OEI and TERPS

400A

New Member
This may cause a crap storm. I am curious, how have those of you that fly 135 (121 probably doesn't apply because of airport analysis) are taught at your school houses (SimCom, Simuflight, etc) that you must comply with published SID/ODP gradients OEI?
 
So you depart out of somewhere like Tucson.

The weather is IMC. You need to be able to meet the required climb gradient depending on the ATC requirement (SID) or TERPS (OPD). Generally it is taken off of Second Segment climb, which will be calculated to the top of the DP. And besides, SIDS and OPDs are only really critical single engine. If you cannot meet the req. of the SID or OPD, then don't accept. If there is no other option then you cannot depart IMC.

I guess I am not understanding the nature of the question.

Also, I know of some 135 Ops that use APG to determine their course of action should en engine fail during their departure.
 
I'm glad you answered the way you did. (This isn't a neener neener, I gotcha either, I used to think like you did, until I was educated).

First thing, a TERPS SID or ODP IS NOT for an engine failure (check out AIM chapter 5-8-2, use the online version to get the latest verstion)

Second, you cannot calculate V2 at the to of the ODP, the performance data doesn't work like that (even though it is a moot point).


I really feel this is something we need to aggresively get changed in all training programs (especially at the 142 houses, and even in primary training).

SIDs and ODPs are normal all engines operating procedures, using them as OEI procedures could very well lead you into colliding with an obstacle when you think you have the gradient to clear them, specifically on the fourth segment or final segment climb.

You are correct about APG, those types of services are the only realy way to satisfy part 135.379(d)(2) (assuming you are using a relatively modern turbine powered part 25 aircraft).

The FAA has let this fall through the cracks. Part 25 certification only requires the aircraft meet certain OEI performance parameters, thus only those performance numbers are published. The approach climb is the only all engines performance numbers required to be published (landing config go around, all engines operating climb requirement of 3.2%)

The FAA has the stance that there is no requirement that an operator has to show they can meet SID/ODP climb gradients all engines, i.e. there is no chart that gives you your normal climb performance. The assumption is that all part 25 aircraft can meet the most aggresive climb gradients currently published all engines operating. This is in the process of changing since the new RNAV type SID require climbing turns and some fairly aggressive altitude restrictions for both traffic and obstacles. Keep in mind SIDs and to some extent ODPs are designed for ATC workload and traffic management first and obstacle avoidance second, third, fourth, etc.

Check out AC120-91 and read AIM chapter 5-8-2 carefully. Please don't think I am trying to be a smarty guy either, I believe most of us have been educated wrong and have been sent down a path, that at first glance passes a logic test, that is incorrect. While I know of no CFIT accident that resulted from an engine failure at V1 and a crew following a SID or ODP with the failed engine, it doesn't mean one couldn't happen. If we can come together as a pilot group and change the way we are taught and start using APG type services (75 dollars a month per tail, BTW), it makes all much safer.

-Rob
 
I'm not quite sure I'm following what you're asking / saying either.

I just came back from recurrent on the Lear 40/45, and did recurrent in the Citation Bravo 6 months ago. Both were done at Simuflight/Bombardier. It was made pretty clear to me that there are places that we cannot accept the SID/ODP in low weather conditions because the aircraft cannot meet the climb requirements single engine. Even the Lear, which is a pretty darned good performer, is limited in some cases. We cover the same thing in our in-house ground school every year (we're primarily 135 ops). There are numerous airports that we will not depart IMC. If we can't visually pick our way out of the valley or around the hills if we lose an engine, we don't go.

We're trying to convince the company to spend some money on APG procedures for some airports, but to be honest the few number of times that we don't do flights in a year's time makes it hard to justify the cost.
 
You, and me, are being taught wrong. Please, go online and read the AIM paragraph and take a look at the AC. Take a step back for a moment and open to the idea that you may have been taught incorrectly.

APG costs 75 dollars a month per tail, hardly cost prohibitive.
 
You, and me, are being taught wrong. Please, go online and read the AIM paragraph and take a look at the AC.

