Inertial Navigation Old Timers/Experts
 

Hi all,

Just a quick question and i apologize if this has been asked in the past.

If the INS was not updated at the gate first flight of day, can the INS reset itself to be accurate after departure, in flight, with an FMS (no GPS on board)? If so, can it 'right' itself within 10, 20 mins after departure? In other words, if the INS can 'right' itself while underway, how long would you anticipate accuracy? And/or can INS be considered accurate if not reset in a stationary position?

In further words, if the pilots werent able to update their INS at the gate (whether too lazy or behind schedule) and the INS was off by 1,000-2,000-3,000 ft at the gate, can they rely on the INS to update itself in flight and/or even perform an update in flight to be accurate?

Please explain for either answer.

Thanks in advance...

The old INS and their accuracy
 

I flew for PanAm - started flying early 1969, until their end 1991.
And I flew with their early INS systems - on the 707s...
And later flew the 747s, also equipped with the INS.
xxx
Early INS systems were the "Litton LTN-51" or "Delco Carrousel C-IV"
We had a maximum of 9 waypoint memory... then had to enter subsequent waypoints.
Their accuracy degraded with time "in navigation", if not updated.
As an example, 1 NM error per hour of flight could be expected as average.
JFK to LHR - some 7 hours in "NAV" - INS generally would drift 7 NM off...
xxx
We sometimes updated the INS over a known point, such as a VOR.
As an example, on the way from JFK to Europe, we would do it overhead Gander.
Then again maybe, overhead Shannon VOR, if our destination was still far.
Then later, INS were improved with "automatic update" operating with DMEs.
xxx
With the 747, we had 3 INS systems (only 2 with the 707 generally).
So we could "triple mix" the navigation between the 3 systems.
It averaged the "inherent error" of the 3 systems and resulted in more NAV accuracy.
xxx
I still fly the 747 - We have the Litton LTN-92s now.
Our 3 INS are extremely accurate now, in "triple mix" updated by 2 GPS receivers.
:)
Happy contrails

Thanks for your reply BelArgUSA,

So if i have your above reply correct, INS can update in flight, even if the INS was inaccurate at the gate and never updated until in flight? And this is done by DME? Mainly due to the fact DME is extremely accurate, at least more accurate than INS... (from my experience with DME anyway...)

??

Thanks again for your reply...

INS "position update"
 

You can do a "manual update" - i.e. overflying a known point (a VOR)...
That was the first form of update that existed on early models.
xxx
A couple of years later, started to have capability of tuning a DME and be updated.
So in the mid-1970s until mid-1990s, no DME = NO update (over the ocean).
And since some 15 years ago, GPS is now used, instead of DME update. Works everywhere.
xxx
Hope that answers your questions...
:)
Happy contrails

P.S.
Warning - DME updates - i.e. France.
Numerous DME not co-located with VOR (are not "VORTACs")
Be careful entering DME (TACAN) coordinates, not the associated VOR...
Danger of updating and entering some 5 or 10 NM error at some locations.

Much of the position updating and velocity correction ability of an INS is determined by the firmware/software in the system. That also determines whether a manual "position update" or re-alignment can be accomplished in flight. For example, the LTN-92s in current 747 Classics cannot be re-aligned in flight, and have no effective manual position updating capability.

The closer the Initial Position is to reality when the INS is aligned, the better it will navigate. While I don't know how far off the IP has to be to adversely affect one of these newer Ring Laser Gyro INS systems, some of the old spinning gyro types would start building stray velocities (nav error) if the IP was more than about 1/2 mile off.

AFAIK, the newer INS systems can be dynamically updated using GPS or VOR/ILS/DME, but the update applied by the radio nav instrument will "fade away" after the update signal is lost. Our LTN-92s will occasionally drift as much as 3 miles crossing the Atlantic or Pacific without updates (measured by the track error when updates are regained on the other side), even when there was "zero error" while updating at the start. Triple-Mix helps to smooth out these errors.

Litton LTN72R's could be position updated in flight using DME/DME data.
Works good, lasts a long time.

Just to add for 320 he may not understand that the on board navaids are auto tuning, ie seeking the nearest aids enroute for INS update
The priority used to be DME/DME, DME/VOR

I quit flying before GPS came along so guess thats one handy gadget now?

Greetings,

There is a big difference between updating a position :ok:, which is not a problem at all, and Re-Aligning an IRS in flight which is not possible :=.

Not so fast, the LaserRef V in the Challenger 605, among others, can be aligned in-flight. Haven't tried it and is supposed to take awhile.

