J-20 5th Gen Fighter Thread VI

[Air Force Brat]

I'll address this part of the reply, because I think it is most pertinent to our previous discussion.

I don't think anyone disputes that Chinese military rumours can be wrong, sometimes even deliberately so.


No one can prove those claims to be definitively true or definitively wrong, I agree. I think what we're left with is whether the 15 ton number should be kept as part of the conversation about J-20 overall in some manner, which is dependent on how we judge the credibility of the evidence.

So I do agree that the last bullet point you make is probably the biggest area of contention.
Fundamentally, I think there is enough evidence put together such that it is safer to keep the number in circulation, but with the important caveat of acknowledging skepticism towards its validity, with the understanding that we do not know what the true empty weight number actually is.

Unless further information comes to light regarding the credibility or lack of credibility of the magazine, as well as any further information or lack of regarding whatever advancements in weight reduction that AVIC has actually achieved or not (which up to this point have been alluded to but far from solid).

Actually there are a great many people who could prove the J-20 is 20 tons and NOT 15 tons,,, the idea that a combat ready J-20 is 15 tons is nonsense...

LockMart went so far as to make the main fuselage bulkhead on the F-35B from Aluminum rather than Titanium in an effort to save a fractional amount of weight, then they were rewarded with an abundance of cracking in that same lighter bulk head...the Alpha and the Charlie retained their Titanium bulkheads and avoided that circus side show.

Aircraft are categorized by weight classes,, the J-20 is a heavy fighter, it is likely stressed to at least 7 Gs, possibly even to 9+, so when posters resort to asking why we should put aside our "incredulity" at the 15 ton suggestion??? because honestly its not only your credibility, but that of the Sino Defense Forum that the 15 tonners are hurting..

but as Jeff Head always states, we will have to weight for Chendu to set the story straight, in the meantime, believe what you wish, but please refrain from insulting the intelligence of sincere and knowledgable SDF posters.

 

[Air Force Brat]

What's qualitative about an EM scatter diagram?



We've been rather generalizing "strength" so far. Doesn't specifically mention bulkheads (though fatigue is certainly a major concern), and careful control of the grain flow in a forging also helps a lot in that regard.



What I meant is that even the legacy F-15 bulkhead has its rim and stiffeners machined down to a thickness which looks every bit as thin as those on the J-31 bulkhead, so I fail to see anything about the latter which would defeat the newer (single-piece) F-22 process.



The rear one is the unfinished forging again, and unlike the J-31 and F-15 pictures this one is of such low resolution that I don't think we can really gauge the thickness of the final part in the foreground.

Furthermore, jobjed seems to have identified where inside the J-31 that bulkhead goes here and it lives in the rear fuselage. That F-22 bulkhead OTOH appears to be one of the forward wing attachments (it includes cut-outs for the main bays but is positioned sufficiently far aft that it lacks openings for the side bays and the inlet ducts are already close to the centreline). If that is true, it's not surprising that it might be thicker.



The rest would be in wing spars & tail boom structures, perhaps a couple of longerons, I think. Wing spars and longerons would be essentially like really long beams/plates, so similar considerations apply as with bulkheads (no weight benefit from 3D printing without novel geometries).

Where I could see significant savings even with conventional part structure is the tail booms, which are EB welded together out of several pieces on the F-22 - EB welding is a very high-performance process, but 3D printing out of a single piece offers interesting opportunities. Then again, this is an area where the analogies between the F-22 and J-20 unravel altogether, with the former having conventional h-stabs against the J-20's canards



That's the central part of our disagreement, because I think it would indeed have to be for a substantial weight saving. How else is there anything about that bulkhead which could not have been done with a machined forging already for the F-22?

It's single piece, so is the F-22 part, they're both titanium and the thickness of the stiffeners/rims was apparently doable back on the F-15, grain flow may be less favourable on the 3D printed version - where's the weight saving coming from unless it's hollow with internal stiffeners? I suppose it's a long-winded way of saying that...



