New FDR Decode

Originally posted by mikelee
reply to post by weedwhacker

 

Per the NTSB -

Its either not working or unconfirmed.

Now your argue with me over what the NTSB criteria states?

Yeah, I should probably put together a .CSV with the functioning but unconfirmed parameters for that flight data since obviously he does not comprehend the meaning of the word "OR".

Edit to add - But that does mean sorting through 751 parameters, so it may take a while.

2nd edit to add - A great example is the "Aft Fuel Pump - R" as it shows "Lo Press" for frames 113941, 113933, 113937, and 113929... And "Normal" for every other time it was polled.

According to the lat/long data for those subframes, it was on the ground being started ( Power just came on ) at Chicago's Ohare Airport when that occured.

That parameter is in the "Not Working or unconfirmed" list, yet he would have you believe it was "Not Working" when clearly it was working.

[edit on 7-1-2010 by JFrickenK]

All this talk about parameter function and FDAU connections is really a
non -issue since proving that documentation is required to support a
theory of a recorded parameter.

What I find even more bothersome is blaiming the NTSB for data which
was NEVER included in their animation, or CSV file. That's all I will say
about that.

Getting back to a question I posted a few pages ago:

Where is the 'fault line' for the PA measurement in the NTSB data?

I have checked near Vmo and also trended / graphed a few parameters
and I cannot find a point where Pressure Altitude readings begin to show
signs of ADC failure.

Is anyone able to provide a time and/or frame count where the Pressure
Altitude begins to jump off trend?

Originally posted by turbofan

[snip]

Where is the 'fault line' for the PA measurement in the NTSB data?

I have checked near Vmo and also trended / graphed a few parameters
and I cannot find a point where Pressure Altitude readings begin to show
signs of ADC failure.

Is anyone able to provide a time and/or frame count where the Pressure
Altitude begins to jump off trend?

Can you describe what exactly you are looking for Tino? I don't think it would be anything that would jump out at you. It would most likely be a slow linear progression, as the calibration slowly started to drift. It would be like trying to pinpoint when and where an IRU starts to drift. It's a gradual process.

For example, I did an ADC accuracy check tonight for a 767, and even during a good test, between 0 and 40,000', the error moved from +20 at sea level to -60 at altitude. The error just sort of migrated.

reply to post by turbofan


Would there be any value in extracting the VSI instrument data and comparing that to the incremental changes in PA reading?

Hey Guys,

I'm not really sure how to find the failure point, but I compared the
accelerometer data near Vmo (and onward) in hopes to find a glaring
delta, or drifting trend.

Another idea I had was to check RAD ALT below 4000 feet with a known
ground elevation to confirm the Pressure Altitude reading at a given point.
(I believe RAD ALT kicks in around 4000'?).

As for VSI, I'm not sure if that's a valid comparison as Tom stated because
it's measured from the ADC...oh and don't forget the pitot tube is attached to it!

Originally posted by turbofan
All this talk about parameter function and FDAU connections is really a
non -issue since proving that documentation is required to support a
theory of a recorded parameter.

What I find even more bothersome is blaiming the NTSB for data which
was NEVER included in their animation, or CSV file. That's all I will say
about that.

True, the blame rests with American Airlines in this case for failing to provide a valid DFL.

The latest "error" discovered is that the "Corrected AoA" parameter was described as "Tank Densities" in the DFL provided to the NTSB by American.

Prior to this "error" was basically all of word 256 in regards to superframe numbering, and that has still not been completely deciphered.

Or perhaps it was not American's plane from which that data originated ?
< shrugs >

reply to post by JFrickenK


"Close, but no cigar"???

Listen, JFK...it is not my problem that some people can't understand what the NTSB meant by categorizing some of the FDR parameters as "Not Working or Unconfirmed"

IN A NUTSHELL, (and Tino or Jay, correct me if I'm wrong) "Unconfirmed" merel means that the parameter was not deemed vital to the decode, as pertained to the crash investigation, and the flight path study.

I do not know how to make that any more clear.

Besides, a simple look at ALL of those items on the "Not Working or Unconfirmed" list, and it immediately jumps out at anyone who knows anything about airplanes that there is NO WAY that ALL of those items would be "Not Working", since many of them would have been important in other instances, such as when the exact cause of the crash is unknown...

If you'd like help understand the abbreviations.....?????

reply to post by turbofan

(I believe RAD ALT kicks in around 4000'?).

Uncertain as to the exact range of the RAD ALT, but it is a fact that the display on the EADI does not appear until 2,500 feet, or less.

reply to post by JFrickenK

2nd edit to add - A great example is the "Aft Fuel Pump - R" as it shows "Lo Press" for frames 113941, 113933, 113937, and 113929... And "Normal" for every other time it was polled.

