737 rudder hardover

[Gabriel]

Hold on... The FARs require... From max overswing (assuming you got there from an engine failure with a neutral rudder) you can still apply the maximum amount of rudder available to correct for that yaw and remove it as needed to cancel out the yaw effect of the thrust asymmetry.

You are wrong. The FARs don't have provsion for anti-yaw / anti-sideslip rudder input, and it doesn't allow a single rudder reversal. Not at overswing, not at equilibrium sideslip, not at... well: NOT. (and the little point is a period)

Let me go again:
1- You start from zero sideslip
2- Suddenly apply rudder to the stop (but don't need to exceed 300 lb of force on the pedal)
3- Keep full rudder (stop or 300 lb) until plane achieves max overswing.
4- Keep full rudder until the plane stabilizes at the equilibrium sideslip. (note that from point 1 to 4 you have held constant full rudder deflection in the same direction).
5- From the equilibrium sideslip, suddenly neutralize the rudder.

Done. Test finished. Passed. Congrats. Let's go test something else now.

Not only anti-sideslip rudder deflection is required at overswing, it is not even required from equilibrium sideslip.
In fact, it is not required that the rudder/fin must withstand ANY corrective rudder input. If the plane yaws to the left for whatever reason (even pilot inadvertently stepping a bot on the left rudder pedal) and you apply any little bit of right rudder to correct, you are doing something that the certification doesn't require.

Anti-yaw / anti-sideslip means that you apply rudder in a way that the forces generated by the rudder deflection would cause a yaw moment contrary to the direction of the yaw / sideslip. So it is a corrective input. Which is a very reasonable rudder input if you ask me. And one the FAR doesn't require that the fin withstands.

[Gabriel]

Now there is still the phenomena of aircraft-pilot coupling that needs to be addressed. Should it be done by pilot training or by making the airframes more robust?

I would start for designing a control system that is less susceptible to PIO.
(Hint: that just a couple of pounds over the friction-break force and just a couple of inches is all it takes for full pedal deflection, goes in the wrong direction)

[Evan]

Do I have to post the Airbus/Boeing pedal travel/pedal force tables again? You're going to indict a lot of Boeing aircraft into this "danger" zone...

But the elephant-in-the-cockpit question is:

If the plane yaws to the left for whatever reason (even pilot inadvertently stepping a bot on the left rudder pedal) and you apply any little bit of right rudder to correct,

...why can't pilots learn rudder? You don't apply a little left rudder pedal (not that that would overstress the airframe) , you remove the rudder pedal you indvertently stepped on. It's not a bicycle, it's a rudder.

[Gabriel]

Do I have to post the Airbus/Boeing pedal travel/pedal force tables again? You're going to indict a lot of Boeing aircraft into this "danger" zone...

Whomever. I don't care. I am not in an A vs B war business.

...why can't pilots learn rudder? You don't apply a little left rudder pedal (not that that would overstress the airframe) , you remove the rudder pedal you inadvertently stepped on. It's not a bicycle, it's a rudder.

I said for whatever reason for God's sake!!
Inadvertent rudder input
Engine failure
Lateral gust
Lateral windshare
Natural turbulence
Wake turbulence

Do you really expect the pilot NOT to use the ruder to correct unwanted sideslip? (and no, I don't consider "step off the rudder" as a form of "use the rudder").

[3WE]

I said for whatever reason for God's sake!!
Inadvertent rudder input
Engine failure
Lateral gust
Lateral windshare
Natural turbulence
Wake turbulence

Do you really expect the pilot NOT to use the ruder to correct unwanted sideslip? (and no, I don't consider "step off the rudder" as a form of "use the rudder").

Yes, he really expects the pilots to not use the rudder pedals.

Evan is extremely binary... never touch the pedals in flight (perhaps we need an automatic pedal lock out)... maybe it's ok for crosswind landings, but we need to incorporate a built in system to pause the rudder at neutral and training to only make steady-state inputs.

[Evan]

Do you really expect the pilot NOT to use the ruder to correct unwanted sideslip? (and no, I don't consider "step off the rudder" as a form of "use the rudder").

Of course I expect them to, carefully, in a single direction, if the disturbance is not quickly corrected by the aircraft's natural directional stability. I don't want pilots getting into PIO overcontrol situations that might exceed structural design loads however. I want pilots to be fully educated on what the rudder is there for and how it should be used. If they learned it wrong bck in their aggie days, I want them to relearn it before becoming CTPL holders.

Alternating full inputs on the control wheel and rudder pedals, such as those made by the first officer, should not be necessary to control a transport-category airplane under any circumstance.

There is a procedure that advises the use of deliberate alternating sideslips to force gear downlocks in an unsafe gear situation. Even that procedure requires the aircrft to be stabilized between sideslips:

The sideslip should be initiated using the rudder on the same side of the aircraft as the unsafe gear indication, i.e., if the right main landing gear is unlocked, slowly apply right rudder up to full deflection if necessary while maintaining wings level to generate sideslip. If the gear still fails to lock, then slowly return the rudder to neutral, allow the airplane to stabilize, and then slowly apply opposite rudder.

