Split from: More trouble in LC paradise. JohnDoe takes his research and goes home

[Arkan_Wolfshade]

Since the original thread is fragmenting into something resembling a jigsaw puzzle I have spun this thread off for Skeptic4Sure, or whoever else wants to present it, to present JohnDoeX's FDR analysis for scrutiny.

[Arkan_Wolfshade]

This appears to be the relevant post It seems the JREFer's dont want to post this on their site and debate the facts. Mainly because it shows how one JREFer's Calculations have blown up in his face working back from the impact hole. I have asked many JREF'ers to post this on their site and they refuse. I wonder why... hmmm...
Alternate Analysis working back from impact hole Final Draft: New Elevations/Calculations from the USGS (also posted on pilotsfor911truth.org)


Pole 1 - 43'MSL ground elevation + 31.5 pole impact height = 74.5 MSL Total height above sea level.
Pole 1 aircraft height - 87'+38'ground elevation of pentagon+10.38' imapct hole height = 135.38 MSL.
Aircraft was 60.88 feet above Pole #1.
(not sure of exact reported impact height so i'll use the full 40' for pole length)
Pole 2 - 43'+ 40' = 83' MSL Pole Height
Pole 2 Aircraft Height - 74' + 48.38 = 122.38 MSL
Aircraft was 39.38' above pole #2
Pole 3 - 42'+ 40 = 82' MSL pole height
Pole 3 Aircraft Height - 60' + 50.38 = 110.38 MSL
Aircraft was 28.38' above Pole #3
Pole 4 - 42' + 40 = 82' MSL
Pole 4 Aircraft Height - 50' + 50.38 = 100.38 MSL
Aircraft was 18.38' above Pole # 4
Pole 5 - 41' + 40 = 81' MSL
Pole 5 Aircraft Height - 39' + 50.38 = 89.38
Aircraft was 8.38' above Pole #5
Numbers in black bold above represent the height above impact hole at pentagon based on descent rate of 66 ft/sec and distance from impact hole at pentagon to pole with a forward speed of 784 ft/sec.
To draw your own line.. go here... USGS Seamless Data Distribution
As a reminder, the above an alternate analysis based on working back from the pentagon impact hole. For a more accurate Aircraft height, please visit here for the Flight Data Recorder analysis mid page.
http://www.pilotsfor911truth.org/pentagon.html
http://s15.invisionfree.com/Loose_Ch...howtopic=15306

[Arkan_Wolfshade]

To clarify, for purposes of this thread I think JDX's experience, or lack thereof, is moot. I do not see any reason his claims can not be judged on their own merit.

[Arkan_Wolfshade]

From original thread:
Welcome Skeptic4Sure, I hadn't realized his analysis was being discussed in this thread too.

The important questions are these -- since the FDR measurements are based on pitotstatic measurements (sampling air data), and the actual altitudes of the light poles are not well described, there are numerous potential sources of error:

1. Was the altimeter calibrated?
2. To what density altitude was it calibrated?
3. What corrections were applied for the aircraft speed?
4. How does his derivation account for aircraft attitude?
5. How much offset is there between probe data and the physical bottom of the aircraft?
6. How much "time smear" did he account for in the FDR records?
7. How accurately is the altitude of the poles known?
8. How were these answers computed?
9. How were these errors combined?
10. What is the final error estimate of aircraft altitude at the points mentioned above?

Since the entire argument rests on discrepancies of under 15 meters, the requirements for error analysis are very tight. Convince me.

[Arkan_Wolfshade]

From original thread:
1. Was the altimeter calibrated?

Of course.. it was in an airplane.. right? The altimeter shows 10 feet too low at take-off. We'll within FAR regs.

2. To what density altitude was it calibrated?

Altimeters arent calibrated to density altitude. They are calibrated to pressure altitude at sea level. Then a "Kollsman" window is added to adjust the altimeter to local pressure to get a proper altitude above sea level. (i got the spelling right.. how bout that.. )

3. What corrections were applied for the aircraft speed?

None. These pitot-static systems/Air data computers are in all types of aircraft with all different types of Vmo's. Also. What good is an FDR that doesnt take accurate measurements outside the aircraft envelope. This is where most accidents happen.. right?

4. How does his derivation account for aircraft attitude?

See above.

5. How much offset is there between probe data and the physical bottom of the aircraft?

airliners.net. Go look for the static port. Its on the bottom side of the fuse...

