LCA News and Discussions

Lalmohan · Post by **Lalmohan** » 13 Jul 2010 21:14

at landing speeds there will be no shock waves to worry about
splitter plates are used for sub to transonic flow transition control (mig23... phantom, etc.)
the raked intake is for high transonic to supersonic speeds (f15, su 30...)

Arunkumar · Post by **Arunkumar** » 13 Jul 2010 21:51

IIRC angled intakes are a economical way of regulating shockwave into intakes for aircrafts(Concorde, Mig 25) that spend most of their time in the supersonic regime. SR-71 solved the supersonic shockwave problem by moving the engine cones back and forth.

Kartik · Post by **Kartik** » 13 Jul 2010 22:17

Dileep wrote:
What confuses me is which happens first? Airframe stall, or engine flameout? IMLMO, a well designed plane should have the airframe stall before engine flames out. I know for a fact that all these things are paranoiacally simulated to death at the initial stages of design, and till now, that effort paid ample dividend. Maybe the simulations did not guarantee non-flameout at, say 33.6 degrees boundary limit. Hence the scare.

In the past, aircraft could have both (airframe stall and engine flameout) happen to them independent of the other limits depending on how the pilot treated the aircraft. For instance, it wasn’t unheard of to have aircraft engines flame out when the throttle was slammed all the way back and then forward even when the aircraft was within its AoA limits. The quoted part below highlights this as spoken by a Viggen pilot.

Q- You have flown most of the Viggen versions, how do they differ performance wise?

Ans- The difference between old recce/attack/EW and the fighter version is tremendous. Let’s break it down to just engine and manoeuvrability to limit the answer some. The modified RM8B was paramount for the Viggen to be effective in air combat. Not so much the extra power, but the carefree handling the modification allowed. You could yank the throttle from idle to full burner and back, while in high AoA, without complaints from the RM8B. It was also stronger, giving better acceleration. Especially at high altitude, going for high supersonic speed with an unlimited engine that we had a few rare examples of.

In air-to-air combat the engine, RM8A, was more than likely to end up in engine stall if you even touched the throttle at some high AoA or altitudes. You could not pull more than 6 G and you had no audio warnings while at the same time the alpha meter and G meter were located ridiculously far away from each other in cockpit. You had to be aware of transonic speeds due to the steering system and so on... So, the AJS was less than a good airplane at air-to-air combat.

It was for this reason (flameouts being relatively common) that engine oxygen systems and engine relights were so important, especially for single engined fighters back in the 60s and 70s..Even the fact that the engine oxygen system (which could be used for a fixed number of times during a flight) was needed for both relights as well as afterburner engagements meant that the pilot would only be able to go for a certain fixed maximum number of afterburner engagements to keep some engine oxygen in reserve in case a relight was required.

The advent of modern turbofans with FADEC has made life a lot simpler for pilots who can now be truly relaxed while pushing the engines close to their limits without worry of them surging and flaming out. Nevertheless, that requires that the aircraft’s air intake and the intake channels be well designed, with measures to prevent engine surging. Excess air or turbulence in the air intake channel can lead to compressor failure and a measure adopted to deal with this (for e.g in the Su-7) was to introduce bleed doors that would automatically bleed off some air and prevent a surge when the fighter was flying at supersonic speeds. Now with CFD analysis, much of these issues can be dealt with at the design stage itself.

But the true problem that the Tejas is facing is most likely explainable by taking a long hard look at its diffuser design and the fact that it was designed FOR THE Kaveri Engine, with its volumetric flow requirements in mind.

Most modern 4th gen fighters (including Tejas) use an S-shaped diffuser duct and the main job it has to do is to ensure that the engine gets proper air supply under all conditions..This diffuser duct’s design is very important since an engine requires air at a moderately sub sonic speed and it has to convey air from the intake to the engine compressor by decelerating the air flow velocity and consequently increasing its pressure head (basically converting kinetic energy into pressure along its flow direction) along the duct length. But since small or medium sized fighter aircraft have short diffuser ducts (due to the airframe limitations of size and weight) the S-shaped diffuser is used to increase the distance over which the pressure head can be created from kinetic energy while keeping the total length of the diffuser within airframe limits.

The advantage of this S-shaped diffuser design is that the engine compressor face is hidden as a result, leading to lower RCS..But, the drawback of this S-shaped design is that since the centerline is not straight, it leads to pressure gradients in cross-wise directions, which is detrimental to the pressure head increase in the flow direction.

Also, these cross-wise pressure gradients lead to the creation of vortices at the exit of the diffuser duct. And what that means is non-uniform, turbulent flow is imparted right at the compressor face, precisely the opposite of what is desired.

So, the combination of pressure loss, and non-uniform turbulent flow leads to reduction in pressure recovery of the duct, which means that the engine performance (installed thrust) is not as good as the advertised thrust may be. So, whereas the F-404 IN20 engine may give a certain thrust when installed on a bench, it may not give the same when installed in the engine bay of the Tejas.