The structural issue again is specific to rapid rudder reversals, for which there is no reason or need.

From a structural capability standpoint, the pilot does not have to be concerned about how fast or how hard to push the rudder pedal in one direction (from zero to full available pedal deflection) throughout the normal flight envelope. However, it is important to emphasize that limiters do not protect against the structural loads or excessive sideslip angles that can be generated from rapid full deflection flight control reversals.

However I do agree with you that CFR Part 25 should be modified to include design requirements that provide protection against this level of pilot error and aircraft-pilot-coupling (APC). So does the NTSB:

Rudder control systems with a variable ratio rudder travel limiter may provide better protection against high loads from sustained rudder pedal inputs at high airspeeds than systems with a variable stop rudder travel limiter because variable ratio rudder travel limiter systems require more physical effort from a pilot (in terms of force and displacement) to produce cyclic full rudder inputs. For airplanes with variable stop rudder travel limiter systems, protection from dangerous structural loads resulting from sustained alternating large rudder pedal inputs can be achieved by reducing the sensitivity of the rudder control system (for example, by increasing the pedal forces), which would make it harder for pilots to quickly perform alternating full rudder inputs.

The Safety Board concludes that certification standards are needed to ensure that future airplane designs minimize the potential for APC susceptibility and to better protect against high loads in the event of large rudder inputs. Accordingly, the Safety Board believes that the FAA should modify 14 CFR Part 25 to include a certification standard that will ensure safe handling qualities in the yaw axis throughout the flight envelope, including limits for rudder pedal sensitivity.

It's just that, as your mantra goes, airmanship should come first.

[Gabriel]

Evan, I am not arguing against the fact that the way the rudder was used in that accident was way off. In fact, it all started with a bad understanding of upset recovery principles, so yes, the simple fact that he used rudder was wrong to begin with, not to mention the repeated use of full rudder in opposite directions. I agree with you and the NTSB regarding the need of a better prescription of yaw handling qualities in the FAR.

But in my last posts I am not arguing that. Please answer without mentioning repeated alternate rudder reversals. Because that is NOT what I am talking about.

Do you agree that there is a big gap in the FAR by not prescribing any structural requirement whatsoever related with any a single rudder input in the direction opposite to any sideslip?

Again, your left engine fails and your plane yaws strongly to the left and you are a little bit sloppy and let the sideslip increase a bit too much. What do you do now? Idle the right engine and let the natural directional stability handle it? Because if you apply right rudder at that point, the rudder or fin can legally fail, since there is no requirement in the FAR that it should withstand such an obvious and rational rudder input.

[Evan]

Do you agree that there is a big gap in the FAR by not prescribing any structural requirement whatsoever related with any a single rudder input in the direction opposite to any sideslip?

Yes, I think I just said that, although I don't know how big of a gap that really represents. When you take in both the design requirements and the certification tests, I don't see a structure that cannot withstand the corrective inputs you describe. (the A300 certification tests showed the attachments failed around 2x the load limits, well beyond certification requirements.)

Nor, despite all such corrective inputs performed over the years, is there evidence of this:

The FAA airframe engineer stated that , since the 1953 implementation of Civil Aeronautics Regulation 4B (the predecessor to the FAR that described the maneuvering conditions for the design of the vertical stabilizer), no historical evidence would lead the FAA to believe that the design loads envelope for the vertical stabilizer was inadequate.

But on November 12th, 2001, fin separation occurred. So where does the problem lay:

The airframe engineer stated that a maneuver with alternating rudder inputs was an extreme maneuver and that, if the maneuver were performed, loads would build that would exceed the current requirements. He further stated that, if two sets of alternating rudder inputs were performed, a series of dynamic maneuvers would start that could be benign or “could lead the airplane into a severe dynamic situation where, at the proper frequency, this continued application of this surface would allow the motion of the airplane to build up to the point where the sideslip would become excessive and overload the airplane.”

That is, a repeated, cyclic abuse of rudder, amplified by aircraft-pilot-coupling due in part to phase lag, resulting in a PIO building to the point of overload... caused the fin to separate. Nothing short of that caused the failure.

Conclusion: a repeated, cyclic abuse of rudder can cause the fin to separate. There is a structural danger there. There is no evidence to conclude that a similar structural danger exists in the conditions you describe.

But I agree out of an abundance of caution that it should be added to the certification testing requirements. The NTSB does as well. (and if this is added tomorrow, it will affect the aircraft introduced into service around 2030. In the meantime, it is far more important that pilots are made to understand the non-aerobatic nature of rudder on transport aircraft before they are certified to fly them.) Agreed?

[Evan]

There are four known notable rudder reversal events in commercial aviation history, two of which occurred as a result of a wake encounter, and the commercial airplane fleet has more than a combined 500 million flight hours. Assuming an equal probability across all airplanes in the commercial fleet, the probability of a rudder reversal that reaches or exceeds design limit loading is approximately 10-8 per airplane flight hour. For this reason, it was determined that use of an ultimate load condition (factor of safety = 1.00) was deemed appropriate.