GF

Greetings
Galaxy Flyer, the subject is OLD INS :}

The INS in the OLD A-6 could be re-aligned in flight, UNTIL a newer/better software package removed the capability. However, we didn't miss it too much, because it made the initial alignment MUCH more reliable...

kijangnim:

Yes that's what I was going to say

The IRS/INS/ADIRU whatever you use does not update after alignment.(on the ground you can get a fast align ( manual or automatically on the 777 )to remove the GS errors ) It's the NAV SYSTEM that updates. In my case the on the 777 the ADIRU drifts and the FMC uses the best info to update itself and that is always the GPS. Unless on App then it's LOC-GPS or LOC-DME

Old INS position updates
 

With the older INS (such as Delco Carousel C-IVA) - 1980s
Using triple mix and DME update...
On typical NAT track between YQX and SNN, was nearly 4 hrs without DME update.
About the same time between NHA (Noronha) and SAL (Cape Verde) DMEs
From US West Coast to HNL was somewhat closer to 5 hrs.
xxx
It took the GPS update to make these old INS compliant with RNP-5
:)
Happy contrails

This is a technology that should never be allowed to fade away. One day...or maybe night, there will be a solar flare that destroys all the satellites within a line of sight. We need that back up science, especially now that it can be contained in a small device.

I used to feel more confident about inputting an update slap-bang over a VOR, cos the vibration and basic acceleration during take-off made me wonder if the system would be affected.

I didn't think the ADIRU in a 777 was going to "fade away" as you put it. All the GPS systems could shut down and we still could find our way around. except without RNAV arrivals and departures, it would be like it was around 7 to 10 years ago.:ok:

When all else will fail...
 

An old E6-B stays aligned for quite long...
If in metal, can even put it for cleaning in a dishwasher.
:{
Happy contrails

As kijangnim has pointed out, the 'experts' are tending to confuse INS/IRS update with position update. Two separate issues. Probably confusing A320Slave too!

Most 'old' systems were indeed capable of position update. The alignment for a platform requires steady conditions over several minutes which can rarely be achieved in flight. Normally the best 're-alignment' in flight gives a fairly crude attitude reference. Often it requires a repeat at intervals.

I do not see any reference to 'OLD' systems in the post (just 'no GPS'), so for A320Slave - it is by no means unusual for an 'oldish' (no GPS) IRS/INS to be aligned at start of day and left untouched all day for several sectors. No loss of navigational (or indeed attitude accuracy) will normally occur in a 'rich' environment where the position update is easily obtained and kept, and most systems can update position at take-off on a defined runway as well. Even with the 'errors' IGH mentions, the resulting accuracy of navigation is still streets ahead of the old star/sunshot/Doppler drift/LORAN/piece of seaweed systems. The only time when it is really important to align accurately is before a longish flight in a region where updates are not available - eg ocean crossings. Of course, now-a-days the REQUIREMENTS for navigation accuracy have tightened, so frequent re-alignments are more advisable.

The really 'old' INS on the early Harriers was position updated by 'pickling' over a map point, but once the platform had lost its alignment and had rushed off into 'enemy territory' at 200kts, there was little point, so we just turned the map off and got lost manually.:)

Must go back to the original question. For the first flight of the day you must turn the INS on. It would never be left on on an unattended aircraft. If the ground power failed then the INS would so go pear shaped with no cooling, and would be running on the aircraft (or its own) battery.
When you turn on an INS you must enter the correct coordinates or the ALIGN light will start flashing at you.

But I bet most short haul pilots never fly on the INS. There is even a Caution message on the A320..INS NAV.. to tell you that you have lost DME updates.

Also on an A320, there is no INS. There are three IRS sensors that supply position to the FMS, along with the 2 GPS and the 2 DME and 2 VOR. The FMS decides which sensors to use and starts with DME/DME.

Hey............cough cough...............read post 12 again. :ok:


The 777 starts with GPS, never seen it use DME DME in flight EVER.

Only on approach it may use LOC DME or LOC GPS

Greetings,

FMS selects the update source from the most precise to the least, i.e., and in order please, ILS, GPS, DME DME, DME VOR (VOR VOR is a NO NO situation) then IRS.
On the Smiths FMS world they use Kalman Filtering (So does Litton on the Flag Ship AIME) this enables an error modeling from which the position will be corrected :}
IRS Alignement on the ground it is highly recommended (even for a fast align) NOT TO MOVE THE AIRPLANE :ouch:

An error in position input during initial alignment, a part from obvious bias, will give the wrong "G" initial value and induce wrong vertical acceleration (delta between the sensed real one and the erroneous one) :eek: throughout the flight dixit Korean Air B747 :(

Thanks for all the replies everyone.

Maybe if i narrow this down with an example it will help...

Here is a lat/long plot from the gate departing IAD runway 30, first flight of day. Obviously the INS was never updated/realigned at the gate.

http://i47.photobucket.com/albums/f1...60/INS_IAD.jpg


Could a pilot expect the INS to realign itself in flight? If so, Could it be done within 10 mins of departure/climbout?

What airplane and navigation system are you talking about? Real airplane or some kind of flight simulator?

