Hominid

WRONG. The early version recorded on metal foil on which overwriting was impossible. Attached is from the NTSB document Smokin provided link to.

How is it that you know what FDRs all planes had in '71?

The recorder Smokin's referring to is not the voice recorder.

Years ago in Cooperland the FBI (one Larry Carr) gave the public (via respected researcher "Sluggo") one or two JPEG graphic images of the Washington state part of a flightpath plot. The files are still available at Sluggo's site. If you look closely at the image you can see little pale red crosses which are the markings of the plotted positions. The person(s) who plotted the points used a "thin" red-lead mechanical pencil. If you're not seeing the pale red crosses, and you're looking at an image with red and blue "balls" and a big green line, you are NOT looking at the FBI flightpath scan. The image(s) also show that someone had sloppily annotated times for some of the plotted points, using long arrows to connect the times to the points. These appear to have been done with simple pencil before the pieces of the original paper charts were stuck together. For unknown reason a well meaning person made one or more copies of the FBI flightpath file(s) and used them to creat new graphics, also in JPEG format. Big red and blue stickpin symbols (appearing superficially like "balls") were placed on top of the position plot crosses of the original FBI graphics, and a broad green line was added from "ball" to "ball." The "balls" (and maybe the green line to some extent) totally obscure almost all the original red crosses. And misleading annotations were made regarding the plotted points, indicating that some were plotted and others were only "estimated." This was a faulty attempt to indicate that the TIMES marked for some of the points were the times that were on the original plot and that the other times were determined by extrapolating point-by-point from the ones that were on the original. The plotted positions were actually all equally accurate and equally "estimated." You would kind of expect that the "balls" would be accurately centered on the original crosses, or that the "pin" ends of the stickpin lines would be. But the stickpin ends weren't so placed. And a few of the original crosses were not covered by "balls." So, at least in some cases, the "balls" were not accurately centered on the crosses. And it is impossible to tell the accuracy just by looking at the graphic since the "new" position plots totally obscure the original ones. Add to this the fact that JPEG files have a nasty characteristic of accumulating "smear" or blurring with every time the file is saved after revision. So, if you want to try to analyze anything about details of the flightpath, you would be foolish to use the red-ball blue-ball graphic instead of the original data from the FBI. In fact, a basic of evidence analysis is to analyze the original evidence rather than some derived product someone has done who-knows-what with.

I've attached a copy of a bit of the bottom half of the 65th page of the transcripts pdf (marked "-190-" at the bottom of the page). This is immediately before takeoff at SEATAC. It's telling 305 that "the" wind is 10 knots from 180°. My assumption has been that this would be the wind at the standard altitude (10m or whatever) above some spot at the airport. Does anyone think this might or would have meant otherwise? And, would it have meant magnetic?

I had never checked on where the North Bend station (just N. of Coos Bay) could have tracked the flight, since we've been concerned with N. of PDX. On checking, I see that there are lots and lots of relative low mountains between the station and the flightpath. So, rather than checking for blockage by individual mountains, I used Google Earth to check on coverage of all three stations that would have covered the flight. The way to do it is to accentuate the terrain by a factor of 4/3, mark the radar site locations, then fly the path at 4/3 of the actual altitude (13,333 ft) while searching for your site markers. I found that the North Bend site should have been able to track the flight starting at about the area we're mostly concerned with--Merwin to Vancouver. The Hebo site could have tracked along with the North Bend site to about Eugene. In the intervening path, there would be some triangulation--especially for latitudes between the two site. The North Bend site tracking while the flight was up around Vancouver would not have helped much because the azimuths from the two sites would have been close to the same and the angular resolution from the North Bend site would not have been nearly as good as for the Hebo site. The North Bend site could have tracked south to about Grants Pass. The "Keno" site over by Klamath Falls could have tracked starting at about Myrtle Cr. From there to Grants Pass the flight could have been tracked simultaneously by both sites. The range capability of the radars would never have been an issue. Given high enough siting and target altitudes, they could track out to 220nm.

I haven't found the exact sweep rate. It was 5-6 rpm. The PRF was 360 pulses per second. The resolution really wasn't determined by these. The digital data was not actually produced by the radar. They fed video like was used for the display scopes into a computer-like device. Each time the antenna swept past a target several pulses would go out and bounce back. The device looked at the pattern of returns and computed where the center of the pattern was and at what time. The programming was different for the two antennas involved, one being the dish and the other being the radiating slot "trough" antenna mounted on the top of the dish for transponder tracking. The device computed an azimuth word and a range word. It would then send out a signal on two parallel telephone lines telling a device at the direction center that data was available relating to target "x", then would actually transmit the target data to the center if and when the center sent a signal to request it. They also regularly transmitted data for a calibration target at surveyed location. The range data resolution was basically constant. The position resolution attributable to the azimuth data would reduce with distance from the antenna. That is, the lengths required for resolution, perpendicular to the range coordinate, were bigger as range increased. Not as "fine" at longer distances. As it happens, the N-S and E-W errors caused by the range and azimuth errors would have been about the same in the area we're concerned with. The range error would cause about the same errors for both N-S and E-W, and the azimuth error would as well. The resolution was better the farther south the plane flew until the flight got down to the latitude of Mt. Hebo. Just the opposite of what he told you. The angular resolution would correspond to smaller distance the closer you get to the radar. After passing Hebo's latitude, it may not have degraded because the site at North Bend, OR may have already been able to track the flight even at that point. At some point heading south both sites would have been able to track the flight. The North Bend site was on the top of a high peak with 360° view just like Hebo.

