Bandeirante crash at Leeds/Bradford

[Alty Meter]

I heard today that KnightAir lost their court case against Embraer and Goodrich, and had to pick up a massive bill for everyone's legal costs. Pilot error blamed.
Crew and all on board were killed when an artificial horizon failed in the climb.

Anyone able to confirm or correct this story?

[t'aint natural]

Is it unfair to ask Flying Lawyer, of this parish, who I'm led to believe acted for one of the parties?

[HugMonster]

Embraer, Goodrich and CSE were not found to be at fault. The judge was clearly reluctant to blame the pilots, but (as I read it) found there was sufficient doubt about their performance to prevent the claim from succeeding.

[The Nr Fairy]

nom :

The AAIB accident report can be found here (http://www.aaib.detr.gov.uk/formal/goeaa/goeaa.htm).

[widgeon]

Did not realize that a std by horizon with power pack was not a requirement on all IFR certified transport aircraft.BTW where is the STDBY horizon located on EFIS equipped large jets . I have a recollection ( maybe false) of an incident in South America where after a dual EFIS failure the std by horizon was in some out of the way position.

[Dagger Dirk]

Widgeon
see http://www.iasa.com.au/IFALPACommittee.html

and
http://www.iasa.com.au/NEXR.html#siting

[Ignition Override]

Gang: Pardon me for switching this comment to US C-130 transports (with four powerful Allison turboprops), but years ago, the C-130 A through H models had NO standby horizons, and if you lost all generators, you might not make it. There was also no battery back-up source for the AC or the copilots' horizons.

My father lost all electrics twice, in night IMC (instrument weather), during stable climb-out or cruise conditions, with no prior electrical faults. The first time in an A model, and later in an E. Thank God they had highly-experienced Flt. Engineers who could reset generators etc quickly, with a flashlight focused on the Aircraft Commander's ADI (horizon). Those cheap penny-pinching SOBs back then in our Pentagon could not "justify" the expense for many years. Another such Air Force Reserve C-130 (Colorado Springs) crew supposedly recovered control from a spiral in IMC over Hawaii only after they broke through an undercast-very nice job, John and crew.

Even civilian Lockheed Electras were "grandfathered in", per our wonderful FAA (safety first...well...), with no back-up horizon needed, which could have had a battery source. Zantop lost such an Electra and the crew when a mistake was made during an electrical fault over Utah in IMC. There must have been, maybe there still are many passenger/cargo planes in the USA which have no such back-up power sources.

[ 05 November 2001: Message edited by: Ignition Override ]

[Flying Lawyer]

The case was heard in the High Court in July. It lasted three weeks, there was a great deal of technical and expert evidence called by all parties. Judgment was given in October.

I should declare an interest - I represented BF Goodrich.

Knight Air claimed that the Bandeirante crashed because of the failure of one or both the artificial horizons shortly after departure.
The aircraft entered a spiral dive from 3000' only four minutes after takeoff.
KnightAir claimed that the crash was caused by the failure of the artificial horizons - as a result of the negligence of the three defendants.

It was clear on the evidence that at least one horizon suffered a 'soft' failure. ie It gave inaccurate readings, but did not fail completely - no flag warning.

KnightAir claimed that both must have suffered 'soft' failures although conceding that, statistically, the chances of both failing in a four minute period was extremely remote.

All the evidence pointed to the Captain being the handling pilot - the FO operated the radio throughout.

KnightAir sued (1) Embraer, (2) BF Goodrich Avionics Systems which now owns the company which manufactured the instruments, and (3)CSE who maintained the instruments.

(The case against CSE was abandoned during the first days of the trial when the Judge refused Knight Air's application to change the entire basis of the case against CSE at the last moment.)

In very brief summary, KnightAir's case was:
(1) against Embraer
that the AH's installed by Embraer were unsuitable for use in commuter aircraft;
that Embraer became aware that they were unsafe and failed to warn operators.
(2) against BF Goodrich
that the CMM was inadequate in that it failed to specify precisely what work should be carried out when the AH's were repaired/serviced;
that Goodrich knew their recommended maintenance procedures resulted in horizons being unreliable and unsafe, but did nothing to rectify the problem.

Both companies denied negligence and further alleged that, even in the extremely unlikely event of both horizons failing, the crew should have been able to fly the aircraft safely on the remaining instruments all of which were functioning normally.

