• Hi Guest! Forum rules have been updated. All users please read here.

DSI Intakes, Ferri, and China's J-20

Inst

CLEARANCE: Restricted
Joined
Oct 18, 2019
Messages
21
Reaction score
1
Lastly, working with new calculations, the MFR of AL-31 at Mach .9 and 35000 ft is about 65 kg/sec, when, if we assume the AL-31 is roughly an analogue to the F100-PW-229 engine, it should be about 81 kg / sec requirements. This implies a thrust loss of about 20%, ignoring factors like the fuselage diverting increased air into the inlets. The thrust loss in the chart, however, is roughly 50% from the chart you posted, so simply blaming MFR for thrust loss is inaccurate.
 

sferrin

CLEARANCE: Above Top Secret
Senior Member
Joined
Jun 3, 2011
Messages
12,198
Reaction score
393
If you decelerate air to Mach 2 from Mach 1, you still have the same mass.
How are you going to do that without compressing it? And if it's compressed the mass per capture area goes up.
 

Trident

CLEARANCE: Top Secret
Senior Member
Joined
May 8, 2006
Messages
846
Reaction score
27
That's not the class room and no one has to grade others posts and make entries to the class register !
Fair enough, it was rather a rather arrogant thing to say. Sorry.

Think of a hypothetical pitot tube that flows through the air. You can make the pitot tube go as fast as you want, but the air capture rate won't exceed density of non-moving air * area of the tube inlet * velocity.
Actually it will, that's why pitot air speed measurements need a compressiblity correction for high speed flight: https://en.wikipedia.org/wiki/Equivalent_airspeed

The crux is that the air IS moving with respect to the pitot tube and constant density along its path into the pitot tube is merely an approximation that holds for low subsonic (as in <0.3) Mach numbers. Once supersonic, with the mass flow also crossing shock waves between free stream and inlet opening, you have to apply a new set of equations. They too conserve mass, but work rather differently in some regards, for example a pipe diffusor to decelerate air is now convergent rather than divergent.
 

kaiserd

I really should change my personal text
Senior Member
Joined
Oct 25, 2013
Messages
745
Reaction score
80
It is a shame to see some contributors posting what appears to be little better than warmed up 90’s Russian-fan-boy “stealth is rubbish” narratives mixed in with some legitimate and interesting technical detail on this topic.
The problem is trying to unpick one from the other while the former acts to undoubtedly undermining the credibility of the latter.
Literally no one is saying that DSI are ”stealth perfection” or purporting the permanent magical powers and unquestionable immortal invulnerability of low-observatory aircraft.
But some of the generalised arguments presented above by some contributors are almost equally absurd in the opposite direction.
 

pegasus

I really should change my personal text
Joined
Mar 23, 2019
Messages
263
Reaction score
135
I




And going back to the actual topic here; what on earth does your argument actually have to do with DSI inlets?
DSI inlets are understood to be be superior to other more traditional “fixed” inlet designs, including but not limited to from a “stealth” perspective, as part of the overall integrated design of the aircraft in question.
And while you are correct a fully variable inlet with moving parts etc. should still have some advantages over a DSI in some parts of an aircraft’s performance envelope it also has other disadvantages such as weight, complexity and a greater difficulty in managing signature management.
As such statements like “DSI are very visible a short range” make little real sense.
More visible than an equivalent variable geometry inlet? More visible than an equivalent more traditional fixed inlet?
As I previously mentioned
I never said they are not effective i know when we write is hard to express ideas, DSI are good intakes, economic with good performance either in aerodynamics or stealth, F-35 has the ideal DSI intake, simple and effective, are they limited? yes to 1.6 Mach that is the ideal speed and pressure recovery, So basically it is cheap and easy to maintain, i only said that any technology has advantages as well as disadvantages, why JF-17 has porous holes for its bleeding system? i find it as a disadvantage, simpler is better on DSI intakes, they are fixed and inferior in subsonic speeds to traditional intakes in performance.