Can't right now - heading out for a long weekend. If you'd take a few minutes to summarize what you are trying to say I'd appreciate it though.


Take a step back for a moment and open to the idea that you may have been taught incorrectly.

I'm always open to that thought. It's happened before and it'll happen again. :D


APG costs 75 dollars a month per tail, hardly cost prohibitive.

$75 x 13 x 12 = $11,700 per year. We might have a couple of flights per year that are affected, and typically it means using a different airport not cancelling and losing revenue. That's the tack that management takes on it, but I think we might be gaining ground in convincing them otherwise. Right now though, with flights down 20% or more, is not a good time to try to pry additional cash free.
 
Can't right now - heading out for a long weekend. If you'd take a few minutes to summarize what you are trying to say I'd appreciate it though.




I'm always open to that thought. It's happened before and it'll happen again. :D




$75 x 13 x 12 = $11,700 per year. We might have a couple of flights per year that are affected, and typically it means using a different airport not cancelling and losing revenue. That's the tack that management takes on it, but I think we might be gaining ground in convincing them otherwise. Right now though, with flights down 20% or more, is not a good time to try to pry additional cash free.

My dyslexia is showing 5-2-8, my bad

From the AIM

2. ODPs and SIDs assume normal aircraft performance, and that all engines are operating. Development of contingency procedures, required to cover the case of an engine failure or other emergency in flight that may occur after liftoff, is the responsibility of the operator. (More detailed information on this subject is available in Advisory Circular AC 120-91, Airport Obstacle Analysis, and in the “Departure Procedures” section of chapter 2 in the Instrument Procedures Handbook, FAA-H-8261-1.)

From AC 120-91

7. TERPS CRITERIA VERSUS ONE-ENGINE-INOPERATIVE REQUIREMENTS.

a. Standard Instrument Departures (SID) or Departure Procedures (DP) based on TERPS or ICAO Procedures for Air Navigation Services—Aircraft Operations (PANS-OPS) are based on normal (all engines operating) operations. Thus, one-engine-inoperative obstacle clearance requirements and the all-engines-operating TERPS requirements are independent, and one-engine-inoperative procedures do not need to meet TERPS requirements. Further, compliance with TERPS all-engines-operating climb gradient requirements does not necessarily assure that one-engine-inoperative obstacle clearance requirements are met. TERPS typically use specified all-engines-operating climb gradients to an altitude, rather than certificated one-engine-inoperative airplane performance. TERPS typically assume a climb gradient of 200 feet per nautical mile (NM) unless a greater gradient is specified. For the purposes of analyzing performance on procedures developed under TERPS or PANS-OPS, it is understood that any gradient requirement, specified or unspecified, will be treated as a plane which must not be penetrated from above until reaching the stated height, rather than as a gradient which must be exceeded at all points in the path. Operators must comply with 14 CFR requirements for the development of takeoff performance data and procedures. There are differences between TERPS and one-engine-inoperative criteria, including the lateral and vertical obstacle clearance requirements. An engine failure during takeoff is a non-normal condition, and therefore takes precedence over noise abatement, air traffic, SIDs, DPs, and other normal operating considerations.


 
I've re-read this a few times now and I'm still not sure I get what you're getting at here? Is the argument that when you're OEI you don't fly the SID/ODP? Ok, but weren't you planning performance so that you'd be able to make the 200 ft/nm climb single engine? I know I'm mising something somewhere because I'm just not getting your point.
 
I've re-read this a few times now and I'm still not sure I get what you're getting at here? Is the argument that when you're OEI you don't fly the SID/ODP? Ok, but weren't you planning performance so that you'd be able to make the 200 ft/nm climb single engine? I know I'm mising something somewhere because I'm just not getting your point.


In a roundabout way, yes. Read the AIM excerpt and read the AC excerpt.

Keep in mind, I am not saying this, the FAA is, more importantly the TERPS guys at the FAA are.

Now, there are plenty of guys at the FAA who have been taught wrong and now implement it as policy and the correct way to teach. There really isn't anything to look up, it is right there in black and white from TWO FAA sources.

