AA77 FDR Data, Explained

[jaydeehess]

IIRC, One example of several seconds of missing data that MacGyverS2000 gave was a block erasure being implemented on the block that had just been written to in which only 4 seconds worth of data was present.

This would depend on the amount of data written during a write command. If the FDR writes one word at a time then any length of time could be missing. If holds the same amount of data in the write buffer as is erased during a block erase then only that amount of data could be missing.

However in either case there is data in the write buffer which would be lost and/or corrupted during an unwanted powerdown or data bus spike or clock spike or a power line spike above that which the filter network can handle or even due to a severe physical blow that severs a conductor trace on a printed circuit board. (It is the memory chip itself that is hardened the greatest against physical damage)

 

[jaydeehess]

In addition to my last post about the FDAU of AA587:



AA 587 experienced in-flight failures and lost both engines prior to crashing.
The FDR continued to record while the DFDAU stopped outputting data as
per report.

This is not a similar case to AA77.

R.Mackey was forgetting that you have the goalposts mounted on a truck.

Given that you are going to state that all cases are not similar to AA77, I asked you for an example of an airliner crash that would be similar to the crash of flight 77, high speed into a vertical concrete wall with all systems intact prior to impact.

Demanding that we show you missing data on an FDR from a crash that may never have occured is somewhat disingenuous.

 

[DavidJames]

R.Mackey was forgetting that you have the goalposts mounted on a truck.

Those idiots won't even admit they have goal posts, yet demand you kick the field goal, then declare each attempt is wide right.

 

[Turbofan]

R.Mackey was forgetting that you have the goalposts mounted on a truck.

Given that you are going to state that all cases are not similar to AA77, I asked you for an example of an airliner crash that would be similar to the crash of flight 77, high speed into a vertical concrete wall with all systems intact prior to impact.

Demanding that we show you missing data on an FDR from a crash that may never have occured is somewhat disingenuous.

I think anyone with an ounce of technical ability is going to agree that AA 587
and AA 77 are totally differenet cases.

This not about goal posts, it's about finding a similar case to compare against.

Where can you draw a parallel between these two recorders in our sense of
the discusison? AA 587 experienced mid-flight failures, lost both engines
and crashed several seconds later. The FDR recorder wrote data interrupt
markers as a sign that it was powered and functional while the DAU stopped
responding.

AA77 allegedly hit a wall at 530+ MPH, and lost 4 seconds of data due to
transients (according to theories on this site) upon impact.

Please, go ahead and draw the parallel. I don't see it.

 

[Jonnyclueless]

Flight 77 also experienced mid-flight failures as it struck objects prior to impact. Every single plane crash is going to be different. But I think it's only fair that you present an example of a crash similar to flight 77 that did not result in data loss. It wouldn't be impossible. But your basic argument is that unless something has happened the exact same way before, then it cannot happen. Thus nothing can ever happen for the first time.

But Jay's point being, how many planes have been in this experience before to make such an assumption on your part?

 

[Turbofan]

At this point, I'm not sure what you are trying to prove/disprove. Quotes are getting cut too short or post snippets are being randomly quoted and muddying the issue. I'll go back to my "Make the question clear and concise" statement to avoid further muddying the waters.

Sorry, there are several sub discussions and theories happening within this
thread so it's tough to keep up.

A quick overview:

The truth movement believes AA77 did not hit the Pentagon based on FDR
analysis, and physical evidence. The FDR data shows several parameters
out of context from the official story but mainly altitude being too high to hit
the building, and the final frame time stamp of which indicates too high at the
alleged impact time.


Those who believe the official story claim that since the impact time and
last recorded info cannot be true at once, offer the following theories:

- The clock sync is off 4+ seconds (spec is .010 seconds as per documentation)

- DME (distance measuring equipment) is out of tolerance and recorded an
erroneous value placing the plane closer than it actually was by about 1 nautical mile
(spec is 0.1 nm for DME as per manufacturer site)

- Data was held up in a buffer for several seconds producing a delay
(of 4+ seconds). This would require the clock sync to be off by 4 seconds
as well.