So if F-16 basically flies most of the time bellow Mach 1 then why Lockheed will do what Chengdu did with J-10, start building F-16s with DSI intakes, it makes no sense, J-10B/C unless it flies most of the time at supersonic speeds does not have any advantage in performance over F-16 with its traditional intake with boundary layer diverter.

On F-35 makes sense you need an aircraft with stealth, on J-10B/C is only price since it will not fly most of the time at supersonic speeds.

On J-20 if it has 2 porous holes nets for its bleeding system, it added weight and complexity, so tell me what is better for a supercruising aircraft? variable geometry or Fixed? if i want an aircraft to fly well between Mach 1.6 to Mach 2.4 i prefer a Caret with variable geometry, yes it is more expensive, but it is more effective; if the aircraft will fly below Mach 2 and supercruise at speeds in the region of Mach 1,6, then i will take DSI intakes like J-20
 
Last edited:

Ronny

CLEARANCE: Confidential
Joined
Jul 20, 2019
Messages
87
Reaction score
27
Your theory is incorrect no matter how many times you repeat it. MiG-31 maximum speed is Mach 1.23 at low altitude - there is no compromise in low altitude high speed performance involved here, that's about as fast as any aircraft ever managed.
Would you mind telling me where that value come from? I only have the chart for Mig-25 but not for Mig-31
 

Inst

CLEARANCE: Restricted
Joined
Oct 18, 2019
Messages
21
Reaction score
1
That's not the class room and no one has to grade others posts and make entries to the class register !
Fair enough, it was rather a rather arrogant thing to say. Sorry.

Think of a hypothetical pitot tube that flows through the air. You can make the pitot tube go as fast as you want, but the air capture rate won't exceed density of non-moving air * area of the tube inlet * velocity.
Actually it will, that's why pitot air speed measurements need a compressiblity correction for high speed flight: https://en.wikipedia.org/wiki/Equivalent_airspeed

The crux is that the air IS moving with respect to the pitot tube and constant density along its path into the pitot tube is merely an approximation that holds for low subsonic (as in <0.3) Mach numbers. Once supersonic, with the mass flow also crossing shock waves between free stream and inlet opening, you have to apply a new set of equations. They too conserve mass, but work rather differently in some regards, for example a pipe diffusor to decelerate air is now convergent rather than divergent.
When I say conservation of mass, what I mean is that you have free stream air with a MFR of roughly density * area * velocity moving into the inlet. The fixed variable for a given altitude is density. We can also treat velocity as fixed for the purposes of our thought experiment.

The compressibility is extremely important for decelerating flow, but because you're limited by the MFR of the free stream air, it's hard to see how compressibility can get you something out of nothing. You can decelerate the air as it comes into the inlet by compressing it, but the change in density does not generate mass and thus does not increase MFR.

The only way I can see MFR increasing as a result of supersonic velocity would be treating area as a dependent variable; i.e, as you go supersonic, the area of free stream flow capture increases because of supersonic effects.

But we have a few problems here. First, if the compression is being done by a normal shockwave, I don't think there's any possible capture here. If it's an oblique shockwave, if you visualize the vector diagram, the only increased flow could come from upstream, not downstream. And upstream relative to the inlet won't come from outside the inlet, leading to no increased flow.

The area increase we COULD obtain incidentally has nothing to do with the inlet itself. It has to do, instead, with the air captured by the fuselage and the resulting boundary layer. But how much of that boundary layer needs to be diverted off, and how much of that boundary layer is recoverable?
 