Also, this static pressure recovery at the end of the S-diffuser duct reduces with an increase in the angle of attack. It can drop by more than 50% at an AoA of 30 degrees. THIS is what can lead to an engine flameout.

It can be improved by adding some mechanical devices (spoilers, fences) inside the inlet, or by having auxiliary inlets. I did inform some members of BRF through mail some time ago that the auxiliary doors in the Tejas intake were actually meant to re-energize the separated flow with the inflow of free-stream air. They were not meant to increase air flow to the engine as Ajai Shukla explained. I can understand that a basic explanation is more understandable, but this is the real reason for why the Tejas has those auxiliary intakes, which were added much later and must’ve been the result of flight tests that revealed less than expected thrust.

Two aircraft that I know don’t suffer from this compromise duct design, are the Su-27 and MiG-29 designs since both are large to medium sized fighters. Both have nearly straight diffuser ducts, where you can literally see the engine compressor face almost in a straight line if you look into the intake. The advantage of their design is that even at high AoA, with no cross-wise pressure gradients leading to turbulence and vortices, the pressure recovery at the duct exit is good. That is why these aircraft are so forgiving of high AoA flights. The problem is of course that since the compressor face is not obstructed by anything, the RCS is high. The Russian way to deal with that is to have RAM coating on the inside of the intake and compressor blades, but this is fraught with danger (in case the RAM coatings peel off, they’re basically going to cause FOD) and is very high maintenance.

The Super Hornet uses an intake blocker but this is a complicated design that has to take into account the flow requirements of the engine and should not cause any more pressure loss.

And that may also explain why Sukhoi kept the PAK-FA design with minimal S-shaping of the diffuser duct, so that the high AoA capability is retained, at a certain trade-off with RCS increase.

I’m also afraid that the level of discussion shown here by some posters completely underestimates the actual complexity of the design of an intake channel. It is a FAR MORE complex and complicated subject and several decades of research has gone into how S-shaped diffuser ducts can be optimally designed for least amount of pressure loss at the duct exit.

A simple increase in the volumetric flow of the Tejas Mk2 design won’t do the trick. They will need to work on a larger intake (due to the higher air flow requirements of the F-414 and Ej200) but will also have to do plenty of R&D as well as experimentation to arrive at the best solution of auxiliary intakes and/or fences, vortex generators inside the duct to get the optimum pressure recovery at high AoA.

Rahul M · Post by **Rahul M** » 13 Jul 2010 22:40

wonderful explanation kartik, thanks a lot.

Kartik · Post by **Kartik** » 13 Jul 2010 22:57

And I forgot to mention that for preventing engine surging in flight due to the large swirl angles at the duct exit, variable inlet guide vanes are very useful in engines. The Kaveri has variable inlet guide vanes, and so does the F404, F414 and M-88-2 engine. But the Ej200 doesn’t feature any guide vanes. This is a factor that is quite important in the eventual selection of an engine for the Tejas Mk2..Since PS Subramanyam is on record as stating that both Ej200 and F414 meet their requirements, they must have some alternative for IGVs on the Tejas Mk2 to prevent compressor stall.

negi · Post by **negi** » 14 Jul 2010 00:37

^ The EJ200 has wide chord low aspect ratio blades and it does have VGV for the HPC section as well as for the nozzle. In addition there are bleed valves controlled by FADEC to allow for recovering from compressor surge. It is obvious that MTU must have achieved the desired surge margin figure for their compressor maps during simulations and testing without having to incorporate the IGVs for LPC stage so they saved on weight .

Fwiw EJ200 even had variable stators in the HPC section in its prototype version aka 'XG.40' here :

http://www.flightglobal.com/pdfarchive/ ... 01786.html

ajay_hk · Post by **ajay_hk** » 14 Jul 2010 01:24

Israel, EU in contention to co-develop radars for Tejas

Manu Pubby
Tags : Tejas Light Combat Aircraft, European Consortium EADS
Posted: Wed Jul 14 2010, 23:54 hrs

India is close to finalising a developmental partner for a next generation radar that will be the eyes and ears of the Tejas Light Combat Aircraft (LCA) in the future. With other contenders falling off the race due to different reasons, the race now is between European Consortium EADS and Israeli company Elta that are vying for the initial contract to co-develop 10 prototypes of an Active Electronically Scanned Array (AESA) radar with India.

While the initial contract is for 10 prototypes, industry estimates put the requirement of the Indian defence forces at close to 600 radars for different types of fighters, making the deal potentially worth over $ 3 billion over the next decade. The tenders for co-development were issued by DRDO’s Electronics and Radar Development Establishment (LRDE) in December last year for the LCA Mk II that has already been approved by the government.

Sources said that LRDE is close to short listing its partner for the project and the competition now is between Elta and EADS, down from the initial bid by five companies that were vying for the potentially large contract. While EADS is showcasing its X band technology, Elta specialises in L band technology and is promoting its new generation X band antenna.