6. How much "time smear" did he account for in the FDR records?

None. Official impact time is 09:37:45(Re: NTSB Flight Path Study of AA77). Recorder stops at :44 . Vertical acceleration column is recording 8 times per second. It is the only column that records up to supposed impact. They omitted all other parameters such as altitude for the final second. Im begining to understand why.

7. How accurately is the altitude of the poles known?

Very. Look at the USGS site linked in the "Alternate Analysis" working backwards from the impact hole. The analysis which shows how shoddy Billzilla was when he worked backwards from impact hole. I have also compared Google Earth to Jeppesen Airway Manuals in and around the DCA and IAD Airports. Error showed to +/-5 feet. However, i used the USGS for this analysis which is more accurate.

8. How were these answers computed?

The math is shown. The references are shown.


9. How were these errors combined?

There arent any errors. The analysis is a hypothetical in which Billzilla tried to use (working back from impact hole based on descent rate). I am now showing him the proper way to do the analysis.

10. What is the final error estimate of aircraft altitude at the points mentioned above?

None. It is based on Vertical speed and not indicated altitude. If you want to get a more accurate altitude data, please see full analysis on pilotsfor911truth.org/pentagon.html.

Again. .the Alternate Analysis is a HYPOTHETICAL working back from the impact hole based on descent rate. If you increase that rate due to speed errors and/or altimeter lag, the height above the poles increase. This analysis was to show Billzilla the propr way and accurate elevations in working back from the impact hole at the pentagon.

[Arkan_Wolfshade]

R. Mackey responds
Welcome back, Skeptic4Sure. Thank you for finding answers to my questions. I'll assume these come straight from John Doe X, though you didn't specify.


Originally Posted by Skeptic4Sure
1. Was the altimeter calibrated?

Of course.. it was in an airplane.. right? The altimeter shows 10 feet too low at take-off. We'll within FAR regs.

So if I understand you correctly, you've shown that (a) it was calibrated but not reset after takeoff, where the local conditions were different, and (b) we can expect 3 meters of possible error on top of that. Fine, I'll accept that.


Quote:
2. To what density altitude was it calibrated?

Altimeters arent calibrated to density altitude. They are calibrated to pressure altitude at sea level. Then a "Kollsman" window is added to adjust the altimeter to local pressure to get a proper altitude above sea level. (i got the spelling right.. how bout that.. )

I'll also accept a calibration to pressure altitude, but you didn't say what that altitude was. I'm assuming the calibration was given for the airport of departure as listed in 1. above.


Quote:
3. What corrections were applied for the aircraft speed?

None. These pitot-static systems/Air data computers are in all types of aircraft with all different types of Vmo's. Also. What good is an FDR that doesnt take accurate measurements outside the aircraft envelope. This is where most accidents happen.. right?

Actually no, the vast majority of aircraft accidents happen on takeoff and landing, at fairly low speed. Assuming that the probes work beyond Vmax as you suggest here is an argument to incredulity, and wrong. If the probe doesn't work beyond a particular speed, it's because the laws of gas dynamics say it doesn't, not just some flaky spec.

Anyway, your answer is insufficient. Pitotstatic probes are different depending on geometry and placement on the aircraft, and all suffer compressibility errors beyond a certain velocity. I have read elsewhere that Flight 77 was flying faster than the altimeter's rated Vmo, guaranteeing that a correction is needed. This is an error that you have not accounted for, and it can all by itself invalidate your results.


Quote:
4. How does his derivation account for aircraft attitude?

See above.


Again, insufficient. Pitotstatic probes are not terribly accurate that close to the deck. Your claim that it was off by 10 feet at takeoff, and that was within spec, is proof enough of that.


Quote:
5. How much offset is there between probe data and the physical bottom of the aircraft?

airliners.net. Go look for the static port. Its on the bottom side of the fuse...

Attempt to shift the burden of proof noted. I'll assume you don't know.


Quote:
6. How much "time smear" did he account for in the FDR records?

None. Official impact time is 09:37:45(Re: NTSB Flight Path Study of AA77). Recorder stops at :44 . Vertical acceleration column is recording 8 times per second. It is the only column that records up to supposed impact. They omitted all other parameters such as altitude for the final second. Im begining to understand why.

I'll assume you don't understand my question. What I mean is that in the air data system, there is a time lag between the measurement and when it is logged by the FDR. This is caused by the sensor response time, compute time in the aircraft flight computers, and message packet delay in the databus. This will add additional uncertainty, and since the aircraft was descending, make the FDR read high. Rate of descent was about 20 meters per second if I recall correctly.

Your answer says that John Doe X didn't account for this either.


Quote:
7. How accurately is the altitude of the poles known?