I figured that. Why I said minor point. I just hope that people get to realize that those charts with the red and blue stickpins are not the data from the FBI.

Doesn't mean they WOULD depart much from V23. They were under some control by NWA and some by the wishes of the hijacker. Also, they were not accustomed to VOR airway flying. One might believe this if one did not understand the info on the FBI flightpath charts. If you put in ±.5 minute tolerance zones centered on the plotted points, a line can be drawn through all the zones with no wandering laterally between SEATAC and the Mayfield intersection other than an approximately 6° drift to the NW entirely consistent with them flying by compass and relying on the erroneous wind report given to them just before departure. (Someone at SEATAC told them from 180 magnetic when it was actually from about 140 magnetic.) This put them a bit N of the airport at Toledo rather than at the intersection. I've studied them. It is not evident to me. Seattle center was quite considerate of the crew's plight and said they would monitor the comms with the company. I think the comms just switched over to the company channel and we have never seen transcript of those comms. I'd almost bet the FBI hasn't either. What appears to be random changes in ground distance covered is caused by the fact the positions that were plotted were rounded to whole minutes of latitude and longitude. Unfortunately, such low precision of the data is significant for our problem. It is known precisely enough to know that it did not pass near Tena Bar. It would be simpler to change course earlier if they weren't being distracted by finding a nice quiet corner to get Coop to jump, and if they knew well ahead of time that they were just going to bypass their first alternate landing site. They did not plan their entire flight out ahead of time without regard to cooperating with Coop.

One minor point, Farf. The chart you showed is not the chart from the FBI. Someone took the chart from the FBI, added red and blue stickpin symbols and a big blue line that obscured the plotted points on the original from the FBI, plus some misleading notations about "estimated" positions, then saved it in lower resolution than the original from the FBI. The actual charts from the FBI are at Sluggo's site along with these garbage charts.

I haven't been here while stumbling over myself, but let me offer a possibility since someone now seems interested in the flightpath topic. I say the "official" flight path is the one shown in the plot Sluggo obtained from the FBI and made available on his website. It is not the "red point, blue point" (actually, stickpin symbol) image you analyzed. That image was produced from the FBI-provided original. In producing that image, essentially all the actual plots were obliterated/obscured. The part of the flight path we've been concerned with was produced from data from just one radar site: Mt. Hebo, OR. Two other sites were the sources for parts of the flight over central and southern Oregon. The site at Blaine, WA was too far away since it was at a low elevation and flight 305 was also too low when it was closest to Blaine, just out of SEATAC. The site at Othello had its view blocked by a ridge to its west. The data coming from the radar site came in increments of .088° (1 4096th of a full circle) for azimuth and .25nm for range. If the likely errors were more than two times these values, the designers would simply have decreased the resolution of the data. There is no point in transmitting data to a precision of 1 if the data is only accurate to 8, for example. The direction center at McChord could have smoothed the data to get better accuracy. But the coordinate data used to plot the FBI plot didn't necessarily take full advantage of the accuracy of the data the airforce had available. The data from which the plot was made was rounded to whole minutes of latitude and longitude. So the actual accuracy of the plots is ±.5 minutes of latitude and longitude. This was the source of the incorrect info about the '72 searchzone map to the effect that the accuracy was ±.5 nautical miles (should have been .5nm for N-S, but .347nm for E-W). The positions were determined by computations, almost certainly by the airforce 84th radar evaluation squadron, using the coordinates and associated time stamps obtained from the Mt. Hebo site, plus the surveyed location of the Mt. Hebo site and the earth model of the time. The 84th had, and still has, a detachment at McChord and doing such analyses was part of their official functions and still is today. The McChord direction center just used the SAGE system in their normal function of tracking practically everything bigger than a piper or cessna. They did not analyze the system or data from it. The times would have come from the 84th's output data listing, but someone made a mistake in annotating times onto the charts. The flightpath range is only .7 nautical miles wide. The positions are quite reliable within the limits of the rounding errors, but better accuracy would be very helpful. Unfortunately, the system that produced the data didn't need any better. All they needed was to be able to get planes close enough to see "bogies" visually or on their own radars. The times are a bit more problematic. The times and positions on the FBI charts are reliable if you are aware of the limitations of the data. Analyze the actual charts, not the graphics someone made based on the charts. Not having been there and participated in the decision with the crew, I don't know why. Since you were there, tell us what conditions the crew was operating under. Had they just been considering (with NWA management) whether or not to land at PDX (their first alternate landing site) since they thought they had just rid themselves of Cooper? What was the short path back to V23 south of Portland? It doesn't locate down to yards at seconds, but it would take a monumental stretch to put them near Tena Bar.

