Estimating dead-stick glide performance.

[Blip]

I've been thinking about this for a while and just want to confirm my suspicions.

Flying B737-400.
Flaps Up.
210 knots.
Rate of descent = 1000 fpm.
Flight-Idle thrust fuel flow = 350 kg/h per engine.

I'm trying to estimate how much the idle thrust is contributing to the glide performance and how much the glide path angle will increase with two dead engines.

The other day while in this configuration (the first one mentioned above!) I pushed the thrust leavers forward until the fuel flow doubled to 700 kg/h per engine and noted that the rate of descent reduced from 1000 fpm to 500 fpm.

I'm just wondering. Does that mean that the idle thrust of 350 kg/h is contributing 500 fpm in this configuration and speed?

If I lose both engines, can I expect the rate of descent to be 1000 + 500 = 1500 fpm while on approach at 210 knots? Doesn't that mean then that if the rate of descent increases by 50%, that I have to increase my altitude gates by 50%.

So at the start of final approach where I would normally be at 3000 ft, I need to be at 4500 ft. Or another way of looking at it, where as I normally allow around 3 nm per 1000 ft, I now have to allow around 2 nm per 1000 ft.

Does that sound about right?

Also the fact that at the higher levels, idle thrust fuel flow is only about 150 kg per hour or less, does that mean that loss of thrust on both engines does not have such an affect on glide performance at higher levels than at lower levels where the idle thrust fuel flows are around twice this?

By the way, I've done some dead stick landings in the simulator without the APU supplying electrical power for the electric hydraulic pumps. With only the windmilling engines supplying hydraulic power, you do NOT want to get any slower than about 190 kt with any more than Flap 5 for two reasons. Firstly, any slower and the engines are no longer turning over fast enough to run the engine driven pumps and you start getting that manual reversion feeling through the control wheel. Secondly, with any more flaps, the rate of descent becomes excessive and you will be unable to arrest the rate of descent in the flare. And if you try to arrest the rate of descent by flaring a little higher than normal, all you do is bleed off the airspeed and encounter the stick shaker and stall it while the high rate of descent remains.

Anyway can someone please confirm or deny my estimations of the idle thrust contribution to glide performance?

Thanks.

[BOAC]

No idea, but I'm surprised your airline has not passed on the Boeing guidance for 2x engine out 737 - roughly double the 'lower-down' height 'gates' -ie 5 miles 3000' etc. Time for a whinge at your training dept, I think In the UK it used to come round as a sim exercise in the 3 year cycle. I even believe there are topics about that here?

[Blip]

Don't mean to answer my own question because I am no way certain of the physics involved however...

If rate of climb (or descent) is determined by power produced by engines, and the power is proportional to thrust (and therefore fuel flow) for a given airspeed, then it seems to me that at 210 knots, a fuel flow of 350 kg/h does indeed equate to 500 fpm.

In my example I increased the fuel flow to 700 kg/h per engine and the rate of descent reduced to 500 fpm. If I were to add another 350 kg/h to make the total fuel flow 1050 kg/h per engine the rate of descent should reduce by another 500 fpm to, well, zero.

Level flight at 210 knots with fuel flow at 1050 kg/h per engine sounds just about right to me.

BOAC I'm suggesting the performance reduces by 50%. You say Boeing says it reduces by 100%. Interesting. And yes I mean no, there is no such information in our training manuals.

[HarryMann]

-ie 5 miles 3000' etc.

thats almost exactly a glide angle of 10:1

Maybe just work from a known L/Dmax (15:1 ?) and degrade it for overspeed

[Tee Emm]

I'm trying to estimate how much the idle thrust is contributing to the glide performance and how much the glide path angle will increase with two dead engines

In a simulator operation we assume the aircraft has arrived (5 miles from airfield) early downwind at 9000 ft after gliding (both start levers cut-off for the exercise) at 210 knots. Then execute normal flap extension to Flap 5 and glide at 170 knots. Turn base at 3-4000 ft. Flap 10 next. When certain of reaching field, gear down manually or hydraulically, and now ensure you are 10-15 knots faster than min maneouvre for flap setting to allow for flare. Dive the aircraft if required to land a respectable distance in and plant it firmly. The rationale behind selecting Flap 5 and 170 knots once you are in the circuit, is to reduce radius of turn required for 210 knots clean and makes judgement of base and final turn easier. .
There are certainly several other flap/speed/ combinations but to simplify for judgement and training practice it was decided to settle for one procedure. Caution: This is purely a personal opinion.

[747dieseldude]

Not based on any solid info, but from a couple of jets I flew, it's around 1 degree steeper than the idle power descent.
If your descent with idle power is 3 degrees (~300'/nm) then glide would be 4 degrees (~400'/nm)

Again, this is only a theory.

