Why don't they transmit flight data?

[mikep777]

Any idea why airplane flight data is recorded and not transmitted somewhere? Is this just historical, or is there some technical reason why it’s difficult?

I would imagine that whatever technology enables in-flight phones would also work for flight data, and it would make black box damage not an issue.

Thanks,
-Mike

[Ethilrist]

1. History. We’ve never done it that way.
2. Money. That’s a lot of transmitters.
3. Bandwidth.
4. IM Interference. Just as you can’t use stuff during takeoff and landing, they’d want to shut down the transmitters, but they could burst-transmit from time to time during flight. Which probably wouldn’t help much, as they usually only need the last few seconds of the flight…
5. Possibly not necessary. Are there any figures for what percentage of investigations came up empty, and had no data because they couldn’t find the black box? If it’s a really small number, it just wouldn’t seem too worthwhile.

[Tranquilis]

• Data transmission rates.
• Quantity of data (it’s fargin’ huge!).
• Storage considerations (where you going to put it all?).
• Cost: BIG buckaroos! Ticket prices will have to be adjusted upwards a long way to pay for the infrastructure.

[Running_with_Scissors]

Funny you should mention this; there was a recent proposal by Iridium to transmit voice and flight data using their satellite network. Since the satellite network is already built, the additional cost would not be prohibitive.

[Whack-a-Mole]

Realtime tranmittal from all planes all of the time may be a bit much but why not a data dump in the event of a disaster?

The ‘black boxes’ could be modified to keep, say, the last minute of data in an electronic storage buffer as well as having the data written to tape. In the event of trouble the black box could dump what it has in its memory to ground stations or satellites and begin realtime transmitting of data till the plane crashes or hopefully lands.

I’m not an engineer but I don’t see why that should be too difficult or expensive.

[Tranquilis]

Even with Irridium, you’d still have a cost / infrastructure issue: First, I doubt the Irridium structure can handle the necessary through-put (we’re talking about loads aof data coming from loads of aircraft), but even if it could, it still has to go somewhere, and that somewhere has to be built. The quantities of data would be staggering, and the storage and processing power to recieve, process, store, retrieve, and validate it would be stupendous. Not that it’s necessarily a bad idea, just that it’s not economically feasable at this time.

As for the last minute data-dump, in many cases, it would the disaster happens before a useful dump can be made. Adding built-in dump capability to the “Black Box” would only make them more complex, which isn’t a good idea when you’re building for maximum durability.

A combination of ideas might be doable, however:
Include the ability to manually or automatically start transmitting data in case of an inflight emergency, such as sudden depressurization, setting the transponder to “inflight emergency” or “hijack”, or any of a number of pre-programmed scenarios. It wouldn’t capture all the data, necessarily, nor in all cases, but it would give some clues, such as perhaps what happened to flights that go down at sea without warning, but don’t actually break-up until they hit, and the data recorders aren’t readily recoverable.

The through-put would be far less, and the necessary storage and processing infrastructure could be greatly reduced. Further, this could capture data in cases of near-disasters that don’t result in crash, but take longer than the recording time of the data recorders to reach the ground. When a dangerous event takes place in mid-air, if it takes longer than the recording span of the FDR/CVR to reach the ground, the data is over-written, partially, or completely. A manual in-flight dump could store that data while allowing the recorders to remain fairly simple and inexpensive. The transmitters for this could be phased into service at a fairly leisurly pace, reducing cost burden, and limiting the resultant rise in ticket prices.

Hmm…

[bathos]

Check this url http://www.trip.com/trs/trip/flighttracker/flight_tracker_home.xsl

Basically it will show you the flight, it’s co-ordinates, altitude and flight speed.

=Bathos

[thinksnow]

I’m not sure you appreciate what the black boxes record.
For your edification: How Stuff Works mentions this about the recorders on Flight 261 (the Alaska Air flight that had a bad jack-screw and plunged into the Ocean)

FDR contained 48 parameters of flight data, and the CVR recorded a little more than 30 minutes of conversation and other audible cockpit noises.
…
Airplanes are equipped with sensors that gather data. There are sensors that detect acceleration, airspeed, altitude, flap settings, outside temperature, cabin temperature and pressure, engine performance and more. Magnetic-tape recorders can track about 100 parameters, while solid-state recorders can track more than 700 in larger aircraft.

The NTSB has this to say about FDR’s:

By regulation, newly manufactured aircraft must monitor at least twenty eight important parameters such as time, altitude, airspeed, heading, and aircraft attitude. In addition, some FDRs can record the status of more than 300 other in-flight characteristics that can aid in the investigation. The items monitored can be anything from flap position to auto-pilot mode or even smoke alarms.

With the data retrieved from the FDR, the Safety Board can generate a computer animated video reconstruction of the flight. The investigator can then visualize the airplane’s attitude, instrument readings, power settings and other characteristics of the flight. This animation enables the investigating team to visualize the last moments of the flight before the accident.

