Cypriot airliner crash - the accident and investigation

[CaptainSandL]

Just for info, an EQUIP COOLING fan OFF light does not necessarily indicate that the fan has stopped, ie an electrical problem. It merely indicates that there is no airflow downstream of the fan.

I have done dozens of (controlled) depressurisations in the 737 and on many of them the EQUIP COOLING - EXHAUST fan OFF light illuminates during this test. This is because of the reduced airflow in the depressurised condition. The flow (& OFF light) is always restored when the aircraft is re-pressurised.

The EQUIP COOLING fan OFF light does not necessarily indicate an electrical fault, but if other faults were observed a crew could be forgiven for making that assumption.

[BOAC]

"However, 10,000 feet is the limit if APU supplied both bleeds and electric."

Do we have ANYTHING to suggest both or one engine generators were u/s before take-off? Why else would the APU be supplying 'electrics'? Was it dispatched with one gen?

"if not to check that pressurization is back to normal after bleed reconfiguration"

It should have been obvious if it had been done 'correctly' as the cabin altitude would probably start descending as both packs gave normal flow. No need to level off. To level off (at 16000ft or whatever) would be essential if the altitude horn had sounded, and should have been done when it did sound, followed by a descent.

"Wouldn’t they switch it to Manual"

That would be the normal drill having tried 'ALTN'

"The conversation lasted through 28,900 feet"

There is a huge gap here between what happened at 16000 ft and 28900ft. How long at 16? When were the various problems ("air conditioning, cabin lights and equipment cooling fans") reported to Maintrol?

"A failure of one or both APU busses"

It SHOULD be impossible to power both busses with the APU in a 300 in the air UNLESS the air/ground sensor is stuck in 'Ground' mode

"if one pack fails, loss of pressurization is possible"

No, it SHOULD be possible to climb to 33000ft and maintain normal cabin pressure.

"Failure of the outflow valve to close completely"

With battery power it should still work. Surely any significant electrical failure would be evident on the FDR? Which "cabin lights" were a problem? The whole topic of this 'reconstruction' is too vague for me. WHAT did they 'reconstruct' - and why?

A no-engine bleed take-off, no reconfiguration plus APU failure?

What parameters were used to 'reconstruct' the abnormal cabin pressure at 12000ft?

What were the defects it carried at dispatch regarding electrics and pressurisation?

[Geoff Smith]

Just a shot in the dark, others more knowledgeable will know. I offer it from the depths of my memory.
On the B747 Classic (just a Boeing) if you forced the APU electrics on in flight by closing the APU electrical field switches and selecting the APU electrics ON you would place the airplane in GROUND MODE (in flight). i.e pressurisation system would drive outflow valves open, take off configuration would sound with thrust levers forward etc. Therefore there was a procedural limitation to forbid this.
For info this characteristic was provided for engineering use on the ground during checks with the airplane on jacks.

[scarebus03]

Hello all,

firstly why would both APU busses be in operation at 34000ft? If we know the aircraft had previous enviromental system problems how would we not know at this stage whether the A/C was dispatched acc. the MEL for pack,bleed or electrical problems? If there was a problem the A/C would have been altitude restricted below FL340 as required by the MEL if the APU was involved.
It has been a few years since I worked on the 'classics' but I remember (I could be wrong) the e&e fault light would also illuminate in flight if cabin diff. press. went below a certain level, 2 or 3 psi I think. But this would only be indicative of a bigger problem.
I wonder will the investigating team ever really know what happened,


Brgds

SB03

[big fraidy cat]

Although the exact connection to the APU electric supply is beyond my knowledge, I know that a group of Greek and Cypriot pilots and engineers also believe that a bleeds-off takeoff was done, with a ground-air sensor malfunction leading to the pressurization failure at a much later stage than the official explanation would admit. (By the way, I have an engineering discussion paper on this, which is circulating at the Athens airport … I’ll find it, type it up neatly, and post it later.)

Previously on Prune, both Kaptin M and Rockhound had suggested that the ground air sensor might have mistakenly been left in air mode after the pressurization tests the morning of the flight. If I misinterpreted your meaning, please correct me, Kaptin M and Rockhound. Also, please tell me if this would be picked up on the FDR.

