DB Cooper

[Sluggo_Monster 0]

All,

Well, I go away for a few hours and you guys let the gypsies (Please, don’t call the PC Police!) take over the castle!
Here are some responses to comments of the last 14 hours:

Quote

georger

Well then I must have misinterpreted everything that
has been said during the last several months because
I thought 'flexibility' in the flight path was an issue.

The flight path (FP)… NO Wait! The actual flight path(AFP) is a major issue (for me). Not that there isn’t documentation, but because 37 years later the technical details behind its determination are occult. The best source I have is Ckret, and frankly, he wasn’t there and he doesn’t think it is very important. The most documented segment of the FP is the LZ area. Why am I so obsessed with the AFP? Because, I have my own agenda to push. Without boring you all (that’s not Southern, Y’all is Southern) the gist is that I think the Cooper profile (as currently described by the FBI), is wrong. I think he was smarter than the FBI thinks. I think he knew where he was at all times. To prove it, I need more historical data. But, back to the FP, there was a minimum of these individuals involved in its determination: A USAF Officer responsible for radar at McChord, the Meteorology Chief at NWA, A representative from The Boeing Company who was an expert in free-falling bodies (human), a NWA Pilot who was an expert parachutist, and a senior management representative from NWA.

I wasn’t there when they made all the decisions about the FP, however as an expert in Human Performance Improvement, I’m sure if I was there I would have seen mistakes being made. Do I feel that the mistakes were major? No!

This is Conjecture:
I think it went something like this (Highly simplified).
Flt Ops to Flt 305 Transcript says: “19:48 Pilot reports beginning climb (from 7,000 ft to 10,000 ft.” Data from FDR says: “Climb started at 19:47” Team says: “Hmmm”
Flt Ops to Flt 305 Transcript says: “19:53 Pilot reports plane level at 10,000 ft.” FDR says: “Plane level at 19:52.” Team says; “Hmmmm the transcript to real-time difference is about (minus) 1 minute.”
End of Conjecture:


Therefore, there is probably no major inaccuracies in the timeline with regard to actions that have corresponding “real-world” physical parameters associated with them.

However, there is one area where there was no corresponding real-world parameter, the “pressure bump” (Not the oscillations, the bump.) It was determined, not by a watch on a pilot’s wrist, not by a bit-bucket on an FDAU, not by a corresponding radio or TTY communication, but by an estimate from one (or both) of the pilots. They estimated the time from the last call back to Cooper (20:05) to the bump as 10 to 15 minutes. Does this mean they didn’t know where the plane was at 20:11? No, they knew where the plane was at any time (within the previously discussed tolerances). They did not know exactly when he jumped. So the issue boils down to, not the flight path, but the “bump” time. Does this help clarify my position?

Quote

snowmman
It'll be interesting to see what data Sluggo decides is valid for this period. None of the other data matters much without this period. So Sluggo needs to put something?

Rock, paper, scissors make about as much sense as asking Sluggo.
I don’t know how to improve on a pilot’s (or pilots’) estimate of time. But, I have flown (as in piloted) in that area at night (never in a 727) and I know that if the pilot says “we were not to Portland but difinetely in the suburbs” it was probably around 20:14 to 20:15”. [Ref: Attached Photo, Pilot View 727-1.jpg]

If they were at 10,000 and is straight and level flight (the crab angle was only about 5 degrees) the pilot could not see objects on the ground that were less than Seven miles away (with the obvious assumptions made). [X = 10,000/Tan 15 deg. = 37,320 ft. = 7.06 SM] So (in 1971) Vancouver would give the light pattern that I woild call the suburbs of Portland, so (in my opinion) they were at least 7 SM north of Vancouver.

Quote

snowmman
Sluggo seemed confident of all of this, but I didn't really understand why.
(edit) it seems like maybe he's just using a constant delay of 1 minute that the FBI came up with. In that case I would say the FBI number is probably just a guess, and should really be a range.

I read a lot… think very slowly… and pay attention to what I read

Quote

georger
So who is doing the typing?

It had to be one of these:
Flt-Ops KC = NWA Flight Operations in Unknown Location (possibly Kansas City)

Flt-Ops MSP = NWA Flight Operations in Minneapolis, MN

Flt-Ops SEA = NWA Flight Operations at SEA

My best guess is: Flt-Ops MSP = NWA Flight Operations in Minneapolis, MN

Quote

georger
I ADMIRE THE HECK OUT OF SLUGGO.