While US companies did not participate in the tender — apparently after they could not gain permission from the government to share the high end technology — Russia’s largest radar company Phazotron and France’s defence giant Thales were dropped due to technical reasons. Italian Selex did not make it to the next round after failing to deposit the earnest money specified in the tender.

The radar will also be considered for the SU 30 MKI upgrade and modernisation projects for front line fighters of the Navy and Air Force.

Kartik · Post by **Kartik** » 14 Jul 2010 02:32

negi wrote:^ The EJ200 has wide chord low aspect ratio blades and it does have VGV for the HPC section as well as for the nozzle. In addition there are bleed valves controlled by FADEC to allow for recovering from compressor surge. It is obvious that MTU must have achieved the desired surge margin figure for their compressor maps during simulations and testing without having to incorporate the IGVs for LPC stage so they saved on weight .

Fwiw EJ200 even had variable stators in the HPC section in its prototype version aka 'XG.40' here :

http://www.flightglobal.com/pdfarchive/ ... 01786.html

I did say that Eurojet must have some alternative for IGVs on the Ej200 to prevent compressor stall, so thank you for pointing me in the right direction. I wasn’t aware of this on the Ej200.

The basic premise behind those bleed valves that you're referring to, must then be similar to the method (but of course controlled by FADEC and hence much more reliable) that I referred to in my earlier post, where I mentioned the Su-7. In that Al-7F-I turbojet engine, the 7th and 8th stages of the engine had bleed valves to bleed off excess air and prevent surge.

A published paper on the Ej200’s very initial design stages describes how its design was arrived at, and the deletion of IGVs is mentioned as a very attractive feature for the designers of the Ej200 because it is complicated, expensive and sensitive to bird strikes and removing it brings weight savings (which you rightly mentioned). There is a nearly 100 kg or more weight difference between the F414 and Ej200, which is significant. However, this decision to go without IGVs was not made for some time after the Ej200 design was initiated. It is also cited as one of the reasons for the reduction of 2-4% of overall life-cycle costs for the Ej200, which translates to millions of $ over the engine fleet’s lifetime.

In that same paper it also mentions that the “Ej200 has been exposed to severe intake distortion testing with different types of distortion gauzes simulating distortion levels upto DC 60 ~ 1.0. The engine is highly responsive and has achieved all the specification handling times. During its total flight testing the engine never experienced a surge in flight. The engine does have a very good track record and the pilots like it as reliable, responsive and powerful.”

So that settles the question I had in my mind about the issue of surge resistance for non-IGV Ej200 engines.

vina · Post by **vina** » 14 Jul 2010 09:51

maybe the soln is to insert two such rafaelish scoop intakes ?

Gentlemen. Remember, the Tejas has a shielded inlet . The inlets are behind the leading edge of the wings! . From basic fundamentals we know that he air behind the leading edge is subsonic (that is why you sweep the wing!). So, no problemo. If Tejas is flying at 0 kmph or at it's designed speed of mach 1.8, the inlet sees only an ultra low SDRE subsonic airflow!.

So why scoop and this and that and variable ramps and all that saar?

And who says that the inlet is badly designed ? Shook Law ?. I tend to agree with the folks at ADA/HAL who pooh pooh it. The auxiliary doors are for take off where the engine will be at full power and the air demand is at it's highest. It will be largely closed in all other conditions. See. either you take in extra air and vent it off /bleed it off at all other conditions , or you take in right amount of air at all other conditions and open a door for extra air at take off!. Twiddledee or Twiddledum. All same-same onree.

Dileep · Post by **Dileep** » 14 Jul 2010 10:30

Well, the test pilots do look at everything very skeptically. After all, it is their LIVES that are at risk.

I mean, look at the testers here who sit on a chair and risk nothing (other than a RSI to the wrist) complain! I do saashtaanga pranaamam to the Test Pilots!!

rohitvats · Post by **rohitvats** » 14 Jul 2010 10:36

It is not without reason that the TP job is considered as the most dangerous in the world....

Singha · Post by **Singha** » 14 Jul 2010 10:57

remember that german invention in WW2 known as the "bachem ba349 natter" . the TP for that was a scary job.

http://en.wikipedia.org/wiki/Bachem_Ba_349

The Bachem Ba 349 Natter (Adder) was a World War II era German experimental point-defense rocket-powered interceptor aircraft which was to be used in a very similar way as manned surface-to-air missiles. After vertical takeoff which eliminated the need for airfields, the majority of the flight to the bombers was radio controlled from the ground. The primary mission of the (inexperienced) pilot was to aim the aircraft at its target bomber and fire its armament of rockets. The pilot and the main rocket engine should then land under separate parachutes, while the wooden fuselage was disposable. The only manned test flight, on 1 March 1945, ended with test pilot Lothar Sieber being killed.

Dileep · Post by **Dileep** » 14 Jul 2010 13:05

BTW, let me clarify. The testers I mentioned are (OT Alert) the guys who right now gives me hard time when ONE switch gave a watchdog reboot ONE time, inside an oven, and after power cycling every two minutes for two days.

tsarkar · Post by **tsarkar** » 14 Jul 2010 14:32

No Vina, not Ajai, the test pilot said so and was quoted verbatim.

