It seems that there is a hell of a lot of work to be done in this area.

You're right autoflight, there is a helluva lot of work on a day to day, runway by runway basis to develop these OEI routes. The crazy thing is that the same thing is being replicated at every airline's offices all over the world (most come up with the same or very similar answers). There will be a helluva lot more work to convince the ICAO states to produce standard, optimised, OEI escape routes for all runways. It would add immeauserably to safety, not only in the performance area, but in the other areas of concern regarding conflicting traffic, airspace etc. Australia has made a good start in providing, in the aeronautical documents forum, obstacle-clear gradients for all runways at all licenced aerodromes, a good and well established start, but a long way to go.

I'm going to buy my pair of sandals, grow a beard, and burn my wife's bra and an ICAO flag on the steps of ICAO headquarters until I get my way. Well, I can dream can't I?

When delaing with the posibility of having to complete a missed approach after a return from a OEI departure, a number of factors need to be considdered;

1. The OEI procedure often works on the worst case scenario........engine failure on the surface, last second airbourne and making the screen height and no more. I don't think that even with a CAT3 missed approach (from the surface in the touchdown zone) would be quite as critical..........even if the weight was the same as departure i.e. no fuel dump.

2. The minima selected for the procedure I believe should (if necessary) allow for the reduced climb performance during the missed approach and if necessary, the minima increased...........this should be in the ops manual. The standard missed approach requires a minimum of 2.5% climb which is 152ft per nm.

---------

Old Smokey,

Yes I am aware that in most cases the OEI procedure is done following a detailed survey by qualified people. However, I am amazed by the statements here that some companies will try to make some quick home-made procedure for such a critical situation while at the same time never dream of constructing an approach procedure unless they had qualified procedure designers and a complete survey.

My comments regarding the chief pilot were designed to counter the argument made that the people who put the procedure in place don't know what they are doing - No matter how qualified (or not) the designer, it is the Chief Pilot who approves the inclusion of the procedure in the Ops Manual........and of course the National Authority (inspector) also get a copy.

Regards,

DFC

I don't think that I would be pendantic about missed approach flight paths being able to be overlaid on the departure OEI flight path.

Like many things in this game .. it all depends .. especially if one throws in a failure during the early missed approach.

Consider that the approach may be at, say, full flap and an approach speed appropriate to that flap deflection. If the takeoff is based on a small flap deflection, then the clean up distance OEI will be significant and the achievable gradient OEI will be not very much ... could put the aircraft in an unenviable position. This is why the more thoughtful airlines look at the situation analytically rather than using the finger in the wind technique ...

DFC,

A very good post addressing some very valid concerns. In my line of work, OEI is my major concern at the Takeoff stage, but whilst adequate and safe obstacle clearance during a OEI missed approach DOES bother me, not nearly so much as in the Takeoff case. A bit of rationale is necessary -

A 2.5% Gross gradient with 100 feet obstacle clearance is provided in PANS-OPS missed approach procedures. The worst case Takeoff OEI aircraft (2 engined) requires to achieve 2.4% Gross, slightly less than the PANS-OPS 2.5%. As I've stated elsewhere, I discarded the "Approach Climb" data for the (FAR25) aircraft that I'm responsible for, and substituted 2.5% data instead. That's one way to do it, no accounting for Gross to Nett though.

The 2 major Airlines that I've worked for both recommend that MDA be increased for OEI. Yes, Cat 3 is the worst case, but even so, the screen height for the departure would occur some considerable distance before the end of TODA, thus the overshooting aircraft has a 'head start' in altitude over the aircraft taking off on the same runway. If all of the above fails, abandon the IAL missed approach and use the Takeoff OEI procedure instead. Some may raise concerns regarding lateral displacement from the RWY centreline for this latter argument, but this should clearly NOT be so for a Cat 3 missed approach.

The above holds good if the missed approach Flap setting is the same as that used for Takeoff, if greater, a whole new box of tricks is opened - Poorer climb Gradient / Longer OEI acceleration etc. I have overcome this by mandating a lesser flap setting for OEI approach (with proper certification of course), and using the normal Takeoff Flap setting as that for missed approach.