Okay georger… let’s get one thing straight… here… and now, damn it!

Flattery will get you everywhere!
One other thing:
I think we (this community) can make a significant contribution to the solution of NORJAK. We have to work together. We each have our own agenda (and I see nothing wrong with that), but if we are “intellectually honest” we can still figure some of this out.

As much shit as I give Ckret, he is the best thing that has happened to NORJAK in the last 20 years. He has made it clear that he cannot release “FBI Work Product” (so don’t hold your breath waiting for a copy of an interview with Tina), but he will answer our questions and supply data. Please… please…. PLEASE… put pressure on him to get me the info I need to complete my FP Model. I think it will be a helpful tool for all of us. It won’t solve the case, but (like my time-line”) it will make visualizing what happened, when it happened and where it happened much easier. Also, let’s try to get Ckret (yes, Ckret I know you are busy) to do some “roll-ups” of the FBI work product (something acceptable to the Bureau) so we can gain information. I will make a space on my web site to act as a repository for such information (un-altered).

Thanks for listening to the ramblings of an old man who has nothing better to do that pour over the minute details of a 37 year old crime.

Sluggo_Monster

Web Page
Blog
NORJAK Forum

[snowmman 3]

I was just thinking that the "XXX" sequence used to indicate a typo in the transcript can tell us something about how this whole TTY thing worked.

The keyboardist would use "XXX" when he "saw" that he made a mistake. He wouldn't just visualize "mistake" in his mind.

So the local TTY console has keyboard and a roll of paper. This is the local transcript, but is probably not the transcript we have, cause as Sluggo notes, the one we have has at least one interspersed message not connected to NORJAK.

It's unclear whether the local TTY console is generating the timestamp and date info when the operator hits send, and includes it as part of the message.

GUESSING: I guess not, since it would then be skewed by the transmission delay at the final printer it gets sent to. Then you would have messages at the final printer that potentially have out of sequence timestamps. I'm thinking of the case of many different keyboard guys sending messages from different places.

So, let's say the timestamp is created at the final printer when it receives a message.

Then the "delay" is how long it takes to type the whole message locally, hit send, and have it be transmitted at the TTY baud rate, probably 45 baud to the final printer. (maybe 60 wpm)

So the delays from utterance on radio would be
1) mental cognition of audio to fingers
2) typing till a full message is created
3) hit send and see TTY baud rate limited transfer
4) it starts printing at the final printer. When the full message is printed a timestamp is printed.

This seems like a variable length process to me, so working back from the timestamp to the radio audio output, it seems we need to allow for a range of time, depending on message length, errors, and possible variable TTY transfer delays due to congestion?

It's so fuzzy, that rather than analyzing each message, we maybe could just assign a range of time.

But saying "1 minute" doesn't seem perfect to me.

[Sluggo_Monster 0]

snowmman,

I disagree on one point:
No baud rate other than the operaters fingers.
As I understand it each TTY was on an open line. When terminal "A" hits a key it registers on each terminal "B" , "C", etc. on the circuit (including "A") at the speed of electrons (sonewhere close to 186272 miles per second).

No modem, no CPU, just a telegraph line.

Now I remember in about 1974 some terminals that were modified TTYs that had an acoustical modem attached so you could communicate with a main-frame, but that wasn't TTY. That was telephone.

Any experts out there? 377? Guru312?

Sluggo_Monster

Web Page
Blog
NORJAK Forum

[Ckret 0]

I never thought about this, untl Sluggo just referenced it, but another clue we have that the oscillation and bump were two different incidents is the visual of the crew. At 8:11 when the crew reported the oscillation there is no way they could see the lights of the suburbs.

As far as Cooper knowing where he was, I think Uncle Sluggo is hittin the sauce again. I took the day off, but Monday I will scan the item you want. If I can stop what i am doing and donate four hours to the Weber mess I guess I can give some in a direction that may go somewhere.

[snowmman 3]

re: no modem
excellent point (if true!) Sluggo.

The interesting thing from all this, is that for short messages, the delay may be <1minute?

But I like your idea of saying that given all the fuzziness, that maybe we should work back from other information for predicting a jump site.

or maybe have two predicts, and use them double-check each other. I'm not sure they can "refine" each other.

