Malaysian Airlines MH370 contact lost

[Blake777]

Dr Phillips

QUOTE "If there is only confidential (military) information available, its existence (or confirmed absence of contact) should have been reported even if no details were."

Pardon?

It took the Captain of BA Flight 9 (volcanic ash over Indonesia) eleven years to find out from declassified information that satellite monitoring stations in Australia and Guam were following his flight.

Capt Moody's comment? "We don't know what they're watching."

I'm not sure why you would be expecting anything different in this case if more were known from confidential sources.

[Shadoko]

Originally Posted by mm43

US Patent #6008758 will provide an insight into the mysterious "D1/D2".

The ADIRU also supplies data to the SBSU/SDU, and if certain parameters are exceeded, say momentary loss of the P Channel occurs, the AES may initiate an "I am here" handshake with the SAT.

In the patent link you give, we can read:

Originally Posted by US Patent #6008758

.../...However, in this situation, conventional AES terminals continue to reserve one channel for receiving P Channel signals for the sole purpose of monitoring Doppler shift. P Channel signals are conveniently used for this purpose since these signals are continuously available.
.../...
In general, customers who need only circuit-mode services are forced to order an AES terminal equipped with one more channel than the number of telephone calls the system must support. The additional channel needed is a costly, wasted resource.
.../...
One method that has been used in an attempt to address this problem is to pre-compensate the transmit frequencies based on a calculated Doppler shift rather than a measured Doppler shift. The method calculates the Doppler shift using the satellite location, the aircraft location, and data which defines flight characteristics (e.g., velocity, ground speed, heading pitch, roll, etc.). Much of the data needed to calculate the Doppler shift comes from the aircraft's Inertial Reference System (IRS). This method has been difficult to implement due to the many different types of IRS systems which are made by many manufacturers and provide inconsistent flight data. Thus, it has been difficult to produce an AES terminal which calculates Doppler shift and is compatible with different aircraft.
.../...
It would also be advantageous to provide a method for measuring Doppler shift in an AES terminal or communication system independent of P Channel inputs..../...

So, if MH370 used the "precompensation method", the wobbling of 3-F1 can't be included and thus the Doppler is not "perfectly" compensated as it could have been if the P channel was continuously monitored, the "total" Doppler measured and then used to transmit a frequency the sat will receive as the right one.
In this case, it could be useful to calculate the Doppler from the fixed points where MH370 might have been located at the times published on the Inmarsat Doppler graph. Could this may also explain why all the values on the graph are positive?

The words "This method has been difficult to implement due to the many different types of IRS systems which are made by many manufacturers and provide inconsistent flight data" could also explain why the tests made on other flights to improve the data have been made on B-777s.

[Derfred]

Do you really think they wouldn't have bothered searching for AF, or indeed MH if they had the black box data already? Try telling that to the relatives of the deceased...

You are missing the point, they would have known WHERE TO SEARCH.

On the contrary, I was proposing that instead of spending a lot of money on SAT bandwidth and onboard equipment for live FDR/CVR transmission, that money (or less) would be better spent mandating frequent ADS-C via SATCOM on all aircraft.

Then they would know WHERE TO SEARCH.

It would also improve ATC efficiency, reduce airspace congestion, improve day-to-day safety, a whole host of tangible benefits.

[Ian W]

If you read the rest of my post that you quoted you will find that is just what I suggested. ADS-C contracts of reports once every minute were suggested by BEA after AF447. if there is ever agreement on the use of ADS-C EPP there will be even more information available and as a standard 'out-of-the-box' system. Add an ACARS over SATCOM that reports exceptions which AF447 did have or health monitoring systems over SATCOM, all existing capabilities and a reasonably coherent picture of what is happening on the aircraft is continually available.

All of this could be done now - and by the generosity of INMARSAT it could be 'free' if considered tracking information.

[mm43]

@ Shadoko;

Re: patent #6008758;
The method calculates the Doppler shift using the satellite location, the aircraft location, and data which defines flight characteristics.

If you are going to use the AES location and ADIRU data to cover its manouvering at any point in time, then the logical conclusion is that the AES will do the same for the SAT using a look-up table for the ephemeris data associated with the same time.