DFC, if you're amazed that some operators are using quick "home made" OEI procedures, then add me to the list, but edit the amazed to horrified. About all that can be said in their defence is that some attempt is better than none at all.

Regarding your last remark - "No matter how qualified (or not) the designer, it is the Chief Pilot who approves the inclusion of the procedure in the Ops Manual........and of course the National Authority (inspector) also get a copy" - You wish! My experience with a VERY respectable regulatory authority is that I design policy and technique, they approve it (or request amendments), examine a few samples very thoroughly, and then allow all other procedures to be produced autonomously. Of course they check one or two at random during audits. My Chief Pilot's role in all of this is to hand the CD I've prepared with the Special Procedures and RTOWs to the Publications Office to print. I'm confident in my work, but it would be nice for my developing ulcer to have a second opinion.

John_T, sorry, I think there was a bit of simultaneous posting going on. No need to edit as I think we said much the same thing.

Regards,

Smokey

Old Smokey,Yes, but where does this 2.5% gradient with 100' end? As someone else put it, it's the gap between 1,500' and MSA which is the problem: could be 10,000'+ at some airports.

I know the tech. guys have checked a V1 cut up to MSA on takeoff, so I have no problems following the SID or OEI procedure. The same can not be said of the MAP - unless I have my facts wrong about the checking of same. I'm not a performance guru like you so am standing by to be educated...

Not quite. It's actually the minimum nett climb gradient. There is no allowance for performance degradation from gross to nett, so the aeroplane has to be able to achieve at least 2.5% all the way through the missed approach.

FullWings...It depends on how long the Intermediate Missed Approach segment has to be. If there's a lot of obstacles around then, yes, it might apply up to 10,000+, as you say. The Final Missed Approach segment only guarantees 164 feet of obstacle clearance, which isn't a lot more than 100 feet. In this country, however, if the obstacles have an elevation in excess of 5,000 feet, the MOC is multiplied by 1.5. If obstacle elevation is 10,000 or above, MOC is doubled.

I don't think this is done anywhere else in the world, but I could stand to be corrected.

FullWings,

Thank you for the compliment, but a performance guru I am not.The only gurus that I know were my mentors and teachers back at Douglas and BAe, I hope to live up to their expectations.

If you're looking for some 'guruship' in the missed approach area, read up OzExpat's post, on checking his profile he is an Instrument Procedures Designer. My area of expertise is in Takeoff performance, and the OEI procedures that go with it. Naturally, ensuring safe levels of performance in the missed approach phase are also one of my responsibilities, compliant with the PANS-OPS procedures used to design them. I have a few current posts floating around with respect to increasing the standard gradients to comply with the minimum PANS-OPS 2.5%, or greater IF REQUIRED. It does concern me that Gross data is used, as opposed to degraded Gross to Nett data used for the Takeoff case. This degradation continues to much increased vertical obstacle clearances during the acceleration phase.

One of your statements concerns me, just in case you misunderstood -

Maybe not everyone does this, I do all the way to MSA, and 'contain' the procedure to within the MSA coverage, but others may well cutoff at a point where the procedure enters a published holding pattern (safely), but well below the MSA. Numerous procedures I've seen from other designers take care of the initial problems, e.g. Right turn at 7.0 DME to Track 270 degrees.......and then what? Some 1st/2nd/3rd segments can take up to 40 to 50 miles - WELL beyond the MSA coverage, now we're concerned with En-route obstacles - Oh my God! One procedure that I saw (no names, but the airline began with "V") had the aircraft taking up a track of 010, to what, infinity? In doing analysis for my own operator from the same airport agreed that that was a good INITIAL route, but a 1620 foot hill lay exactly on the 010 track against an acceleration altitude of 1000 feet and during what would be his 3rd segment. So, when you say "I know the tech. guys have checked a V1 cut up to MSA on takeoff", I worry - DO CHECK the criteria by which your procedure designers have created the procedure. Your statement of "I have no problems following the SID or OEI procedure", OK, I mentioned possible limitations on the OEI procedure, but don't think for a moment that with OEI you can follow the SID. Where reasonable loads are achievable with OEI following the SID, I create secondary data for the SID as the OEI route. This is the exception, and more often than not, actual loads make the optimised OEI procedure the only option.

