777236ER From , joined Dec 1969, posts, RR: Posted (14 years 1 month 3 weeks 3 days 15 hours ago) and read 3442 times:
There have been a large number of aircraft crashes where the cause has been indeterminate, purely because of one thing: the loss of power to the CVR/DCVR and DFDR.
Now, these tend to be powered solely from engine generators and the APU, not batteries or an APU (if fitted). This really does seem stupid -- the flight recorders will be of most value in an emergency, and the most serious emergencies are engine failures.
TWA800, Swissair 111, the Transat A330 that ran out of fuel and lots and lots of smaller, but potentially fatal incidents such where an engine has failed and power has been lost to the DFDR/DCVR.
Now, a few years ago the biggest problem were 30 minute, out of date CVRs. This was resolved by changing the certification regulations to a solid-state 2 hour CVR. This is all well and good, but the biggest problem was ignored! That solid-state CVR is of no use if it doesn't have power.
A small battery could do the job. Independant from all other electrical systems, it would have enough power for, say, 10-15 minutes of DCVR/DFDR operation. 10-15 minutes could provide that bit of vital data that solves a crash. Just think, we could know once and for all what caused TWA800. The issue is further helped by the introduction of the solid-state DCVR, which requires less power than the older tape models.
The NTSB, Britain's AAIB and scores of other accident investigators have made recommendations again and again for DFDR/DCVRs to be fitted with independant power supplies, yet nothing has been done. As far as I know, there haven't even been any fesability studies into such an idea.
The ironic thing is that even QAR (quick access recorders), ussed primarily inter-company to monitor crew performance, sometimes have independant power supplies! Yet NOT the most informative investigative tools of them all, the DFDR/DCVR.
When is this issue going to be resolved satisfactorily?
TechRep From , joined Dec 1969, posts, RR:
Reply 1, posted (14 years 1 month 3 weeks 3 days 9 hours ago) and read 3420 times:
This question would have received much better response had it been posted in the Tech/Ops forum but I came across it and will attempt to answer your question.
The industry has taken steps to improve the probability of 100% data recovery. First, it had been noted that in the majority of cases where the data has been unrecoverable after an aircraft disaster was due to a post-impact fire not power loss. It was also concluded that in many cases, the post-crash fire was not the short, high intensity scenario covered by qualification to the existing TSOs, but due to a much lower intensity, with substantially longer duration fire. Therefore EUROCAE, included the low intensity fire requirement (10 Hours at 260°C) into ED-56 Rev. A. The NTSB and FAA have also acted in increasing the existing high intensity fire survivability from 30 minutes to 60 minutes (50,000 BTUs, 1100°C) in TSO-C123a and TSO-C124a which superceded existing TSOs. in August of 1998.
Another major cause of unrecoverable data (or a "poor" recording) after an aircraft disaster, is due to broken and/or damaged recorders which were unable to record the necessary information due to an internal or system level fault.
Historically, electro-mechanically-based CVRs and FDRs have suffered in terms of overall reliability and performance due to their basic design. Recording head and tape wear, motor belts and bearings, etc. will degrade over time in the severe environments encountered on commercial air transport aircraft. This cause has been dramatically improved by the introduction of all solid-state crash survivable recorders, where non-volatile memory devices have replaced the electro-mechanical tape based recording systems. Since there are no longer any moving parts in Solid State recorders, basic reliability (as expressed in Mean-Time-Between-Failures) has improved dramatically by at least 5-fold. Additionally, since there are no moving parts in the recording system, there is no degradation over time in the quality of the recording. In the past, the quality of the recording has been greatly effected by where in the "maintenance cycle" the recorder is when data is extracted from it (i.e. if it has been a long time since the last overhaul of the unit, head/tape wear, motor bearing wear, etc., will naturally cause a lower quality recording).
Several recent accidents of commercial aircraft have demonstrated the effect of the Solid State Recorders. The AlliedSignal SSFDR and SSCVRs have performed flawlessly with 100% recovery and no errors in the data. In one particular investigation the NTSB started with a download of the SSFDR contents at 9:00 am and by 1:00 pm of that same afternoon a full animation of the final moments of the flight were available for review.
As a result of the survivability and maintainability of Solid State recorders the NTSB have released Safety Recommendations to mandate solid-state cockpit voice and flight data recorders by 2005, i.e. retrofitting of all tape based recorders to solid-state. Many commercial airlines are already doing this on their own based on economic (cost-of-ownership) arguments.