If the sensor was left in air mode:

(1) How would that affect the T/O configuration and the pressurization settings during climbout?

(2) How would that affect the operation of the APU during a bleeds-off takeoff?

(3) When the captain spoke to Irwin and discussed the air conditioning problem, what CB was pulled at 29,000 feet? On the CB panel behind the captain’s seat, the T/O Aural Warning and T/O Configuration CBs are next to one another. Will this also be picked up on the FDR?

(4) For pulling CBs in flight, I excerpted the following from Chris Brady’s technical site. Emphasis added is my own: From the QRH CI.2.3 March 29, 2004 "Flight crew reset of a tripped circuit breaker is flight is not recommended. Unless specifically directed to do so in a non-normal checklist. However, a tripped circuit breaker may be reset once, after a short cooling period (approximately 2 minutes), if in the judgement of the Captain, the situation resulting from the circuit breaker trip has a significant adverse effect on safety. A ground reset of a tripped circuit breaker by the flight crew should only be accomplished after maintenance has determined it is safe to reset the circuit breaker. Flight crew cycling (pulling and resetting) of circuit breakers to clear non-normal conditions is not recommended."

[BOAC]

First and foremost if left in 'Air Mode' the APU would not power both busses before start-up so that should have alerted the crew. There would be several other OBVIOUS clues. It happened to me on landing a 737 a few years ago and it was immediately obvious all was not well!

[GotTheTshirt]

Not familiar with B737 but can you select "ground mode " in the air?

Most aircraft do this through the gear sensors.

[big fraidy cat]

Poljot, you've asked this question 3 times now. Why do you think his DNA was on two masks?

The bleeds-off takeoff was normal procedure for Helios’ crews and was the case in the five previous incidents where air conditioning problems were reported in less than two months, since July 2005. The Chief Investigator has already released the plane’s recent history on that (Flight International 24 Aug.). I am not suggesting that this is a direct causal link to the crash, but it’s part of it.

If I understand the technical argument of the Greek and Cypriot skeptics correctly, an apparently insidious and increasingly recurrent electrical problem in the wiring and electronic interfaces of the ground-air sensor system, occurring in the first minutes of the climb, could have damaged the electronic card which triggers or cuts the T/O Warning Horn from the Aural Warning Generator (located above the captain’s left knee).

This would have the effect of retaining the T/O Warning Signal to the Aural Warning Module instead of canceling it when the aircraft is in the air. This would explain both an early sounding of the Cabin Altitude Horn without an actual pressurization failure, as well as the readiness of the crew to pull the CB of the T/O Aural Warning circuit. This would have rendered the pressurization warning system inoperative with a later decompression at 34,000 feet.

[BOAC]

Interesting. That last para, if relevant, would mean that the signal to close the outflow valve and pressurise the a/c would not be there either, with a DCPCS system. Not being a techie I do not know whether a failure after t/off of the a/g 'air' signal would actually open the valve or leave it closed. Any techs know?

This 'failure' would, I presume, be instantly identifiable on the FDR, since several other circuit functions would change also. So, unless the crew were actually monitoring the cabin altitude, it could all have gone un-noticed. Do we know at which point the 'horn' was audible on the R/T?

[BOAC]

poljot - it is indeed an important question, and will, I hope, be addressed in the course of the investigation. What is more important right now is finding out WHY the cabin pressure was abnormal. In that answer may lie the answer to your question. There are several other questions as well that may become clearer as the investigation progresses.

With your enormous flying experience I'm sure you can see why this is the necessary course of action? May I suggest that if you have not already done so, you read the original thread from start to end as well as this one. The original thread is here.

Do you have a theory as to why control was lost? I believe all sensible inputs can help. Do you have experience of hypoxia?

[Woodman]

For general heads up - our marketing department has been approached by a film company wanting to make a TV programme about this incident and they were looking to rent a 737 for a day. Not something to be associated with so our aircraft were 'busy'.

[big fraidy cat]

From today's Cyprus Mail online:

Tsolakis returns

THE PRELIMINARY draft of the investigation report into the Helios crash will be ready next month, Akrivos Tsolakis, the chief investigator into the accident said yesterday.