Trident

CLEARANCE: Top Secret
Senior Member
Joined
May 8, 2006
Messages
846
Reaction score
27
But we have a few problems here. First, if the compression is being done by a normal shockwave, I don't think there's any possible capture here.
You still have the air moving at much higher speed through the same area upstream of a normal shock than downstream, leading to a substantial increase in density downstream. At a free stream Mach number of 2.0 compression is about a factor of 2.7, giving virtually sea level density at 30k feet altitude.
 

overscan

Administrator
Staff member
Joined
Dec 27, 2005
Messages
11,600
Reaction score
1,068
But we have a few problems here. First, if the compression is being done by a normal shockwave, I don't think there's any possible capture here.
You still have the air moving at much higher speed through the same area upstream of a normal shock than downstream, leading to a substantial increase in density downstream. At a free stream Mach number of 2.0 compression is about a factor of 2.7, giving virtually sea level density at 30k feet altitude.
Which is presumably why in afterburner at high speeds, thrust can exceed the value at sea level, despite the thinner air it has to work with.
 

Inst

CLEARANCE: Restricted
Joined
Oct 18, 2019
Messages
21
Reaction score
1
But we have a few problems here. First, if the compression is being done by a normal shockwave, I don't think there's any possible capture here.
You still have the air moving at much higher speed through the same area upstream of a normal shock than downstream, leading to a substantial increase in density downstream. At a free stream Mach number of 2.0 compression is about a factor of 2.7, giving virtually sea level density at 30k feet altitude.
But we have a few problems here. First, if the compression is being done by a normal shockwave, I don't think there's any possible capture here.
You still have the air moving at much higher speed through the same area upstream of a normal shock than downstream, leading to a substantial increase in density downstream. At a free stream Mach number of 2.0 compression is about a factor of 2.7, giving virtually sea level density at 30k feet altitude.
The issue is, what is compared to what?

This is basically a quantitative discussion. For instance, Mach 2.7 at 35k feet gives you roughly the same MFR as Mach .9 at sea level, assuming the inlet's pressure recovery is about the same at the different regimes. On the other hand, if you want to do Mach 1.2 at 35k feet with the same level of thrust as Mach .9 sea level, you'll need a larger inlet.

As an alternative, we could instead do a comparison between Mach .5 and Mach 2, which is a factor of 4 in terms of velocity. In this case, since air density drops by about 70% from sea level to altitude, you get a total factor of about 1.2x sea level at Mach .5.

So inlet area in fact becomes a tuning factor when it comes to engine performance, too small, bad performance at a variety of altitudes and speeds, too big, bad performance at low speeds.

====

This originally came up in discussing as to whether the J-20's different inlet could allow the J-20 to supercruise on Al-31. The answer was ultimately no, even with the best case inlet area calculations (which could very well be wrong), except that it could hit a pseudo-supercruise (dive or AB into supersonic) at 35k ft of Mach 1.2-1.4 if the best case inlet area calculations were used. So that's an important flight regime, i.e, MFR at Mach .9 and when breaking the Mach barrier, so that's a flight regime we should be concerned about.

===

And TBH Trident, I was more concerned about puncturing your claims regarding speed being able to unconditionally overcome air density issues at altitude. If you look at real thrust / altitude diagrams for engines, as with the one you showed me, it's hard to pinpoint MFR as the only factor limiting engine performance at altitude. Likely, there's also factors of total pressure recovery at different altitudes and speeds that limit the airframe. I'll point out that with your Al-31 chart, at quite a few altitudes the Al-31 thrust just flatlines despite increasing Mach. Another possible factor is that the charts we've seen discuss pressure recovery as a function of speed. But does pressure recovery also vary as a factor of freestream air density? That could be an interesting factor, but quite likely dynamic pressure overwhelms static pressure when it comes to inlet TPR, even if dynamic pressure is supposed to be near-constant if the inlet does it work correctly at the compressor.

===

I will also note that I'm very happy that we're having a serious conversation about inlets. This is, as overscan's book discussed, an oft-overlooked element of fighter design. Stuff like wing designs, wing loading, engine designs, and so on, these are all way sexier for enthusiasts than the inlet. I didn't quite manage to catch your insulting statement, so I'd like to point out I enjoy our conversation.
 