I'll leave it

Yep, any knowledgeable honest engineer or poster will agree with your rather outstanding arguments and valiant attempt to bring this thread back to the "real world"??

 

[Tirdent]

@Tirdent Can you tell me what numbers you found for the volume of the two planes?

I didn't work out the actual volumes in cubic meters, just the ratio based on component scaling. In isolation the volume figures wouldn't be all that meaningful given the crude approach to deriving them - I rely on any errors which may be inherent in the method to cancel out to a large extent by measuring both in the same way and then taking the ratio only.

 

[manqiangrexue]

Actually there are a great many people who could prove the J-20 is 20 tons and NOT 15 tons,,, the idea that a combat ready J-20 is 15 tons is nonsense...

Nonsense is an old retired pilot thinking he knows what 2017 classified 5th gen fighters weigh. Nonsense is thinking that the F-22 can be the best forever (if it ever was, which was never proven) or that its old tech could keep down a country with the highest research funding, greatest number of engineers, and the fastest supercomputers on earth.

Prove? LOL Now I know you clearly don't know what it means to prove something but just humor me. What's the "proof?" Proof isn't what sounds likely or sounds reasonable; proof something is 20 tons means it cannot weight anything but 20 tons, case closed, no debate. Not 19, not 19.5, not 20.5, not 21. Proof of weight is when you put an object on an accurate scale and you read the number; there can be no challenge to it. Show me the "proof." Let's see how funny you can get.

 

[Tirdent]

If you bisected the conical camber of the F-22, you'd get a thinner wing with a shallower conical angle. To be *very* clear, I'm not actually saying this is the case with the J-20.

I'm not quite sure I follow your logic there, it seems you may not be completely aware what conical camber is?



Note how the leading edge part of those airfoil sections droops, and how a progressively greater fraction of the airfoil chord is affected by the droop the further you move outboard? That's (leading edge) conical camber - "conical" because it amounts to wrapping the leading edge round an enormous cone with its tip sitting in the apex of the wing leading edge. This is what's seen on the J-20 (and also J-10), but you can also wrap the entire wing round the imaginary cone as on the F-22 and Concorde, although this is much more difficult to manufacture - in this case the wing looks like a Rogallo glider in a spanwise cross section:



The image above is from a Su-57 patent as I didn't have an appropriate F-22 drawing handy, but it does the same thing to an even more marked degree (presumably because its wing isn't swept back as sharply?).

In both cases (J-20 & F-22) there is also spanwise twist superimposed, making it pretty much impossible to accurately measure wing thickness from a photo (but particularly so for the F-22).

Totoro estimated roughly 20% greater in volume than the F-22. Assuming the same density (which he hasn't claimed to do) , that would make a weight estimate of 23.64 tonnes. I'd say that's quite a dramatic difference, all things considered, well outside of 5% of each other.

Those numbers were in (metric) tons - he provided a "gut instinct" weight based on his volume estimate in an older post.

I think an error of 500 kg plus minus might be a tad overconfident...for example, if you missed average frontal cross section by say 0.25 m^2, then that would be a 5 m^2 difference over a 20 m length. That's not small, and while I haven't done the math yet I have trouble thinking that won't throw a weight estimate off by more than 500 kg.

Works out to just about 500kg (Whew!)

Thanks for making me check BTW, it's so easy to get complacent!

 

[latenlazy]

Part 1

Part 1.

Fine. There's no point discussing happiness, as it can't prove anything regarding the underlying debate. Like I said, they can be happy with many things about the J-20, including its overall capability and flying characteristics, even if it's 20% more voluminous and heavier than the F-22. Even if they aren't happy about something, I doubt the public would hear about it.

Why would they be happy about flight performance if it didn't meet their requirements?