Tsolakis was speaking on arrival at Larnaca airport to begin a 24-hour visit where he will conduct another round of interviews related to the August 14 crash in Greece in which 121 passengers and crew died when the plane slammed into a mountainside north of Athens.
“We are at the end of the investigation and the preliminary conclusions will be ready in March,” said Tsolakis.

“There are a number of urgent points left to clarify,” added Tsolakis who had a meeting later in the afternoon with Transport Minister Haris Thrasou. The investigation into the crash so far points to human error.

[big fraidy cat]

BOAC, as promised, here’s an excerpt from one of the engineering discussion papers that I have. I’ve had it translated from Greek, and I’ve removed any direct references to the Cypriot maintenance team.

I know that some of the technicians involved in the engineering discussion are readers of PPRune, though not necessarily posters. It would be extremely valuable to our discussions if they, or others who have raised similar issues directly with Capt. Tsolakis and his team, would post to this thread, rather than simply read it.

Also worth noting is that the last news article that I posted states that the Preliminary draft report will be released before the end of March, but what was not said is whether it will be released to the general public at that time. I have a sinking feeling that, unless some pressure is brought to bear by those outside the investigation (particularly by the press), we will continue to be kept in the dark. And, if the cause of the crash is directly related to some technical occurrence, such as that outlined below, then withholding the report would serve no one.

AIR-GROUND SENSING

1. This is a very important system because it affects many other components of the aircraft, from the brake system to the engine power system, the APU electrical supply, as well as the pressurization controls. Ground-air sensing derives from two sensors, one on the right landing gear and another on the nose landing system. Electrical signals from the two sensors are distributed through 10 electronic cards from where they lead to various electrical and electronic aircraft modules.

2. It is possible to have an air-ground sensing problem as a secondary effect from related electrical wiring or electronic cards, rather than from the sensors themselves. A problem in the sensors leads to several malfunctions, which are immediately noticeable, so it’s more likely that there was a chronic and insidious secondary malfunction in the ground-air system, which may occur intermittently during flight and is difficult to trace by the engineers when the aircraft is on the ground [text removed].

3. Let’s assume that the electronic card, which prompts or cuts the T/O Warning Horn from the Aural Warning Generator (located above the captain’s left knee), had a malfunction, which retained the T/O Warning Signal to the Aural Warning Module instead of canceling it when the aircraft is in the air. One possible reaction of the flight crew, who has experienced the problem before, would be to pull the T/O Aural Warning CB. But, because this fuse is located right next to the CB of the ground-air sensor system, they could pull the latter by mistake. This could then cause serious problems to the pressurization system. I have seen this happening many times and the reverse is also likely: instead of the ground-air CB, they pull the T/O Aural Warning CB. This move would silence most of the flight deck’s aural warning systems, especially the cabin altitude horn, which sounds when cabin altitude passes 10,000 feet.

4. The situation becomes even more complicated given that the crew carried out a bleeds-off takeoff, which is often done in hot weather, with full loads and transit trips with no enroute refueling, as was the case with Helios Flight 522 to Prague via Athens. This was also the case in the five previous incidents of electrical malfunction in the aircon packs, which the same plane had experienced before its final flight. In this case, the APU compressor normally supplies bleed air for the initial cabin pressurization to allow the engines to develop full thrust in the early stages of climbout to a maximum of 17,000 feet. But the APU electrical generator is one of the systems directly affected by electrical malfunctions in the ground-air sensor system. So you can have a situation where APU electrics are also left ON together with APU bleeds without the crew knowing about it. More likely, however, is that the plane was left in AIR MODE on the ground after the pressurization test carried out three hours before the last flight. Combined with a secondary electrical fault in the ground-air sensor, this would affect the pressurization AUTO controls, outflow valves and take-off configuration warning systems in all sorts of unpredictable ways.

[BOAC]

BFC - many thanks and I echo your 'hopes' for comments.

More likely, however, is that the plane was left in AIR MODE on the ground after the pressurization test carried out three hours before the last flight

- on a quick read through your post only, this would have been instantly noticeable to the pilots as they would not have been able to select the APU electrics onto both generator buses before start, as that function is 'ground mode only' on the Classic 737. This would have caused several problems with systems etc during the pre-flight preparation and taxy.