[vodmor]

Engine thrust isn`t strictly proportional to fuel flow, especially at low power levels since you`re using some power to sustain engine work and to power external units (electricity, air condition etc). Another thing is when the engine is working, even at idle level, it produces some thrust, while when the engine is not working, it produces only drag force, actually quite a lot of it. I think this matter is quite complicated, and definetly the thrust/fuel flow relation isn`t linear. To solve your problem you`d need more data regarding engine aerodynamics

[Mad (Flt) Scientist]

Other issues that come into play are the nature of the engine failure - is the fan and/or the core windmilling, or have they seized? (Are they even still there - if the fan rotor disk failed there may not be much left of the front of the nacelle at all.) All those possibilities can drastically affect the drag produced by the engine(s) and make a nonsense of any engine-out performance you may have.

Also, if you start losing multiple hydraulic systems, you may have to contend with things like spoiler panels upfloating if they are no longer being positively held down - and thus potentially yet more drag.

[Pugilistic Animus]

oh Mad(Flt)Scientist always spoiling the party with your correct real life facts:ok:

PA

[misd-agin]

18:1. Altitude x 3 = gliding distance. 10 thousand feet = 30 miles.

Boeing lists 767-200 at 17.9:1
737-800 manual says - approx. 30 miles at 10,000'. (18:1)
757W, with engines at idle, is about 19.5:1 (personal observation)

[Jimmy Hoffa Rocks]

A-320 once you put the gear down the glide ratio changes a lot.
Anyone know by how much ?

from 2 to 1 to



Anyone done dual engine no relight in the sim lately

zero wind
60tonnes

[galaxy flyer]

Hate to be agreeable, but MfS has it right--go with what you have when the poop hits the impeller, way too many variables.

[Blip]

misd-agin. It looks to me you are quoting normal glide figures, not all-engine-out figures which is what this thread is about. I might be wrong though. Can you confirm either way please?

galaxy flyer and others, there is no need to complicate (and subsequently not answer) the question with a million "but what if" scenarios that would adversely affect the drag qualities of what's left of the aeroplane. A simple engine flame-out will do thanks.

So I asked a simple question. What contribution to the glide performance do engines operating at flight idle have? What happens to the glide performance when the energy equivalent to 200-400 kg/h per engine is taken out of the equation?

I've expressed my ideas based on a hunch really, and wanted some confirmation that my reasoning was at least sound, if not strictly scientific.

Thanks for the replies so far.

[misd-agin]

Blip -

The 767-200 figure is from the Boeing manual. It did not state if it was with no engines or engines at idle. I tend to think a glide ratio would be with no engines running.

The 737-800 was with 'windmilling engine drag'.

757-200W was inflight observation(checked on several flights, 19.5 was lowest observed)

[MainDude]

Did double engine fail in sim, and double the alt gates was a good figure to follow when in clean config, (the 5NM x 3 + 1500'... double it = 3000') when on the ILS, then stay one dot above the GS in gear dn, flaps 15. Take more flaps when it's clear you're overshooting/accelerating.

Windmilling turbines, vs damaged engine standing still was hardly noticeable- perhaps it was poorly modeled in the sim software - don't know.

[Mad (Flt) Scientist]

Windmilling turbines, vs damaged engine standing still was hardly noticeable- perhaps it was poorly modeled in the sim software - don't know.

My money would be on poor modelling. Depends on the qualification level and age of the device, but I don't think there's likely much in the way of validation data for a damaged/stopped engine. Depending on the nature of the damage (core or fan makes a big difference) it should be very noticeable... many thousands of feet in ceiling capability is possible.

edit to add its not the sim manufacturer's fault; in many cases the data just doesnt exist to accurately model these events.

[john_tullamarine]

The other way to give the crews exposure .. and that's the aim with this sort of simex ... is to look at the simple double failure at one end of the spectrum .. and then, say, the double failure but with the gear down at altitude to provide for a high drag situation which might resemble an engine disintegration scenario .. one might even wind in some mistrim on the rudder to introduce a yawing problem .. whatever.

If the pilot can deadstick these in the sim, then his/her ability to read the signs in whatever the real world failure might present are reasonably well enhanced. Surely it's not too much different, philosophically, to a PFL in a lightie .. start with some notional keys for aiming at .. and then massage the descent to suit the picture ?

I think back to an exercise with a bunch of ex-MIL FJ Chinese pilots .. didn't matter WHAT I threw at them .. they all got in just fine, albeit with a modicum of variation in technique amongst the group ... and that was on the first attempt for each of them. I was quite impressed.

[galaxy flyer]

Certainly, US Airways and Air Transat have given Airbus some modeling data on the A320 and A330 that might be useful. :p

Seriously, the important thing is to quickly, but accurately, size up the performance available, which won't be a "by the book" number, and using the energy available create a trajectory to a survivable landing.

GF

[johns7022]

I am probably wrong, but I thought that losing 1000 ft for every NM forward was considered a kinda of minimum for certification when flying at best glide airspeed.

[john_tullamarine]

but I thought that losing 1000 ft for every NM forward

On the classic that's about as steep as you can readily get (FI, gear down, flap 25) for descent onto final ... makes a circuit a doddle .. over the upwind threshold at 10,000ft and a circuit to end up nicely in the slot and spun up as you roll out onto final ... regular practice into Cairns from Darwin in bumpy conditions.