Very. Look at the USGS site linked in the "Alternate Analysis" working backwards from the impact hole. The analysis which shows how shoddy Billzilla was when he worked backwards from impact hole. I have also compared Google Earth to Jeppesen Airway Manuals in and around the DCA and IAD Airports. Error showed to +/-5 feet. However, i used the USGS for this analysis which is more accurate.

So you're claiming +/- 5 feet elevation? Call it two meters? I'll accept that.


Quote:
8. How were these answers computed?

The math is shown. The references are shown.


Shown where, please? I haven't seen them here.


Quote:
9. How were these errors combined?

There arent any errors. The analysis is a hypothetical in which Billzilla tried to use (working back from impact hole based on descent rate). I am now showing him the proper way to do the analysis.

Ummm... yes there are. Any measurement has error. If you're criticizing a different approach, please present that work.


Quote:
10. What is the final error estimate of aircraft altitude at the points mentioned above?

None. It is based on Vertical speed and not indicated altitude. If you want to get a more accurate altitude data, please see full analysis on pilotsfor911truth.org/pentagon.html.

Then you haven't done your homework. Bring your evidence here, please.


Quote:
Again. .the Alternate Analysis is a HYPOTHETICAL working back from the impact hole based on descent rate. If you increase that rate due to speed errors and/or altimeter lag, the height above the poles increase. This analysis was to show Billzilla the propr way and accurate elevations in working back from the impact hole at the pentagon.


I'm unaware of the details of you (or John Doe X, not sure who is who here...) and your conversation with Billzilla. I was asking about how you conducted your analysis, not Billzilla's. And you did not show him the "prop[e]r way," because you have not shown that your conclusion -- that the aircraft was too high -- is a statistically significant result.

If your estimate has as little as 20 meters of error, then your results are totally inconclusive. It is clear from your answers that your error is plausibly much higher.

Adding the FDR logged error at takeoff plus the height uncertainty of the poles alone, added directly since these errors are uncorrelated, is five meters right off the top. Since the claimed rate of descent is about 20 meters per second, a time lag of even a quarter second can add another five meters, and this time lag is not unusual, could be higher. We further know the altimeter was not corrected for local conditions, was operating outside its calibrated performance envelope with respect to speed, and was pushed into the ground effect -- ten meters error for these factors is conservative unless you can show otherwise. And there are many more possibilities that you didn't account for.

It's entirely plausible for the FDR to read 20 meters higher than reality under those conditions. It is therefore possible for the FDR data to be consistent with the claim -- backed by evidence of wreckage and eyewitness accounts -- that Flight 77 did indeed hit those poles on the way to the Pentagon.

If you would, please inform John Doe X that he has not proven his case. If he would like to address these sources of error and sharpen up his estimate, this should give him a good place to start.

--------------------------------------------------------------------------------
Last edited by R.Mackey : Today at 02:41 AM. Reason: Responding to speculation that probes work "outside the envelope." Missed it at first.

[Arkan_Wolfshade]

From original thread:
Is the NTSB time synchronised with the FDR time? If not, what is the error range?

Even if the time is correct, an aircraft travelling at 463 knots (the last recorded computed speed in the CSV file) will cover ~29.8 meters in 1/8th of a second.

If you are working backwards from the impact point to extrpolate the position of the aircraft, the above information is absolutely vital, and is one of the reasons why you shouldn't work backwards from the impact point.

If the aircraft location was taken using the recorded latitude/longitude, how where they obtained by the FDR? GPS? If so, how was the GPS error accounted for. Some GPS units can be out by meters and given that the JDX analysis measures things to fractions of a foot, this would seem important to me.

[Arkan_Wolfshade]

From original thread:
Some things to note:

1. The altitude data DOH! is working with is pressure altitude data from the DFDR, not the altimeters. The altimeter setting has absolutely nothing to do with the DFDR altitude data. The DFDR records what QNH was set on the altimeters, so we can then work out what altitude they were seeing in the cockpit. It has ZERO to do with what altitude was recorded by the DFDR. What the cockpit altimeters were reading has nothing to do with altitude above ground level or pressure altitude.

2. I'm told that most, if not all, FDRs have a G-activated "crash" switch to stop them recording after an incident. This prevents vital data being over-recorded. That's why the 'official' impact time is given as 09:37:46.

3. The last valid pressure altitude recorded was at 09:37:44, two seconds before impact. That was 173 feet. Then you have to account for the change in pressure. The standard pressure at which the DFDR is recording altitude data is based on 29.92in/Hg or 1013.2 Mb. The QNH at Ronald Reagan airport was 30.21in/Hg or 1023 Mb. At close to sea level, the rate of change is about 29 feet per Mb of pressure difference. 10 Mb change equals 290 feet to be added to the altitude data (173 + 290 = 463 feet).