The SDU knows which satellite it is coupled with, and to my mind there is no reason for the above not to take place, i.e. the Sat is not about to do a "barrel roll" or look over the neighbors fence as it goes about its predefined orbital parameters.

[Robin Clark]

BFO and Doppler........

So this is the updated chart , a bit less cluttered , showing the original BFO curve in blue , and with it a brown line illustrating how the values can change when the satellite motion is removed ......
...on the first part of the chart the values are increased , and after 19:40UTC , the opposite effect , reducing the values to bring the BFO required to a much flatter curve .....
We need two data points to get the correct match...the satellite motion should be part of a sine curve although it looks like a fairly straight line in the previous chart . One obvious data point is at the satellite apogee , close to 19:40UTC ......ie , no vertical satellite motion , so the value depicted on the chart is all down to the aircraft motion........
..the second reference point I used is at 16:30UTC , thought to be when the aircraft was stationary from the cockpit transcript .............but if they were already taxying then it is possible that the satellite contribution was higher.........meaning the first part of the chart should be in turn higher values , and the second part even flatter.......

[David Bass]

Forgive me for interfering in the detailed analysis of the doppler and related satellite engineering parameters, but there is a comment from Robin Clark to the effect that at apogee the satellite will have no "vertical motion".

I'm not sure which vertical is being referred to here, but assume that the author means the change in satellite altitude is zero. This seems correct to me, but that does not mean that the satellite will produce a non-zero doppler shift - there are x and y components to the motion as well as z - the satellite moves in a 3-dimensional figure of eight.

Apologies if the x and y components are irrelevant to this discussion - I don't pretend to understand the detail.

[flight191]

Pardon?

It took the Captain of BA Flight 9 (volcanic ash over Indonesia) eleven years to find out from declassified information that satellite monitoring stations in Australia and Guam were following his flight.

Capt Moody's comment? "We don't know what they're watching."

I'm not sure why you would be expecting anything different in this case if more were known from confidential sources.

"It took 11 years to find out how Boeing was able to ring a ground engineer in Jakarta who got to us after about 15 to 20 minutes after landing."

Indeed, they contacted Moody within 15 minutes of landing. I wonder if remote telemetry/diagnostics of systems has been in heavy A/C since the late 70s?

[The Ancient Geek]

Forgive me for interfering in the detailed analysis of the doppler and related satellite engineering parameters, but there is a comment from Robin Clark to the effect that at apogee the satellite will have no "vertical motion".

Satellites orbit around the mass of the earth but the earth is not a sphere and does not have uniform density. This means that any movement with respect to a fixed point on the surface of the earth will result in a change of gravity and thus a change in altitude.

Radio waves are not affected by gravity [1] so a change of absolute vertical distance caused by the earth's "lumpy" gravitational field will result in a doppler shift.

Apogee is the point in the orbit of maximum altitude so at this point there will be no vertical motion.

[1] Actually radio waves are affected by gravity due to relativistic effects but the change is rather small.

[Lonewolf_50]

Originally Posted by James7

Someone pointed out many pages ago that MH370 will be found by a back packer. He just might be right.

That makes more sense that GIGFY's nonsense about Diego Garcia. Note from history: with a far better estimated last position, and some floating wreckage, it still took a long time to find AF 447.

OK, so you don't think it's in the IO near where they are looking.

Where do you want to go looking for it? I am sure the Malaysian Government, the Chinese Government, and the folks at Malaysia Airlines are keen to hear your better suggestion.

Lat and Long of your suggested search datum, if you please.

Fill in the blank _______________________________________ .

[mm43]

Burst Frequency Offset

Originally Posted by mm43

US Patent #6008758 will provide an insight into the mysterious "D1/D2".

Following the Hyperveloce & Shadoko posts around the same time, I have come to the conclusion that the MH370 AES wasn't using the P Channel to determine and correct for the SAT / AES Doppler, but was using hardware/software operating to the principles described in the patent.

As hamster3null has pointed out, the requirement is that the AES assigned Tx frequency when corrected for the perceived Doppler shall be within 335Hz of the SAT assigned channel frequency when received at the SAT. So if the AES was using the P Channel method to create the correction, there would be close to zero BFO at the SAT, and obviously that was not the case. It would be safe to assume that the AES SDU software is using a fairly loose algorithm to correct for the SAT position and motion then compute the Doppler associated with its own (AES) position and motion. The AES Tx frequency arriving at the SAT is "offset", which seems to leave the satellite reporting the residual Doppler shift (BFO) to the GES.