Last edited:

Deino

CLEARANCE: Top Secret
Staff member
Top Contributor
Senior Member
Joined
Jan 2, 2006
Messages
2,527
Reaction score
89
Interesting image

J-20 vs F-22 intake.jpg
 

overscan

Administrator
Staff member
Joined
Dec 27, 2005
Messages
11,600
Reaction score
1,068
Agreed. Ronny or Pegasus's posts aren't helpful to the topic so I've split to a new topic. I'd missed Deino's post in a sea of irrelevancy.
 

galgot

CLEARANCE: Top Secret
Senior Member
Joined
Jul 6, 2006
Messages
528
Reaction score
215
Website
galgot.com
Looking inside that air intake, one can see rows of rivets or holes ? if these are holes , then the exit could be the hexagon shapes ?
And the first hexagon shape does not project in the inside, so if it's an inlet then the airflow must get out further back in the main intake or elsewhere.

Sans titre.jpg
 

Trident

CLEARANCE: Top Secret
Senior Member
Joined
May 8, 2006
Messages
846
Reaction score
27
I noticed that too, they look like rivets to me. A BL bleed inside the inlet would probably be on the fuselage side (i.e. the compression surface), as that's where the adverse pressure from the shocks would affect the boundary layer. Frequently the BL air removed by the bleed is ducted some distance away to a convenient location for an outlet, so the fact that there is nothing visible on the interior of the duct directly opposite the hexagons doesn't mean they aren't bleed openings. For example, on the Su-27 the BL bleed (slot + perforations) are on the ramps inside the roof of the intake while the outlets are on the exterior of the left and right side wall.
 

overscan

Administrator
Staff member
Joined
Dec 27, 2005
Messages
11,600
Reaction score
1,068
I agee they look like rivets not bleed perforations.
 

Inst

CLEARANCE: Restricted
Joined
Oct 18, 2019
Messages
21
Reaction score
1
I think pegasus's points on optimizing DSI for supersonic performance are highly relevant, however.

A big on-going mystery is the apparently lackluster nature of the J-20's maneuverability. Chinese fanboys usually claim that the PLAAF isn't going to give away the goods as to how agile the J-20 is, but we've been given a strip tease on the J-20's capabilities for years. Likewise, considering that the J-20 is scheduled to be re-engined, why not give enemies a false impression of the J-20's agility under the older Al-31 or WS-10 engines when the WS-15 is waiting in the wings?

Moreover, we know from the J-20's other characteristics than the J-20 SHOULD be relatively maneuverable; i.e, the wing loading is good and the T/W is around the same level as a F-35 at 60% fuel.

One possibility is that the DSI is just optimized for high speeds, i.e, the DSI bump is overly large and reduces air pressure at low subsonic speeds.

The VTech student analysis suggested that with 21.77 meters length and about 13.1 diameter (most accurate numbers are currently about 20.4m length with 12.9 meters wingspan) a shock probe forms at about Mach 2.1, and their estimate is that the J-20 is a Mach 2.1 to Mach 2.5 fighter.

Optimizing a DSI for middling Mach 2 regimes, on the other hand, implies slouch at subsonic speeds.

Something I've pointed out is that with moderate J-20 weight numbers, the J-20 could end up having absurd 1.45 to 1.6 T/W ratios at 60% fuel if WS-10s come to the 180kn range. That implies that the J-20 could simply rely on surplus power at subsonic speeds instead of focusing on high pressure recovery now.

If you're stuck working with poor engines in the 125kn-145kn range when you want a 180kn engine, you have two workarounds. First, you can optimize for low-speed maneuverability a la the Su-57, while sacrificing high-speed performance. When the upgraded engines come in, you now improve both in maneuverability and high-speed performance. Second, you can optimize for high-speed performance while sacrificing low-speed maneuverability.

The question then comes to, when you're running a modern stealth fighter in an arena where most competitors are running 4th generation aircraft, do you optimize for high-speed performance or do you optimize for low-speed performance? Remember, at the end of the day, the J-20 is currently slated to be equipped with only 6 missiles, two of them dogfight missiles. If you assume it'll take 2 missiles to knock down an enemy plane, would you prefer to dogfight or would you prefer BVR combat, where 5th generation aircraft excel in?
 

kaiserd

I really should change my personal text
Senior Member
Joined
Oct 25, 2013
Messages
745
Reaction score
80
I think pegasus's points on optimizing DSI for supersonic performance are highly relevant, however.

