBAR carriers, each carrying some 30 manned fighter aircraft.
So, I think 350 carrierborne J-XYs is fine.
Re: production bottleneck and money
Of course there is a relationship between investment and future production capacity.
My argument is that money can only move things along so fast -- that is to say in the first 3-4 years of production I do not expect J-XY production to be that impressive. Say, 6-10 aircraft per year for the first couple of years, then 10-20 aircraft per year in years three and four.
However I fully expect them to seek to build a large number of J-XYs per year at peak production rate -- your 20-30 aircraft per year IMO is actually lower than what I envision.
I would aim for a peak annual production rate of over 60 aircraft per year for J-XY.
I.e.: I'm just saying that in the first few years of J-XY production, the production rate will be low despite significant investments into production capacity because it will take time for those investments to bear fruit.
Perhaps we may have misunderstood each other slightly in this regard.
Re: the role of J-20, future stealth vs stealth, and quality vs quantity, and WVR
So I think this is where our views diverge most significantly.
I see the future of air to air warfare and stealth vs stealth A2A combat as being dictated by things like: VLO, networking, number of simultaneous sorties in the air, unmanned aircraft as MUMT (both as sensors and as shooters), and still heavily emphasizing superiority in BVR relevant domains.
Specifically, if your aircraft are of similar stealthiness with broadly similar weapons systems and sensor quality, then your goal should be to have more aircraft in the air that can actively emit and sense for your enemy and shoot at your enemy in a more aggressive and voluminous manner than your enemy.
Having "more aircraft in the air" can of course be achieved by having more manned stealth fighters in the air -- and that is why IMO the F-35 is such a potent weapon, because the sheer number of them and the quality of their VLO, sensors, weapons and networking makes it a veritable swarm of mutually supporting sensor/shooter network -- however in future that will likely be multiplied by a magnitude through use of UCAVs with sensors and weapons of differing sizes and configurations.
The goal would be to still achieve the situational awareness and first look/first shoot superiority that has typified air combat for the ages, and would be enabled by having attritable unmanned systems acting as distributed sensors operated in a much more aggressive manner than a manned aircraft can do. All of this of course would also be done in context of strikes against opfor air bases, carriers etc with long range missiles, bombers with stand off/long range missiles, submarines with SLCMs etc, but the enemy will also be conducting those same sort of strikes against yourself as well, meaning you still need to endeavour to be able to achieve air superiority the best way you can.
For a manned fighter aircraft of the future, kinematic characteristics like range/endurance, speed, will be important -- kinematic maneuvrability will still be relevant, but likely significantly less emphasized than on aircraft like F-22, Su-57 or J-20.
There may still be a chance that a manned stealth fighter may stumble upon a manned stealth fighter at ranges whereby they need to engage in WVR combat -- however I believe most engagements will be decided at BVR ranges and "system of systems" level.
When two manned stealth fighters do engage in WVR in rare occasions, WVR weapons will of course still play a role. But traditional dedicated SRAAMs like PL-10 or AIM-9X may not be the future -- indeed, I expect weapons like SACM (like Cuda or Peregine), or even MSDM, to be more important. That is to say, having a "dedicated SRAAM weapons bay" may not be that important, because you can fit two SACMs in the same space as a single AMRAAM or AIM-260 in your primary weapons bay. I expect such weapons to be viable by the end of this decade.
Re: operating costs of J-20, and procurement of land based J-XY
I agree that obviously in a high end conflict you will be using all available resources to support your war effort. In great power competition you will also be using as much resources as you can bear to support your developmental preparations.
But seeking better efficiencies and better opportunity costs is still desirable, because that enables you to free up resources for other efforts/procurement, or to expand procurement of the same type of system etc.
In the case of land based J-XY versus J-20, my rationale for why I think the PLA should procure of a large number of J-XYs to complement J-20 procurement, rests basically on two primary premises:
- land based J-XY will be comparable or superior to J-20 in terms of VLO, sensors, networking and weapons systems, with sufficient range and minimum kinematic performance (even if both are inferior to J-20)
- land based J-XY will be meaningfully cheaper to procure and operate than J-20 by virtue of J-XY's design and production methods that J-20 cannot implement without a substantial investment to create a new dedicated variant
(If either of the two above premises are not true, then my support a land based J-XY procurement to complement J-20 would also decrease)
The end goal is that procurement of land based J-XY + J-20 can result in a meaningfully larger total 5th generation fleet than procuring J-20s alone, and this is of course predicated on my aforementioned vision for how future stealth vs stealth air combat will go, where the need is to put more stealth aircraft into the air, to achieve a system of systems level superiority vs the enemy's system of systems.
