Annexe B Thoughts on seeing the Yak-38 and Yak-141 aircraft at Farnborough 1992 The most obvious difference between the Yak and the Hawker way of achieving VSTOL is the number of engines. Yak use three on both the 38 and the 141. Two vertically mounted lift engines are just behind the cockpit with one horizontally mounted thrust vectoring engine further aft. On the other hand Hawker chose a single engine with four interconnected swivelling exhausts to vector the total installed thrust for both horizontal and vertical flight.
The Yaks need all three engines to work in the hover and the failure of any one will cause a violent and immediate change in pitch attitude since all the engines exhaust at some distance from the aircraft centre of gravity. The Harrier engine is virtually at the CG so, should it fail in the hover, the aircraft falls but with no great attitude change. With over 800 Harriers delivered, this has happened only once since 1960. (That USMC aircraft hit flat bending the wings and gear. The pilot did not eject but was unhurt).
Yak deal with VSTOL failures by automatically ejecting the pilot should the aircraft attitude exceed certain limits. Vladimir Yakimov, Chief Test Pilot for Yakovlev quoted these for the Yak-38 as: 15° nose up, 10° nose down and 20° in roll. These values are increased to 25° nose up, 20° nose down and 30° in roll in the Yak-141. Yakovlev say the seat has a 100% success record with 21 auto and 13 pilot initiated ejections.
If we assume all engines have the same reliability, the Yak concept is more than three times as likely to suffer an engine problem than the Harrier. This is because a three engine installation is inherently more complex than a single one and such installation aspects can contribute to thrust loss causes. There is also the matter of lift engine starting system reliability at the end of every sortie, assuming that there is no runway within range that is long enough for a conventional landing.
Combat damage of powered lift and control systems also needs to be considered. A Harrier pilot only needs high rpm, a variable nozzle angle and reaction controls to carry out a safe VL. These items can be checked quickly at any speed below 500 kt by opening the throttle fully (to check rpm response) then momentarily pulling the nozzle lever back (to check nozzle rotation) and, while the lever is still back, moving the stick and rudder (to check the reaction control air supply or ‘duct pressure’ behaviour is normal).
A Yak pilot suspecting damage will have to suck it and see when in the circuit. To be fair, this is no worse than the way that a conventional aircraft pilot has to wait to see if such things as his gear, flaps, weight on wheels sensors, thrust reverser, braking chute, hook, wheel brakes, tyres and nosewheel steering will do the necessary to stop his aircraft on the narrow, cross-wind, 500 metres strip that the brochure says is OK.
The Harrier concept has major problems not present with the Yaks. They stem from needing one very large engine close to the CG and include aerodynamic distortions due to exhaust flows passing close to the airframe; a short air intake that is poor for efficiency, surge margin and head on signature, one compromise engine and intake design to deal with the conflicting requirements of low and high speeds and a general lack of flexibility in overall layout.
The Yak engine configuration has problems beyond the engine failure case. For example, the total installed thrust is not available throughout the flight envelope, the lift engines take up valuable volume that could be used for more internal fuel and their weight effect is similar to carrying around a couple of large but ‘dud’ weapons.
The blast, vibration and temperature environment under the fuselage centreline during V/STOL limits high tech stores carriage to the wings (unlike the Harrier where it is possible to splay the nozzle efflux 15° either side of the vertical viewed from behind). Once a Yak has its lift engines running for takeoff, any holding point delays will warm and soften asphalt surfaces (or concrete block seals) so that they erode when the aircraft later moves forward. This actually happened at Farnborough when the Yak-38 had an unexpected 20 minute hold before lining up for its first pre-show qualifying sortie.
Despite such configuration induced problems, Yak were quoting a 141 internal fuel load of 4,400 kg for an 11,650 kg empty weight (37.7%) which compares with early Harrier Is at 5,000 lb for 12,500 lb (40%) rising to some 43% for later Harrier IIs. However, given the better optimisation of the Yak cruise engine, wingborne range and endurance are likely to be in the same ballpark for both concepts. This was certainly not the case with the earlier Yak-38 which was observed to fly sorties of only about one third the endurance of the Harrier.
Intake ingestion of foreign objects is a problem for both concepts. However, careful pilot control of thrust vector angle versus ground speed has shown that Harriers can operate from loose surfaces. The Yak engines require a non-erodable surface for routine operation. A ship’s steel deck would be ideal, especially if it was fitted with a gridded area at the deck edge.