And finally to your question of "Yes, but where does this 2.5% gradient with 100' end?. Absolutely no guarantee that it will end at the MSA. Typically, it will take you to a Holding Pattern with it's associated Minimum Holding Altitude. This may well be well below the MSA. On the flight I operated yesterday, the Missed Approach Altitude was 3,000 feet in a Holding Pattern, meanwhile, the MSA was 5,700 feet. Safe for now, climb to MSA/MEA in the holding pattern if you have to divert, or commence another approach. At least you don't have to worry about obstacle clearance in the climb within the holding pattern, only to ensure that you have the aerodynamic performance available for the required further climb. This does have implications in Flight Planning to allow for additional fuel for diversion.

Ozexpat,

This was always the case in New Guinea (was Port effin Morbid the giveaway?) due to reliability of topographical information. It may well apply in other regions also, but don't forget the factor of Altimeter accuracy as vertical displacement from the QNH source increases.

Said too much already, I'm no guru.

Fly safe, and check the criteria used to create your data.

Old Smokey

I wouldn't go so far as to say "always", but it was one of the first things I implemented in PNG. That would've been back around 1986 or so. It was my response to the Pans Ops requirement to cater for "mountainous areas" in a country that is basically all mountains.

I started off by looking at what other countries were doing to meet the requirement. I checked the AIP for New Zealand, Oz, the USA, Singapore and the UK. My conclusion after all that reading was that everybody was copping out on the problem (ie taking the easiest option). I couldn't do that because it would've been much too pessimistic in PNG.

I undertook a review of TAF QNHs over a 5 year period all over the country and found that our QNHs are pretty stable in a fairly narrow range for most of the time. Yes, we get some massive differences in QNH between coastal and highlands regions, but I found that they were pretty well graduated from the low QNH in coastal regions, typically around 1010 mbs, to around 1017 mbs in the Highlands.

This work resulted in the progressive increase in MOC, as explained in my previous post, to account for progressively higher obstacles.
I account for this as well. I adopted the Australian practice of adding a value of 100 feet for QNH accuracy. This value results from the requirement for an Area QNH to be within plus or minus 3 millibars of the local QNH for all aerodromes covered by the Area QNH.

3mbs = 90 feet, which is simply rounded up to 100 feet to make the calcuation simpler.

I apply this 100 feet to all minimum altitudes, regardless of whether it is an Area Minimum Altitude, LSALT, MSA, or a limiting altitude on an approach. It is also applied to the MDA and DA for all procedures and pilots are permitted to subtract 100 feet from the MDA or DA upon receipt of the local QNH.

Of course, in common with other countries, we also require pilots to make an appropriate adjustment for PEC.

The topo charts that are available in PNG are not updated very often. In fact, all of them continue to show "Edition 1", even after an update, so I am constantly worried about that.

Thus, whenever I'm in doubt as to the accuracy of the topo data, I go out and take a look. I always do that for approach and departure procedures, right up to MSA. I will only do it for LSALT or AMA if I'm uncertain - that doesn't happen very often because, after 15 years of procedure design in PNG - and a lot of flying around the country myself - I already know most of the areas that are shown appropriately on the charts.

If engine failures at V1 or on missed approach were more common, there would be an unacceptable number of aircraft accidents due to insufficient regard for extended flight path.
After 37 years of military, airline & various odd overseas airliner contracts, my opinion is that few operators or pilots pay more than lip service to worst case engine out flight path. If the pilot can hack a standard V1 cut and a quick radar circuit to a 15 NM final with a position freeze on downwind, they think they've done it! Maybe throw in a couple of no brainer emergency turn procedures. Handling skills? No contest, must meet requirements. Thinking? Just demonstrate the required procedures. We can get a satisfactory write-up and all go home and do it all again in 6 months.
Nobody ever intentionally forced me to consider the extended flight path. Even when one new destination airfield grid MORA was around 27,000ft, there was no discussion / memo / anything.
Nobody wants to rock the boat. Anyway, we can always find a combination of circumstances that contributed to the accident. How many investigators will be critical of their own employer? Who will sacrifice their promotion by pointing the finger at their own authorities or their own major airlines?
Why won't this subject be a wake-up call for the majority of aviation authorities, operators & pilots? Many hundreds of deaths can result from failure to act.
" status quo good - me like!"