The reason why I do not support the idea of continuing J-20 production for too long -- i.e.: beyond 2035 -- is because I expect CAC and its production facilities to be given the 6th generation aircraft, and investing more money, personnel and facilities to expedite and scale 6th generation procurement is a better use of money than developing a new J-20 variant.
====
In summary, I agree with you that the PLA's production run of total 5th generation fighters will not ever reach US F-35 procurement in totality.
However, I think the PLA should still seek to maximize as many 5th generation aircraft as they can procure to reach as favourable a force ratio in the region as possible.
That of course would in turn be complemented by PLA efforts to use strike/offensive counter air methods to try to reduce the number of air bases that F-35s in the region can effectively operate from, to achieve favourable "in the air" force ratios that can be exploited to achieve more longer lasting air superiority in the region.
Underlying this, is recognition that stealth vs stealth air combat into the future will likely still be decided at BVR ranges, where a greater number of networked sensor and shooter aircraft (whether it be manned, or a combination of manned and unmanned aircraft) will dictate the outcome of an air contest, with WVR encounters likely only forming a small minority of the overall air war.
Last edited:
plawolf
Lieutenant General
You are making some oddly inconsistent arguments. On the one hand you stress the importance of numbers, yet on the other, you want the PLAN to top out at half the number of manned 5th gen as the USN. And yes, I am lumping the marines in there since to separate them out from the USN smacks of a distinction without a difference.
Firstly, the F35 procurement numbers for the USN and USMC are done in relation to each other and not in total isolation. I dare say the USN would be buying considerably more airframes if the Marines air arm didn’t exist or their purchases were strictly limited to UCAVs for CAS.
The USMC has some ‘independent’ naval aviation capabilities in their LHAs, but they don’t have anything like the ability to justify a 300+ F35 fleet if only intending to operate from them. That means that operating off of navy carriers is baked into the procurement decision, which in turn will affect how many airframes the navy needs given embarkment limitations on carriers.
And secondly, which is more importantly, in a real shooting war, it doesn’t matter what is painted on them, those are still assets and numbers the PLAN would need to face.
While I agree with you that China would not try to match US F35 numbers in total, in naval terms, I would say China has no choice but to go toe to toe with America numerically speaking. Because unlike with the Air Force, geography is not an issue, and there isn’t anything stopping the USN from bringing the sum total of its available aviation assets into play if it wants to.
The whole point of carriers are fighters, having the same number of carriers as the USN but only half the fighters just makes no sense at all.
Something else you might wish to reconsider in your calculations for carrier fighter numbers is that while you have ‘typical’ peacetime general purpose air arm numbers and compositions, you might want to take a very different force composition if you are going to go head to head with the USN. In which case I would think J15s would be largely considered useless, and a carrier captain would much prefer a couple more J21 squadrons in their place. Same with UCAVs.
I also think you are being far too optimistic about UCAVs in terms of air combat, and the potential for leaps in radar technology and not fully appreciating the fully ramifications of 5th gen on 5th gen combat.
Even today, with legacy on legacy, BVR is only practically effective in the 30-60miles range as a maximum engagement distance. Things like PL15s and ASEA can push that out, but only against legacy targets, or else J20s would not dominate so much against legacy PLAAF fighters with the latest AESA.
Stealth vastly reduces that distance still further, and even modern AESA doesn’t really push practical detection ranges out anywhere like far enough to make BVR the decisive phase that it used to be. Moreover, even if you can get a lock at BVR ranges against a stealth, the radar seeker of your missile operates in the precise frequency stealth is most optimised to counter, so your KP is going to be significantly reduced. Weapons like CUDA were floated when American 5th gens were the only game in town, and were designed to help them curbstomp legacy fighters harder. They are actually a step in a very silly direction if you want to counter 5th gens since the lack of a warhead will further reduce your KP by making everything but a direct hit a miss. Interesting we haven’t really heard much about that since the J20 appeared.
As such I do not expect 5th gens to stumble upon each other only once in a blue moon as you seem to, but rather think that will be the default result of 5th gen on 5th gen air combat, where the overwhelming majority of 5th gens survive BVR, so the battle will have to be decided in WVR.
Yes, having drones operate as a vast distributed radar network could work against stealth, and I myself have suggested that. But such networks are going to be complex and delicate. They would be good for peacetime, opening stages of conflict and counter enemy stealth bomber/missile attacks, but they will not survive even the opening stages of massed air combat.
On drones, you also seem to be a little too optimise on both their pace of development and cost effectiveness.