But it's good insight. Can we triangulate from any other ways of going from some kind of info to predicted DZ? Are there just these two ways?

[Sluggo_Monster 0]

ooopppss

"Earlier Teletype machines had 3 rows of keys and only supported upper case letters. They used the 5 bit baudot code and generally worked at 60 words per minute. Teletypes with ASCII code were an innovation that came into widespread use in the same period as computers began to become widely available."

So there was a transition period that resulted in a sort of baud rate.

Sluggo

Web Page
Blog
NORJAK Forum

[Sluggo_Monster 0]

Ckret,

Thanks. Remember if I'm wrong... I owe you dinner. Or would that constitute a gift of greater than "nominal value"? [Big, silly, shit-eating, thank God I don't work for the Guv'ment anymore GRIN]

(Uncle) Sluggo

Web Page
Blog
NORJAK Forum

[snowmman 3]

I actually have a hazy memory of a TTY device back in college. I think we used it for printing output on job status or printer status, in the local terminal room for the mainframe. It was always chattering away.

they had round keys that went straight up and down?
it wasn't like typewriters. Hard to type fast.
I seem to remember a yellow roll of paper

web says asr-33 was most common. attached some pics

[snowmman 3]

nice video here
http://www.youtube.com/watch?v=E4IztV7M3jI

you can get a feel for the typing speed of the asr-33...
and how it automatically types
(it's hooked to a computer in this case, but doesn't matter)

Hey close you're eyes and maybe you're at FLT OPS on 11/24/71 with the TTY rattling away.

I had forgotten about the bells! brings back memories. Dinging the bell on like a decwriter used to be our fun...you could get the whole terminal room dinging.

[snowmman 3]

here's a nice video of the SAGE WDC display and the vacuum tube stuff..(from the Computer Museum)
http://www.youtube.com/watch?v=o7ssZyRl6MU

cool! now you're tracking flight 305 too!

hey they show the cigarette lighter/ashtray ..

[377 21]

Quote

As for the flaps at 15 degrees, now i am confused. In the crew interviews they listed Coopers demands as I laid them out. Could it be that when a 727 flys at 10,000 feet with wheels down you would only fly at 15 degrees. If so, maybe they reported 15 degrees by default.

As for the RTTY time stamp, the person communicating with 305 kept a handwritten log with time entries at each line. I'll compare the two logs to see if there is any difference

I wondered about this too some time back. From my memory of a 727-100 flight manual (same basic plane as the one Cooper was on) you could fly gear down at 10,000 ft without any flaps, BUT... if you impose an airspeed constraint then you might need flaps, the lower the speed, the more flap deflection you would need to maintain level flight. I'll get out my manual and confirm over the weekend. There is an upper speed restriction if gear is extended, but it is not an issue here since at all times of interest the 727 was flying at speeds far below the gear extended limit.

2018 marks half a century as a skydiver. Trained by the late Perry Stevens D-51 in 1968.

[377 21]

Quote

snowmman,

I disagree on one point:
No baud rate other than the operaters fingers.
As I understand it each TTY was on an open line. When terminal "A" hits a key it registers on each terminal "B" , "C", etc. on the circuit (including "A") at the speed of electrons (sonewhere close to 186272 miles per second).

No modem, no CPU, just a telegraph line.

Now I remember in about 1974 some terminals that were modified TTYs that had an acoustical modem attached so you could communicate with a main-frame, but that wasn't TTY. That was telephone.

Any experts out there? 377? Guru312?

Sluggo_Monster

Back then TTY machines were synchronous electromechanical beasts and used Baudot coding, not ASCII. There was a speed limitation. Common speeds (set by gears linked to a synchronous AC motor) were 45.45, 50, 75, 100, 150 and 300 baud. The vast majority of ones running in the late 60s early 70s were running below 100 baud. The common TTY speeds in the early 70s would probably have been equivalent to about 50 words per minute typing.

2018 marks half a century as a skydiver. Trained by the late Perry Stevens D-51 in 1968.

[377 21]

from Wikipedia:

Speed, intended to be roughly comparable to words per minute, was the standard designation introduced by Western Union for a mechanical teleprinter data transmission rate using the 5-bit baudot code that was popular in the 1940s and for several decades thereafter. Such a machine would send 1 start bit, 5 data bits, and 1.42 stop bits. This unusual stop bit time was actually a rest period to allow the mechanical printing mechanism to recycle. Since modern computer equipment cannot easily generate 1.42 bits for the stop period, common practice is to either approximate this with 1.5 bits, or to send 2.0 bits while accepting 1.0 bits receiving.