A big on-going mystery is the apparently lackluster nature of the J-20's maneuverability. Chinese fanboys usually claim that the PLAAF isn't going to give away the goods as to how agile the J-20 is, but we've been given a strip tease on the J-20's capabilities for years. Likewise, considering that the J-20 is scheduled to be re-engined, why not give enemies a false impression of the J-20's agility under the older Al-31 or WS-10 engines when the WS-15 is waiting in the wings?

Moreover, we know from the J-20's other characteristics than the J-20 SHOULD be relatively maneuverable; i.e, the wing loading is good and the T/W is around the same level as a F-35 at 60% fuel.

One possibility is that the DSI is just optimized for high speeds, i.e, the DSI bump is overly large and reduces air pressure at low subsonic speeds.

The VTech student analysis suggested that with 21.77 meters length and about 13.1 diameter (most accurate numbers are currently about 20.4m length with 12.9 meters wingspan) a shock probe forms at about Mach 2.1, and their estimate is that the J-20 is a Mach 2.1 to Mach 2.5 fighter.

Optimizing a DSI for middling Mach 2 regimes, on the other hand, implies slouch at subsonic speeds.

Something I've pointed out is that with moderate J-20 weight numbers, the J-20 could end up having absurd 1.45 to 1.6 T/W ratios at 60% fuel if WS-10s come to the 180kn range. That implies that the J-20 could simply rely on surplus power at subsonic speeds instead of focusing on high pressure recovery now.

If you're stuck working with poor engines in the 125kn-145kn range when you want a 180kn engine, you have two workarounds. First, you can optimize for low-speed maneuverability a la the Su-57, while sacrificing high-speed performance. When the upgraded engines come in, you now improve both in maneuverability and high-speed performance. Second, you can optimize for high-speed performance while sacrificing low-speed maneuverability.

The question then comes to, when you're running a modern stealth fighter in an arena where most competitors are running 4th generation aircraft, do you optimize for high-speed performance or do you optimize for low-speed performance? Remember, at the end of the day, the J-20 is currently slated to be equipped with only 6 missiles, two of them dogfight missiles. If you assume it'll take 2 missiles to knock down an enemy plane, would you prefer to dogfight or would you prefer BVR combat, where 5th generation aircraft excel in?
Do we currently have any reliable up-to-date sources re: the J-20’s current maneuverability and subsonic (and/or supersonic) performance?
If we do then such conjecture around the role of the DSI in those performance parameters may be warranted.
If not then this conjecture appears somewhat premature, with a lot of “...and if...” accumulating.
 

Inst

CLEARANCE: Restricted
Joined
Oct 18, 2019
Messages
21
Reaction score
1
I think pegasus's points on optimizing DSI for supersonic performance are highly relevant, however.

A big on-going mystery is the apparently lackluster nature of the J-20's maneuverability. Chinese fanboys usually claim that the PLAAF isn't going to give away the goods as to how agile the J-20 is, but we've been given a strip tease on the J-20's capabilities for years. Likewise, considering that the J-20 is scheduled to be re-engined, why not give enemies a false impression of the J-20's agility under the older Al-31 or WS-10 engines when the WS-15 is waiting in the wings?

Moreover, we know from the J-20's other characteristics than the J-20 SHOULD be relatively maneuverable; i.e, the wing loading is good and the T/W is around the same level as a F-35 at 60% fuel.

One possibility is that the DSI is just optimized for high speeds, i.e, the DSI bump is overly large and reduces air pressure at low subsonic speeds.

The VTech student analysis suggested that with 21.77 meters length and about 13.1 diameter (most accurate numbers are currently about 20.4m length with 12.9 meters wingspan) a shock probe forms at about Mach 2.1, and their estimate is that the J-20 is a Mach 2.1 to Mach 2.5 fighter.