I don't know what to make of this part. That the J-20 is longer than the F-22 is a fact. I assume you agree? That it has a larger cross section has been found by the two members who have performed a measurement. Apparently you don't accept it. I could try measuring it myself, but I suspect you'd reject it just the same (if it turned out consistent with the other two). The only measurement you probably wouldn't automatically reject is your own, which you refuse to provide. Reject here means refuse to use and claim it could be significantly wrong, even if your stated position is 'we don't know / can't know.'

Golly, what happened to your disdain for imputing motives? If this is turning into a "I don't trust you exercise" then don't ask for my thoughts. As I said earlier If you don't want to approach what I have to say with an open mind and consider that it might be right, and if you're just looking for reasons to affirm your own preferred conclusions, then let me know so I can save my time.

Volume can be deceptive. Planes aren't one dimensional objects. If this were merely a matter of length then the F-22 should be a lot lighter than the Su-27. We now know the J-20 probably has a shorter wingspan. I've already mentioned the point about control surface size, so I won't get into it here. The F-22's aft fuselage seems more blended with the wing than the J-20's which might add volume. We don't have a good measure that includes how much each fuselage tapers towards their tails. These two planes aren't congruent shapes. We should be more scrutinizing than simply assuming length tells us much.

I don't accept their cross sectional measurements because I've done my own rough estimates, and found the errors to be too unsatisfactory to be sure (I generally don't share more than some thoughts and considerations for judging volume because I don't post detailed work I can't be happy with the precision of. I have my reasons). When someone submits some work but says the methods are rough and there are potential errors, you should take them seriously. This is in part because any cross sectional measurements require getting an accurate estimate of fuselage width and height, and that's proven to be tricky with the resolution of the pictures we have on hand, and partially because it seems the two fuselages *taper* at different rates, so we don't know the extent to which their frontal cross section area reflects average cross section along the length of the plane.

If you *really* want to see what I mean, here are two different estimates I did last night of the J-20's cross sectional area compared to the F-22's, caveats that the pictures I used aren't perfect. I just picked up whatever I could find on my hard drive.

(Another caveat*—you may notice I didn't draw in all the frontal for the J-20 in the picture I'm using. This is because it seems this photo was shot with a slight upward looking angle, so you ended up getting a bit of the belly too. Side views of the J-20 also tend to show that the plane slightly tilts up when on the ground, and doesn't rest perfectly parallel to it, so this shouldn't be surprising. You can judge for yourself here

Please, Log in or Register to view URLs content!

)

Assuming the F-22's fuselage was 3.85 meters wide (and I couldn't be sure, because I got a range from 3.80-3.85) and the J-20's were identical (similar range, but I couldn't be sure either), my estimates showed that they should have roughly the same frontal fuselage area.

However, *if* the J-20 were 3.8 meters wide and the F-22 were 3.85 meters wide, the J-20's frontal area becomes smaller by about .2-.25 m^2. Multiplied over the length of its fuselage, and you get a 4-5 m^2 discrepancy, which can contribute anywhere between a 5-8% error on the J-20's volume estimates. That's not small.

Or take for example, the wing thickness argument. A very rough estimate of the J-20's wing root shows it to be about .36 meters in thickness. The F-22's wing root thickness, with a very rough estimate, seems to be about .5 meters. I don't advertise these numbers because I'm not happy with their accuracy. Getting good estimates for thin objects can prove tricky without very high resolution pictures that are shot under the right perspective conditions. Let's for the sake of this mental exercise assume that average wing thickness is 1/4 wing root thickness. The J-20, if it has about a 12.9-13 meter wingspan, would have between a reference wing area of 74 m^2 to 76 m^2. The F-22 is reported to have a reference wing area of 78 m^2. (76*0.36)/4=6.84 m^2. (78*0.5)/4=9.75 m^2. That would make for a 3m^2 difference, but these small differences can add up. If average wing thickness is 1/2 wing root thickness instead, then that's 13.68 m^2 vs 19.5 m^2. Not so small difference anymore (I'm aware that there are some problems with this mental exercise, such as reference wing area including the fuselage, which you'd normally exempt from volume contributions to the wing as they're already accounted for in the fuselage, but this exercise was mainly to illustrate a point).

