PDA

View Full Version : 737 FMC position


xetroV
20th Feb 2007, 19:10
Hi all,

Question: consider a 737 classic with one FMC, two IRUs, two GPS receivers, and a single frequency-scanning DME radio. Should it be possible for a single malfunctioning IRU to cause a noticeable shift in the FMC position (thus a mapshift on the EHSI) if all other sensors are working fine?

I know it is possible, since this scenario happened during an actual flight, but I wonder if this is normal system behaviour. Based on the information from the Boeing FCOM, I'd say the FMC position is derived from GPS, DME/DME, VOR/DME or Localizer/DME in that order of priority, while IRS position is used as a reference only if external position updating is not available (of course, that raises the question: which IRS? Left or right?). Did the impossible occur?

Any ideas would be appreciated!

alexban
21st Feb 2007, 07:03
Are you sure you have GPS receivers on the classic. We don't have on any of the classics,but maybe it's an option.
Without the GPS it is quite posible to have a mapshift,but this is caused by a wrong DME position update,most often.
The IRS won't influence the FMC calculated position,maybe it was just a coincidence.Or maybe someone tried to update from the faulty IRS?:ugh:

BOAC
21st Feb 2007, 07:31
Like Alexban, I also am not aware of any 'GPS fits' in Classics other than a GPS card which can be fitted for EGPWS, but has no input to navigation.

Still, there are miriad versions of 737's around, so who knows (Captain Sandl?) but I suspect you have a 'standard' nav kit with a rogue IRS. Line selecting to the 'good' IRS would ease the problem.

If you do have GPS navigation it may be a 'patch' installation and the system may not default to the usual degradation rules?

xetroV
21st Feb 2007, 11:50
Yes, we do have GPS navigation; it was retrofitted several years ago, and my degradation list comes straight from our 733/734 FCOM (we also operate NG's, so I double-checked today to be sure I had used the correct manual). Your comments are interesting, though, because I wouldn't rule out a discrepancy between the operation manual and the actual retrofit status.

But even without GPS, I wouldn't have expected an FMC map-shift to occur with one faulty IRS: the degradation rules continue with DME/DME, VOR/DME, LOC/DME, LOC, and I am pretty sure DME/DME or at least VOR/DME would have been available at the time of occurrence. (A faulty DME update is possible, but highly unlikely, since the FMC position shift occurred at the same moment the IRU started drifting.)

Now, this map-shift could very well be normal system behaviour, but based on the FCOM only I wouldn't have thought so. :confused:

BOAC
21st Feb 2007, 13:17
Interesting!

In the event of a loss of radio (?GPS?) update the FMC position will drift slowly towards the default IRS position, thus giving you gradual 'map shift'.

Seat1APlease
21st Feb 2007, 13:54
With so many different options and retrofits etc. it's very difficult to give definitive answers, but isn't the FMC position a calculated compromise between the 2 IRS positions and radio and other positions? it is on the basic 300-400 kit.
If say the radio position is temporarily lost the FMC position continues to move relative to the IRS positions rather than give a huge instantaneous shift.
If that is the case then if one IRS goes faulty and starts drifting about then the calculated position could also wander about presumably causing a shift.

SR71
21st Feb 2007, 14:02
The IRS won't influence the FMC calculated position,maybe it was just a coincidence

I believe thats incorrect.

In a dual IRS configuration, the FMC uses 5 positions to arrive at the system position.

You've got left IRS position, right IRS position, radio position, best position and system position.

The best position is heavily biased towards the radio position, which of course may create problems if its inaccurate.

The system position is updated using the computed best position which is the navigation position used by LNAV.

This is all from Bill Bulfer's FMC Guide.

IMHO then, any malfunctioning IRS has the potential to cause a noticeable shift in FMC computed position.