- Data was corrupted at the bit level (3072 bits worth)

- The last four seconds of data was erased due to a transient(s) upon
impact.

The last two points are the most recently debated and when you were invited
into the discussion.

We are now close to weeding out the second last point because of truncation
of the raw file. It may also be argued that power availabilty and duration of power at a max of 400 mseconds would be an unlikely cause of rendering
3072 bits inaccessible, however we are free to leave this topic open for
further scrutiny.

Finally, we are down to erasure by transient. 4 seconds worth of data.

Here's the monkey wrench:

L3 Communications was contacted and on record stating that sensor data
must be recorded to crash protected memory no later than 500 milliseconds
after being transmitted by the sensor (IE: altimeter, limit switch, pitot probe, etc.).

Some people say that because we received this 500 millisecond value from
a salesman of L3 Communications (the MFG of the Flight Data Recorder) that
it is considered hearsay and not admissible to the discussion.

Having said that, we can carry on with our debate.

Another possibilty is that the NTSB was in error about the impact time
according to the FDR, however their flight animation and flight path study
will indicate otherwise.

If I've missed anything, or incorrectly conveyed the points please correct me.

 

[beachnut]

The truth movement believes AA77 did not hit the Pentagon ...

- DME (distance measuring equipment) is out of tolerance and recorded an
erroneous value placing the plane closer than it actually was by about 1 nautical mile
(spec is 0.1 nm for DME as per manufacturer site)

Here's the monkey wrench:

L3 Communications was contacted and on record stating that sensor data
must be recorded to crash protected memory no later than 500 milliseconds
after being transmitted by the sensor (IE: altimeter, limit switch, pitot probe, etc.).

Some people say that because we received this 500 millisecond value from
a salesman of L3 Communications (the MFG of the Flight Data Recorder) that
it is considered hearsay and not admissible to the discussion.

Having said that, we can carry on with our debate.

Another possibilty is that the NTSB was in error about the impact time
according to the FDR, however their flight animation and flight path study
will indicate otherwise.

If I've missed anything, or incorrectly conveyed the points please correct me.

It is a fact 77 impacted the Pentagon. You have zero evidence to stop the truth.

Your fantasy lie omits the facts and evidence.

The DME is stored in the FDR every four seconds at a resolution of X.0, X.2, X.5, X.8.
The 0.1 NM accuracy is out to lunch if only p4t math experts could understand what that means. The accuracy of DME in the aircraft is not 0.1 NM in practice but gives you a usable accuracy close to the station of 0.23 NM for the equipment 77 had on board, not some new unit used by cherry pickers. That being said, who cares, I will give the false information experts their 0.1 NM accuracy and the storage resolution error; It mean 1.5 DME stored in the FDR can be off as much as 0.25. 77 was as far as 6 or more seconds from impact when the FDR data ends for unknown reason. Even if 77 is closer, the data ends with seconds to go. OOPS

It is funny getting the p4t group to use 1.5 DME as an absolute clearly points to their cherry picking ways as they twist 9/11 into a paranoid conspiracy.

The impact of 77 could introduce transients to mess up the FDR. OOPS; but I am only an electrical engineer with a masters degree and understand explaining how and giving examples take time and someone who has more practical experience has already explained THIS! But you have to listen and try to understand.

Why are you unable to get 9/11 right? When will you break this big story to the press and when will L3 support your attempt at pushing fantasy ideas. How are those DVD sales of false information going?

You have no idea what the 500 ms means! Not a clue. When was the 500 ms standard in effect and why? F2100? oops

You left out the facts and evidence which make your 77 did not impact the Pentagon a lie. How did you do that? Miss all the evidence?

 

[WildCat]

I'd just like to point out that 4 seconds of data need not be erased, just that 4 seconds of data was either erased, not written to memory in the first place (due to delays from the time the sensor sends the signal to the time it is written to memory), or some combination thereof.