For example, a 60 speed machine is geared at 45.5 baud (22.0 ms per bit), a 66 speed machine is geared at 50.0 baud (20.0 ms per bit), a 75 speed machine is geared at 56.9 baud (17.5 ms per bit), a 100 speed machine is geared at 74.2 baud (13.5 ms per bit), and a 133 speed machine is geared at 100.0 baud (10.0 ms per bit). 60 speed became the de facto standard for amateur radio RTTY operation because of the widespread availability of equipment at that speed and the FCC restrictions to only 60 speed from 1953 to 1972. Telex, news agency wires and similar services commonly used 66 speed services.

2018 marks half a century as a skydiver. Trained by the late Perry Stevens D-51 in 1968.

[snowmman 3]

I'm thinking about how we won't be able to resolve the data to a fine enough resolution, and rather than getting into endless debates that can't be resolved, we can target a goal that captures all information...

1) a jump window that's 100% confidence..i.e. a window where it's guaranteed Cooper had to jump somewhere between these two points. This will be a wide window. I don't even know what it would be, right now. That's wrong. I should know.

2) A window that represents a best attempt at narrowing the window....i.e. narrow it until everyone starts yelling NO!...so it's a gut check window of some sort...fuzzily defined by "seems right"

Now if we have two methods or predicting jump zone, then that creates 4 windows.

I think it's important to resolve this before grinding the numbers. I think a big "human performance" issue, back in '71, to use Sluggo's thinking, was not tracking confidence intervals along with the numbers. So all numbers were considered black and white when they were transferred to personnel who didn't know where the numbers came from.

By including some notion of confidence intervals, we can propagate "quality" along with each number we use, so if we re-calculate with the number down the pipeline, we can see how the re-calculate should affect a range prediction. Most of the calcs are linear, so there's no strange effects, but hey! I'm waiting for someone to introduce calculus here!

[Orange1 0]

Quote

Thank for verifying my James Earl Ray story. I may not have understood it within in its entirety and still do not, but at least it did happen and you were able to find that it did happen. The FBI told me it never happened.

If you would be so kind pm or post the reference you found for my records. Thank You, Orange.
I am in your debt.

I'm not sure what you don't understand about it? Ray got his fingerprints changed at Jefferson, but that was as far as it went. I'd be careful about saying things like the FBI told you it never happened, because we've all seen ckret post on here that changing prints at Jefferson couldn't have affected the integrity of the master prints that the FBI keeps. The story backs that up absolutely, so the reference is not going to help you one iota - in fact, it will prove the opposite of what you want - unless you insist on using it out of context, of course. But it's hardly hidden away anyway - just do a google search.

Some people have been, um, less than complimentary. While I do not actually think that you suffer from a mental disorder, I do think the kindest advice anyone could offer you right now is to find a therapist - not because you are mad, but because you need to find out why you have this obsession with proving Duane was Cooper against overwhelming evidence to the contrary.

Skydiving: wasting fossil fuels just for fun.

[quade 3]

Quote

As for the flaps at 15 degrees, now i am confused. In the crew interviews they listed Coopers demands as I laid them out. Could it be that when a 727 flys at 10,000 feet with wheels down you would only fly at 15 degrees. If so, maybe they reported 15 degrees by default.

Certainly not by "default"; they aren't going to deploy themselves to 15° automatically. One of the pilots would have to take his hand, put it on the control and set it to 15°.

That said, a fairly normal thing to do for a pilot would be to fly the airplane as close to normal as possible. Being asked to fly; below 10,000 feet, at lower than cruising speed with the wheels down, is exactly what every 727 does when it's configured on decent for landing. Further, wheels would normally be extended at the "outer marker" (the navigation marker five miles away from the runway) and it would be my guess that a change in flap setting is normally called for as well.

In my estimation, the flap setting was probably just input without a whole lot of thought, was mentioned in a debrief and is now being blown way out of importance by people looking at minutiae.

Think of it this way. Let's say you owned muscle car from the '60s; 500 hp, 4-speed manual transmission, capable of going very fast. You're carjacked by someone who, as a crazy option thinks that his escape -may- be made just by hopping out of the moving car, so he tells you to only drive 30 mph.