Kit d'Rection KG
21st Feb 2007, 18:39
Yes, there will be drift. The position is resolved according to some very complicated algorithms. Best keep an eye on the raw data!

arba
22nd Feb 2007, 00:50
With GPS equipped .. I never got "IRS NAV ONLY"

Or my colleague was very fast to clear that message ?

xetroV
22nd Feb 2007, 11:37
Thanks for the replies so far. It is pretty hard to understand what happened without knowing the exact formulas used by the FMC. The FCOM doesn't provide any details (but that's what the crew in question had available).

Thanks SR71 for your suggestions! I probably need to get myself a Bill Bulfer FMC guide. :)

CaptainSandL
22nd Feb 2007, 18:23
Should it be possible for a single malfunctioning IRU to cause a noticeable shift in the FMC position (thus a mapshift on the EHSI) if all other sensors are working fine? The short answer is no it should not, but it may.

On non-GPS equipped aircraft, FMC position is biased approximately 80:20 towards radio position and the position determined by the L IRS. So if the L IRS was drifting it could influence the FMC position.

On GPS equipped aircraft, FMC position is calculated from GPS and ADIRU inputs. The FMC uses its onside GPS (or the offside GPS if the onside GPS data is invalid). No radio updating occurs, unless both GPS signals are lost or data is invalid.

BTW if you suspect the integrity of an IRS, you should select the POS REF 2/3 page after flight when you have come to rest and see what the residual groundspeed is of the IRS’s. The limits are laid down in the AMM, but as a rough guide any more than 6kts per hour of flight (after a full align) is unacceptable. Below is a photo of one I snagged last year after a 2hr airtest.

http://www.b737.org.uk/fmc_irs_drift.gif

S&L

(Edited for typos and to add photo)

xetroV
25th Feb 2007, 15:26
Thank you too, CaptainSandL. In the occurrence I described, the left IRS first started to drift; then after several minutes it failed altogether: the IRS FAULT light came on. Hence my strong belief that the mapshift was directly related to the IRS failure.

What's puzzling is that in this incident, after the left IRS had been switched off according to the IRS FAULT procedure, the FMC position still remained incorrect, despite two (retrofitted) GPS receivers being available (supposedly not only as stand-alone devices or for EGPWS purposes, but actually integrated in the FMC navigation logic).

BOAC
25th Feb 2007, 16:56
Ok - a quick stab at an answer from a snowy Newfi:)

Could it be that loss of one of the primary nav systems (an IRS) renders the whole system 'defunct' in terms of failure logic and that for 'safety' your system simply says ' you are on your own, matey'? As Captain S says, to answer another of your queries, it is normally the left IRS in the Classic which is the key (in your case not available), and I wonder whether line selection of the right as 'default' in your incident might have restored the pos?

Your tech pilot should be able to establish the system logic for your fit, I would have thought?

I'm not sure Bill B will help in this case as you appear to have a 'one-off' installation as far as I can see, although it is an excellent series to own for everything 737.

SR71
25th Jun 2008, 17:15
Had a similar situation to xetroV a few days ago....

IRS L suddenly developed a huge error - approximately 12nm.

The warning was the CDU message in the scratchpad.

Radio position was good, ANP below RNP and IRS R sensible.

I expected some map shift...but none to my knowledge.

I wondered whether if the IRS output is discontinuous, the FMS dumps that input into the system position computation, just like it does if the radio position from a suspect navaid falls outside the Circular Error of Probability (CEP).

Anyone?

Denti
25th Jun 2008, 19:47
The classic usually only uses the left IRU for reference, try to manually switch from left to right inflight and have your ND in the lowest range. You will allways see a mapshift and a small course correction if in VNAV. So if the left develops an error and you switch to right one you might see a rather large mapshift indeed.

ANP should however return to normal once you switch away from the faulty IRU. However if you got a "VERIFY POSITION" scratchpad message you will have the position error and pointers on the ND for the remainder of the flight even if navigation performance is ok. At least thats how i observed it flying the classics so far (no GPS on ours except for EGPWS).