JFK, are you a pilot?

Nah, didn't think so.....

JFK, do you not understand WHAT the "Press" light on a fuel pump on/off switch is telling us??

It means either, the pump switch is selected to OFF (in which case, of course, the pump output pressure is "low"...), or, if the switch is selected to ON, then illumination of the light (and its associated EICAS message) indicates a low output pressure, below the threshold set by the pressure switch.

I'd think this would be obvious, but I forget that non-piots won't immediately understand this concept.

FWIW, the CENTER tank pump "Press" lights and EICAS warnings are inhibited when the associated switches are selected to OFF position.


Here's more fuel system knowledge, just a bit for you:

Look to see, in the historical data that you seem to be examining, when the APU switch is in the ON position. (This would normally be the case when the APU is operating, however the switch could be in the ON position, and the APU not running if, for instance, it had been placed ot ON, but not to START, or the APU had died for some reason after being started).

Anyway, when normal AC power is available, the LEFT FWD fuel pump activates automatically, with APU switch ON. (Regardless of fuel pump switch position). IF no AC power, and IF on the ground, then there is a dedicated DC fuel pump, also in the LEFT MAIN tank, for the APU.

See if your data accurate depicts those various scenarios, since you seem keen to try and find any niggling mistake or anomaly.....

Originally posted by weedwhacker
reply to post by JFrickenK

 

"Close, but no cigar"???

Listen, JFK...it is not my problem that some people can't understand what the NTSB meant by categorizing some of the FDR parameters as "Not Working or Unconfirmed"

Actually it is since up to this point you would be describing you.

Originally posted by weedwhacker
IN A NUTSHELL, (and Tino or Jay, correct me if I'm wrong) "Unconfirmed" merel means that the parameter was not deemed vital to the decode, as pertained to the crash investigation, and the flight path study.

I do not know how to make that any more clear.

Gee that sounds very familiar... Oh yeah, scroll back to my response to mike someone.

Originally posted by weedwhacker
Besides, a simple look at ALL of those items on the "Not Working or Unconfirmed" list, and it immediately jumps out at anyone who knows anything about airplanes that there is NO WAY that ALL of those items would be "Not Working", since many of them would have been important in other instances, such as when the exact cause of the crash is unknown...

Do ya think ?

Originally posted by weedwhacker
If you'd like help understand the abbreviations.....?????

Nope, but I would accept a .PDF version of the ARINC 573/717 documentation titled "Flight Data Aquisition and Recording System Characteristics" Published by Aeronautical Radio, Inc.

Originally posted by weedwhacker
reply to post by JFrickenK

 

2nd edit to add - A great example is the "Aft Fuel Pump - R" as it shows "Lo Press" for frames 113941, 113933, 113937, and 113929... And "Normal" for every other time it was polled.

JFK, are you a pilot?

Nah, didn't think so.....

Oh My what a short attention span you have.

Should you scroll back through this very thread you would come across a post where I have said that I am not a pilot and the closest I have ever been to being a pilot was an emergency landing of a Cessna several decades ago....

Provided the mods here have not removed that post also.

I stopped reading your demeaning post at the end quote and have no more desire to converse with you so you shall also join those in my ignore list now.

reply to post by JFrickenK

....so you shall also join those in my ignore list now.

It's an honor.

Because, whilst ignoring logic, you might as well ignore people who hactually know what they're talking about.

Embrace ignorance.


PS....I linked to the ARINC website, just to one aspect.

Surely the arcane and mysterious workings of the Flight Data Recorder can also be examined via other sources as well, such as maybe with what the manufacturer may have to offer?


But, regardless, have fun (or not) with a fruitless search, since it's your time that belongs to you, to waste.

It would be nice, though, when the results continue to prove the veracity of American Airline flight 77 FDR data, that someone will come back and report those findings, instead of burying the results because they don't agree with a "pet theory".

reply to post by JFrickenK

but I would accept a .PDF version of the ARINC 573/717 documentation titled "Flight Data Aquisition and Recording System Characteristics" Published by Aeronautical Radio, Inc.

How about a sales brochure from a company called "Ballard" that complies with those requirements???

Includes tester, carrying case,
and manual. Specify cable separately.

(I expect batteries are sold separately, your mileage may vary**)

www.ballardtech.com...

~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~

Well, I had ten minutes to kill, so just hunted through more of the ARINC site...

For 100 clams, you can buy this:www.arinc.com...

Maybe this will guide you: www.aviation-ia.com...

I don't know....maybe ARINC just doesn't want to put out a .pdf labeled "Flight Data Acquisition and Recording System Characteristics" for free???