If the Present Position is incorrect when the INS is initialized, it will NOT "re-align" itself! It should be shut down and re-initialized with the correct Present Position.

Though the FMS or other navigation system MIGHT be capable of radio updating via GPS, DME, or other means, it will NOT "re-align" the INS! All it will do is put an appropriate correction into the position output. Depending on the navigation system, that correction may or may not persist after the updating is no longer available.

Greetings
No IRS realign in flight, :=alignement realign requires the platform to be stable. no acceleration, no movement, sometimes the movement created by loading container is enough to screw up the align process, on a modern Airliner.
Now looking at your plotting :hmm: there is a shift (bias) between ploted and aircraft position, WGS84 coordinates aside, cross-check your initial position values, specificaly the last digits (sometimes gates coordinates are wrong :ouch:) verify with a hand held GPS.
That's all I can think of :O

The above are coordinates from an American Airlines B757 aircraft, manufactured 1991, no GPS.

From what i understand, it cannot be realigned in flight. And thats pretty much what i already thought....


Thanks again all...

Greetings A320Slave

How did you insert the pix in the thread? :eek:

15 years ago we used to have the coordinates of every gate in the FMS database. Not any more. Just one position for each airport. The pilot can choose to enter this position, or type in the actual coordinates of the gate. Now that IRS position is only a back up, and not the main nav position source it doesn't really matter. In your example I expect this is what happened. Even though the IRS is showing a position to the left of the runway, the FMS which will have been updated with DME (no GPS fitted) will show the actual aircraft position on its display, and on the EHSI.

By the way, I use the term IRS on purpose. B742 and L1011 era had INS fitted. FMS did not appear until the late 1970s. So the INS was the main navigation system. All modern aircraft are navigated by the FMS using inputs from various sources. You cannot navigate with a modern IRS (Inertial Reference System) unless you have a map to plot on. The FMS does the navigation, the IRS only tells it where it is.

- to be strictly accurate, I think you'll find an 'IRS' merely "tells you which way up you are". Of course, the software etc that control an IRS can easily be adapted to give motion and therefore position information. To give one example, in the 60's the English Electric (BAE) Lightning used a basic inertial platform taken from the Blue Steel ICBM as an 'IRS' which fed the instruments and weapons system computers, but fed TAS from the aircraft ADC into the platform to allow correction for pitch reference as it flew around the globe (not for very long, obviously:)).

As soon as you add any output from an 'IRS' platform to give "where you are" it becomes an 'INS'.

I concur with BOAC to a certain degree.

IR(S) System
IR(U) Unit
I(NS) Navigation System

All different.

Think of the Unit as outputting accelleration values

System putting those values to some use.(perhaps using them to calculate a position relative to a given starting position)

Navigation System using a calculated postion to output stearing comands between given waypoints.

The Navigation function is usually done by FMS these days, hence the use of IRS rather than INS to describe the unit.

So the best description of an IRS is a navigation (position) sensor.

The IRU has to align itself in-order to accurately diferrentiate between aircraft accelerations and earth accelerations.

Some newer IRS can do this inflight. Honeywell do a nice micro IRS system with this capability so after a power interupt the unit will automatically re-align itself using GPS position to calculate aircraft track values and subtracting this from its total acceleration to calculate earth related accelerations then use these to re align to true north.

Way too complicated for my small brain but very very nice. Align In Motion will take between 15 and 30 minutes.

If the correct runway was entered in the FMS and the pilots hit the TOGA switch when going into position the aircraft position should have been updated at that point. If not you could be be looking at "map shift" perhaps?

The IRU isnt exactly a navigation sensor: it is more like an error reporting unit.

When you start in the morning, you tell it where it is. Either by telling it the gate position or by giving it the GPS position.

What it then tells you is where it isn't, as it measures accelerations away from that known position.

Sounds like George Bush made the last post!

Yes pushing TOGA updates the FMC position and does NOTHING to the IRS

Except for SaudiArabian aircraft.
These aircraft were fitted with three Litton ISS units (not capable of stand alone navigation), which in turn provided data to two Hamilton Sundstrand FMS units (far more useful than their Smiths cousins), and this whole kit and caboodle also included a 12 inch diagonal (CRT) moving map, all in one neat package.
The first and finest available, anywhere, circa 1976.
Aircraft type?
Lockheed TriStar.:ok:

On the B777 the FMC determines present position from these navigation systems; ADIRS, GPS and navigation radios. When receiving reliable GPS data, the primary mode of navigation is from a GPS updated FMC position. If GPS data is not available, cannot be validated, or is inhibited, the FMC position is updated using navigation radios. When navigation radios are available or no reliable, the FMC position comes from the ADIRU.

FMC position may be manually updated to any of the navigation system positions and this is accomplished on POS REF page 2.

slim shady
 

Yes and no.