What gear are you driving in?

Further, does it matter?

quade -
The World's Most Boring Skydiver

[Ckret 0]

By default I meant that when the pilots fly with the wheels down they always fly at 15 degrees so when they reported Coopers demands they just said flaps at 15 degrees, not flaps down. I realize they didn't "miracle their ass down"

Like asking for a kleenex to blow your nose.

To a large degree it doesn't matter, however, there is not much on this case to draw from so every little bit counts in some way.

[JSBIRD 1]

I'll wait for 337 to return with the 727-100 flap and gear speeds before commenting further, but I will say that if Cooper requested 'any' flap setting other than "0", he likely had the aircraft speed reduced to a manageable exit speed for someone with much experience.

359

"Now I've settled down,
in a quiet little town,
and forgot about everything"

[snowmman 3]

some random pics I had gathered up.

First is a black/white satellite image that includes the money find location at the top.

I like it because the channel area at the top of Caterpillar Island is almost choked off. Might give georger some thoughts on water flow/eddy currents. Also, I think you can intuit water depth a little right near the money find, compared to the rest of the Columbia.

Also note labelling of current man made structures. You can see all the houseboats and dock stuff in the channel in the Kadows Marina area.

The current Shillapoo Boat Ramp (public access I believe) is also labelled.

The lower channel is labelled "do not attempt passage"
But I found a hand drawn map people used for an annual get together, that shows "with care" you can navigate that channel if you do it correctly. (plus I like how it looks like a treasure map ! :)

I included that also...to stimulate georger's thinking.The Quinn's Cove stuff is the south end of caterpillar island. Remember, it's not the money find location for those but maybe relevant.

Interesting note on left side of the hand written map that says tides affect the water levels and currents. Surprising, if true, this far from the ocean.
Again this is current. Not sure what 1971 looked like.

(edit) georger had wondered about the opposite shore from the money find. You can see that Reeder Beach (the RV park) is over there. It's used a lot and has been there a long time (pre Cooper I believe).
My prior post included commentary on when spoils were last used on that beach ('60s?). One can surmise that the heavy human traffic would have served as a "excavation crew" over the years to discover anything buried as shallow as the Ingram find was? Just guessing.

[quade 3]

Quote

By default I meant that when the pilots fly with the wheels down they always fly at 15 degrees so when they reported Coopers demands they just said flaps at 15 degrees, not flaps down. I realize they didn't "miracle their ass down"

I didn't assume you did. I did, however, think some might be confused and think that the gear lever and the flap lever might be somehow interconnected and that by deploying the landing gear the flaps would automatically deploy as well.

There was at least one small aircraft in the past that automatically deployed the landing gear if the airspeed dropped below a certain limit. This actually turned out to be a horrible idea.

quade -
The World's Most Boring Skydiver

[snowmman 3]

I was quizzing Sluggo on what kind of data might be on the flight recorder that was used in conjunction with the radar data to generate the '72 map, cause I couldn't envision how the radar data could be tweaked.

He replied with a list of probable events. I just noticed the last he listed was

* Time of each radio transmission either to or from air traffic control.

If this was true, and the data was available somewhere, it would eliminate some imprecision in the transcript log for the exact time of various radio transmissions. If we also knew how skewed the FDR time was relative to the transcript time, which we're using as the "golden" standard I guess?

[snowmman 3]

This may have been obvious to everyone, but I just realized it. I like to belabor the obvious to make sure we all agree.

I was reading an article from way back, that said the lite went on, and it implied that meant the stairs were fully extended. I thought, well that's wrong, and was musing about whether FBI agents back then mentally connected it that way, or the writer.

Then I realized WE KNOW FOR SURE it's wrong, because of the 1996 case where the stairs opened in flight. They had aft stair lite on in that case. Also the '95 case where I didn't grab all the info, but it was while the plane was landing.

i.e. we have testimony that tells us about inflight stair opening to just 2' (or less?) and the lite being on.

In the 1996 case, we know the stairs opened like 2' based on the aero.com summary, which is confirmed by news reports from that day.

So, my thinking is, the lite is on when the stair is unlatched. And as we know, too much air pressure will keep it closed (Cooper's complaint about getting the stairs down was after the aft lite when on, correct?)

So: in terms of timeline, the "aft lite on" message provides no data to us, other than leading to other things.