Turbofan will continually misrepresent the above, because his case is extremely weak. In fact, there's not a single FDR expert in the entire world who agrees with the pretend FDR experts at PfffT.

 

[lapman]

- Data was corrupted at the bit level (3072 bits worth)

It could take only one bit to be corrupted for software that is reading the data to think it's at the end of the file.

 

[Turbofan]

The DME is stored in the FDR every four seconds at a resolution of X.0, X.2, X.5, X.8.

This is incorrect.

The NTSB data clearly shows that DME recorded a value of 1.7 and 1.5 DME.

1.5 DME was recorded at :43

The aircraft then flew an additional TWO SECONDS at 530+ MPH, and
therefore would be even closer.

The 0.1 NM accuracy is out to lunch if only p4t math experts could understand what that means. The accuracy of DME in the aircraft is not 0.1 NM in practice but gives you a usable accuracy close to the station of 0.23 NM for the equipment 77 had on board, not some new unit used by cherry pickers. That being said, who cares, I will give the false information experts their 0.1 NM accuracy and the storage resolution error; It mean 1.5 DME stored in the FDR can be off as much as 0.25. 77 was as far as 6 or more seconds from impact when the FDR data ends for unknown reason. Even if 77 is closer, the data ends with seconds to go. OOPS

If the aircraft was 1.7 DME at :43, then it would have recorded 1.7 DME
because it has that capability. Also keep in mind the time stamps along with
the DME data.

It is funny getting the p4t group to use 1.5 DME as an absolute clearly points to their cherry picking ways as they twist 9/11 into a paranoid conspiracy.

DME is more accurate than INS.

The impact of 77 could introduce transients to mess up the FDR. OOPS; but I am only an electrical engineer with a masters degree and understand explaining how and giving examples take time and someone who has more practical experience has already explained THIS! But you have to listen and try to understand.

I don't recall.

You have no idea what the 500 ms means! Not a clue. When was the 500 ms standard in effect and why? F2100? oops

According to L3 communications, the sensor data must be recorded to the
FDR within half a second. We have the e-mails and recorded telephone
audio to support this.

Wildcat, 4 seconds of data cannot be suspended in a buffer.

 

[WildCat]

According to L3 communications, the sensor data must be recorded to the
FDR within half a second. We have the e-mails and recorded telephone
audio to support this.

And L3 does not support the conclusions of PfffT. Do you understand why that is?

Wildcat, 4 seconds of data cannot be suspended in a buffer.

Never said it was. What we do know is you do not understand the spec the salesman at L3 gave you.

 

[jaydeehess]

TF, how much data gets written to the FDR during a write command?

 

[dtugg]

You know what, you convinced me Turbofan. The NWO tricked all the eyewitnesses into believing a plane hit the Pentagon and successfully planted a lot of very convincing physical evidence. But for some explicable reason, they forgot to record a few extra seconds on the phony FDR. And this incredibly damning evidence bothers nobody in the entire world, expect for an extremely small set of people with no expertise that most people would regard as insane morons. I can't figure out why.

 

[Turbofan]

And L3 does not support the conclusions of PfffT. Do you understand why that is?


Never said it was. What we do know is you do not understand the spec the salesman at L3 gave you.

Here is the e-mail exchange between Calum Douglas and Ed Santana.
YOu tell me where the interpration is lost, and then tell us what the
0.5 second value is for:

4: What would be a typical time lag between the sensor signal being
generated (for example aileron angle) and the data being logged to the
protected memory of the recorder?

L-3 Response: Per ED55, it shall not exceed 0.5 seconds,



5: Is the size of this recording delay regulated by industry or just
minimized by good design?

L-3 Response: Regulated per ED-55, Minimum Operational Performance Specification for Flight Data Recorder Systems.



6: In the case of a major accident like CFIT (controlled flight into
terrain) how much data (in terms of seconds of flight) is typically
lost? (For example signals still being processed by the DFDAU).