[edit on 8 January 2010 by weedwhacker]

Originally posted by turbofan
Hey Guys,

I'm not really sure how to find the failure point, but I compared the
accelerometer data near Vmo (and onward) in hopes to find a glaring
delta, or drifting trend.

Another idea I had was to check RAD ALT below 4000 feet with a known
ground elevation to confirm the Pressure Altitude reading at a given point.
(I believe RAD ALT kicks in around 4000'?).

As for VSI, I'm not sure if that's a valid comparison as Tom stated because
it's measured from the ADC...oh and don't forget the pitot tube is attached to it!

"VVI" or "VSI" isn't a recoreded parameter anyways, so it's a moot point. However, since IVSI's use accelerometers to help reduce lag in climbs and descents, such a parameter might have helped pinpoint if there was any lag or not.

Originally posted by turbofan
Hey Guys,

I'm not really sure how to find the failure point, but I compared the
accelerometer data near Vmo (and onward) in hopes to find a glaring
delta, or drifting trend.

Another idea I had was to check RAD ALT below 4000 feet with a known
ground elevation to confirm the Pressure Altitude reading at a given point.
(I believe RAD ALT kicks in around 4000'?).

As for VSI, I'm not sure if that's a valid comparison as Tom stated because
it's measured from the ADC...oh and don't forget the pitot tube is attached to it!

Warren tried something over at PFT and came up with something interesting. It's over my head, and he could have missed something crucial to his calculations, but he came up with 2 separate altitudes based on aircraft pitot static data and METAR data.

Link

I don't know if there is really a good way to check PA accuracy using other parameters. RA is useless because you don't know where and what the exact elevation is when that "snapshot" is taken. Vertical acceleration would be probably your best bet, since VVI isnt recorded. Maybe use one of the other 11 flights, when the entire flight was conducted inside the flight envelope to get a good baseline for vertical accel/s to PA change/s - as we know everything was within the ADC calibration range during a prior flight. Take that baseline and apply it to the final 7000' or so of the data for the 9/11 flight.

[edit on 8-1-2010 by 767doctor]

Originally posted by JFrickenK

Originally posted by turbofan
All this talk about parameter function and FDAU connections is really a
non -issue since proving that documentation is required to support a
theory of a recorded parameter.

What I find even more bothersome is blaiming the NTSB for data which
was NEVER included in their animation, or CSV file. That's all I will say
about that.

True, the blame rests with American Airlines in this case for failing to provide a valid DFL.

The latest "error" discovered is that the "Corrected AoA" parameter was described as "Tank Densities" in the DFL provided to the NTSB by American.

Prior to this "error" was basically all of word 256 in regards to superframe numbering, and that has still not been completely deciphered.

Or perhaps it was not American's plane from which that data originated ?
< shrugs >

It's not AA's fault. How many times have I said that these frame layouts are generic, and all parameters don't apply to a given accident airplane? As much as you and Rob want this process to be an exact science, it isn't.

My airline has the largest 757 fleet in the world, with 5-6 different major configurations and 15 minor configuration differences. Can you guess at how many 757 DFL's we have to support this fleet?

Well perhaps the system needs to be changed then.

Just out of curiousity, just exactly how do you keep track of which FDR stores what variation without specific DFL's ?

Or don't you care ?

Originally posted by JFrickenK
Well perhaps the system needs to be changed then.

Just out of curiousity, just exactly how do you keep track of which FDR stores what variation without specific DFL's ?

Or don't you care ?

I don't have to keep track of anything, it's not my job. I trust that when there's an accident or incident, the NTSB will research the FDR and not come to any flawed conclusions based on a misunderstanding of what the FDR was and wasn't recording, as Pilots For Truth has.

[edit on 8-1-2010 by 767doctor]

And JFK, on the AoA issue...I can admit my error of not knowing that the values you used were already scaled. I saw 15.6, looked at the DFL, did some quick math and saw a result of 2.7 degrees that made sense for a 757. This isn't an error due to fundamental misunderstanding of what AoA is, though.

It baffles me that Rob apparently has articulated well enough what AoA is...but at the same time thinks a 757's AoA can be -15 degrees in level flight. What we are looking at doesn't represent AoA, by definition, but an engineering value used by all the systems that require this input.

Since this matter is enough to start another PFT "shake up", I'll look in the Component Manual for the vane and see what -15 degrees is relative to the index point.

reply to post by 767doctor


No, Rob is saying that the AoA angle displayed is based on that -15 degree figure.

I disagree because if that were the case there would be notes stating such in the generic Boeing DFL and there are none, and indeed the calibration units in the custom DFL are degrees.