Russian Su-57 Aircraft Thread (PAK-FA and IAF FGFA)

Engineer

Major
True, but not in the way you insinuate. Theory says pressure recovery of a DSI with one conical oblique shock should beat a ramp inlet with one planar oblique shock, and lo and behold - it does. The reason why it isn't attributed to conical shocks is because that would be trivial - what makes it worthy of a paper is that not having conventional means of handling the BL (diverter, bleed) means it is a challenge to translate that theoretical advantage into practical applications.

Conical flow alone doesn't address the BL diversion aspect of DSI, but it establishes an upper limit for the achievable pressure recovery (which was the subject).
A discussion on DSI is pretty much a discussion on boundary layer diversion. Saying "conical flow" tells us nothing more about DSI than saying "air flows around solid objects." Saying "conical flow" doesn't give any information in comparing Su-57's variable geometry inlet to DSI.

So?

Let's take a step back here and return to the basic arguments - we're starting to digress again by getting bogged down in too much detail.

You claim "DSI has equal or better pressure recovery than conventional fixed intakes".

I say this is both a moot point to a degree (in the context of the Su-57 we're dealing with a variable intake, after all) and too broad an argument (not all DSIs are created equal, and nor are all fixed inlets of conventional design).
Actually, I said "DSI is known to have higher performance than traditional fixed inlets, and that includes better pressure recovery." As to whether the claim is broad, it is less broad than your generalization where you stated "Does a DSI have better pressure recovery than a variable intake? No. Period." At least my claim is based on the following:
gsZOWea.png


Discussion about fixed inlet isn't moot at all, as a variable geometry inlet operates at all times similar to a fixed inlet working at designed point. We get bogged down on details because you can't accept a simple fact, and you made more assumptions to rationalize the discrepancies between your original assumption and reality.

The reason why I say so is that the sources you've cited to support your claim are unsuitable to prove it - the fixed inlet configurations they use for comparison with the DSI designs studied are of a nature which is at a disadvantage in theoretical pressure recovery from the outset. For the conventional counterparts to come out ahead in the more detailed analysis would therefore require the DSI to suffer larger penalties in translating the pressure recovery potential from theory into a practical design.

Since achieving acceptable BL diversion without conventional means of doing so (diverter, bleed) is a challenge, the result is indeed not a foregone conclusion, making the investigations performed sensible and pertinent. However, for them to act as proof of your point, DSI would have to be compared and found superior to a conventional fixed inlet configuration which is on par in theoretical pressure recovery potential. The results presented in the papers you've presented merely confirm the theoretical advantages over the specific conventional intakes chosen as benchmarks - they prove nothing about the universal superiority over all other fixed inlets which you keep postulating.

For the same deflection angle (ramp or cone body angle) conical shocks have shallower shock angles and better pressure recovery than oblique planar shocks - that's indisputable (in [6], using a ramp of the same angle as the DSI compression surface will actually result in a normal shock!). Another scenario relevant for practical purposes is same *shock* angle as the ramp equivalent (allows same overall compression surface length, used in references [4] & [5], possibly also [6] as far as the rather poor OCR translation indicates). This gives the same total pressure recovery as the ramp at higher static pressure rise and hence lower Mach after the normal shock, meaning a weaker terminating normal shock with better pressure recovery.

Run the numbers with the shock and cone angles at the design Mach of 1.6 quoted in [6] if you don't believe me (NASA has some nice online Java apps).
This concern of yours has both addressed by me earlier, and by reference [4] and [5]. Creating shock waves thickens boundary layer which gets ingested thus reducing pressure recovery. This phenomenon occurs on all inlets, including conic inlets, but only DSI can naturally diverts this boundary layer. In any case, this concern of yours is red herring, because the Su-57 uses ramp in the variable geometry inlet.

If you leave the BL bleed off the ramp inlet it'll be outperformed by DSI, sure. Not an entirely realistic scenario (nobody would implement a ramp intake without bleed unless the ramp angle is REALLY shallow), and it only adds to the choice of shock system already being an apples to oranges comparison. It's a reasonable means of demonstrating that unlike a ramp intake DSI gets away with out it, but the pressure recovery comparison is unrealistic and moot in the context of fixed intakes *in general* (as opposed to the singular case examined in these papers).
No. The comparison is made between fixed inlet without bleed system and DSI without bleed system — an apple to apple comparison. Suction holes can be incorporated into DSI to further remove boundary layer, as done on JF-17, which improves pressure recovery even more. DSI has an advantage by starting off from a higher pressure recovery base-point.