I have been hearing bits and pieces since 2006 on this issue, however, Kartik’s hypothesis seems to be the most plausible one.

Isn’t it possible to discover the differential pressure gradients caused in the duct during development via simulations or wind tunnel scale model testing of the duct to assess airflow behavior? Or is actual flight testing the only way to find out?

vina · Post by **vina** » 14 Jul 2010 15:12

No Vina, not Ajai, the test pilot said so and was quoted verbatim.

I have been hearing bits and pieces since 2006 on this issue, however, Kartik’s hypothesis seems to be the most plausible one.

The IAF/HAL for whatever reason seem to have some "inlet" hangups. Why they even went and put those inlet doors in Jaguars, while the same Jaguar in French, Brit and Oman and other services don't have them!. Only the IAF made HAL put them on as an "innovation".

Inlet flow distortion at high angle of attack have nothing to do with those auxiliary doors, which only serve to increase size of the inlet, increasing mass flow rate and work best in slow/zero speed/take off when there is no air rammed into the inlet and the thrust requirement is at it's highest. In all other scenarios, there will be ample air and those doors will be firmly shut.

The inlet business really is a hangover from the Kaveri days. You see, the PV-1 is a "Kaveri" prototype (I can't but imagine the dunderheads who actually made the PV-1 with a Kaveri fuselage, when the Kaveri is nowhere near ready.. Heck all they need to do is pick up the phone and ask the GTRE folks how the engine was coming along, classic project management goof up). The Kaveri has a lower mass flow rate than the GE-F404. So when you put the GE404 in the Kaveri fuselage/inlet, you need extra air and hence the auxiliary inlet makes sense.

But I guess the LSP etc have probably no need for the Kaveri inlet and fuselage. They are designed to take the F404 as in the other prototypes. Maybe the IN20 version needs more airflow, hence possibly auxiliary doors made the appearance in the final version (if they did, can anyone check the LSP series prototype photos?)

But all that is in no way related to high alpha /side slip and other distorted flow conditions, that is a separate problem and the inlet would have been suitably designed and tested in wind tunnels to give acceptable performance (having a shielded inlet makes it easier.. you don't need a supersonic wind tunnel and all the testing can be done in the subsonic and transonic tunnels available in India..we don't have a supersonic tunnel as far as I can recall).

tsarkar · Post by **tsarkar** » 14 Jul 2010 16:05

Yes, the auxiliary doors are open in LSP-4 and Tejas Leh takeoffs. They do add to airflow during takeoff as hypothesized. Whether they help regain pressure needs to be checked by observing them during maneuvers.

How about the duct? Wasn’t flow in the duct end assessed during development, simulation or scale model testing? Why is the issue being raised during the fag end of flight testing just near expected IOC? Why wasnt it raised earlier?

Kanson · Post by **Kanson** » 14 Jul 2010 16:23

Dileep wrote: What confuses me is which happens first? Airframe stall, or engine flameout? IMLMO, a well designed plane should have the airframe stall before engine flames out.

I guess it is so. IOW, as engine cannot provide lift in conventional sense, aircraft designers has to extract every juice from the aero structures (aerodynamic efficiency) and the design will be done in such a way so that wing/frame stalls before the engine, provided it is backed by good engine. I think all modern engine has enough surge protection.

I know for a fact that all these things are paranoiacally simulated to death at the initial stages of design, and till now, that effort paid ample dividend. Maybe the simulations did not guarantee non-flameout at, say 33.6 degrees boundary limit. Hence the scare.

Boundary seperation doesnt leads to flameout, if i'm not wrong.

Kanson · Post by **Kanson** » 14 Jul 2010 16:29

Dileep wrote:Well, the test pilots do look at everything very skeptically. After all, it is their LIVES that are at risk.

cheta, if the test pilot is not comfortable, anytime he can quit from the mission/test. He has every right to be skeptical but no need for paranoia the extent shown throu the words of Ajai or misunderstood by Ajai.

Kanson · Post by **Kanson** » 14 Jul 2010 17:07

Kartik wrote:Most modern 4th gen fighters (including Tejas) use an S-shaped diffuser duct and the main job it has to do is to ensure that the engine gets proper air supply under all conditions..This diffuser duct’s design is very important since an engine requires air at a moderately sub sonic speed and it has to convey air from the intake to the engine compressor by decelerating the air flow velocity and consequently increasing its pressure head (basically converting kinetic energy into pressure along its flow direction) along the duct length. But since small or medium sized fighter aircraft have short diffuser ducts (due to the airframe limitations of size and weight) the S-shaped diffuser is used to increase the distance over which the pressure head can be created from kinetic energy while keeping the total length of the diffuser within airframe limits.