I'm still not aware from any reports as to when the horn began sounding? Do you know?

To add that anyone wishing to contribute 'off-line' is welcome to PM me and I will de-identify and post if I consider it suitable material.

[big fraidy cat]

The following article by today's Cyprus Mail online edition goes into some detail about the plane's history of inspections; this information follows the political aspects of the investigation.

Can some tech people comment as to whether this plane's maintenance history is standard for the industry, or indicative of an underlying technical problem that no one could find?


‘A deafening silence’
By Alexia Saoulli

Helios victims’ families still waiting for independent investigation six months on
DISY DEPUTY George Georgiou yesterday criticised the government for failing to appoint an independent investigation committee into the fatal Helios Airways crash last August.

Speaking to reporters at a press conference, Georgiou questioned why President Tassos Papadopoulos remained silent on an issue the relatives of all 121 passengers killed on board wanted answers to.

“President Tassos Papadopoulos promised the formation of an Independent Investigation Committee following the initiative of the Democratic Rally (DISY) for the issue to go to the House,” he said.

“Everyone accepted DISY’s suggestion and the House unanimously officially put the suggestion to Tassos Papadopulos. The president accepted the suggestion. What happened from then until today? Six months following the accident and five months after the president’s acceptance of the committee’s appointment there’s been a deafening silence.”

Georgiou also pointed out the Greek Air Accident Investigating Committee had nearly concluded its investigation into the crash and yet its Cyprus counterpart had not made any noises.

When journalists pointed out the Air Accident Investigation Committee in Cyprus had no jurisdiction over an accident which occurred in Greece, Georgiou said he was well aware of that, but that the government should be upfront that the local committee’s role was null and void.
The DISY deputy questioned how the government system worked and if its aim was for “time to go by so responsibilities and those responsible are forgotten”.

Speaking for his party he said the House had been shown no respect as its call for an independent investigation committee had been ignored and said the House Communication and Works Committee had been refused a discussion on the tragedy with regard to the Civil Aviation Department nor had been given access to the findings of two European experts on the crash.
He said: “It is clear some people in the government and coalition parties are trying to bide their time so as to reduce public reaction and to smooth over the damage.”

‘Plane was given all-clear on August 5’

COMMUNICATION Minister Haris Thrasou confirmed the fatal Helios Airways Boeing 737-300 which crashed in August last year had undergone continuous inspections from the first day of its registration.

Thrasou’s written confirmation was given to journalists by DISY deputy George Georgiou yesterday, during a press conference to highlight the government’s failure to implement an independent body to investigate the crash.

During the press conference journalists heard the House Communication and Works Committee, of which Georgiou is a member, has repeatedly been refused a discussion concerning the tragedy with regards to the Civil Aviation Department.

Nevertheless Georgiou gave journalists a letter Thrasou had addressed to the House regarding what mechanical inspections the aircraft had undergone.

Before the plane’s registration two Civil Aviation Department inspectors travelled to a German maintenance company in Munich where the aircraft’s annual mechanical inspection (maintenance) was conducted, he said.
In the inspectors’ presence it was determined that any repairs and changes that had been made to the plane on prior occasions had been in accordance with European Union regulations. Also all necessary systems, specified by European regulations, were installed on the aircraft and it underwent a thorough visual inspection, he said. A test flight was then carried out during which the aircraft’s systems underwent technical inspection.

Following the plane’s registration, Helios Airways appointed the aircraft’s maintenance to ATC Lasham, a British Boeing aircraft maintenance, repair and overhaul company. The company maintained the aircraft according to approved maintenance programmes, the Minister said.
In November 2004 three Civil Aviation Department inspectors attended a scheduled maintenance inspection of the ill-fated plane.

On December 7, 2004 a mechanical inspection of the plane was conducted by the maintenance company’s specialised mechanics in the presence of a Civil Aviation Department inspector in Larnaca.

In March last year a Civil Aviation Department inspector travelled on board the aircraft and carried out an in-flight inspection. On June 8 and 9 inspectors from the Civil Aviation Department inspected the private airline company’s technical department and, among others things, the specific aircraft’s log book, said Thrasou. A month later Civil Aviation Department officials met with the UK company to discuss Helios Airways’ aircraft maintenance matters.