In 1992, an industry-wide meeting discussed ways to improve the probability of 100% post crash data recovery for both the cockpit voice and flight data information. AlliedSignal proposed that one way was to have combined recorders in a dual-redundant installation (i.e. two separate crash survivable recorders storing both the cockpit voice and flight data information) on the aircraft. Further, these combined recorders should be installed in vastly different locations on the aircraft, such that in a worst case scenario at least one of the recorders would be subjected to a less severe post-crash environment. Aircraft manufacturers agree with this basic philosophy, but have yet to implement a dual-redundant recorder installation on any commercial aircraft being currently produced.
To avoid recurring changes to aircraft interfacing and the recorder itself; a new recording system philosophy is required. This new architecture can be an extension of the existing "Data Acquisition System" as the central processing component, accompanied by multiple high-speed serial interfaces to dual Solid State Digital Data Recorders (SSDDR).
In this proposed new system the crash survivable recorders are reduced to simply recording digital information received on high-speed serial interface(s). The recorders need not know the specific source or type of information being recorded, but simply recording the digital data as it is received under a yet to be defined rule set. The processing of information and digitization would therefore take place in other avionics within the aircraft and transmitted to the redundant recorders. If developed suitably, this architecture can provide the following benefits:
1) Dual redundant Crash Survivable Recorders - Reduces airlines spares and logistic support and provides improved probability of 100% data recovery.
2) Crash survivable recorder need not require modification to meet changes in recording requirements.
3) Lower system cost - data processing and digitization process moved to highly integrated avionics subsystems (instead of the recorders themselves, which are subject to more severe operating environments)
4) Minimizing Installation costs by reducing wiring required for growth systems.
5) Eliminating the need to add other crash survivable recorders for additional information storage
6) Minimize cost impact for the recorder itself. The price of two redundant recorders would only be on the order of 50% more than the total price of today's separate FDR/CVR combination.
777236ER From , joined Dec 1969, posts, RR:
Reply 3, posted (14 years 1 month 3 weeks 3 days 4 hours ago) and read 3402 times:
TechRep, thanks a LOT for that, but that's not the point I was making Maybe I didn't make it clear, sorry. As I said, solid-state recorders are vastly superior to the traditional tape recorders. However, incident after incident has shown that the recorders need an independant power supply. I was wondering if there was any reason why regulators were trying to bring in solid-state recorders as if they were going out of fashion, yet doing nothing about the constant accident board recommendations for independant power supplies, even if that supply was from a small non-rechargable battery, lasting only minutes.
TechRep From , joined Dec 1969, posts, RR:
Reply 4, posted (14 years 1 month 3 weeks 3 days 2 hours ago) and read 3393 times:
Recorders from an independent power source is not an issue, but the major problem arises as to what other equipment must also still be operating in order for the CVR or FDR to be receiving data. I personally think that battery life/size is a non-issue here and will explain in more detail.
For the SSFDR, the case is completely different. Since the data is being obtained from a multitude or aircraft systems and sensors (with a lot of wire in-between) through the DFDAU (or equivalent), the likelihood of retaining a lot of the flight data for recording purposes after a major catastrophe (like TWA 800) is much more remote. However, like the CVR, the probability would be greatly improved if the SSFDR were also located near the Electronic Equipment Bay (where the DFDAU is located as well as most of the other LRUs with which it interfaces). In other words, the SSFDR is a very small piece of the flight data recording system, and that many other systems must be powered (with intact wiring, etc.) to get useful information.
So in retrospect it doesn't matter how long the power source can supply back up power, if the associated Data Acquisition System (DFDAU) becomes inoperative due to catastrophic failure.
System requirements are being drafted by EUROCAE (WG50) to include cockpit video recording of the main instrument panel. The number of cameras and coverage (2 or 3 may be needed to view the front instrument panel, and another may view the overhead panel) are being considered for future FAA/JAA considerations. This may allow for the cameras to remain fixed and recording the instrument panel and human factors of the crew after the Data Acquisition System has been interrupted due to failure.
So in effect your wishes are being carried forth however the means are different and much more effective than adding a backup power source. Again however we run in to difficulty, pilot's unions do not as readily accept this. Pilots and safety officials agree that privacy legislation similar to current voice recorder laws would be needed to keep video footage protected from court cases and Freedom of Information Act requests.
TechRep From , joined Dec 1969, posts, RR:
Reply 5, posted (14 years 1 month 3 weeks 3 days 1 hour ago) and read 3387 times:
Now if that wasn't sufficient enough you can go to the source and discuss this with EUROCAE Working Group 50. EUROCAE is a European organization, which prepares minimum specs for airborne equipment. These specifications are then considered by JAA/FAA/CAA. These guys are basically a think tank of experts and industry geniuses. I will include the e-mail for WG-50 and you may ask this question but I think I have more than answered it.