4. At takeoff at Dulles, the DFDR pressure altitude was reading between 49 feet & 74 feet (it changes during the takeoff run). Dulles QNH was 30.22in/Hg or 1023 Mb. Again, that is a correction of 290 feet. That means the DFDR was recording the pressure altitude as between 339 feet (49 + 290) and 364 feet (74 + 290). Dulles field elevation is given as 313 feet. If someone can find out (because I don't have the plates) what runway AA77 took off on, and what the elevation for the threshold of that runway is, I can be more accurate with this. For the moment, though, the DFDR pressure altitude appears to have had an inaccuracy of between 26 feet and 51 feet, reading higher than actual altitude AMSL. This error may or may not have been constant throughout the flight, so lets stick with the lowest error - 27 feet high.

5. Assuming there was a constant error with the DFDR, and using the lowest assumed error, the revised pressure altitude two seconds before impact is now 436 feet.

6. The last recorded valid ground speed of the aircraft, two seconds before impact, was 465 knots or 237 metres per second.

7. The last recorded valid rate of descent, two seconds before impact, was 3960 feet per minute or 66 feet per second.

8. The last recorded valid pitch angle of the aircraft, recorded less than two seconds before impact, was 5.9 degrees nose down.

9. So two seconds prior to impact (or 474 metres from the Pentagon), the aircraft was at a pressure altitude of about 463 feet and descending at 66 feet per second.

Someone might do the math to determine about how much G-force would have been encountered to level off the aircraft just above the ground, match it up with the boom gate security camera footage (distance from impact), and also match it up with the light poles.

Correction to my last post - the Flight Path Study of AA77 does indeed give 09:37:45 as time of impact.

You'll have to adjust the distance figures accordingly.

Just a minor correction, the AoA was -4.9 at 0937:44



I dont know about the g-forces invloved, but a gentle takeoff rotation is about 3 degrees/second. I wouldn't be suprised if the AoA flattened to 3.0 - 3.5 from 4.9 between the 937:44 plot and impact.

[TjW]

Actually, I think JDX will find, if he looks, that altimeters are calibrated to a pressure altitude in a theoretical "standard atmosphere". This is sometimes referred to as the "standard day". I've never seen a real one.
I'm not going to look all the details up, but IIRC, that would be +15C at sea level, with a specified lapse rate up to certain break points.

If the temperature was warmer than 15C at the Pentagon, then the absolute altitude displayed by the altimeter was not correct. That's not to say it couldn't be corrected to be closer to reality if one knew the actual temperature, but it didn't seem to be included in the calculations.

Generally the altimeter will read higher than true altitude on days warmer than the "standard day".

This is all part of the U.S. written exam for private pilot.

[weedwacker]

I haven't read this whole thread. But I know pilots check their altimeters prior to take-off to see if its within tolerances. I think I remember hearing something about an animation out there which shows the altimeter was within 10 feet of actual (or True Altitude), at Dulles.

From what I understand, there is a .csv file and an animation? The animation shows True Altitude at departure, but pressure altitude at end of data (wasn't set on descent through FL180). The csv file shows pressure altitude only.

I'm still checking on it.


edit, absolute altitude is radar altitude. I dont think this parameter was available.

[Arkan_Wolfshade]

I haven't read this whole thread. But I know pilots check their altimeters prior to take-off to see if its within tolerances. I think I remember hearing something about an animation out there which shows the altimeter was within 10 feet of actual (or True Altitude), at Dulles.

From what I understand, there is a .csv file and an animation? The animation shows True Altitude at departure, but pressure altitude at end of data (wasn't set on descent through FL180). The csv file shows pressure altitude only.

I'm still checking on it.


edit, absolute altitude is radar altitude. I dont think this parameter was available.

Is this the one you for which you are looking? http://www.mikejwilson.com/911/

[weedwacker]

No, i haven't seen that one. Looks good.

The one I'm thinking of is an animation produced by the NTSB. JDX has it on his site to a youtube video. Just saw it this morning. I'd like to get the whole thing.

I'll post the url after my next post. Need 15. haha

youtube NTSB animation. The comments are funny.

http://www.youtube.com/watch?v=DzR-q0ijbV0

[weedwacker]

number 15

[Bell]

I believe this (http://www.youtube.com/watch?v=DzR-q0ijbV0) is it.

ETA: beaten