DME/DME correction of INS & France?

[IO540]

How does this work over say France, where most of the VORs don't have a DME?

I realise most ILS airports have a DME on the ILS but I would have thought those would be too short range to be useful for this - I can pick up TDMEs from only about 50nm max even when clearly high enough to be line of sight.

[flyboyike]

Unless you're flying a non-GPS airplane, like I did on my last trip.

[IO540]

Interestingly, Concorde had no GPS and used single-DME corrections to the INS. This is what the crew told me after I went on one of the last flights.

I wonder how that worked. I can think of a few ways, all relying on solving simultaneous equations involving a previously computed GS and track.

[eckhard]

Maybe the single DME used agility tuning (a la 737-300 in 1987) to get two distances to work from. Any CCDE pilots out there who can give us the real answer?

[reynoldsno1]

a DME on the ILS
I believe most are unidirectional and are not used to update FMCs
used single-DME corrections to the INS
I suspect it used VOR/DME updates - 'tis the most basic RNAV process for INS's...

[Pub User]

The FMC on a 737 finds plenty of VORs with DMEs over France, and will quite happily use an ILS DME, as well as a TACAN.

[Intruder]

The 744 FMS uses ILS DMEs as well.

[Denti]

Never had a problem with updating over France in our 300/500s. Actual navigation performance usually stays below 0.3NM unless we switch the NAVs to manual for the approach setup.

[Nubboy]

In my innocence I've always beleived that DME paired with an ILS had an offset introduced to cater for the difference between the beacon and the runway threshold. Therefore 0 dme wasn't at the beacon, but rather a circle around it.

[Jaguar Pilot]

Corret Nubboy.

JP

[exflyboy]

Remember also DME reading is slant distance so will always vary as a function of altitude.

[IO540]

I had some PMs from someone here, saying (basically) that the INS/IRS system is separate and its error is never adjusted all the time the a/c is airborne, and when over land the FMS derives its position from navaids but not from the IRS.

Is this true?

If it was, how could a jet transport meet the BRNAV requirements for example? Quite a bit of the time in Europe, one is out of the DOC of any navaid (maybe not at FL350 but definitely lower down). I asked this chap this Q but he didn't get back to me.

Surely the FMS *always* gets its position from the IRS, and the IRS error is corrected by navaids (DME/DME usually) when these are available.

So if e.g. you have crossed the Atlantic and have picked up a 2nm track error on the IRS, then pick up a few DMEs, that will fix up the error and if then you lose all navaid contact for a bit, you will still be flying along with (essentially) zero error. Your nav will *not* revert to the previous 2nm error.

Is this right?

(ignore GPS for now)

[BOAC]

Possibly a little confusion there. The FMS does indeed compute its postion from navaid updates whilst they are available. When no longer available, it reverts at a controlled rate to INS derived position which can, as you say, be 'significantly' in error. Hence the normal restriction on use of RNAV for approaches in such systems. What the PMs may have been telling you is that no update from navaids has ANY effect on the position that the INS thinks it is at. That is set at initial alignment and adjusted for manouevre thereafter by the acceleration derived inputs.

In 'mainland' Europe it is extremely unusual to be out of some sort of update coverage.

[Permafrost_ATPL]

Just to finish the confusion clearing job from BOAC, if you also have integrated GPS (e.g. NGs), the FMC will:

1) Determine its position based on GPS (which means you don't have to enter a take-off shift for intersection departurers, like you do on the Classic)

2) If GPS not available, FMC will determine its position from both navaids and IRS (left), with a bias towards navaids (I think it's 80% bias towards NAVAIDS, but I'd have to check).

3) If no GPS or Navaids available, the FMC will come up with a "most probable" position based on what the IRS units are saying at that time. It determines the error margin based on the IRS position error it calculated when it last received a navaid update.

Again, this is for 737 NG. And again, no the IRS boxes never update their position.


P

[matt_hooks]

The DME works by a transmitter on the aircraft interrogating a ground station which then sends a reply. I would assume the reason your range seems to be limited is that the DME ground station is at capacity, so it is incapable of dealing with your request. They have a fairly limited capacity, and in busy airspace at an airfield this might well be reached fairly quickly.

[IO540]

When no longer available, it reverts at a controlled rate to INS derived position which can, as you say, be 'significantly' in error.

Why does it work that way? Why not simply fix up the IRS when you have a verified DME/DME fix? Maybe there is a subtle safety reason why the IRS position is never modified.

Referring to the "controlled rate" above, let's say you have an IRS that is 5nm out, and you have been in navaid contact, then lose the contact. Your position will gradually drift to say 4nm off track. Then you get within range again, and your track error is again reduced. Does this actually happen, and if so doesn't it result in your track being a bit "wiggly"?

[hoey5o]

IO540

It is impossible to update an IRS when moving. That is the whole point of IRS. You tell it its position before you start moving. It can then ony work out its position by knowing how far it is moving from that original point. If you are flying at 500kts at what point in time would you be able to say your exact position ? It would always be changing.

Just to clear up the confusion, on big boeing's 757/767/777/747 and almost certainly 737's, the FMC position is derived from radio aids and/or GPS. When these are lost (IRS NAV only) the FMC compares its last known position (with the aids) to the IRS position and then knowing the error, it maintains a parallel course with this error until it acquires more aids to fix on.

The FMC does not navigate using IRS it simply uses it as a reference.

[IO540]

It is impossible to update an IRS when moving

That may be true in practice but is definitely incorrect technically. The system comprises of x/y/z gyros (laser ring nowadays) whose signals are integrated to give position in space. The lat/long is then calculated from that, noting the original (known) position it started from, and nothing prevents an offset to be added to that calculation.