Then on August 5, nine days before the crash, the Boeing 737-700 underwent a ground inspection by inspectors and officials of the Civil Aviation Department (flight operations). “Any problems that arose during the inspection were dealt with and the flight was deemed flight worthy,” he said.

In his letter the Communication Minister assures routine checks were carried out on the plane by licensed mechanics authorised by ATC Lasham. These inspections were carried out two or three times daily from the day of its registration till its crash.

He said: “The results of the mechanical inspections are recorded in the aircraft’s log book, which is in the possession of the Greek Air Accident Investigating Committee. It should be noted the Civil Aviation Department has no access to these records.”

Meanwhile according to European regulations on continuous air worthiness, an aircraft’s owner is responsible for its constant airworthiness and entrusts its mechanical inspections to an approved maintenance organisation.

According to the same regulation, the competent authority (the civil aviation department) inspects the maintenance programme and carries out inspections to determine the aircraft’s continuous air worthiness, said Thrasou.

Copyright © Cyprus Mail 2005

[Danny]

Doesn't sound like anything out of the ordinary. Civil Aviation Department inspectors from whatever country have a duty to regularly check up on the airlines. The timescale indicated in the report don't appear to be anything out of the ordinary, particularly for a small airline in a small country.

It appears to be more of a report of a politician grandstanding and looking for soundbites. The facts about the operational inspections do not appear to be unusual in frequency or results. Often, after an operational inspection there will be some things highlighted that will need to be reviewed at the next inspection but these are not usually things that affect safety.

At this stage, they may as well await the results of the investigation which will have taken into account all the previous history from previous inspections. It may eventually surface that there was a particular recurring problem that was reviewed during the operations inspection but it is likely that on its own it didn't pose a particular safety issue. Unfortunately, opening statements such as

COMMUNICATION Minister Haris Thrasou confirmed the fatal Helios Airways Boeing 737-300 which crashed in August last year had undergone continuous inspections from the first day of its registration.

are ambiguous enough to imply that the fact that 'continuous' inspections were carried out somehow implies there was a problem when in fact it is perfectly normal that aircraft are inspected on a regular basis as part of normal operations. An aircraft is inspected in one way or another after and before every flight. Does that imply that there is a problem?

Again, without specifying exactly what type of inspections or by whom they were carried out the statement is nothing more than waffle and hyperbole by the Minister. Fairly typical of politicians wouldn't you say?

[big fraidy cat]

I think that what called my attention to the frequency of the daily inspections was the use of the words two or three times per day. Two times made sense to me, but three times seemed excessive. My attention was also drawn to this statement since it has already been reported (yet not confirmed) that the plane had been fitted with a 128-parameter FDR, more than required by law for a plane manufactured in 1997.

[GearDown&Locked]

My attention was also drawn to this statement since it has already been reported (yet not confirmed) that the plane had been fitted with a 128-parameter FDR, more than required by law for a plane manufactured in 1997.

Two questions:
-If true, was it installed to trace an eventual chronic problem with the a/c?
-and by whom? Boeing? or maintenance own decision?

GD&L

[Plane View]

As an independent consultant who is in regular contact with the official investication on behalf of some of the victims’ families, I can only comment on issues I have discussed with Captain Tsolakis, the chief investigator, the NTSB and Boeing, as matters of public record.

Regarding the aircraft’s inspection history as presented in parliament by Communication Minister Thrasou, it is important to differentiate between regular or routine checks (which may be carried out two or three times daily, depending on the number of scheduled flights per day) and special inspections undertaken following a report logged by the crew of possible technical problems during a previous flight.

Mr Thrasou’s confirmation that the plane was undergoing continuous inspections - since its registration with Helios Airways in April 2004 - is hardly reassuring as a clean bill of health about its airworthiness.

On December 7, 2004 a mechanical inspection of the plane was conducted by the maintenance company’s specialised mechanics in the presence of a Civil Aviation Department inspector in Larnaca.

Note that the above inspection took place barely nine days before the aircraft suffered a serious incident of Decompression at 34,000 feet on December 16, 2004.