Optimizing a DSI for middling Mach 2 regimes, on the other hand, implies slouch at subsonic speeds.

Something I've pointed out is that with moderate J-20 weight numbers, the J-20 could end up having absurd 1.45 to 1.6 T/W ratios at 60% fuel if WS-10s come to the 180kn range. That implies that the J-20 could simply rely on surplus power at subsonic speeds instead of focusing on high pressure recovery now.

If you're stuck working with poor engines in the 125kn-145kn range when you want a 180kn engine, you have two workarounds. First, you can optimize for low-speed maneuverability a la the Su-57, while sacrificing high-speed performance. When the upgraded engines come in, you now improve both in maneuverability and high-speed performance. Second, you can optimize for high-speed performance while sacrificing low-speed maneuverability.

The question then comes to, when you're running a modern stealth fighter in an arena where most competitors are running 4th generation aircraft, do you optimize for high-speed performance or do you optimize for low-speed performance? Remember, at the end of the day, the J-20 is currently slated to be equipped with only 6 missiles, two of them dogfight missiles. If you assume it'll take 2 missiles to knock down an enemy plane, would you prefer to dogfight or would you prefer BVR combat, where 5th generation aircraft excel in?
Do we currently have any reliable up-to-date sources re: the J-20’s current maneuverability and subsonic (and/or supersonic) performance?
If we do then such conjecture around the role of the DSI in those performance parameters may be warranted.
If not then this conjecture appears somewhat premature, with a lot of “...and if...” accumulating.
Maneuverability evidence:


Watch 0:21.


Start around 1:11:21 for the J-20 in action. Watch the nozzle positions to check for when afterburners are on.

Pilot statement:

"The J-20 is good subsonically, but it's outstanding supersonically."

Song Wencong design document ended up being substantially incorrect, insofar as he proposed an aircraft with caret intakes instead of DSI.

However, he stated that short-take off, supermaneuverability, stealth, and supercruise were targets of the J-20 program.

As a retort to Chinese nationalists who want to exaggerate the J-20's maneuverability, I point out that supermaneuverability refers to post-stall maneuvering, which can also be understood as ITR, as opposed to the sustained turn rates that are a mainstay of dogfighting.

====

When I discuss the Su-57, part of the stealth objective seems to be stealthy enough to get into WVR combat, which is typically dogfighting, as supersonic maneuvering has corner radiuses measured in miles. In my view, low-speed (sub-Mach 1) and dogfighting are synonymous, as by the time you get to Mach 2 9G turns are around 2 degrees / second.

That said, I wouldn't bash the supercruise on the Su-57 either. It has variable intakes; it's compromising stealth, but not engine performance at speed or altitude.
 

overscan

Administrator
Staff member
Joined
Dec 27, 2005
Messages
11,600
Reaction score
1,068
Song Wencong design document ended up being substantially incorrect, insofar as he proposed an aircraft with caret intakes instead of DSI.
Aside from the entire aerodynamic layout being nearly identical. But with slightly different intakes.
 

pegasus

I really should change my personal text
Joined
Mar 23, 2019
Messages
263
Reaction score
135
Maneuverability evidence:



Pilot statement:

"The J-20 is good subsonically, but it's outstanding supersonically."

Song Wencong design document ended up being substantially incorrect, insofar as he proposed an aircraft with caret intakes instead of DSI.

However, he stated that short-take off, supermaneuverability, stealth, and supercruise were targets of the J-20 program.

As a retort to Chinese nationalists who want to exaggerate the J-20's maneuverability, I point out that supermaneuverability refers to post-stall maneuvering, which can also be understood as ITR, as opposed to the sustained turn rates that are a mainstay of dogfighting.

====

When I discuss the Su-57, part of the stealth objective seems to be stealthy enough to get into WVR combat, which is typically dogfighting, as supersonic maneuvering has corner radiuses measured in miles. In my view, low-speed (sub-Mach 1) and dogfighting are synonymous, as by the time you get to Mach 2 9G turns are around 2 degrees / second.