However that is being reconsidered. Gabriel, read this:

https://www.faa.gov/regulations_policies/rulemaking/committees/documents/media/TAEfch-rpsrr-3282011.pdf

Read especially: Attachment B – Proposed New Regulation 25.353 Section 25.353 Rudder control reversal conditions (Version 1 – Single Doublet Condition). I think it is wht you are proposing.

Regarding the potential danger this represents:

The work of Table 1 was repeated, but this time with the pedal reversal coincident with the peak overswing sideslip. Results are presented in Table 2 and show the sensitivity to timing of the rudder pedal reversal input. It should be noted that for most aircraft types there was no more than 2% difference.

The results from Tables 1 and 2 show that the ultimate load single pedal doublet condition defined herein does not generate higher airframe loading than the current FAR design ultimate load level.

[Gabriel]

However that is being reconsidered. Gabriel, read this:

https://www.faa.gov/regulations_policies/rulemaking/committees/documents/media/TAEfch-rpsrr-3282011.pdf

Read especially: Attachment B – Proposed New Regulation 25.353 Section 25.353 Rudder control reversal conditions (Version 1 – Single Doublet Condition). I think it is what you are proposing.

Regarding the potential danger this represents:

Yes, that condition is what I am proposing, and what Boeing has been proactively using with a safety factor of 1.2 (ultimate load = 1.2 * limit load).

Now, this work of the FAA is a piece of... ok, I will skip the adjectives.
It is designed to yield the desired result: that the current FARs are ok and that no change is needed.

How do they do that?
Well, first, disregarding any other abnormal condition simultaneously with the rudder doublet.
- Weight, CG and flight envelope conditions are not most critical, but "typical" (meaning that about 50% of the time it will be worse than that).
- Aircraft with FBW was considered to be in normal law, and yaw damper was considered on if that was the expected condition as per the AFM. Didn't it ever cross their mind that a deviation from these typical conditions may cause a change in the airplane handling characteristics that can be a factor in how the pilot reacts and uses the rudder in the first place?
- Now, the cherry of the cake...

The results from Tables 1 and 2 show that the ultimate load single pedal doublet condition defined herein does not generate higher airframe loading than the current FAR design ultimate load level.

Combined with this:
For this reason, it was determined that use of an ultimate load condition (factor of safety = 1.00) was deemed appropriate.

Do you get it? They are comparing the ultimate load that results from the current certification standards that includes a safety factor of 1.5 over the actual loads produced by the maneuver in the certification standard (design load) with the ultimate that results of the new proposed maneuver that includes a safety factor of 1.0 (i.e, no factor of safety) over (or below, or neither over nor below) the design loads produced by the new proposed maneuver.

Let me be more clear: The design load is 66% of the ultimate load. Whenever in the table below you see a number bigger than 0.66, the design load was exceeded:

Aircraft Type Doublet/Ultimate
Small/Medium Business Jet 0.40 up to 1.0 (approx)
Medium/Large Business Jet 0.34 up to 0.75
Fuselage-Mounted Engine Regional Jet 0.42 up to 0.79
Wing-Mounted Engine Regional Jet up to 0.80
Single Aisle Passenger Jet 0.61 up to 0.73
Widebody Passenger Jet 0.50 up to 0.81

Do you see a number greater than 0.66 for any row? Because if you do, then know that the design loads were exceeded.

By the way: The AA A300 was subject to loads 2 times the limit load. That is 33% above the ultimate load.

And then we have this:

Both airplanes survived these events due to having strength in excess of that required by the current standards

This means that by the current standards these 2 fins were legal to fail and all on board (plus maybe some on the ground) legal to die.

[Evan]

Now, this work of the FAA is a piece of... ok, I will skip the adjectives.
It is designed to yield the desired result: that the current FARs are ok and that no change is needed.

I don't think you're being entirely fair. Did you read through that document? I think they are trying to find a balance between structural requirement and practicality.

So you think the proposed ultimate loads (safety factor of 1.0) should be the limit loads (safety factor of 1.5). The practical argument is as follows:

These conditions are anticipated to occur very rarely, and so these are considered ultimate load conditions, and no additional safety factor is applied.

Because:

For some types of aircraft, overly severe criteria, including multiple full-stroke doublets, would lead to structural strengthening, a weight penalty, and/or system changes that could be detrimental to normal operations.

Especially efficiency, which is both a business concern and an environmental one.

Whereas:

Enhanced training (as recommended by FCHWG) is the single most effective countermeasure to inappropriate rudder control reversals.

The most effective (and the only effective measure in the short-term) also comes without a penalty for efficiency. Although I would replace the word 'enhanced' with the word 'fundamental'.

Meanwhile, aircraft certified under this new requirement are all within the ultimate load envelope in these very rare instances you described (and most likely with some additional safety factor provided by the airframer).

Gabriel, I would LOVE to see pilot-proof planes that can withstand anything thrown at them, but I do recognize the point where providing for extreme situations becomes detrimental to progress.

I think the industry sees this as well.

Teach pilots how to use rudder.