Keep in mind, this is a legality and worse case scenario type discussion as well. The chances of hitting something because you follow a SID or ODP after an engine failure, I would argue, are small. That doesn't change the fact that performance OEI is being taught incorrectly. The problem is at first glance it looks correct, but dig a little deeper and see what your single engine performance charts are really giving you and how the climb profile is different all engines vs OEI. Think about what happens to your OEI performance when you turn, what charts do you have in your AFM that give you the performance in a turn (I bet there isn't any, in fact I bet all of your OEI charts state the aircraft is flown straight ahead). Now, the BeechJet has charts for single engine climbs at V2 for 10 minutes and even lables them SID performance charts, BUT there is no account for turning and the performance is pretty weak.

Take the step back, be open to the idea that this is being taught incorrectly across the board, especially at the 142 school houses. Dig out your AFMs (part 25 airplanes) and read what the conditions are that validate the data you get from the charts and ask yourself, what happens when you change those conditions.
 
Well, the second segment net climb gives the the % SINGLE ENGINE that we will receive under certain conditions, such as flap setting, temperature, altitutude, winds, etc.

In the AFM we really don't care about two engines operating as per climb requirements.

If you look at an XL AFM, it says that you subtract something like 1.2% if you make a turns during your second segment climb, single engine.



If you accept a SID, and it requires something like a 480ft/nm climb to 9900, and that comes out to a climb gradient of 6.7% then to accept that STAR and fly it IMC then you need to be able to meet 6.7% second segement climb as per the AFM, and obviously some planes cannot. Therefore, your options either include NOT accepting the star or alternative means reducing load, fuel, waiting for different conditions or using an alternate ODP or STAR.

I still do not see what this argument is about.
 
What if you lose comms IMC, experience and engine failure during climbout and you are departing with a STAR?

You better be able to meet the climb req. single engine to ENSURE that you can follow that STAR provided there are no alternative means.
 
You're still missing it. Two seperate sources say that SIDs and ODPs ARE NOT for OEI procedures. 2nd segment only goes to a given height, some airplane its 400 feet, others it 1500 feet, that all depends on what is in the AFM. You cannot, I say again, cannot, look at the top altitude of a SID/ODP and calculate your V2 and 2nd segment climb data and say you will just fly V2 to the top of the SID/ODP. If it were that simple, your AFM would have procedures for doing that.

Like I said, the BeechJet has got a new chart (relatively new) that allows for a 10 minute climb at V2 (10 minutes is the limit for take off power OEI). That chart still doesn't give any real performance, it isn't FAA approved and the foreword still limits the user to restrictions on 3rd and final segment climb data.

You are making the assumption that if your 2nd segment climb gradient is equal to or greater than the SID/ODP requirement you will miss the obstacles and that is incorrect.
 
From the AIM
2. ODPs and SIDs assume normal aircraft performance, and that all engines are operating. Development of contingency procedures, required to cover the case of an engine failure or other emergency in flight that may occur after liftoff, is the responsibility of the operator.

From AC 120-91

7. TERPS CRITERIA VERSUS ONE-ENGINE-INOPERATIVE REQUIREMENTS.

a. Standard Instrument Departures (SID) or Departure Procedures (DP) based on TERPS or ICAO Procedures for Air Navigation Services—Aircraft Operations (PANS-OPS) are based on normal (all engines operating) operations. Thus, one-engine-inoperative obstacle clearance requirements and the all-engines-operating TERPS requirements are independent, and one-engine-inoperative procedures do not need to meet TERPS requirements. Further, compliance with TERPS all-engines-operating climb gradient requirements does not necessarily assure that one-engine-inoperative obstacle clearance requirements are met. TERPS typically use specified all-engines-operating climb gradients to an altitude, rather than certificated one-engine-inoperative airplane performance. TERPS typically assume a climb gradient of 200 feet per nautical mile (NM) unless a greater gradient is specified. For the purposes of analyzing performance on procedures developed under TERPS or PANS-OPS, it is understood that any gradient requirement, specified or unspecified, will be treated as a plane which must not be penetrated from above until reaching the stated height, rather than as a gradient which must be exceeded at all points in the path. Operators must comply with 14 CFR requirements for the development of takeoff performance data and procedures. There are differences between TERPS and one-engine-inoperative criteria, including the lateral and vertical obstacle clearance requirements. An engine failure during takeoff is a non-normal condition, and therefore takes precedence over noise abatement, air traffic, SIDs, DPs, and other normal operating considerations.