Most of the foregoing is by way of background. What was new to me came from talking to Vladimir Yakimov and Alexander Dondukov, Yak's Chairman and General Designer, from watching videos of their operations back home and seeing both aeroplanes fly.
A cockpit brief in both aircraft showed that, so far as pilot operation is concerned, the Yak's VSTOL controls are perhaps even less demanding on the pilot than the Harrier's simple nozzle lever. Understanding how this is achieved, given the inherently complex Yak three engined concept, is to appreciate the remarkable hydro-mechanical control engineering achievement on the 38, which has none of the 141's fly by wire facilities. However, such comparison presumes familiarity with Harrier piloting techniques.
The Harrier pilot has two engine controls mounted side by side on the left console: a normal fighter type throttle outboard and alongside it the special ‘nozzle lever’. Pushing this nozzle lever fully forward points the jets aft while pulling it back progressively increases the downwards deflection of the exhausts, up to a maximum angle of 18° forward of the hover position.
This nozzle lever is sometimes used as a configuration change selector and sometimes as a flying control. Examples of the selector type of use being pulling the lever back to select the nozzles down, for example prior to doing a VTO, a hover, or a VL. The actual height control of these three manoeuvres is then done by throttle adjustment. An example of the lever being used in the control sense is when it is moved progressively forwards, from its hovering position, to control speed and flight path angle during an accelerating transition, the throttle remaining fixed throughout. Another example is the use of the nozzle lever to control the airspeed during a fixed throttle slow approach when weight precludes a VL. Such control/selector uses (and habits) must not be confused (like accidently selecting the nozzles aft in the hover!)
On the other hand, despite his three engines, the Yak-38 pilot has only a single throttle handle operating in a fore and aft slide on the left cockpit wall. On the left hand console, in a similar position to the Harrier's nozzle lever, is a lever which is used only as a selector for choosing the VSTOL or conventional mode of the flight control system. This lever never has a control function.
In VSTOL mode, use of the single throttle gives height control through thrust modulation of all engines, while stick forwards and back gives pitch attitude control by differential thrust modulation of the lift and main engines. A trimmer-like switch on the top of the stick controls the vectoring nozzle angle and the limited rearwards component of the lift engines thrust. As on the SC1, this switch is a case of push forward to go faster, pull back to go slower. It has no other function or mode so confusion is unlikely. Additionally, the display of the vectoring nozzle angle is clear and shown to the pilot as it would appear to an external observer of the side elevation of the aircraft, with the needle indicating the direction of thrust. This gauge has a line to mark 60° on the Yak-141 from which angle the rest of the transition can be automatic if desired.
Playing such mode change tricks with the control systems in order to simplify the piloting task is nothing special in the world of fly by wire, but to make it all happen in the Yak-38 is an astonishing achievement. Vladimir Yakimov suggested I would "find the control system very interesting". He went on to repeat the point. However I believed him the first time. Watching videos of the Yak-38 doing short takeoffs and rolling landings and operating vertically from a mobile hide completed my education on Yakovlev’s achievements with a configuration which is intrinsically problematical.
The mobile hide was a lorry like container, the sides and ends of which folded down to become the pad showing the benefits of folding wings again. After this briefing the flying rehearsals and demonstrations were just as I expected: routine wingborne operation of both types, including takeoff and landing but with hovers thrown in mid-sortie. All slow flight was carried out with remarkable attitude steadiness but then you don’t bow to the President at the end of your show with a Yak unless you want to eject and plant the aeroplane in his tent.
So what conclusions should we draw? It is very dangerous to oversimplify what are complex issues but, at that stage of VSTOL history, it was difficult to avoid the conclusion that the Harrier concept represented the best way to do a military sub-sonic VSTOL job.
Supersonic flight, with its requirement for high exhaust gas velocities, is a very different matter and it simply would not work to immerse the existing Harrier configuration in such velocities and temperatures. On the other hand, the Yak solution to powered lift, proven in the 38 and developed further in the 141, is suitable to drop in to the plan view of a Mig-29 or a Su-27/35. Looked at like that, putting the Yak-38 in service becomes a very sensible stepping stone towards an ultimate supersonic goal and a far cry from being a poor attempt at doing ‘a Harrier’ as suggested in the 1970s by many who should have known better.