The advantage of this S-shaped diffuser design is that the engine compressor face is hidden as a result, leading to lower RCS..But, the drawback of this S-shaped design is that since the centerline is not straight, it leads to pressure gradients in cross-wise directions, which is detrimental to the pressure head increase in the flow direction.

Also, these cross-wise pressure gradients lead to the creation of vortices at the exit of the diffuser duct. And what that means is non-uniform, turbulent flow is imparted right at the compressor face, precisely the opposite of what is desired.

So, the combination of pressure loss, and non-uniform turbulent flow leads to reduction in pressure recovery of the duct, which means that the engine performance (installed thrust) is not as good as the advertised thrust may be. So, whereas the F-404 IN20 engine may give a certain thrust when installed on a bench, it may not give the same when installed in the engine bay of the Tejas.

Also, this static pressure recovery at the end of the S-diffuser duct reduces with an increase in the angle of attack. It can drop by more than 50% at an AoA of 30 degrees. THIS is what can lead to an engine flameout.

Dear Kartik ji

What you say about S-duct could be true but i fear that you havent taken note of Tejas duct as two semi S shaped (small bent at the tip) Y-duct. In all practical purpose it is a Y-duct. The small curve at the beginning can no way be called as S-duct.

Let's leave alone the Tejas. You described abt the S-duct very informatively. In your opinion, what happens to the variation in the pressure head across the cross-section, as you described, if there is two S-duct that are connected at the end with an angle.

Correct me if i'm wrong, in all practical purpose, Tejas has straightened two S-duct connected with an angle, which is nothing but Y-duct. So i feel the observation made here may not apply to Tejas. If i'm not wrong, F-404 IN20 is a custom built for the Tejas with its own FADEC.

shiv · Post by **shiv** » 14 Jul 2010 18:38

Kanson wrote: Correct me if i'm wrong, in all practical purpose, Tejas has straightened two S-duct connected with an angle, which is nothing but Y-duct.

Yes - seen in green

Singha · Post by **Singha** » 14 Jul 2010 18:59

> we don't have a supersonic tunnel as far as I can recall

Hypersonic Flight and Ground Testing Activities in India
does any of these have it? http://espace.library.uq.edu.au/eserv/U ... _16_07.pdf

Kartik · Post by **Kartik** » 14 Jul 2010 20:53

Kanson wrote:Dear Kartik ji

What you say about S-duct could be true but i fear that you havent taken note of Tejas duct as two semi S shaped (small bent at the tip) Y-duct. In all practical purpose it is a Y-duct. The small curve at the beginning can no way be called as S-duct.

Let's leave alone the Tejas. You described abt the S-duct very informatively. In your opinion, what happens to the variation in the pressure head across the cross-section, as you described, if there is two S-duct that are connected at the end with an angle.

Correct me if i'm wrong, in all practical purpose, Tejas has straightened two S-duct connected with an angle, which is nothing but Y-duct. So i feel the observation made here may not apply to Tejas. If i'm not wrong, F-404 IN20 is a custom built for the Tejas with its own FADEC.

Dear Kanson,

What I've stated is based on what I read in a published paper by 3 Indian scientists. It clearly mentioned the above facts (as well as the use of auxiliary intakes to re-energize the flow and improve pressure recovery at duct exit) as well as the fact that the Tejas has an S-duct. I cannot find that paper now, but maybe RahulM or Austin will have it..

Besides that, I also had a picture from ADA that someone sent (which I cannot share as he said so) that clearly illustrated the S-shaped duct from the side. The LCA's diffuser duct is S-shaped when viewed from the side, so the centerline is not straight. the Y-duct that you refer to is in the plan view.

Rahul M · Post by **Rahul M** » 14 Jul 2010 21:05

marten ji, by all means, it's an open thread. please do the honours.

here's the paper and the paragraph in question.

http://www.jafmonline.net/modules/htmla ... 415-bp.pdf

(1983) investigated the swirl in an S-duct of typical aircraft intake proportions at different angle incidences. The static pressure recovery (C
SP) reduced with the increase in angle of attack (CSP = 0.89 at 0°angle of attack and CSP = 0.37 at 30° angle of attack) and it could be improved by incorporating several mechanical devices at the inlet, such as, spoiler, fences etc. They studied two methods in order to reduce the magnitude of swirl by means of a spoiler and to reenergize the separated flow with the inflow of free stream air through auxiliary inlets.

Kartik · Post by **Kartik** » 14 Jul 2010 21:11

Thanks Rahul.

shiv · Post by **shiv** » 14 Jul 2010 21:12

I can claim zero expertise about the flow of fluids in pipes. But there are a couple of general points that are quoted regarding fluid (blood) flow in medical topics. Two of those things seem relevant

1) A small narrowing of a blood vessel will have no effect initially. Fluid will just flow faster across the narrowed segment and no loss is felt downstream.

2) In situations where blood vessels are already dilated to the maximum possible level. no increase in flow can be expected by drugs that purport to increase the diameter of the blood vessel

So what am I trying to say?