When the plane is on the ground, and the stair deployment is handled by gravity, the aft stair lite on might be quickly followed by full deployment.

So: I suspect there is nothing, other than maybe oscillations that tells us when the stairs are open far enough to jump.

People may say "what about the placard in Toutle". Well maybe aft lite on is useful for resolving that. We can say the placard should not have gone out before the aft lite on..It's small enough that it may have gone out thru a 2' or less gap.

(edit) Maybe we can say that since oscillations were not reported at initial "aft lite on" that means that stair opening to just 2' is not enough to create the reported stair oscillations. So it was not a simple depressurization event. Although, we're not sure we got a 2' opening at aft light on. It is interesting that there's a substantial time gap between aft light on and oscillations. I think it justifies the Pucker Walk delay idea Ckret initiated.

[377 21]

Quote

Quote

By default I meant that when the pilots fly with the wheels down they always fly at 15 degrees so when they reported Coopers demands they just said flaps at 15 degrees, not flaps down. I realize they didn't "miracle their ass down"

I didn't assume you did. I did, however, think some might be confused and think that the gear lever and the flap lever might be somehow interconnected and that by deploying the landing gear the flaps would automatically deploy as well.

There was at least one small aircraft in the past that automatically deployed the landing gear if the airspeed dropped below a certain limit. This actually turned out to be a horrible idea.

Quade is sure right about that being a bad idea. Poor inexperienced pilot gets too slow, starts to stall and then his gear drops making things MUCH worse.

Some pilots put autofeather in the same dumb idea category. Autofeather is armed prior to takeoff and automatically feathers the propellor on an engine that loses power. It has its advantages and disadvantages. I flew right seat in a Learstar (modifed go-fast Lockheed Lodestar, 300 mph plus) that had autofeather. The pilot cursed its very existance and NEVER armed it despite it being recommended. He said: "nobody but me ever makes a feather decision, period." He thought even an engine on fire and losing power might be contributing some thrust and you don't throw away ANY power if you are in a critcal flight situation. What he was worried about is max TO power on one engine and an unexpected autofeather of the other resulting in MAJOR thrust assymetry during takeoff.

Here is an idea that might have saved some lives: automatic take off flaps settings deployed after an engine start and with landing gear "squat switches" activated. More than one airliner (including a 727) has crashed trying to take off with zero flaps. The pilots and FE got distracted during checklist reading and skipped it, more than once.

A squat switch changes state when the landing gear strut is compressed indicating ground contact. The spoilers you see pop up above the wings immediately after touchdown are triggered by the squat switches. The pilot arms them during approach. They activate when the plane puts some weight on the gear and the landing gear strut compresses.

I always thought Cooper's specific request for 15 deg flaps indicated a lot of plane knowledge. Now, I am not sure if he made the request. Sluggo, is Cooper's flap request in the fact or fiction bin?

2018 marks half a century as a skydiver. Trained by the late Perry Stevens D-51 in 1968.

[Orange1 0]

Quote

I always thought Cooper's specific request for 15 deg flaps indicated a lot of plane knowledge. Now, I am not sure if he made the request. Sluggo, is Cooper's flap request in the fact or fiction bin?

The transcript clearly states that Cooper requested to fly with gear down and flaps at 15 deg. (and, btw, that after underway all lights to be turned out in aircraft - which sounds like someone familiar with a night jump?)

Skydiving: wasting fossil fuels just for fun.

[georger 197]

Quote

Quote

Once air rushes out of the plane
it probably equalised, unless some change in altitude.

But I have always wondered how far into the plane
turbulence was a factor. You skydivers would know
something about this -

As stated in numerous posts before, the aircraft was not pressurized...Therefore, there would be no 'rushing out of air' from the plane when the rear door was cracked open.

Also, stated before, there is no turbulence at the back of the plane with the door open. Or at least, very little that would perhaps remove a paper placard poorly affixed to the open door.

ltdiver

__________________________________________

Yes, the plane was not pressurised for altitude, but
there is a pressure difference between inside vs outside due to air flow around the aircraft. Hi velocity air outside vs. no-velocity air inside equals a pressure
difference, ie. the Bernoulli affect. True, the pressure
difference is least at the tail (underside) of the aircraft
but the minute you open the rear door, air from
the inside is drawn out and pressure equalises. Then, you have a more steady state. This is basic physics.