I am not making these points frivolously. You keep accusing me of not doing the work, but if you yourself had done the work maybe you wouldn't be so sure either.

My argument is that just because something was a requirement is not a guarantee that it was achieved and you haven't shown that a weight of 21 to 23 tonnes is unacceptable performance wise.

I agree that doesn't tell us what was achieved. That's not what I said though is it? I said we should presume if they're happy with the plane's capabilities, that should mean it reasonably reached design requirements, which emphasized performance parameters, which means we should be wary of conclusions that might be prohibitive to performance. Perhaps being 23 tonnes isn't unacceptable to performance, but then if it wasn't we wouldn't have people constantly badgering the point about the J-20 being overweight and underpowered.

I concede the point. You've shown that impugning motives and psychologizing is the way to go.

With those three examples I was indeed trying to make a point. I was asking whether people here use the same standards for Chinese claims and claims from elsewhere, particularly when the Chinese claims seem implausible.

Well, at least we're being honest about things now aren't we? As I've already said, given the particulars of this case, I don't find the evidence exclusionary to the claim, no matter how incredulous the claim might sound. It doesn't matter if the shoe were on the other foot. I've learned long ago that incredulity isn't a reliable truth meter, especially when it comes to China watching (or watching any country that's shown itself capable of rapid technological development).

Ah, here we go. I've seen you make this statement before (you're making a warning about underestimating potential enemies), and I don't believe you. This forum is totally irrelevant and you should contact your representative with your warnings, if you're being sincere. I think this is a patent cover for your desire to assume the best about Chinese military developments while not wanting to be accused of fanboyism. Wow, this really is necessary and productive.

As necessary and productive as saying "Just for the record, I'm not American and don't stand for their views or interests"?

I've only engaged you earnestly in this discussion. If you don't believe me then don't believe me, but don't engage me in discussion if you don't think I'm trustworthy or don't care for the substance of my arguments. Though, maybe at least give some consideration that it's not a desire in me to assume the best about Chinese military developments, but a refusal on your part to accept possibilities that exceed your own presumptions?

As for "contacting my representative", that's a load of condescending crock. What does the average Congressmen know about military development anyways? I have other channels. I do have a life outside of forum posts you know.

Word of advice, if you're going to argue from a place of contempt, better not to engage at all.

I will rely on machines translators, since they're very advanced and useful. In any case, there wasn't any problem with their output here. Now we're in a situation where we have to consider whose translation was more accurate, yours or Kenhmann's. You're probably right, the Chinese seems less clear.

You mean the machine translated version of Kenhmann's translation. Machine translators are not that advanced, especially not for a language like Chinese, and especially not for double translations.

I asked you three times because you didn't answer the first time, said there was a grey area the second time and after we both clarified it turns out there might be grey area (the third time).

Or...you could have followed the actual discussion on the JF-17's radar and make up your own mind rather than try to interrogate me.[/quote]

 

[latenlazy]

Part 2.

If I thought you were acting in good faith, the above could make sense. However, I think that in your desire to not have the J-20 be heavier than the F-22, you also don't want it to have a larger volume and thus try to invalidate any attempt at measuring said volume. Estimating dimensions and calculating volume is not something that can only be done well by experts. If we have quality measurements of the J-20's length and wingspan, there's no reason to think we can't also get the same for the cross section.

Okay. Believe what you want.

Of course estimating volume isn't something that can only be done well by experts, *but* it can only be done well with precise methods, which I haven't seen from anyone yet. Experts are experts because they exercise great care in their work. If you can't measure up to the quality of work, then don't claim the work to be as good. Being able to get quality measurements for length and wingspan does not mean its straightforward to get quality measurements for volume. Wingspan and length are singular dimensions, not continuous shapes with curvature.