That said, I wouldn't bash the supercruise on the Su-57 either. It has variable intakes; it's compromising stealth, but not engine performance at speed or altitude.
Check this is a Chinese document


The diverterless supersonic inlet (DSI) of the Lockheed Martin joint strike fighter (JSF), which operates mostly at transonic speeds, has been designed taking whatever is mentioned above into enough account. Fundamental researches on this inlet configuration have been continued since the mid-1990s.
The inlet cowl lips are so designed as to allow most of boundary layer flow to spill out of the aft notch. The DSI structure complexity has been greatly reduced by the removal of moving parts, a boundary layer diverter and a bleed or bypass system thus decreasing the aircraft’s empty weight, production cost, and requirements of maintenance-supporting equipment[1-2].

the effects of the free stream Mach number on the mass flow coefficient and total pressure recovery when D = 0º and E = 0º. As the free stream Mach number increases, the mass flow coefficient decreases, and, after reaching the minimum at Mach number 1.000, it increases. Fig.7 also shows that the total pressure rises and remains constant when the free stream Mach number is up from 0.600 to 1.000, and, afterwards, drops sharply while the free stream Mach number approaches the supersonic.

4 Conclusions A wind-tunnel test of a ventral diverterless high offset S-shaped inlet has been carried out to investigate the aerodynamic characteristics at transonic speeds. Some conclusions can be drawn as follows: (1) There is a large region of low total pressure at the lower part of the inlet exit caused by the counter-rotating vortices formed at the second turn of the S-shaped duct. (2) The performances of the inlet reach almost the highest at Mach number 1.000. This renders the propulsion system able to work with high efficiency in terms of aerodynamics. (3) As the mass flow coefficient increases, the total pressure recovery decreases; the distortion increases at Ma0 = 0.850, but fluctuates at Ma0 = 1.000 and 1.534. (4) The total pressure recovery increases slowly first, and then remains unchanged as the Mach number rises from 0.600 to 1.000. (5) The performances of the inlet are generally insensitive to angles of attack from –4º to 9.4º and yaw angles from 0º to 8º at Mach number 0.850, and angles of attack from –2º to 6º and yaw angles from 0º to 5º at Mach number 1.534.


1572143259764.png


A Ventral Diverterless High Offset S-shaped Inlet at Transonic Speeds Xie Wenzhong*, Guo Rongwei College of Energy and Power Engineering, Nanjing University of Aeronautics and Astronautics, Nanjing 210016, China Received 13 September 2007; accepted 18 December 2007



See it says the cowl position has to be so that the boundary layer stays away from the inlet, check it says as the speed grows beyond Mach 1, the pressure recovery goes down, but mass flow goes up, the original document is a PDF you have to download it but it has a graph that shows the limits of the fixed DSI intake, since it has no bleeding system or bypass doors the pressure recovery goes down, please also check they say F-35 operates at transonic speeds, thus as overscan mentioned those parts in this thread speculate are bleed systems or bypass doors will be logic conclusions from the Chinese document.

As you mentioned chinese nationalist hardly want to see even the evidence by Chinese documents, see it says the intake has its best performance at Mach 1, from that you will see that J-20 like F-35 are to operate in the best ideal speed at Mach 1 and a decent one at Mach 1.7.
 
Last edited:

overscan

Administrator
Staff member
Joined
Dec 27, 2005
Messages
11,600
Reaction score
1,068
The document you cite is specifically a Chinese analysis of the F-35's DSI intake. Its as relevant to J-20 as studying the pitot intake of the AMX and concluding the F-16 must likewise be subsonic.
 