.. which is why some of us, in sim training, throw in some curve ball exercises in spare time .. this sort of thing included. I like to sleep easier after I see the student head off to wherever.

Read someone’s procedure today for HKG 07L/R, straight ahead to pointX, then right turn to heading 190°………………….. If we gave that to our crews, they would be over the Philippines before they decided to turn back :):):)

Which do you prefer when a procedure calls for “straight ahead to the VOR and hold”. Do you select ALTITUDE HOLD when reaching MFRA and accelerate in the hold, or do you maintain a residual rate of climb while holding?

Mutt.

Smokey & Oz,

Thanks for the replies.

It's interesting that you should single out my comment about being "happy to follow the SID or OEI procedure". I think I meant to say "happy to follow the OEI procedure, or the SID if there isn't one..."

In my company we have datalinked performance accomodating environmental conditions and any MEL/perf. restrictions. It generally produces a TOPL and then a small selection of assumed temperatures around the ATOW. It will also say things like: "No restriction if SID followed", "Engine failure turn right TRK 210 climb 3000 before return" or "EO accel. ht. 2400", etc. Maybe I'm putting too much trust in the printout but I was hoping that as it was written down somebody/thing might have checked it...

I have to say I generally brief as I said above, as I share the misgivings around pilot-invented OEI procedures. I do vary this sometimes, in places like Kingston, where the sea is at the end of the runway and our ET takes us back over the hills where the storms usually sit. I'd much rather go and dump fuel 10 miles offshore than hold inside a cu-nim on one engine. :ooh:

Loke’s original question:

“When can pilots legally/safely abandon a S/E procedure? I realize that pilots cannot accept vectors from a controller – at least initially – when performing a Special S/E procedure. The question is – when can a pilot accept a vector (and altitude assignment) from a controller?”

These engine out procedures are conducted on the Captain’s Emergency Authority. This is what allows us to deviate from ATC instructions and/or departure instructions. This means you do whatever is necessary to ensure safety of flight. Once that is accomplished it is the Captain’s decision to continue the procedure or return to ATC control. If the emergency is under control and you are above the MVA why not accept vectors at that point? It all comes down to judgement.

Call me stupid, but.......

If there is no required climb gradient published on my SID departure chart, then where do I hit an obstacle? Where do the design criteria for SID's differ from OEI performance requirements?

And to answer the original question:

Reading my company documents, it says that; "The final segment and the takeoff flight path is complete when all obstacles have been cleared OR the aircraft has reached a minimum of 1500 feet above the airport elevation." This also defines the end of the singe engine procedure. Now I'll never know whether this end of the procedure is due to reaching 1500ft HAA OR I've cleared all obstacles flying in a direction I usually do not fly to (asuming 1500ft HAA is still below MSA).
My experience is that knowledge of airfield and surroundings are the only factor in deciding safe course of action reaching end of single engine procedure. And yes, it would be nice if the single engine procedure would end if MSA or, holding pattern with obstacle clearance assured, is reached. But reading this thread, this not industry standard......

nice thread :ok:

This comes straight from the Ops manual from one US-based Regional-Jet operator who operates in the Western US (ie., the Rocky Mountains)

The most important thing to remember is that the performance evaluation terminates at 3000' AFE.

The system they use creates two types of departures.

All airports will have an engine failure procedure. However, they do come in two different varieties, Standard and Special. Standard Engine Failure Procedures are designed for airports, other than those special airports where climb gradient and/or obstacles are a factor.

The general instructions for a Standard Engine Failure Procedure Takeoff Path are contained in the table below. They are very simple and allow the aircraft to attain an altitude and position where radar vectors can be provided, or to reach an MEA or approach altitude.
The Standard Engine Failure procedures for each departure will be found on the “Takeoff Report” page of the release directly under the runway length.

If the engine failure occurs at 1000 AFE or more, when in IMC, Commence turn to NAVAID or heading as listed on runway analysis page using maximum
bank angle appropriate for aircraft speed.
1. If NAVAID is listed, hold on the inbound radial using a direct entry and standard holding procedures.
2. If heading is listed, fly heading until a minimum safe altitude is attained.
Radar vectors may be accepted when available.