[threemiles]

My understanding is that the sat itself is NOT measuring any BFO or frequency offset.
It is done by the aircraft itself, that reports its frequency correction to the ground station as parameter in the downlink packet.
The frequency correction can be either calculated or measured from P channel.
It is certainly known to the investigator which method the specific 777 sat unit onboard uses.

The 777 "test" flights that have taken place were not to calibrate a specific frequency offset vs. a geo position/ track / speed. This would have required to fly a 777 exactly the same route that is estimated to be flown by MH370. The flights were only to confirm the general logic and pattern.

[Volume]

with a far better estimated last position, and some floating wreckage, it still took a long time to find AF 447.

It took exactly 5 days to find enough debris (including bodies) to answer a lot of questions. For MA370 we have found nothing and answered zero questions...

[henra]

Originally Posted by threemiles

This would have required to fly a 777 exactly the same route that is estimated to be flown by MH370. The flights were only to confirm the general logic and pattern.

Which is a pity. Compared to the cost of this Operation re-flying the route based on all the 'known' would have been a rather cheap but potentially very valuable excercise. They could have even flown different altitude pattern.
Why hasn't it been done?
That way it should have been possible to validate the accuracy of the SAT data number crunching excercise and determine a probably scope of error.
That said, it is not to late to do this. Maybe before starting the next Phase, they could try to validate the numbers.

[Heathrow Harry]

""It took 11 years to find out how Boeing was able to ring a ground engineer in Jakarta who got to us after about 15 to 20 minutes after landing.""

The British Embassy in Jakarta were called at least 45 minutes before the flight landed by Halim Air Traffic saying the plane had lost all four - I met the embassy duty officer a few years later and he said the brown stuff was all over the place well before it landed successfully, The reaction of the Ambassador was basically "oh God - not on my watch"

No doubt Uncle Boeing's man in town had been called as well..................

It probably took the engineer 30-45 minutes to drive out to the airport which was in the sticks (relatively speaking)

[Pace]

That said, it is not to late to do this. Maybe before starting the next Phase, they could try to validate the numbers.

Henra

Why would you need to fly the route in a real aircraft with different winds and conditions ?

A simulator would be far better as the winds and conditions could be loaded to match the conditions on the day!

I would also add i am sure this has already been done in a sim many times to date.

[Keef]

Sometimes I am amazed by the level of unwarranted effort going into this thread! Those folks at Inmarsat know a thing or two about how their hardware and software work, and don't need references to articles about Doppler correction.

Nor do they need to fly another aircraft by the same route: they validated their calculations using the handshakes from a number of Malaysian aircraft flying in that area. That's part of what they were doing in the days between the aircraft vanishing and their conclusions being published.

[ampclamp]

Why would you need to fly the route in a real aircraft with different winds and conditions ?

A simulator would be far better as the winds and conditions could be loaded to match the conditions on the day!

I would also add i am sure this has already been done in a sim many times to date.

Agreed. And the simulation would be better. They know the exact fuel load,where the freight was loaded in the holds, the a/c all up weight, a reasonable idea of the weather and more importantly very detailed history of that particular airframe and its engines' performance.

[GarageYears]

Why would you need to fly the route in a real aircraft with different winds and conditions ?

A simulator would be far better as the winds and conditions could be loaded to match the conditions on the day!

I would also add i am sure this has already been done in a sim many times to date. Agreed. And the simulation would be better. They know the exact fuel load,where the freight was loaded in the holds, the a/c all up weight, a reasonable idea of the weather and more importantly very detailed history of that particular airframe and its engines' performance.

Well, if the point was to validate the satellite track data then the simulator is of absolutely no use at all is it and I believe that was the point of this particular suggestion.

[James7]

Inmarsat

Interesting article, I did not see it posted before, apologies if it has.

Why the Official Explanation of MH370?s Demise Doesn?t Hold Up - Atlantic Mobile

Could go a long way to explaining why not one single scrap of the plane has been found floating in the ocean.