The subsequent investigation by the Cypriot Accident Investigation Board, headed by Costas Orphanos, failed to arrive at a definite conclusion regarding the causes of that pressurisation failure. Orphanos’ Investigation Report, issued one month later, made no mention of any FDR data being retrieved and assessed during that investigation or after the plane was flown to ATCLasham, UK, for further checks and equipment overhaul.
The same cursory reference to later unscheduled inspections raises more questions about the unspecified recurrent problems the Helios crews have had with this particular aircraft in the weeks prior to its final flight:

In March last year a Civil Aviation Department inspector travelled on board the aircraft and carried out an in-flight inspection. On June 8 and 9 inspectors from the Civil Aviation Department inspected the private airline company’s technical department and, among others things, the specific aircraft’s log book, said Thrasou. A month later [July] Civil Aviation Department officials met with the UK company to discuss Helios Airways’ aircraft maintenance matters. Then on August 5, nine days before the crash, the Boeing 737-700 underwent a ground inspection by inspectors and officials of the Civil Aviation Department (flight operations). “Any problems that arose during the inspection were dealt with and the flight was deemed flight worthy,” he said.

These intensified inspections were obviously related to at least five incidents of (reported) electrical malfunctions in the plane’s air-conditioning system, within two months prior to the crash, which Mr Tsolakis had confirmed in his first interview with Flight International on August 22, 2005.

The five incidents include the problems reported in the penultimate flight which necessitated a “full pressurisation test” on the ground three hours before the fateful flight on August 14. Mr Thrasou failed to mention this last inspection as another proof of the aircraft’s flight-worthiness, something which Mr Tsolakis himself has disputed on several occasions.
With hindsight, I don’t see how someone can say that this inspection record doesn't sound like anything out of the ordinary.

[Melchior]

There might have been numerous inspections done on the aircraft and it may have had a lot of technical incidents but I don't believe that this accident happened due to technical failures.

The following is from a previous post which in my opinion is very close to the truth about what happened in the accident:

There was no electrical problem, no radio failure, no toxic gases, no hijacking nor any other kind of actual problem. The aircraft behaved perfectly. What happened is that the outflow valve was left on Manual by the engineers that carried out the pressurization check during the previous night. The valve’s wrong position was missed by the flight crew during the overhead scan and during the checklists. It’s not unthinkable, it happened before and it will happen in the future.

The Equipment cooling light came on and did not go out when 'Alternate' was selected, simply because the intake pressure was low since there was no cabin pressurisation. (The EE light is triggered by a low pressure sensor in the equipment cooling duct.) The QRH says: 'IF THE LIGHT DOES NOT GO OUT WHEN ALTERNATE IS SELECTED, NO FURTHER ACTION IS REQUIRED'.

The flight crew were dealing with this 'problem' when the cabin altitude warning (intermittent horn) came on. They then probably considered that this was a false take-off configuration warning (the two warnings are exactly the same) caused by the equipment cooling problem. The captain probably stood up to look at the circuit breakers, probably to try and reset the EE cooling sensor circuit breakers. While they were trying to figure it all out the aircraft was climbing and they were suffering progressive hypoxia due to the high cabin altitude, which impairs reasoning.

At 14000 ft cabin altitude, the cabin masks dropped. Cabin crew are told not to contact the flight deck immediately in such cases, due to high workload of the pilots. The warning for the oxygen masks in the FD is a light in front of the pilots saying ‘overhead’, same as for the EE cooling light! The oxygen masks light though on the overhead panel is much further aft on the panel than the EE cooling light though. Thus this warning was probably totally dismissed by the crew as a repetition of the EE cooling warning.

Therefore the flight crew were totally unaware that there was a pressurization problem while the cabin crew and passengers were on oxygen masks from 14000 ft onwards. Once the flight crew were unconscious there was no way to enter the Flight deck since the door was locked, unless the crew knew the code. Only the senior CC member would know the code, unless the code was changed and the crew was not informed. Maybe the senior was unconscious or panicked or forgot the code. Who knows. We may never find out. When both engines flamed out and electrical power was lost, the door unlocked and the cabin crew that survived on the portable oxygen cylinders entered the cockpit.

So to sum it up, this accident was caused by a series of human errors alone. The aircraft behaved as if it was brand new.