The gyros themselves are obviously not messed with; they just happily carry on generating the pulses.

the FMC compares its last known position (with the aids) to the IRS position and then knowing the error, it maintains a parallel course with this error until it acquires more aids to fix on.

OK, that's exactly what I am saying above. You add an offset to the inertial-derived position, and carry on flying with that offset. That makes sense to me because it gives you seamless BRNAV capability no matter where you are.

I wish GA has this capability. Presently, if we lose GPS reception (very rare but possible) BRNAV goes down the pan immediately and we fall back to VOR/DME/vectors etc. A couple of years ago I was looking at a project to design an "IRS" for GA but the FOG gyros were still at least $5000 each and aren't likely to come down in the near future. They are also export controlled. A lot of other people have looked at this; the cost of the gyros is the issue. Everything else has been done many years ago.

[reynoldsno1]

It is impossible to update an IRS when moving

This is possible because an accurate determination of the aircraft motion is available based on measurements obtained from GPS. Align In Motion allows initialization of an IRS while an aircraft is moving, in the air or on the ground. This is accomplished using GPS and an inertial reasonableness test, thereby allowing commercial data integrity requirements to be met. Align In Motion has been FAA certified to recover pure INS performance equivalent to stationary align procedures for civilian flight times up to 18 hours.

[BOAC]

The system comprises of x/y/z gyros (laser ring nowadays) whose signals are integrated to give position in space. - NB: Unless technology has changed dramatically, system POSITION is derived by applying integrated accelerometer readings to initial position. The 'gyros' are there to maintain a space attitude reference which is adjusted to an earth frame by knowing where the thing has been since alignment. I am well out of touch with the systems having 'grown-up' with gymballed gyro attitude platforms but I am not aware of the 'gyros' being used to compute position. I await correction!

Align-in-motion is indeed techncally feasible now and produced, I believe, by Honeywell at least. Damned clever little black boxes! I cannot see it ever being applicable to the 'basic' DME/DME updated systems as on the Classic 737s.

[IO540]

I think the general principle in all cases is that a gyro gives you an acceleration-related signal, which you integrate once to get velocity, and integrate again to get distance i.e. position.

The implementation is normally complex due to the need to make small corrections for all kinds of things (e.g. temperature related drift) and then you have to do some more maths (old well established stuff though) to work out where you are in terms of lat/long on the earth's surface.

The current holy grail is to make something useful with solid state gyros but they aren't good enough by orders of magnitude. You might get 10 minutes of navigation out of them :)

[Blacksheep]

BOAC is correct. The laser gyros only measure the aircraft displacement rates about the x, y and z axes. The displacement rates are integrated to obtain angular displacement from the starting attitude - when the aircraft was hopefully parked right side up somewhere upon the surface of the earth. When the operator enters the geographic coordinates into the system the IRS computes the acceleration rates and refines the raw aircraft attitude into a true hoizontal starting point. Once the system has an accurate starting point and horizontal reference, all motion along the x, y and z axes is resolved by mathematical integration of accelerometer outputs into velocity, direction and distance. The tricky bit with "strap down" laser gyro systems is that the three accelerometers are fixed to the aircraft axes, so the mathematical calculations become quite complex. In old "steam driven" systems the accelerometers were attached to a gyro stabilised level platform.

[exflyboy]

Please correct me if I am wrong but the term strap down actually refers to the old INS where there was a set of gyros on a platform that would basically remain gyroscopically erect while the aircraft rolled around this platform. An update to this system was the strap down platform as you mentioned. This system had the gyros attached to a platform that was attached to the aircraft hence it was only the individual gyros that moved. Using the term strap down for the modern ring laser gyro is sort of a misnomer as they are attached to the aircraft regardless. So the term strapdown is a bit old hat really. Again please correct me if I'm wrong.:ok:

[Blacksheep]

Older (civilian) systems had gyro stabilised platforms with accelerometers mounted on a stable platform - that is, they remained in a horizontal plane. The popular Carousel system did complicate matters somewhat, by rotating the platform at 1 r.p.m. to enable any platform leveling inaccuracy to be isolated as a one cycle per minute error signal in the accelerometer outputs. However, all gyro stabilised systems isolated the accelerometers from the aircraft roll, pitch and yaw movements and held them in a level plane.

In a ring laser gyro system, accelerometers are fixed to the aircraft, aligned with the aircraft's roll, pitch and yaw axes. The laser "gyros" are used to accurately measure the roll pitch and yaw movements of the aircraft to enable the spacial alignment of the accelerometers to be determined by the processor. The resolution of the flight path is performed by computing the various vectors obtained from each accelerometer - each of which will simultaneously be measuring an acceleration in each of the three planes. The IRS computer resolves the complex pattern of accelerometer outputs into horizontal and vertical components by reference to the measured attitude of the aircraft and thus of the accelerometers themselves.

This is what I intended to convey by the term "strap-down" - that the accelerometers are not isolated from the aircraft attitude by a stabilised platform, but that the entire reference unit - including both the accelerometers and the ring lasers - is firmly attached to the aircraft with no moving parts.

The IRS measures the movement of the aircraft away from its starting point and in general cannot be realigned in flight. Indeed, for most systems in use today, the aircraft cannot be moved during IRS alignment on the ground and even aircraft loading can disrupt the alignment process. Position updates are performed by adding an offset to the computed position, the IRS continues to use its own computed position, then adds the offset to its output before displaying position to the crew and other aircraft systems.

You can find the value of this correction at the end of any flight by perfoming a terminal error check as per the AMM.

[exflyboy]

Yep, we're talking the same talk. I agree.:ok:

[BOAC]

If anyone wants a pretty thorough run-down on IRS/INS I recommend http://en.wikipedia.org/wiki/Inertial_guidance_system

There is a link in this site to Honeywell and 'Align-in-motion' too.