I don't fly 135, and had just a skosh, at best, about 11 years or so ago.

I've read, and re-read your posts and the above several times.

We used to use a Part 142 school at my previous airline, and flew a model of airplane certified under a SFAR since it needed a 5 segment climb to be legal. (Yeah, I know, right.....:rolleyes:)

Here's what I get out of all of the information you've posted, adding to it other junk accumulated in the grey matter over the years.

1) ODPs, as published, are required 2 engines.

2) If you blow a motor, the operator is required to come up with an alternate procedure.

Now, every one I know plans to reject every takeoff. Then when you hit V1, you plan to lose the most critical engine.

Why do they make ODPs and SIDS (besides the side effect of organizing outbound traffic to the departure gate)? To either a) Not hit stuff b) not waking up NIMBYs.

If it's number (b), and you lose a motor, tough cans alice. Wake up and deal. You are in an emergency situation and need to do what you can do to have a safe outcome.

For (a), as far as I can tell, you have 2 choices. 1) To ensure the airplane can, one engine down, make the performance. If the airplane can make the performance, the operator has a valid procedure - making the required ODP performance.

2) The airplane can't make the ODP/SID gradient OEI, then the operator needs a tailored OEI departure procedure.

Those are the only two options I can think of.

The other thing that people love to forget in these academic arguments is that all takeoff performance ends at 1500 AGL unless otherwise noted, as the data gets you to traffic pattern alt and a box pattern to land is the practical application.

Sure you can take a look at your OEI ceiling for the weight of the plane at take off and make a guess that you'll be able to climb up there. However, you have no data on how long it will take, so there is no terrain clearance.

All airports need some sort of area. There are some airports, out west especially, that are in bowls or are one way in and one way out. In that case, you'd better make sure your plane can make the turn so you don't hit the terra granite. In that case, it'd be worth getting a tailored solution for that airport, IMHO.
 
Polar

You are the closest one to getting it so far. 135.379 is the reg that is controlling here (there is a sister 121 reg, I just can't recall what it is off the top of my head). All .379 says is you must clear the obstacle(s) by 35 or miss them by 200 feet laterally inside the airport boundary or 300 feet outside.

Thats it. The catch here is, how do you show you can do that? As you pointed out, most, if not all, OEI performance data ends at 1500 feet. By the geometry of OEI profiles, you have to level or drastically reduce your climb rate to clean up and accelerate. The point after acceleration, the final segment, is where you get into trouble. The TERPS criteria doesn't assume a level off, it is a constant plane that cannot be penetrated from above, not a plane where a snapshot of peformance is met, as is with V2 climb gradients.

I am not saying it is legal to ignore SID/ODP climb requirements, I am saying that using OEI performance numbers from the AFM to say you meet those requirments is incorrect and that, by 135.379 an operator must have a program to ensure the required obstacle clearances are met. Notice the reg does not allow for the operator to use see and avoid as an option.

The point to of all of this is simple. As a 135 operator you are to have a plan to not hit stuff on a normal take off AND when you lose a motor at V1 and continue the take off (com loss and such are all secondary at this point). The SID and ODP data does not give you that clearance, even if your computed data from the AFM says you can climb at the required gradient. That last point is the confusing one. If you stop and think about it for a second, go and read through part 25 about certification and what the airplane is being certified to do and couple that with what the AIM says and what the AC says, it becomes clear.

We are all taught that a SID and ODP keeps you clear of obstacles, but we aren't taught the geometry and such of where you need to be to be clear. It isn't just a bowl around the airport, there is specific flight paths that must be flown bank angles that are expected, etc. As soon as you deviate from that geometry you are outside what is protected and following the standard OEI profile does that.
 
EX.