Why would engineers make a jet engine intake that is not already wide enough to accommodate the maximum possible flow?

Would airflow be deliberately restricted for any reason? From what I have read about blood flow - one would have to make the intake/ducting pretty narrow before the flow is affected significantly.

If an intake already allows as much air as it can take, how can windows/slits/auxiliary inlets on the side let in "more air" . A person who is short of breath will not get more air by making slits on the sides of his nostrils.

Sorry if the questions are naive and based on irrelevant assumptions.

Lalmohan · Post by **Lalmohan** » 14 Jul 2010 21:22

the side inlets are used to re-energise the air as others have said, particularly the boundary layer (where it sticks to the walls) so that the flow is more effectively able to reach the compressor fans, its not so much to increase the airflow, but to keep it flowing more smoothly.

flow separation, etc. leads to drag and turbulence which can lead to drop of mass flow through the pipe and so reduce the available pressure head to the compressor stage. please note that the compressor can work at zero flow since that is what is experienced at start up (although the starter has to start spinning the compressor fan to create sufficient pressure in the combustion chamber), so the larger problem is about the available pressure at the engine start point (first stage compressor)

as an analogy you'll see tiny strakes just above the leading edge of the wing, which serve a similar purpose of energising the boundary layer and help keep the flow laminar (and less draggy)

as Shiv says, viscosity (reynolds number) is also a factor to be allowed for. inside pipes, it is a very large determinant of whether the flow chokes or not (i am now operating from distant memory so one of the gurus needs to take over)

Telang · Post by **Telang** » 14 Jul 2010 22:56

What I am stating below is absolutely an ignorant guess. As a layman what I understand is, in a theoretically fiction less duct when air flows, pressure increases if the velocity of the air drops. This is basically by conversion of the kinetic energy of the air rammed into the intakes (created by the velocity of the aircraft) in to pressure owing to the enlarged area of the duct. (Bernoulli's principle). So in designing an air intake, we can achieve some advantage in increasing the pressure of the air that enters the air intakes by enlarging the duct at the first stage of the compressor. In other words the area of the air intakes is less than the area of the first stage of the compressor. This advantage is somewhat reduced when the AOA is larger than say 25 degrees. Consequently the volume and velocity of the air that enters the intakes is reduced and the ramming effect is lost and the first stage –or even the other successive stages- of the compressor will have to work as a suction pump also. That is what this stage additionally does at the time taxying on the ground (and at the time of starting the engine - of course aided not by its own power but through external source of energy). The rear driving blades consequently get less energy and fail to provide the compressor shaft with sufficient rpm or energy, and this process decelerates to a grinding halt resulting in a flame out. To avoid this, the scientists would have tested the engine under the simulated conditions and would have taken care of some additional mechanism whatsoever (that I do not know) including a middle-of-the- air starter, and some control measures to stabilize otherwise stalling aircraft. Talking of air intakes, the MiG 21 has variable air intake with a radar cone in the center, and this design does not seem to have been repeated elsewhere, surprising.

In the good olden days of subsonic flights, the first jet to enter IAF, the Vampire had an engine with an equally devilish name – “Goblin”, which was huge, with a single stage impeller type compressor. Those days the jet engines looked so simple and less complicated than the cumbersome piston engines. Now it is the other way round.

negi · Post by **negi** » 14 Jul 2010 23:50

^ Telang there is a catch if you observe the EPR and its distribution along the various sections of a jet engine the moment compressor would encounter an unstable/turbulent airflow the compressed air in aft of the HPC would have this tendency to gush out through the fore section of the engine as the pressure towards the LPC end is lower than that towards the HPT end .

Modern jet engines already have measures to bleed the air from HPC section until the EPR comes down to a value which allows for the LPC section to recover pressure (all controlled by FADEC).

Mig-21 is a Mach 2+ capable AC Tejas is not hence it should not require a variable geometry intake.

Raman · Post by **Raman** » 15 Jul 2010 02:23

Why would engineers make a jet engine intake that is not already wide enough to accommodate the maximum possible flow?

Would airflow be deliberately restricted for any reason? From what I have read about blood flow - one would have to make the intake/ducting pretty narrow before the flow is affected significantly.

If an intake already allows as much air as it can take, how can windows/slits/auxiliary inlets on the side let in "more air" . A person who is short of breath will not get more air by making slits on the sides of his nostrils.

The air flow required by an engine is not fixed in all regimes: it depends on altitude, throttle setting, airspeed, etc. So why not create an oversized inlet to accomodate the maximum required airflow? Because in all other regimes, the required airflow will be less. The "extra" air rammed into the diffuser by the inlet cannot be used by the engine and must be dumped through valves or will otherwise "spill" from the front of the duct itself. (Think of pouring water into a funnel faster than the funnel can drain - the funnel will overflow.) This results in either bypass drag (if dumped after entering the inlet) or spillage drag (if not) and remember that this is for *all* other flight regimes than the one that requires maximal flow (which will be for a very restricted portion of the mission profile). This can result in overall poor performance. Thus it would be better to size the inlet for better performance in the more prevalent case and add extra inlets for the maximal case.

rahulm · Post by **rahulm** » 15 Jul 2010 07:43

Conceptually, the same issues are faced in IC engines.