With drones, you can either have high end, or high numbers. Trying to have both is dangerously wasteful and ineffective.
Moreover, drones face the exact same issues in BVR against stealth as anything else, it’s simple physics. So if a J16 emitting it’s huge AESA on full power could not counter a J20, what makes you think a drone would do any better? Especially when it is likely to only have a much smaller and less capable radar?
To try to build a drone to take on stealth fighter in BVR would see you building something just as expensive as existing manned legacy fighters and be just as ineffective, with the only benefit being you don’t loose pilots when they get shot down. But such drones are going to be orders of magnitudes more expensive than missiles, and no matter how much you spam them, the enemy would be able to make more missiles far cheaper and quicker. This is especially acute since China will most likely be playing defence, so it’s not like they could force the enemy to stand and fight either.
If you throw up such expensive drone swarms, I will just do hit and runs until I whittle your drone fleet to nothing and then press my attack.
Against legacy opponents, things like loyal wingmen would work in BVR, but against stealths, they will not bring anything new that isn’t already on the table, I think the much better approach is to push for WVR for drones.
That means ditching radars and mounting IRSTs as primary sensors, which will pair nicely with LOV designs on the UCAVs themselves, and short range IRAAMs as primary weapons to bypass stealth altogether. On top of that, WVR combat is where UCAVs have the biggest comparative advantage to manned fighters by not being limited to 9Gs, and where advanced AI swarm algorithms can vastly outperform human pilots by having many drones work in unison to kill human pilots through levels of teamwork that is simply beyond human ability. But that is probably at least another couple generations away.
In the meantime, things like loyal wingmen will be limited to glorified missile magazine extensions for manned fighters, or man-in-the-loop remotely piloted drones flown from AWACS and/or 5th gen back seats to keep the lag to a minimum.
Maybe for future UCAVs, but the current trend seems to be emphasizing maximizing range and payload, not manuverability.
@plawolf @Bltizo
As of end-2020 or early-2021 the USN has 30x F-35C; 60x F/A-18C; 25x F/A-18D; 313x F/A-18E; 273x F/A-18F in its Active Naval Aviation branch (+158 EA-18G which are of course combat capable), and an additional 30x F/A-18C; 5 F/A-18D (+5 EA-18G) in Reserve.
So we are talking 701x in Active (859x total including EW) and 35x in Reserve (40x total including EW). The USN therefore has 736x carrier-capable combat aircraft or 899x if you include the EW Hornet variant which is very combat-capable.
Of course not all of these are on carriers at the same time, but they all belong to the same limited variants, and I would expect if China did field the same number of carriers then at worst you'd be talking 10% or 20% less of the combined carrier mix (5th gen + 4th gen), to account for perhaps less FRS-equivalent squadrons in China if they aren't going to have as many squadrons for shore-based training.
Of course if you throw in the USMC numbers too then you're talking about as many as 387 more combat-capable aircraft on top of the up to 899x in the USN. Certainly not nearly as many will be on carriers but you get the point; there's a lot of room to be had in Chinese expansion of the Naval combat-capable carrier-capable fleet. Now, if China ever started producing VTOLs and putting them on amphibious assaults ships then these USMC numbers would become entirely applicable, but of course there is no news to that extent.
As of end-2020 or early-2021 the USN has 30x F-35C; 60x F/A-18C; 25x F/A-18D; 313x F/A-18E; 273x F/A-18F in its Active Naval Aviation branch (+158 EA-18G which are of course combat capable), and an additional 30x F/A-18C; 5 F/A-18D (+5 EA-18G) in Reserve.
So we are talking 701x in Active (859x total including EW) and 35x in Reserve (40x total including EW). The USN therefore has 736x carrier-capable combat aircraft or 899x if you include the EW Hornet variant which is very combat-capable.
Of course not all of these are on carriers at the same time, but they all belong to the same limited variants, and I would expect if China did field the same number of carriers then at worst you'd be talking 10% or 20% less of the combined carrier mix (5th gen + 4th gen), to account for perhaps less FRS-equivalent squadrons in China if they aren't going to have as many squadrons for shore-based training.
Of course if you throw in the USMC numbers too then you're talking about as many as 387 more combat-capable aircraft on top of the up to 899x in the USN. Certainly not nearly as many will be on carriers but you get the point; there's a lot of room to be had in Chinese expansion of the Naval combat-capable carrier-capable fleet. Now, if China ever started producing VTOLs and putting them on amphibious assaults ships then these USMC numbers would become entirely applicable, but of course there is no news to that extent.