No, you said, "Consider also that if the J-20's fuselage cross section were say only .5 m^2 smaller than the F-22's, multiplied across that alone could neutralize half the volume difference if the J-20 shared the same cross section area as the F-22." I refused to consider that because the actual measurement said it was instead 0.6 square meters larger. I don't think there's any reason to think a mistake like getting 5.9 square meters instead of 4.8 square meters occured (5.3 square meters for F-22).

You mean, the *estimate* said it was .6 square meters larger. There are plenty of reasons for getting such a large discrepancy in cross section. Let's say you measured 3.9 meters for a width and 1.7 meters for a height of a rectangle of some unknown dimensions, and someone got 3.8 and 1.6 (given that within one image proportions are preserved, so an over or under estimate of one dimension naturally leads to the same for the other, this problem is actually quite common). Depending on the resolution of the picture, that might be the difference of choosing where to measure by 2-3 pixels (so a resolution of ~.05 meters per pixel). Suddenly you're talking about an estimated area of 6.63 m^2 vs 6.08 m^2. This is why you should *never* take an estimate at face value unless they can prove their estimates to be precise.

Trident stated that he used the most charitable assumptions (that he still found reasonable) for how light the J-20 can be when making his assessment, so you weren't actually arguing in the other direction. Also, the vast majority of your warnings about measurement and scenarios of what the volume could actually be were toward a lower volume before any argument began. When you know the errors could be in both directions but only make the case for one, that's not a desire for accuracy, it's motivated reasoning to protect your preferred outcome. You never said, for instance, that the J-20 could be 25% percent more voluminous, instead always presenting reasons why it could have the same volume as the F-22. In my view, this is simply a result of your bias and not some desire for only the most accurate conclusions. Similarly for your dismissal of any attempts to actually measure the volume.
As for 'incredulity,' there are countless things that all of us find not to be credible, so this isn't any kind of point.

Whether Trident was being charitable or not is besides the point. No offense to Trident, but no matter how reasonable he is I would rather let the merits and problems with his method speak for themselves, and by his own admission, what he did was very rough. I make the case for less volume because the case for greater volume has already been made by others. It doesn't need reinforcing. I don't believe in bs false equivalence. I'm not angling to make people feel like the conclusions they're most comfortable with are affirmed. These are ultimately discussions over technical details. There is one right answer. Instead of trying to demand that people who have inputs over dimensions make platitudes about estimate range to make different sides feel good about themselves, how about let's focus on figuring out more precise methods and get more precise estimates?

That said, if you actually followed my earlier comments, I said explicitly that outside some minor quibbles with some assumptions I think Trident was within the ballpark of where I would expect the J-20 to be, *except* for the question of density, which maintains relevance particularly because we have two different sources that have some degree of corroboration on a lower weight estimate. As I said to Trident earlier, the claims made by the Chinese 3D manufacturing industry and by the trade magazine were rather unambiguous. You can take it or leave it.

 

[latenlazy]

I'm not quite sure I follow your logic there, it seems you may not be completely aware what conical camber is?

View attachment 43867

Note how the leading edge part of those airfoil sections droops, and how a progressively greater fraction of the airfoil chord is affected by the droop the further you move outboard? That's (leading edge) conical camber - "conical" because it amounts to wrapping the leading edge round an enormous cone with its tip sitting in the apex of the wing leading edge. This is what's seen on the J-20 (and also J-10), but you can also wrap the entire wing round the imaginary cone as on the F-22 and Concorde, although this is much more difficult to manufacture - in this case the wing looks like a Rogallo glider in a spanwise cross section:

View attachment 43866

The image above is from a Su-57 patent as I didn't have an appropriate F-22 drawing handy, but it does the same thing to an even more marked degree (presumably because its wing isn't swept back as sharply?).

In both cases (J-20 & F-22) there is also spanwise twist superimposed, making it pretty much impossible to accurately measure wing thickness from a photo (but particularly so for the F-22).