It is the key word, because it indicates a recognition that not all conventional intakes are created equal (something Lockheed-Martin's publicity material in particular never really addresses). DSI pressure recovery (and even that is a bit of an oversimplification, since it can be based on different shock systems) beats some, matches others and is inferior to still others. Isn't it remarkable that you have as yet, over all these pages, not been able to provide an example which doesn't in fact fit the allegedly "meaningless" theory?

Have you notified the USAF yet that they are wasting their money because DSI in fact does provide pressure recovery comparable to the best 4th generation intakes?
USAF already adopt DSI, so did China. In fact, the two countries putting money where their mouth is a premise for my arguments. Neither country uses variable geometry inlet on their new fighters. The Su-57 is the anomaly.

It doesn't explain every little effect (which are beside the point), but it establishes a pretty good picture of the technological limits in terms of pressure recovery (which IS the point). Plenty good enough to prove your sweeping generalizations on that aspect wrong.
Actually, the detail effects that you are avoiding are the points, because discussion of an inlet design is about implementation details. Inlet performance is not just pressure recovery but includes drag and distortion to pressure recovery, and these points are not addressed by conical flow.
 
Last edited:

Gloire_bb

Captain
Registered Member
USAF already adopt DSI, so did China. In fact, the two countries putting money where their mouth is a premise for my arguments. Neither country uses variable geometry inlet on their new fighters. The Su-57 is the anomaly.
Well, best possible answer is initial speed requirement for PAK FA rfp(request for proposals). Even if this task was lowered(or wasn't, or was relaxed only briefly for initial barch, or they hope to achieve it later on) - things were already too far to make any substantial changes worth it.


Btw, interesting point. Earlier I made a claim on m=2.5+ speed for F-22.
I was wrong.
YF-22 was meant to be capable of reaching this speed.
F/A-22A most likely is uncapable of anything far beyond mach 2.

Furthermore, some interesting piece on DSI: while Lockheed proceeded with fixed inlets, production version of its competitor(claimed to have much better supercruise performance and top speed around m=2.3), F-23A, was expected to have DSIs.


Additional piece: there is still no contract for initial 12 articles(it was expected to be signed in late 2017). So, instead of first products this year, they probably will slide to the right. Delays, delays.
Please, Log in or Register to view URLs content!
 

Air Force Brat

Brigadier
Super Moderator
Well, best possible answer is initial speed requirement for PAK FA rfp(request for proposals). Even if this task was lowered(or wasn't, or was relaxed only briefly for initial barch, or they hope to achieve it later on) - things were already too far to make any substantial changes worth it.


Btw, interesting point. Earlier I made a claim on m=2.5+ speed for F-22.
I was wrong.
YF-22 was meant to be capable of reaching this speed.
F/A-22A most likely is uncapable of anything far beyond mach 2.

Furthermore, some interesting piece on DSI: while Lockheed proceeded with fixed inlets, production version of its competitor(claimed to have much better supercruise performance and top speed around m=2.3), F-23A, was expected to have DSIs.


Additional piece: there is still no contract for initial 12 articles(it was expected to be signed in late 2017). So, instead of first products this year, they probably will slide to the right. Delays, delays.
Please, Log in or Register to view URLs content!

not a good sign, and I'm fairly certain the Indians are giving this whole process a "thumbs down",, the SU-57 will be agile and fast, but it won't be stealthy, and while it may very well supercruise, it will be much slower than the F-22 in dry thrust.. the F-22 will do between Mach 1.6 and Mach 1.8 supercruising in dry thrust,,, the F-22 will have much better high altitude performance than the SU-57, and in fact will pull 6Gs at 50,000 ft,, crazy thrust to do that!

we'll just have to wait and see what kind of performance increase the SU-57 has with the article 30 engines???, that will take a lot of trial and testing to get those engines "rung out", it will be very interesting to see how it stacks up against the hype??
 

Gloire_bb

Captain
Registered Member
we'll just have to wait and see what kind of performance increase the SU-57 has with the article 30 engines???, that will take a lot of trial and testing to get those engines "rung out", it will be very interesting to see how it stacks up against the hype??
Bmpd blog quoted above(which is led by CAST team, one of the most known and respected defense think tanks here in Russia) expects to see actual series beyond 2025(probably 2027 or like this).
Sounds far too much for just an engine upgrade. It's known what VKS doesn't want to settle on an inferior plane either, but it's almost 10 years from now.