Are variable geometry inlet systems a possible answer?

There exists a patent for IC engines Variable geometry intake system for an internal combustion engine. Another solution is variable length inlet manifolds

However, the operating regimes of the IC engine and aircraft engines are different.

But, if these were possible for aircraft they would have been tried by now!

I am not an aeronautics engineer, just saw a conceptual similarity.

vina · Post by **vina** » 15 Jul 2010 08:11

Raman wrote:
Why would engineers make a jet engine intake that is not already wide enough to accommodate the maximum possible flow?

Would airflow be deliberately restricted for any reason? From what I have read about blood flow - one would have to make the intake/ducting pretty narrow before the flow is affected significantly.

If an intake already allows as much air as it can take, how can windows/slits/auxiliary inlets on the side let in "more air" . A person who is short of breath will not get more air by making slits on the sides of his nostrils.
The air flow required by an engine is not fixed in all regimes: it depends on altitude, throttle setting, airspeed, etc. So why not create an oversized inlet to accomodate the maximum required airflow? Because in all other regimes, the required airflow will be less. The "extra" air rammed into the diffuser by the inlet cannot be used by the engine and must be dumped through valves or will otherwise "spill" from the front of the duct itself. (Think of pouring water into a funnel faster than the funnel can drain - the funnel will overflow.) This results in either bypass drag (if dumped after entering the inlet) or spillage drag (if not) and remember that this is for *all* other flight regimes than the one that requires maximal flow (which will be for a very restricted portion of the mission profile). This can result in overall poor performance. Thus it would be better to size the inlet for better performance in the more prevalent case and add extra inlets for the maximal case.

Ah, Raman Niyar-al-Kampooteri . Perfectly explained, including the analogy of pouring water into a funnel faster than it can drain.

Yes, Shivji, that is the point, any extra air that the engine cannot you take though the inlet is dumped overboard. So as an alternatively you size the inlet for the most scenarios and use doors to let in more as an when you needed.

Let me try explaining it again in a way you can relate to. Consider ordinary Hyooman (unlike Djinn powered SooperMan, who doesn't need to breathe) . In most circumstances, you breathe through your nostrils and hardly notice breathing(if you do,you are probably short of breath/palpitating/mild asthma/whatever). Now you climb a flight of stairs, the oxygen demand increases, your diaphragm starts doing ooper-neeche faster and you start taking deep breaths. If you are now doing a 400m dash, oxygen demand shoots up, diaphragm does ooper-neeche very very fast indeed and the teeny weeny nostrils are maxed out and so what do you do? Ta-daa.. the mouth opens and you start taking in in massive gulps of air through a direct,shortest and least loss-most efficient path straight to lungs, by passing the looped thin and lossy passages in the nose/head.

In Jihaj-e-Kufr, when the Yin Jinn demands max air (at take off, as I will explain right after this), just like you open your mouth for extra air , the auxiliary doors open to let in more air to feed demand. In all other regimes it will be shut and it will breathe through the inlets,just like Hyoomans do 99% of the time. Just think of the inlets of Jihaj like the nostrils of a Hyooman..(those teeny weeny ones resemble nostrils anyways..

)

auxiliary doors are used to energize boundary layer

This cannot be true for the inlet. If you energize the boundary layer , the resulting airflow will be turbulent and can trip the entire flow to highly turbulent and maybe unpredictably I think and can do very bad things to the engine compressor. That is why I think it is used only during take offs and low speed regimes.

Also, the most use use is during take offs and slow speeds , when the free stream velocity is zero and there is really no dynamic pressure to be recovered when it is slowed down in the diffuser. There is no ram air effect , and the max flow rate through the engine will be at it's highest. Such a scene will have huge pumping losses, so larger than normal sized inlet will be useful (lowering pumping loss) and the auxiliary doors will serve some useful purpose.

However , this door business has really limited utility and probably helps decrease take off distance by a few tens of meters at most. That could probably be the reason why no one else went for this kind of solution. Maybe for IAF, that few tens of meters difference is important and they insisted on it. Dunno. Can be useful if taking off from damaged forward airbases I suppose.

negi · Post by **negi** » 15 Jul 2010 08:42

Vina ji imho as far as the engine is concerned the conditions during the take off are very similar to AC pitching its nose up to operate near its stall speed so the auxillary doors will come in play during such maneuvers too and these conditions can surface even when AC flight profile enters high AoA regime or say encounters strong cross winds or if AC happens to fly into a jet wash , coming to energizing the boundary layer inside the inlet unless there is a way to bleed it before it hits the IGV/fan letting in extra air through auxiliary intakes to streamline the flow makes sense.

shiv · Post by **shiv** » 15 Jul 2010 08:58

Question: Isn't there a huge huge difference between takeoff at sea level (or sea level plus say 1000 meters like Bangalore) and stalling at 5000 meters. Wouldn't air density and pressure make a huge difference in what the engine thinks and feels?

vina · Post by **vina** » 15 Jul 2010 09:54

Question: Isn't there a huge huge difference between takeoff at sea level (or sea level plus say 1000 meters like Bangalore) and stalling at 5000 meters.