Position is only one output from an IRS.

When position output is not valid it may still output accelerations or track and groundspeed.
The FMS can use these outputs to navigate in dead reckoning mode. In this case it is sensing change in position so I guess it is a position sensor. Or at least thats how the manufacturers see it. I'm not going to argue with them.

I always liked this little explanation:

"Airline Inertial Guidance Systems The aircraft knows where it is at all times. It knows this because it knows where it isn't. By subtracting where it is from where it isn't, or where it isn't from where it is (whichever is the greater), it obtains a difference, or deviation.

The Inertial Guidance System uses deviations to generate error signal commands which instruct the aircraft to move from a position where it is to a position where it isn't, arriving at a position where it wasn't, or now is. Consequently, the position where it is, is now the position where it wasn't; thus, it follows logically that the position where it was is the position where it isn't.

In the event that the position where the aircraft now is, is not the position where it wasn't, the Inertial Guidance System has acquired a variation. Variations are caused by external factors, the discussions of which are beyond the scope of this report.

A variation is the difference between where the aircraft is and where the aircraft wasn't. If the variation is considered to be a factor of significant magnitude, a correction may be applied by the use of the autopilot system. However, use of this correction requires that the aircraft now knows where it was because the variation has modified some of the information which the aircraft has, so it is sure where it isn't.

Nevertheless, the aircraft is sure where it isn't (within reason) and it knows where it was. It now subtracts where it should be from where it isn't, where it ought to be from where it wasn't (or vice versa) and integrates the difference with the product of where it shouldn't be and where it was; thus obtaining the difference between its deviation and its variation, which is variable constant called "error".

:)

It still sounds like George Bush said that!

Fascinating thread, you guys, with loads of good gen. Glad that the initial confusion between on-the-hoof INS/IRS position updating (impossible on any system I came across) and FMS position updating (happening all the time) was corrected by kijangmin and AMCS.

In the absence of today’s FMSs, with divers forms of position updating, it was important to align the INSs with the most accurate position possible. This is why, on non-FMS aeroplanes with early INS nav computers, we used to use the actual gate position. This was obtained from our Aerad/Jeppesen and/or the gate marker board at many airports. [In the latter case, beware of confusing seconds with decimals of minutes.] The more accurate the latitude, the less the INS drifted during the journey.

As kijangnim has pointed out:
An error in position input during initial alignment, apart from obvious bias, will give the wrong "G" initial value and induce wrong vertical acceleration (delta between the sensed real one and the erroneous one)
[Unquote]
And FE Hoppy:
The IRU has to align itself in-order to accurately diferrentiate between aircraft accelerations and earth accelerations.
[Unquote]

We could update the INS nav computer position en-route, but BelArgUSA’s method of doing it as you flew over a VOR was not an accurate one, because of the cone of confusion (for want of the correct term) over the station. Provided the VOR had a co-located DME, the best simple way was (and is) to take the DME distance as you fly due-south or due-north of the station (remembering to allow for the magnetic variation at the station). You know the longitude of the VOR, and simply apply the DME distance to the latitude. Try to do it when you are far enough away to minimise the DME slant-range error, but close enough to avoid a big error with the VOR bearing. I think about 30-90nm was considered a good compromise. [In the absence of DME, you can try a 45-degree or 26-degree “running fix”, assuming you know your track and GS; but you need some sort of chart for that, and you wouldn't use it for an update unless desperate.]

That earth-versus-aircraft acceleration business brings to mind the recent thread:
“A320 pitch-up at AP disconnect”

On the last generation of INSes, the position updating used the DME/DME as the most accurate.
The slant range error was minimised by entering both the DME coordinates and its MSL altitude. Some heavy manipulations, there...
As a matter of fact, on those old aircraft, one INS would provide navigational data to either the Captain (INS 1) or the First Officer (INS 2), INS 3 was used as an arbiter in case of vastly differing positions between 1 and 2.
Problem was that on autopilot/FD, the non-flying side would have a meaningless vertical bar after a while as the drifts were different.
Kept crews alert, though.

Hi Lemurian,

Everything you say is correct, but not relevant to what I was dealing with in the post you quote!

In the early days of INS on airliners - like retrofitted dual-INS on B707s, or triple INS on B747s and DC-10s - the associated nav computers (one per INS) were not capable of AUTOMATIC position updating of any kind. What you could do, however, was to update the PPOS in the nav computer MANUALLY, if you were sure it was necessary. That is the scenario I was addressing.

Re automatic updating of PPOS in FMSs, DME/DME remained the most accurate method en-route until the incorporation of GPS around 1990. I say "en-route" because updating at the start of the take-off, and LLZ/DME on approach (where available), were (and are) more accurate.

One day, for one reason or another, Uncle Sam may have to turn GPS off. And he probably won't want to issue a NOTAM until after he's done it...