L-3 Response: With the use of the Solid State Flight Data Recorders,
typically, data is only lost at the point when power to the recorder or
FDAU is terminated. This would be better asked of the NTSB instead of
L-3 Communications.

Contact:

Mr Calum Douglas, Cert Math.(Open).
Oxford Brookes University
United Kingdom
Mech Eng Undergraduate (2nd year).

Ed Santana
Military Marketing & Sales Manager
L-3 Communications Aviation Recorders
Telephone: 941-377-5542

[email protected]
Website: www.L-3AR.com
Pubs Download Website: www.L-3AR.net

PFT also has Ed Santana on audio concerning the 0.5 second value. I can
link that shortly.

 

[WildCat]

Here is the e-mail exchange between Calum Douglas and Ed Santana.
YOu tell me where the interpration is lost, and then tell us what the
0.5 second value is for:

4: What would be a typical time lag between the sensor signal being
generated (for example aileron angle) and the data being logged to the
protected memory of the recorder?

L-3 Response: Per ED55, it shall not exceed 0.5 seconds,

You asked a salesman a technical question. Do you know what ED55 is? Does Santana understand what it is?

I don't care what a salesman says on an ambush phone call. Experts Turbofan... got any? Of course you don't.

 

[Turbofan]

You asked a salesman a technical question. Do you know what ED55 is? Does Santana understand what it is?

I don't care what a salesman says on an ambush phone call. Experts Turbofan... got any? Of course you don't.

This was an e-mail between Calum and Ed. No ambush.

I'm sure Ed is capable of answering tech quesitions about the products he
supports and sells. If he wasn't able to asnwer them, he has a host of
options including a quick conversation with his tech support staff, or
consulting his engineering team.

I have become quite familiar with EUROCAE since researching this subject.

EUROCAE MOPS ED-55 is a performance specification. You can find out
more here:

http://www.validatedsoftware.com/avionics/avfaq.htm#eurocae

Your arguement is not very credible. L3 products are certified to these
standards, so even if Ed did not know the spec, his product has been
certified to meet, or exceed this spec.

Once again Wildcat, you seem to have more knowledge than Ed about the
meaning of 500 milliseconds. How exactly do we apply this value if the above
e-mail exchange is *incorrect (*according to you)?

Jay, the data is written in lumps of 256 words * 12 bits per second to memory.

Perhaps MacGuyver can shed some light on how many bits are written at one
time (x bit(s)/second) as I'm not sure.

 

[jaydeehess]

originally posted by Turbofan

The block is first reset, and then an entire block of data is written at once.

TF, how much data gets written to the FDR during a write command?

Let's recap shall we.

How many bits are reset during a block erase TF?(how big is an erase block)
How many bits are written to memory during a write command TF?(is a write block the same size as the erase block?)
What is the data rate from the DFDAU to the write buffer?

 

[jaydeehess]

Jay, the data is written in lumps of 256 words * 12 bits per second to memory.

Perhaps MacGuyver can shed some light on how many bits are written at one
time (x bit(s)/second) as I'm not sure.

We posted at about the same time.

12 bits per second,, for 256 12 bit words that would take 4.25 minutes.
Better correct that or we need a much bigger buffer as its gonna get ful right quick.

 

[Turbofan]

We posted at about the same time.

12 bits per second,, for 256 12 bit words that would take 4.25 minutes.
Better correct that or we need a much bigger buffer as its gonna get ful right quick.

Jay, that is 256 words per second (12 bits per word) for 3072 bits per second
into CPM minimum.

 

[Jonnyclueless]

The car salesman told me that the car should get no less than 15mpg. But for some reason after being impacted by a tractor trailer the mpg is much less. It must be an inside job.

dtugg,

What's even worse than the absurd notion of faking a plane crash in front of 10s of 1000s of people is not simply pulling it off, but just the absurd notion of someone actually taking that kind of risk and just crossing their fingers and hoping for that one in a billion chance of it actually working.

But this is one of the many reasons people like Turbofan are unable to actually present an account of what happened and have to rely on stats meant for perfect working conditions.