When in VMC:
Return to land visually or complete the IMC procedure.

If below 1000AFE when the engine fails (like in the V1 cut scenario), climb to 1000AFE and complete the IMC or VMC procedure (as appropriate). The weight report gives the heading that should be flown (whether runway heading, or a left or right turn as required).

For airports that require complex OEI departures (such as at KRNO here in the US), they create simple-special, or complex special procedures that define the procedure to be flown.

The Complex-Special Procedure is issued when an engine failure procedure is too complex to fit in the Simple-Special table and/or other considerations must be taken into account. Often, all-engine procedures are also specified in order to ensure that the aircraft will remain in the obstacle protected area until reaching a safe engine failure altitude. The minimum flap retraction altitude remains at 1,000’ AFE unless otherwise specified in the procedure.
2) A Complex-Special Procedure may be issued as a Complex-Special DP that replaces the normal DP or IFR departure procedure. An additional page may also be issued to provide important preflight guidance and restrictions.

Complex-Special
Procedures are mandatory under IMC and VMC.

have another coffee...
If your charts are based on Pans Ops, the standard minimum nett climb gradient for a SID is 3.3%. I don't know about TERPs, but it is probably the same. There is no requirement by ICAO to show the SID climb gradient on the chart, if it doesn't exceed the standard gradient. I think that most countries have adopted this policy. So, to answer your first question, provided that your aircraft can sustain a nett climb gradient of at least 3.3%, you won't hit any obstacles if you follow the SID (and if it's been designed properly, of course).

The answer to your second question is a bit more complex and I think that I'd prefer to leave that one to Old Smokey, or John T. They have far more experience in the OEI scenarios than I do. However, what I can say is that, if your aircraft can still climb at 3.3% or better with one engine inop, you can happily and safely follow the SID.

There are 2 problems with this, however. Firstly, there are a lot of aircraft that can't achieve a 3.3% climb gradient with OEI. Second, the particular SID may require a steeper climb gradient and, if so, it will (or at least should) be highlighted on the chart. I've seen SIDs with gradients up to 4.7% and there may be some that demand even better performance than that. If you're dealing with one of those, then a 3.3% climb gradient is likely to ruin your whole day. :eek:

With those few comments, I'll pass the ball to the OEI experts!

OzExpat

So yes, you can call me stupid..... :uhoh:

3.3% is above the minimum required 2.4% climb gradient for the OEI scenario (2 engined aircraft).

I would assume that the final climb gradient (4th segment) of 1.2% should be enough to climb safely above MSA from the position I am following a single engine procedure. If that's the case, the answer on the orignal question would be a no brainer. And I think it is, as it's the responsability? of the airline to provide me with save climb-out procedures in case of single engine (JAR). I can't think of an example right now where this would not be the case as most emergency turns would lead you to area's of low terrain or in the direction of facilities which have a published holding procedure.

Food for the brain......

Have another coffee,

It really depends on what the airline is trying to achieve and the regulatory requirements, if that goal is to gain the highest takeoff weight, then you might have a straight out departure based upon the Type A aerodrome chart.

If you elect to ignore the straight out departure and follow the SID, then the only protection that you are offered is based on your own assessment of the aircrafts ability to meet the required climb gradient.

Old_smokey and J_T show us all how the OEI assessment should be conducted in an ideal world, however within airlines due to monetary constraints, attitudes or just a pure lack of knowledge, their ideal world doesn’t always exist.

In the FAA world, we do not assess all SIDS, we provide either a straight out departure or a specific engine failure procedure.

Mutt.

“Go-around performance requirements??”
 

ALL PERFORMANCE JUNKYS:

I just can’t stand to see this thread go to Page 2 – so I’ll start another related one. Though there is clearly some ambiguity in requirements for Take Off Performance (especially under FARs), Go-around requirements are even less specific.

I am going to start a thread entitled “Go-around performance requirements??.”

Although Takeoff Performance has a specific FAR – 121.189, which specifies requirements for 121 carriers, Go-around requirements are a bit more vague. In fact some of the guideline documents – AC 120-OBS-11 & AC 120.29A, confuse rather than clarify requirements.

See you there,

LOKE