Yes. They are all different.

Wouldn't air density and pressure make a huge difference in what the engine thinks and feels?

No. Because, the engines and all stuff are designed around "corrected" values, so that everything is referenced back to ISA conditions and designed for that. For eg, Corrected Flow . So all reference generally is to ISA conditions only.

Telang · Post by **Telang** » 15 Jul 2010 10:01

Shiv Jee, there ofcourse is a huge difference. But it requires a PhD to explain and it requires a PhD to understand. This particular one topic is highly complicated. For we commoners, atleast I for one, am assured that my beloved Indian scientists do know about it and in the case of Tejas, they are taking care of all that is required to be taken care of, despite our colonel sahib and his ilk., period.

vina · Post by **vina** » 15 Jul 2010 10:13

Telang wrote:Shiv Jee, there ofcourse is a huge difference. But it requires a PhD to explain and it requires a PhD to understand.

No no.. Modern technology can do wonders in explaining. How I wish there were things like internet and tools like these- Nasa Glenn Corrected Airflow to explain these when folks like me were in college. Would have made things lot easier than sharing 3 copies of textbook for the entire class (that is how skinflint things were back then.. dunno how it is now)

Just click on it to play around with that calculator to get a feel of what I am talking about. I just googled around and found it after Shivji asked the question.

Telang · Post by **Telang** » 15 Jul 2010 10:36

vina wrote: Just click on it to play around with that calculator to get a feel of what I am talking about. I just googled around and found it after Shivji asked the question.

Vinajee, I am a layman, do you think I can make head and tail of this?????

these- Nasa Glenn Corrected Airflow

What level of education is required to understand that mumbo jumbo? 10th, degree, masters??

But I trust when people like you just assure that some one as learned as it requires to be is in the required job, at the required spot, to do the required trick, in the making of Jahaj-e-kafir.

shiv · Post by **shiv** » 15 Jul 2010 17:07

Thanks all for the patient explanations

enqyoob · Post by **enqyoob** » 15 Jul 2010 17:45

If an intake already allows as much air as it can take, how can windows/slits/auxiliary inlets on the side let in "more air" . A person who is short of breath will not get more air by making slits on the sides of his nostrils.

Interesting pooch relating biological and engineering gas dynamics. Same answer as the question: Why don't they make Pakistanis with mouths big enough? Answer is the same in both cases:

The controlling constriction is really at the other end.

The "demand for air flow" is created when more fuel is put in, generating more heat initially, and spinning the turbine faster, and this drives the compressor faster. This, and not how fast the flow is going, the density of the air in front, etc., dictates the pressure in front of the compressor. This typically occurs at very low flight speeds (like takeoff). Making the inlet capable of opening much wider, is impractical, because it will make the whole engine bigger and heavier and more draggy. If the inlet is small, it's not that air won't come in. The problem is that the pressure inside the inlet would be so low that the flow would separate inside the inlet (or diffuser) and instead of a nice uniform flow, the compressor face would see a jet. The compressor would stall.

So instead they size the inlet for best operation at supersonic speeds, and allow other doors to open to allow more air to come in and fill the duct ahead of the compressor.

Anyway these doors are needed. As the aircraft goes through transonic speeds, there is a thing called "starting" of the inlet so that it changes from having a shock out in front of the inlet and all subsonic flow inside, to having supersonic flow coming in the inlet and gradually slowing to subsonic. The way to get around this, is to open those doors, which changes the pressure inside the inlet just enough to "swallow the shock".

Now about the constriction at the other end (of the engine, not the Pakistani): The limiting space is the space between the turbine blades of the last turbine stage. If you look carefully at a picture of a turbine stage, you will see that the total area available there is pretty small, much smaller than the inlet. There is a fatwa relating this mass flow rate to the pressure and temperature ahead of the turbine. It basically says that as as pressure goes up, max flow rate goes up (same as with musharraf) but as temperature goes up, max flow rate goes down as square root of temperature. This is not terribly important to the above discussion (these things depend on other factors such as how much fuel is put in, how much work is taken out in the turbine etc). Point is that this is what limits maximum mass flow through the engine. You can of course beat this somewhat by sending a lot of the flow through the fan duct, but then you need work from the turbine to run the fan, so this has its limits as well.

As the old fatwa goes:

U don't need to be a brain to be the CEO. U just have to be a musharraf.

Disclaimer: I don't try to understand deep stuff such as NASA Glenn K-12.

Note: I c that discussion on LCA is still on why engines don't work.

shiv · Post by **shiv** » 15 Jul 2010 18:18

AoA enqyoobuddin. May flatulent camels be upon your enemies.