In the Citation 650 AFM we have second segment charts that go up to something like 13000 MSL, most likely higher. If I use the charts and come up with a zero headwind, 17000#, and some temperature, it might give me something like 5.2 slope which then would meet a given example only requiring 4.3 slope.

That chart is saying if I fly at V2 with the flaps at approach and OEI with the gear retracted I will meet that climb requirement to that altitude with the given conditions. Most likely I will exceed that requirement down low, but up top, I am guarenteed to get at least that climb gradient.

Are you saying I will hit the mountain/obstacle when I have met ATC climb requirements?

It doesnt matter how LONG it takes to get there, as long as the climb gradient of the airplane is greater or equal to the climb required.

The FAA will come out and say stuff like "You just can't break the plane from above, the plane being the required 4.3 degree slope"

Can you accept a STAR and not meet the requirements? I guess so, but you better have an alternative out. Be it VFR or an ODP.

BTW I am not saying SIDS and STARS are designed for OEI. They are just stating a required climb either by ATC or TERPS. I am saying that if you can't make the ODP or STAR with OEI, then you most likely shouldn't accept them.

Should we get into why some ILS (AKA Burbank) have DH of something like 1200' AGL. Its because of required climb gradients, usually the standard for IFR of 3.3. And most airplanes dont have a problem meeting them two engine. It is the result of a failed engine that is usually a limiting factor.

I really want to see what you are getting at because I think we are coming at it from different angles.
 
I think you're chasing a wild turkey.

TERPS merely set a floor.

The airplane operations side of stuff - including OEI - sets out a procedure and data on how not to get into the floor.

If I remember the numbers correctly, the floor rises at 1.25%. Second segment is supposed to be 2.5%, thus quickly diverging from the 1.25% plane. This will allow your 3rd segment to occur prior to catching the floor.

Again, where there is a high obstacle or gradient that the plane may not otherwise meet, you need a specific runway analysis. I've had that for years in 121. Some runways even have the 3rd segment raised for whatever reason (that's for the nerds in the nerdery to figure out).

There are generic clearance charts built. They take the worst turning OEI performing AC and the best, and use those radii as the bounds for which they set lateral terrain clearance. I forget the vendor's name. However, if you really want to carry the weight, you need aircraft specific performance.

The 142 can only teach what is in the curriculum, and really, only the operator's procedures.

There is no restriction that says the operator can only use manufacturer AFM data to the best of my knowledge.

Don't forget that the TERPS, Part 25 and the applicable operations FARs were not developed in vacuums. The aircraft was certified to meet the criteria designed in the applicable FARs.
 
I find this entire discussion incredible. At my airline, and I would assume all the others, it is well established in our FOM and training that ODP's and SIDS are predicated on ALL engines running and that there is no guarantee to make the gradient if you don't have that.

You need to establish what you want to do if you lose power or performance for some other reason. It falls under your emergency authority. Not being able to make the SID or ODP gradient with an engine shut down does not mean you can't legally fly the departure at all, who came up with that (VERY FALSE) notion? That is so far out of what the TERPS designers had in mind (same for PANS-OPS) it is incredible.

As for some FAA inspector not knowing this, that would be no surprise. AFS doesn't talk to the TERPS branch and neither talk to ATC to a large extent, so the different branches of FAA tend to operate in isolation.
 
You need to establish what you want to do if you lose power or performance for some other reason. It falls under your emergency authority. Not being able to make the SID or ODP gradient with an engine shut down does not mean you can't legally fly the departure at all, who came up with that (VERY FALSE) notion? That is so far out of what the TERPS designers had in mind (same for PANS-OPS) it is incredible.

I understand your point, but my airline, and I'd say yours as well, has engineering data on a terrain critical departure for an OEI situation.

Airport analysis ends up with a specific runway turn procedure.

What the OP is talking about is that they operate under the premise of not having airport analysis and use charts or tables to figure their OEI performance.

All these guys may have data for is takeoff performance, but no runway specific data (although it's all the same math, they don't have an analysis of the specific terrain clearances).

And for the last part, I take it then, that all FARs are written in a vacuum and there isn't a "system" approach?
 
Back
Top