Yeah. Those are real "tells." How much of the advertising in Playboy do you suppose were for tobacco products, booze and cigarette lighters? How likely would it be if none of these were represented in the letter if characters were randomly selected? It would "tell" something if the text was cut from the Raleighs ad, a bourbon ad, and a matchbook ad--things that actually matched the case.

You are joking, right? Have you seen anything about this case that is detailed, official accounting? Something accepted by all? Probably not detailed, but the case agent might find something in the file. I think someone "way back" on this or the preceding thread found something that indicated that bank peoples usually bundle together several packets at a time when packaging money. And shortly after the money was found on Tena Bar, an FBI agent stated to a newspaper that all of the found money was from one bundle even though it seems pretty much accepted that there were parts of at least 3 packets that everyone calls "bundles."

I was thinking more of the time period in which the flight slowed down to try to get the guy in back to leave the plane.

Unlikely, unless the chute was really aged, poorly maintained and the A/C was in the max range of deployment speed. BUT, not impossible. Matt Yeah. Not like the chute would have been left spread out in the sun all the time. Is it true that cotton was used to sew panels together?

A really, REALLY old ex-loadmaster told me recently that the chute would likely have blown out, especially if Coop didn't slow to terminal before pulling, because the nylon panels were sewn together with cotton thread which would deteriorate much more than the panels. I have no idea if he has any knowledge of where the chute was obtained. What do you think?

Following is a position plot for an aircraft flying due south at a constant ground speed: Min Lat 0 - + --- --- --- + --- 5 - --- + --- --- --- 10- + --- --- --- + --- 15- --- + --- --- --- + 20- --- --- + --- --- 25- --- + --- --- --- + 30- --- --- + Notice that the vertical spacing of the crosses is not constant. But, the distances covered in equal time periods are all the same if speed is constant. Without a solid understanding of what you see, you could erroneously conclude from this that the ground speed for the plotted positions could not be constant. But it is constant. The latitudes are calculated from a constant speed of 191 knots starting from a latitude of .13 minutes. The plot would be similar for most starting positions and ground speeds. The plot has the "jumpy" spacing irregularities for the same reason as the flight path plot from the FBI has them. If you don't believe the plot is for a constant ground speed, or if you want to have a better understanding of how the spacing irregularity comes about, calculate a string of positions for yourself and plot them using one minute of latitude per text line space as I did. To make it easy for you, one knot is one minute of latitude per hour. So take your speed in knots and divide by 60 to get the minutes of latitude covered per minute of time. Calculate distance (latitude) covered in one minute, two, three, etc. Then plot the positions. But don't use a ground speed that is an exact multiple of 60 knots, for this is the one condition for which the spacing will have no irregularities. And, if you use a ground speed that is close to an exact multiple, you may have to fly quite a few minutes before the irregularity occurs.

When someone says they knew when something happened, they frequently do not mean they knew what time it was when it happened. The words often mean that, when it happened, they were aware of it happening. Notice in your second hand quote of Rataczak, that he didn't say something like, "it was at 00:13 UTC." I've never found any such claim by any crew member. It is this kind of SPECIFIC that would be necessary to narrow down jump time, which is more of a problem than the drift under canopy since we do have some specifics about the drift. And, since you keep repeating the incorrect info, NWA's head of tech ops training was their chief pilot, Paul Soderlind--not "Sortum." edit: Yeah. What Robert99 said.

I think it more likely the reference to McChord was the reply the crew gave him when he wanted to know why it was taking so long. Either way, though, it appears he knew something about McChord and Tacoma. But that does not mean he said anything about either of them upon actually seeing them. edit: The reference to Tacoma in the transcripts (5:26) was 305 saying he knew TCM was 20mi away. Dan may not have said anything about Tacoma. The crew may have been just trying to stop the stalling on the ground.

Georger, I don't know if you were jesting or not. Farflung explained it in his recent post. If I understand correctly, the Toutle on the old chart marked a rail stop that is no longer used. I think probably just the official location of "Toutle" changed. Still people living around the old location.

You're right. I'm looking at an '08 sectional from Sluggo's (which is slightly tilted) and I'm looking at the original flight path plot from the FBI. Maybe your 09 sectional doesn't show rivers and highways? I've reproduced bits of the two charts below. In each, the little circle represents Toutle location.

Looking at the two charts, in the Toutle area, it is obvious that the Toutle circle moved with respect to nearby landmarks, namely the river courses.

Thanks Orange1. You're right. I do it as many times as I can get away with on each occasion.

I'm sure it had to be at the aft end of the cabin. Anywhere forward that would have something like a "room" (galley, etc.) behind a seat would have to have seat row number much less than "18."