I know what conical camber is. I kinda regretted making that point because it wasn't clear what I was getting at, but couldn't go back to edit. I should have stuck to wing root thickness, because that's ultimately what I'm basing the point about wing thickness on. What I was trying to say was that if wing root is thinner for the same wingspan you will have conical camber that's flatter, which was really just a roundabout way of saying "if we want to get an idea of wing thickness without the confusion of trying to figure out conic look at wing root thickness ".

Works out to just about 500kg (Whew!)

Thanks for making me check BTW, it's so easy to get complacent!

Mind tossing me the volume you arrived at for the F-22? I want to check for myself but I've been busy with comment replies and am too lazy to dig for it.

 

[latenlazy]

What's qualitative about an EM scatter diagram?

Nothing, but Boeing didn't advertise an EM scatter diagram.

We've been rather generalizing "strength" so far. Doesn't specifically mention bulkheads (though fatigue is certainly a major concern), and careful control of the grain flow in a forging also helps a lot in that regard.

Of course not, but where is most of the titanium used in the F-22?

What I meant is that even the legacy F-15 bulkhead has its rim and stiffeners machined down to a thickness which looks every bit as thin as those on the J-31 bulkhead, so I fail to see anything about the latter which would defeat the newer (single-piece) F-22 process.

They look thicker to me. Who has a ruler?

The rear one is the unfinished forging again, and unlike the J-31 and F-15 pictures this one is of such low resolution that I don't think we can really gauge the thickness of the final part in the foreground.

I wasn't referring to the rear one. I agree that we can't actually say anything definitive about the thickness of the final part, so we're reaching a point where it's your eyes vs my eyes.We can leave it at that, I think, since neither of us have anything better going for us on this point.

Furthermore, jobjed seems to have identified where inside the J-31 that bulkhead goes here and it lives in the rear fuselage. That F-22 bulkhead OTOH appears to be one of the forward wing attachments (it includes cut-outs for the main bays but is positioned sufficiently far aft that it lacks openings for the side bays and the inlet ducts are already close to the centreline). If that is true, it's not surprising that it might be thicker.

I'm aware of that, but based on the 3D graphic bulkhead thickness for the fore body doesn't seem thicker in than the aft in the J-31.

The rest would be in wing spars & tail boom structures, perhaps a couple of longerons, I think. Wing spars and longerons would be essentially like really long beams/plates, so similar considerations apply as with bulkheads (no weight benefit from 3D printing without novel geometries).

Given the simple geometry of these parts and the particular kinds of stresses they're meant to handle, I imagine it would be easier to incorporate features such as a hollow structure with trusses for them. I wouldn't presume to know, of course.

Where I could see significant savings even with conventional part structure is the tail booms, which are EB welded together out of several pieces on the F-22 - EB welding is a very high-performance process, but 3D printing out of a single piece offers interesting opportunities. Then again, this is an area where the analogies between the F-22 and J-20 unravel altogether, with the former having conventional h-stabs against the J-20's canards

It's complicated.

That's the central part of our disagreement, because I think it would indeed have to be for a substantial weight saving. How else is there anything about that bulkhead which could not have been done with a machined forging already for the F-22?

It's single piece, so is the F-22 part, they're both titanium and the thickness of the stiffeners/rims was apparently doable back on the F-15, grain flow may be less favourable on the 3D printed version - where's the weight saving coming from unless it's hollow with internal stiffeners? I suppose it's a long-winded way of saying that...

If they were able to achieve thinner walls or incorporate more troughs into the structure, that wouldn't necessarily be immediately obvious based on visual inspection alone. Also, we haven't factored (and I imagine won't be able to) how the manufacturing process can change what they do with structural design of the plane. For example, in the J-31 graphic we've been referring back to it looks like span wise beams for the wings and the vertical tails are integrated with the bulkhead.

I'll leave it

Does that mean we're done here

 

[Tirdent]

We both measured fuselage cross sections in about the same way and our results for the F-22 agree quite well - I got 5.3m² (from the three-view drawing). Length to the tips of the divergent nozzle ramps I made 16.8m from the drawing.