(Emphasis in both quoted sections is mine.)

The amount of information that would need to be transmitted to accomplish this would be staggering. I’ve been wondering if it would be possible, too, so don’t take this as a beratement (word?), but right now, unless they did a burst transmission, there is no way they could transmit all the information and unless they did it every 5 seconds or so, there doesn’t seem to be any way they could maintain an effective level of information in the event of a sudden catastrophic event.

[scotth]

hmmm… lets look at the actual data rates for a single plane and think some more.

1. Lets say we send just the required 28 parameters.
2. I would say that 32bit resolution for each parameter would be suffient.
3. Sampling rate… that is tougher to guess. Does 10 samples per second seem high enough? It does to me, I’ll use that for this example.
4. We need a couple of words for each transmition to identify the beginning of a data cycle and the what plane it is coming from. So, that would be at most a 35 word (of 32 bits) frame 10 times a second. Since this is a special purpose transmission, there is little need for alot of control data.

That gets us a data rate of 11.2 kpbs… seems pretty manageable to me. This could easily be carried by a channel of one of the “onboard cell phones”.

Now, the infrastructure to support those phones exists. How expensive would it be to shunt one channel from each plane to a data recording system.

As for data recording, one 2Ghz PC should be able to handle the load. Obviously, some redundency would need to be added, still pretty cheap. Keep in mind the PC doesn’t need to do much/any processing of the info. Just record it coming in. 1000 commercial jumbo jets in the air would send it not much more than a sustained 10Mbps. Don’t even bother sorting it into seperate aircraft info, just record it in one stream. It will be easy enough to pull it apart if need be later. An 80GB harddrive would yield aprox 80,000 seconds (22 hours) of record time.

This actually seems like a pretty decent idea to me.

I would bet this has never been done, because in the past, it was never really feasible. Clearly it is becoming so today…

[Napier]

I agree with scotth. We’re talking about a manageable stream of data for a pretty high dollar situation. Scotth, you could record a great deal less - 32 bit information for each parameter gives you 10 digit precision in base 10. 12 bit is very commonly used in science and technology. And 3 observations per second is still probably plenty. So you could make do with an order of magnitude less than your generous estimate - all the more reason to agree with your conclusion.
Another way to look at it is that storing and recovering the data is, like just about anything else, much much cheaper to do on the ground. Space and power and maintenance and access are pricey commodities on a jet.
And airliners have another advantage. They are some of the only things around that absolutely always have a clear view of the entire sky, for line of sight transmission of high frequency radio.
If anybody still thinks it seems hard to do, check back in another 5 years when information carrying infrastructure has grown bigger still, and it’ll be a cinch then.

[scotth]

I just put those up as “for instance” numbers to see what we would get. Dropping some bits in resolution and sampling rate (as Napier suggest) could easily allow for more channels of information as well. It wouldn’t take to much study to figure out what the optimal balance between resolutions and sampling rate to number of channels would be.

[gazpacho]

I agree with Napier 32 bits is way to high for most things. I disagree about sample rate. The sample rate must be much higher the a few times a second for some things. Engine parameters would want a higher sample rate. I can imagine flap position oscillating pretty fast under abnormal conditions which is what we are concerned about.

I used to work designing data collection equipment that was used by Boeing to test their aircraft. (Not black boxes.) Our systems would have thousands of channels of various sorts with sampling frequencies from about 5Hz to 30KHz. Not all of the channels or even most of the channels were on the upper end.

You also want the voice recorder phone quality sound is 64Kbits per second. I am not sure if phone quality is good enough and I would think there are a couple of microphones in the cockpit. I don’t think you want to do much compression of the voice because they are always talking about trying to here strange noises in the background which perceptional compression technics would probably loose.

I definitely think that you want more than the 28 required channels but maybe you still have the recorder for those and only transmit the 28.

[thinksnow]

Thing the first: air planes afforded uninterupted communication with the ground? Do they ever have periods of black-out? Say, in a heavy storm? It would seem to me that that would be one time in particular when the data might be very much needed.

Thing the next: the new solid-state recorders (see my previous post) record up to 700 data points. That’s a lot of information to transmit. They are also much more robust than the magnetic tape recorders of old. It’s technology that is already in place and reasonably easy to maintain.

[scotth]

I think the idea here is to record at least some of the important data in the rare event the on board systems are lost or destroyed.

I think it would be acceptable to have onboard system record more channels at higher resolution.

I bet a crash investagator would think 30 or 40 channels of data looks pretty good compared to getting nothing if the onboard system was unrecoverable.

This is a back up system only. Small gaps in coverage would not be too big a deal.

I do agree that squeezing voice onto that stream would be a good idea though.

For example, we would have the voice recordings from the twin towers bound planes right now instead of still hoping to find the boxes in the rubble several months later.