pegasus

I really should change my personal text
Joined
Mar 23, 2019
Messages
263
Reaction score
135
The document you cite is specifically a Chinese analysis of the F-35's DSI intake. Its as relevant to J-20 as studying the pitot intake of the AMX and concluding the F-16 must likewise be subsonic.
Not necessarily, it is more like the intake of IAI lavi and F-16 or Sepecat jaguar, Mitsubishi F-1 and JH-7, consider while there are differences, in that you are correct, the speed ranges DSI intakes are to operate are very close, the speculation you pointed very well is correct why JF-17 has porous bleed system in the DSI intake bump and cowl? and you translated into J-20, are there differences? yes, but the limits of these intakes are pretty similar and the gains they get are not spectacular but reduced to small gains

plus the article is more for J-10B
see the title

A Ventral Diverterless High Offset S-shaped Inlet at Transonic Speeds Xie Wenzhong*, Guo Rongwei College of Energy and Power Engineering, Nanjing University of Aeronautics and Astronautics, Nanjing 210016, China Received 13 September 2007; accepted 18 December 2007
 
Last edited:

kaiserd

I really should change my personal text
Senior Member
Joined
Oct 25, 2013
Messages
745
Reaction score
80
I think pegasus's points on optimizing DSI for supersonic performance are highly relevant, however.

A big on-going mystery is the apparently lackluster nature of the J-20's maneuverability. Chinese fanboys usually claim that the PLAAF isn't going to give away the goods as to how agile the J-20 is, but we've been given a strip tease on the J-20's capabilities for years. Likewise, considering that the J-20 is scheduled to be re-engined, why not give enemies a false impression of the J-20's agility under the older Al-31 or WS-10 engines when the WS-15 is waiting in the wings?

Moreover, we know from the J-20's other characteristics than the J-20 SHOULD be relatively maneuverable; i.e, the wing loading is good and the T/W is around the same level as a F-35 at 60% fuel.

One possibility is that the DSI is just optimized for high speeds, i.e, the DSI bump is overly large and reduces air pressure at low subsonic speeds.

The VTech student analysis suggested that with 21.77 meters length and about 13.1 diameter (most accurate numbers are currently about 20.4m length with 12.9 meters wingspan) a shock probe forms at about Mach 2.1, and their estimate is that the J-20 is a Mach 2.1 to Mach 2.5 fighter.

Optimizing a DSI for middling Mach 2 regimes, on the other hand, implies slouch at subsonic speeds.

Something I've pointed out is that with moderate J-20 weight numbers, the J-20 could end up having absurd 1.45 to 1.6 T/W ratios at 60% fuel if WS-10s come to the 180kn range. That implies that the J-20 could simply rely on surplus power at subsonic speeds instead of focusing on high pressure recovery now.

If you're stuck working with poor engines in the 125kn-145kn range when you want a 180kn engine, you have two workarounds. First, you can optimize for low-speed maneuverability a la the Su-57, while sacrificing high-speed performance. When the upgraded engines come in, you now improve both in maneuverability and high-speed performance. Second, you can optimize for high-speed performance while sacrificing low-speed maneuverability.

The question then comes to, when you're running a modern stealth fighter in an arena where most competitors are running 4th generation aircraft, do you optimize for high-speed performance or do you optimize for low-speed performance? Remember, at the end of the day, the J-20 is currently slated to be equipped with only 6 missiles, two of them dogfight missiles. If you assume it'll take 2 missiles to knock down an enemy plane, would you prefer to dogfight or would you prefer BVR combat, where 5th generation aircraft excel in?
Do we currently have any reliable up-to-date sources re: the J-20’s current maneuverability and subsonic (and/or supersonic) performance?
If we do then such conjecture around the role of the DSI in those performance parameters may be warranted.
If not then this conjecture appears somewhat premature, with a lot of “...and if...” accumulating.
So further to the attempts to show otherwise above it now appears to be very evident that the actual current answer to my initial question is very much “No”.
Instead we have conjecture pilled upon conjecture with very little to really root it to reality.
As for “Chinese-fan-boys” they are not very evident on this particular site so I’m not familiar with what they may or may not be claiming.
But I’m afraid much of what is being claimed by some contributors in this topic may also fall into the “fan-boy” category, just of another persuasion.
 
Top