Disinterested Third Party
Certification standards do nothing more than set a minimum level of performance, in this case with an engine out in a worse-case scenario during takeoff. The earlier in the takeoff the engine fails, the longer the takeoff, the longer the takeoff and climb distance, and as a result, the lower the overall climb gradient shall be for the first segment (and depending on the time of failure or phase of the takeoff, and departure), subsequent segments where a failure may occur).
APG is a Aircraft Performance Group, which provides runway analysis for a given runway on a given day under given conditions. A runway analysis takes into account not just temperature, runway length, and weight, but also obstacles in the departure path, for a specific aircraft performance model. A runway analysis isn't about wringing extra performance or bang-for-the-buck out of an airplane. It's about safety.
It's a good idea to know before you take off that while at JFK on a given day you could use diverse climb criteria for the departure (200'/nm) for the climb, today at XXX airport you need something else. You view the chart and determine what the required climb gradient is, and then determine if you can make that climb gradient at this temperature and weight. In a departure procedure, there may be multiple points in the departure that require a certain climb; can you make them all? Can you make them with an engine-out? That's where the runway analysis comes in.
The runway analysis is telling you that you need to do something other than standard in the event of an engine failure or power loss, and here's what you do. You can view it as a gimmick the company is using to enable them to legally operate at that field, or you can view it as the lifesaving piece of equipment and tool that it is, and incorporate it into your briefing and fly it as briefed, to save your life.
Yes, on a given day with no obstacles, you could fly the minimum certification profile for departure, and there's nothing to hit. Today, not so. Today, as you brief for takeoff, with an engine failure, there's no configuration change until 1750', at which time you will turn left heading 180, level the aircraft, clean up to Vf, and then continue your climb to 10,000. Etc. Today your terrain and obstacles require it. Today isn't a departure without obstacles. Today, if you want to live you'll take advantage of the runway analysis which has been prepared for your departure, and fly it.
Note that even diverse departure criteria for an area departure from a given runway assumes a climb to 400' before turning on course and climbing at a rate of 200'/nm enroute. If a runway as diverse climb criteria, you can do that; takeoff, turn any direction, and climb at the minimum gradient. If it doesn't, then you'll need a departure procedure, obstacle departure, or runway analysis to allow you to know what climb is required. Note that you've got to be able to do this with all engines, or with an engine-out; even if you lose an engine, the obstacles are still there, and must still be cleared. Unless your company has been issued a special engine-out procedure (the proverbial "pink pages" from Jepp), then you will make a determination if you can do it using the runway analysis provided. The runway analysis says "yes, you can do it, but you'll need to do it this way." If you choose not to do it that way, then no, you can't do it.
The alternative, if you don't want to take advantage of a detailed obstacle analysis for your departure, is to get out a sectional and chart and plotter and do all the math in determining exactly what distances to what obstacles in the departure you can make with an engine out, and what climb gradients you'll have at what point. I've seen many departure analysis which tell me I'm fine until obstacle #15 in the departure procedure that I'm requesting, and then I'm in trouble. To do the calculations to that point, assuming I had a way of getting that data at all, would have been arduous and taken a very long time. I had the data quickly, in the cockpit, before engine start, thanks to an obstacle analysis program. Because that program is approved and has all the data necessary to safely fly out, I need to adhere to it. Likewise, if the program tells me that I can't fly the procedure, even though I think I can...I can't. Legally, it's my butt.
A few years ago I was called to the company office to defend an airman, as his union representative. He made a departure in Afghanistan, which contains some very tall rocks. He had made a departure from the same runway the night previous, under nearly identical conditions, with a nearly identical load. When all of his trip data was placed in an envelope at the end of the trip and sealed up for the company, he figured it was done. As it turned out, he was not.
The trip data is occasionally reviewed, and spot checked. It's kept for a certain period of time, and the company would pick packets at random (or not, depending on one's sinister views) for inspection. In this case, they opened his trip packet, and reviewed his takeoff and landing data. It turned out that he couldn't accept the runway assignment. It was a last minute assignment while taxiing, departing opposite direction, and though the crew had already completed all their planning and briefing for the former runway, they felt confident based on the previous night's flying that they'd be fine. As it turned out, there was one obstacle in the departure procedure that they accepted which made it a no-go. All-engines, no problem. The obstacle analysis determined that with an engine-out, however, based on the departure they selected, there would be a problem. It was a technical issue that had to do with the input into the program; had they selected it as a VFR departure rather than an IFR departure in the obstacle analysis, the program would have let them go. They'd run it all previously the night before, and off they went. When the company ran the numbers from the comfort of a stateside office, they caught the error, and we all ended up in a review board, and I ended up in negotiations with the Chief Pilot in order for the captain to keep his job.
He did, but it was close.
Take the obstacle analysis seriously. Run the numbers yourself on some of the departures to see if they fly, and run them using a different profile to see if they don't.
Today, we will continue the climb to 1250' before cleaning up, as indicated in our departure obstacle analysis. Liberal, Kansas, we woiuldn't need to do that. Newark, not a problem. Here, today, tonight, this runway, this weight, this temperature, this departure, we do. We'll leave takeoff flaps set until 1250', then clean up, because we have a professional obstacle analysis that tells us that if we dont', and we lose an engine, we will lack the climb gradient performance to meet the obstacle clearance requirements for the selected departure procedure.