Heck, the Japanese expect to have F-3 in the air by then, and right now it's still a paper plane.
 
Last edited:

Tirdent

Junior Member
Registered Member
Jeff: since Engineer's post didn't get deleted, I suppose it's ok to reply to it? If not, feel free to delete!

A discussion on DSI is pretty much a discussion on boundary layer diversion. Saying "conical flow" tells us nothing more about DSI than saying "air flows around solid objects."

In terms of pressure recovery it does. Nothing you've managed to supply to date contradicts that view.

Actually, I said "DSI is known to have higher performance than traditional fixed inlets, and that includes better pressure recovery." As to whether the claim is broad, it is less broad than your generalization where you stated "Does a DSI have better pressure recovery than a variable intake? No. Period." At least my claim is based on the following:

The point still stands - you're once more trying to make sources which only deal with a very restricted field prove a more general point that they were never intended (and are unable) to address.

The Su-57 has a variable intake - your source does not concern itself with that at all, clearly referencing fixed intakes exclusively. And even there the statement only holds true in specific cases, not as a general rule like you keep asserting (which is also reflected in the wording: "certain favourable characteristics compared to other fixed intakes" doesn't necessarily refer to *all* other fixed inlets). This is borne out by the two sources I've provided which indicate some conventional types provide better pressure recovery than DSI.

There is not actually any contradiction between what your sources claim and what mine say, none of them is wrong - your generalization of fixed/conventional intakes is.

Discussion about fixed inlet isn't moot at all, as a variable geometry inlet operates at all times similar to a fixed inlet working at designed point. We get bogged down on details because you can't accept a simple fact, and you made more assumptions to rationalize the discrepancies between your original assumption and reality.

Exactly - a fixed intake provides optimum pressure recovery only in a narrow set of operating conditions, while a variable intake performs much like an optimized fixed intake over a far wider range. No assumptions, discrepancies or generalizations involved at all.

No. The comparison is made between fixed inlet without bleed system and DSI without bleed system — an apple to apple comparison. Suction holes can be incorporated into DSI to further remove boundary layer, as done on JF-17, which improves pressure recovery even more. DSI has an advantage by starting off from a higher pressure recovery base-point.

As I said, for the purpose of proving that DSI gets away without bleed whereas a conventional intake doesn't (which is what the paper is about) it's a sensible comparison to make, but for the question of pressure recovery in configurations applicable to real-world aircraft (which is what our discussion is about) it's useless. Fitting a BL bleed to a DSI also adds all of the associated drawbacks back in too, not just the advantages.

DSI does start from a higher pressure recovery base point in these specific sources, but it's due to the more efficient shock system (also available to a conventionally designed inlet in principle), not any magical property unique to DSI. Again, run the numbers - the difference in pressure recovery at Mach 1.6 (design point) between the shock system of the DSI in [6] and a wedge with the same oblique shock angle is a pretty good match to the difference in experimental results seen in figure 16b at Mach 1.53 (and it even predicts the increase in difference at Mach 1.8 off-design pretty well too). Both intake designs obviously suffer virtually identical penalties compared to their respective theoretical potential from incorporating the required means of BL diversion (bump or diverter + bleed) - the offset between the two remains constant in practise.

USAF already adopt DSI, so did China. In fact, the two countries putting money where their mouth is a premise for my arguments. Neither country uses variable geometry inlet on their new fighters.

Yes - having adopted it, the US found its pressure recovery inferior to the best available 4th generation intakes and is researching alternatives. Apparently, pressure recovery comparable to a good 4th generation variable inlet is still relevant to requirements they consider relevant for the future, even though it works well for the F-35.

And again, where are all those super efficient, low-drag (yeah, drag is important, but DSI has no fundamental advantage in this regard - we've been through how this follows from the design process before) DSIs on modern SST concepts? Your attempt to support a generalization with a couple of singular examples just doesn't work - obstinately ignoring the counterexamples doesn't make that reality go away.

The Su-57 is the anomaly.

No, the J-20 is the anomaly: canards (playing devil's advocate here).

If the Su-57 is an anomaly in terms intake type that's most likely because it is an anomaly with respect to the requirements driving intake design.
 
Top