KnightAir's case depended heavily on the exceptionally low MTBF noted by the AAIB which, if accurate, would arguably have meant the horizons were unsafe. However, we discovered an error in the data used by the AAIB. The true MTBF was significantly higher. KnightAir's own expert agreed that, on the basis of the true figure, the horizons could not be said to be unsafe.

Although KnightAir alleged that the CMM was inadequate, examination of other manufacturers' CMMs revealed that it was consistent with the industry standard for this basic type of horizon.

Further, KnightAir seemed to have overlooked the significance of artificial horizons being 'on condition' items and, for that reason, are likely to fail in service. The CAA in common with all (or virtually all) aviation authorities in the world treat artificial horizons as 'on condition' instruments. Although it is accepted that an AH failure will significnatly increase workload, it is assumed that a pilot will be able to fly safely on the remaining instruments.

The Judge concluded that KnightAir were unable to substantiate even one of the many allegations of negligence made against either Embraer or Goodrich.
Both claims were dismissed.
KnightAir was ordered to pay the legal costs of all three defendant companies.


In the circumstances, it was not strictly necessary for the Judge to decide whether the crew's handling of the situation contributed to or caused the crash.
However, for good legal reasons which I needn't go into, it is normal practice to do so.
The question in law is whether their actions fell below the standard to be expected of a competent flight crew.
The Judge said that, if it had been necessary for him to decide the point, he was sadly driven to the conclusion that they fell below that standard on this occasion. He noted that both the CAA and the AAIB, together with all expert pilot witnesses except one, told the court that professional pilots with I/Rs should be able to fly safely by partial panel on the remaining remaining instruments. (The one exception agreed if it was a single failure, but not a double failure.)

One expert witness called by KnightAir (a retired CAA I/R examiner at Leeds/Bradford) said it was legitimate practice to continue to use a failed horizon making allowances for the failoure, and to 'interpolate' from it.
His view was not shared by any of the other expert witnesses (on either side) who considered his idea to be bad practice and potentially very dangerous; although it is difficult to do so, a horizon known to be giving inaccurate readings should be ignored completely.


NB:
The case did NOT affect the financial position of the families of the crew and pasengers whose claims had been already been settled.
KnightAir blamed the demise of the company on the crash, and hoped to recover damages from the defendants. Their insurers hoped to recover from the defendants the sums already paid out.

Sadly, what it did mean was that it then became necessary to examine in detail the crew's actions - which would have been better not reopened. That's just a personal view as a pilot.


HugMonsterThe judge was clearly reluctant to blame the pilots, but (as I read it) found there was sufficient doubt about their performance to prevent the claim from succeeding.
Your understanding is not correct.

[MrNosy]

Hi Flying Lawyer.

Is there a summary of this case including the judge's summing up etc published on the web?

[Flying Lawyer]

MrNosy

Not on any free (non-subscription) site as far as I am aware.

I have a copy of the judgment, and would be happy to email it to you if you let me have your email addresss.

FL

[MrNosy]

Flying Lawyer

Thank you for your kind offer to send me a copy of the Knight Air judgement - much appreciated. However, I find that we subscribe to LawTel (you learn something everyday!)and am now printing out all 87 pages (I hope).

We will update our summary of this accident to include the most pertinent bits from the judgement.

[GotTheTshirt]

Widgeon,
I also thought that the self powered Standby A/H was a requirement. certainly on the PT aircraft I have been involved with have them.
One interesting point is that when aircraf come from the US register to the UK register I have had to change the A/H ( usually SFENA or AI manufacture) to meet the UK requirements
The reason for the change ?? the FAA ones have the sky/ground colours ( or colors! :))
Blue and black but CAA require Blue and brown !!
:confused: :confused:

[ 08 November 2001: Message edited by: GotTheTshirt ]

[GotTheTshirt]

BIK,
Commuter category can go to 19,000 Lbs under FAR 23
:) ;)

[Huck]

In many jets, the standby a/h does not have it's own battery, but is wired to the ship's emergency power system, allowing it to be powered by the ship's battery (or the ADG) in an emergency.

I've flown sims on the standby instruments, and I looked like a monkey screwing a greased football!

[Shaggy Sheep Driver]

If that's the accident I think it is, they crashed because they lost an AI in severe turbulence. Didn't this result in the requirement to have a standby AI as well as the two main ones? Prior to that, the recommendation had been to fly limited panel to see which of the two AIs was telling porkies - not easy inside a cu nim.