A number of years ago, I flew for a fractional operator. There was a small hill at one end of a runway, and there was a numbers-oriented individual in the company who took the time to calculate that with an engine-out on the runway at V1, we'd strike the hill XX feet below the top. We couldn't climb over the hill. The solution was simple, really. Go around the hill. Left, right, didn't matter. Go around the hill. Visually, it could be done. Not on instruments, however. A low visibility day, or low ceiling, or both, one couldn't maneuver around the hill, and shouldn't. The obstacle analysis for the hill was different under visual conditions, and under instrument conditions.
You'll note that some departure procedures specify a minimum visibility on departure, for the same reason.
Some departures have a lot of potential obstacles. Aspen, as you've noted, is one of them. A glance at the SARDD3 obstacle departure identifies a requirement of 460'/nm to 14,000,' along with departure minimums of 400' and 1 mile. (Why the ceiling and visiblity requirement? The procedure requires greater than standard/diverse climb criteria, and so places the ceiling in 100' increments above the controlling obstacle on the initial departure, and requires a minimum visibility of a mile)
You could climb to 400' and initiate your clean-up, but you'd still have to maintain 460'/nm in order to do it, while doing it. You can't put the climb gradient on hold while you do it. Obstacles and terrain don't give you a pass or tell you to take your time while they wait. They're still there, and the laws of physics still apply. If you can clean up at 400' and still make the grade, have a ball, but you have a very useful tool in your obstacle analysis which is telling you to change your cleanup altitude for that departure from that runway on that day in that aircraft, to a different number. Ignore it at your peril.
Standard operating procedure means doing something the same way under standard conditions. If you have a runway analysis or obstacle analysis which is telling you that the departure requires something else, then you don't ignore it simply to follow "SOP." Operating conditions are no longer standard. Now you need to do something else.
Going back to the SARDD3, the takeoff obstacle notes include:
Rwy 33: Multiple trees beginning 35' from DER, 386' right of centerline, up to 100' AGL/7722' MSL. Multiple trees, bushes and terrain beginning 4' from DER, 400' left of centerline, up to 100' AGL/7821' MSL. Multiple trees, bushes, and terrain beginning 2484' from DER, 752' left of centerline, up to 100' AGL/8179' MSL.
Do you want to whip out the aircraft flight manual and calculate your departure based on those obstacles, individually, overlay them into the departure procedure, compare them to the obstacle departure as well as your other procedures, make a comparative analysis against your aircraft performance on all engines and with an engine-out, compare that to the standard procedure of cleaning up at 400', and then make a decision? Or would you rather have that all done for you and presented as an obstacle analysis with the data already tied off with a pretty pink ribbon? I'd rather have the full analysis, and use that valuable tool to my advantage in planning my departure, do that I know I can make the grade in the event of an engine failure at any point.
I understand what you're saying. You're conflicted between what seems to be an inflexible and possibly unwise company standard procedure of cleaning up at 400,' on each takeoff, and the question of keeping yourself configured to a higher altitude in the event an obstacle analysis says otherwise. You don't have a training department, but use an outside vendor such as Simuflite or FSI. You're conflicted between being told to clean up at 400,' and following the guidance from the obstacle analysis which is telling you that you need to climb higher before cleaning up. I get it.
First, the obstacle analysis is a safety tool, and it's also got some legal teeth. Ignore it at your peril. If anything happens and you're asked why you had the data and ignored it, you'll be spinning in the wind, music stopped, no available chairs.
Second FSI or simuflite don't have your company's analysis data. That's an analysis service paid for by your company, so there's going to be a certain disconnect between the boilerplate standard that the training vendor will provide, and what your company is doing, unless the company chooses to incorporate their data into the vendor's program.
Third, when you use an obstacle analysis and you're given something that's not standard, then standard operating procedure (eg, clean up at 400') no longer applies. You have non standard conditions; adapt accordingly.
Fourth, design certification criteria is not a hard 400' altitude. 400' is the MINIMUM altitude for flap retraction insofar as the climb gradient schedule is assigned. It can be higher, and that's not something that the manufacturer could know at the time of certification. The question for the manufacturer is strictly one of aircraft performance to meet a MINIMUM generic certification standard. What's actually required in the field, when rocks and hills and trees and a spinning Julie Andrews in a blue gingham skirt intervene, will determine what's actually required of the flight crew in order to stay alive. That's where your obstacle analysis comes in, and that analysis tells you how far from "SOP" you must deviate in order to remain safe and alive.
If SOP says clean up at 400' and considers an engine failure at V1, the obstacle analysis differs by considering an engine failure at any point from V1 thereafter until you've cleared the obstacles, and tells you that you had better keep the climb going to XXX altitude, if you want to live. Remember Arnold the Terminator, "Come with me if you want to live?" That's the obstacle analysis. Ignore it at your risk.
On a given day, we might normally clean up at 400' or a thousand feet, or whatever the company procedure might be, but today we will be cleaning up at a different altitude, thanks to the benefit of the obstacle analysis which not only tells us that we can safely depart, but how to do it. Thanks Arnold, next time Julie.