PPRuNe Forums - View Single Post - Air Canada 767 Problems at Madrid
View Single Post
Old 4th Feb 2020, 02:37
  #40 (permalink)  
fdr
 
Join Date: Jun 2001
Location: 3rd Rock, #29B
Posts: 2,956
Received 861 Likes on 257 Posts
Originally Posted by Dave Therhino

The regulation that determines whether a jettison system is required is 14 CFR 25.1001(a). It has nothing to do with the relationship between max takeoff weight and max landing weight. The requirement for a jettison system is instead based on climb capability at a weight equal to max takeoff weight minus the weight of fuel necessary for a 15 minute flight consisting of a takeoff and return to land. If the climb gradient requirements of sections 25.119 (all engine climb in the landing configuration flaps down gear down) and 25.121(d) (engine out climb in the approach configuration flaps down gear up) cannot be met at this weight, then a jettison system is required by 25.1001(a).

The flow performance of the jettison system, if required by 25.1001(a), is required by 25.1001(b) to be able to get the airplane within 15 additional minutes to a weight that allows the airplane to meet the climb gradient requirements of 25.119 and 25.121(d).

The braking system regulation, section 25.735, sets the energy absorption capability requirements for the brakes. Landing at max landing weight at anticipated speeds must be withstood repeatedly as normal operation with no damage other than normal wear. Landing at max takeoff weight must be withstood without hazard, but parts can be destroyed or require inspection/maintenance. The structural regulations (25.473) set the landing loads that must be met by setting descent rates that must be accommodated as limit loads at max takeoff weight (6 feet per second descent rate at touchdown) and at max landing weight (10 feet per second descent rate at touchdown). A design can be limited by the braking and structural regulations to a maximum landing weight that is significantly less than the maximum takeoff weight, and whether or not a jettison system is required would have nothing to do with this. It's all design choice - how strong do you want to build your gear and brakes, and what climb performance do you want to provide.

Of course, performance information (climb gradient and runway distance) for landings in excess of the maximum landing weight up to the max takeoff weight is required to be provided in the AFM by 25.1587(b)(3).
DTR, quite so.

Many oddities on a takeoff will result in the sort of decision that this crew had to make, and it is definitely in the realms of decisions made under uncertainty.

The F18 off the wing is notable, assume that an airborne inspection was requested by the Captain, or accepted if offered by ATC. That may have provided the crew with the information that would otherwise be missing to them after the entertainment of a clunk-bang event on the takeoff above V1. A tyre failure can result in an engine failure, sometimes, not all that prevalent. the first failure may not be determined to be a tyre failure, the crew may only known they have an engine failure with hints of severe damage, until someone looks out a cabin window and sees excess tinsel and stuff (maybe). The potential for one engine to damage another is relatively low in a pylon mounted engine case from one wing to another. engines on the same side have been damaged by a single failure on a number of occasions. Across the aircraft a HPT/LPT letting loose has caused damage to the other engine on occasion, with a ricochet off the ramp.

So, chance that both engines are damaged is not high, it is also not zero, but it is not high. An immediate landing is possible, but is compromised in this case by the tyre failure, depends on the availability of Edwards or similar maybe, as your stopping is dependent on variables outside of certification. A single flat tyre, not a problem, but if it takes out the other 3 on landing, the things are awkward, 1/2 TRs, 1/2 braking, control authority etc... becoming a bit interesting, and a lot of unknowns, stuff that the OEM pays TPs to play with, without pax.

When would a multi engine failure be an elevated risk?

birdstrikes, with large flocks. The larger fans deal with birdstrikes better than small donks, there is a smaller relative area of intake air to the core and that is protected by larger roots of the blades, birds tend to be centrifuged, still happens though.

An unknown cause to pilots:
  • contaminated fuel. bad fuel can result in losing any or all engines, but isn't a high prevalence, and contamination appears to affect engines more at lower power than higher power.
  • Fuel icing. BA038 sort of deal. Happens later in flight after cooling of a system, and needed the power to be low first up to cause buildup.
  • Airframe icing: SAS MD80 style. not much of a choice left to make. failure is going to be prompt and more or less symmetric for an aft engined aircraft with undetected ice on the wings that sheds into the intakes.
Once the crew know the failure is a tyre, then the urgency to land overweight is reduced. Even on a small jet, such as a Gee whizz, or a Lear, a tyre/wheel/brake disk failure may well cause damage to the flaps, and possibly to the same side engine, but is not likely to cause bilateral damage, unless... a disk happens to spit its dummy at the same time, which is unlikely. A fan blade or two going cross country is possible, but the disk has the energy to hurt and is not likely to part ways. It it does, your decisions are going to come up rather quickly and be made for the pilot.

The OEM on the topic had a single comment of note, that the Autoland function is not certified for the overweight landing. The logic that they operate under would not change due to being a bit heavier, it is a relatively simple geometric progression, and that doesn't care about the start conditions so much. However the suggestion was to manually land. aiming for lower than 360FPM around the touchdown which would rattle teeth anyway. (ground effect is dependent on CL, so a reduced flap setting is going to result in a lower ground effect in the flare, so the flare needs to ensure a reducing in sink rate from pitch change in the flare process. As we always do.... or try to anyway)


fdr is offline