Hi guys,Do you use IRS Drift option? Is it drifting everytime or just when IRS NAV ONLY? I can't decide to use it or not..

Hi BozhanI use it. IRS drift all the time, even when in area of navaid cover - the average position is calculated when this is the case.

Thanks Tom,So IRS updates itself time to time from navaid positions and no need to worry about the drift unless flying oceanic or desert.. I'll tick it tonight. PIC all the way.. By the way, let's say you flying oceanic and after some 2000! miles you reached to a navaid's coverage, what happens next? I think drift may over 3-5 miles..

Hi,The IRS drift all the time, due to several gyroscopic effects (earth rate, transport wander et al). The position given to the pilot is calculated by the FMC as is taken from Navaids (RNAV) and the IRS.When the FMC has sufficient radio aids, it will solely rely on the these to determine it's position on the globe, however, IRS's are still available and drifting.When you get to an area of no operation coverage, the FMC then relies on IRS position, like over the atlantic.Then, once you get to the other side of the pond, you get an updated position from radio aids, and FMC then starts using these again as they are more reliable...Look at the EHSI and you'll suddenly 'jump' a few miles off track as the FMC switches to RNAV.This is the system used in 767 PIC. Real world, you've also got GPS position. Then there's difference aircraft which have different ways of using IRS/RNAV positions. Some take averages and so on.Hope this helpsJoe

Hi, Although I am not a real world pilot, I have worked quite a bit with IRS and positioning systems. I will try to give the explanation I have learned without getting too confusing (:-)) The IRS consists of three IRUs (Inertial Reference Units). Each of these IRUs are essentially accelerometres that can measure acceleration in each of the three axis. If you know your initial position and the acceleration in each of the three dimensions, you can compute your position. However, to do this, you will have to integrate (mathematically) with respect to time once to get the speed and an additional time to get the position. The problem is that each time you integrate, you get an additional constant into the equation so by the time you finally arrive at a position solution, you have an error that is dependent on time (if my memory serves me right, the error will be quadratic in time). This, essentially, is the reason the IRS will drift. I know from my work with some Honeywell equipment that to solve that, the position solution from, for instance, a GPS is used to "force" the IRS position solution back into place. The reason you still want to use the solution from the IRS is that it updates much more frequent than the GPS. For instance, the equipment I was working with had roughly 65000 solutions a second where the GPS only had one. The equipment gave three solutions: A pure GPS, a pure IRS and a blended solution. So with that, we had data at hand that allowed us to interpolate a relatively good position solution in between the GPS solutions. Hope this helps, if not give a holler :-)Boaz

Hi guys thank you ver much for answers, it's clear now.

Boaz,Nice explanation that everyone can understand.It is true that to get the position, one has to integrate acceleration twice (v=ds/dt, a=dv/dt), and this produces two CONSTANTS that are unknown, IF the limiting conditions are not known.I guess the drift does come mainly from the fact, that integration from acceleration will (infact twice) produce uncertainties that are actually the integration constants. We would have to know the situation very thoroughly to be able to foretell what the unknown constants actually are, and this is not possible in a 3D realm that also has ALL the uncertainties that only a REAL WORLD can have ;).So, IRL they use the GPS to verify the position and so on...This is interesting. I'm wondering if the laser-gyros themselves have a tendency to produce measurement inaccuracy to the measuring of acceleration, and how that is taken into account... Ie. just how accurate are the laser-gyros ???Tero

Hi all,I just read Delco Carusel INS System's manual and discovered it's position can be updated in air. ie when overflying a navpoint pilot can update current INS position with lat/long on chart. It's also have a automatic position updating system called "DME UPDATE" So they can fly without drift over navaid coverage then cross the pond and update IRS pos again. But if I get it correct IRS in 767PIC drifts all the time and no position update available. We have RNAV in 767, but what happens if a route contains multiple IRS NAV ONLY segments?

767's use IRS. You read an INS manual, they are different in operation so you can't really relate the two systems.

Hi,There is an important concept you might be missing here. You are right, the IRS's position will never be updated after alignment. BUT the position information you see on the ND or which is used for navigation does not come directly from the IRS but from the FMC. The FMC is one heck of a smart computer when it comes to computing present position. First, it uses the three IRS for position. The data is mixed and an average is calculated. (Note, not all 3 IRS's recieve the same priority). Then you have DME updating. But not just DME, also localizers are used aswell as VOR DME's. The most accurate is DME - DME. When radio aids are available, the FMC will use DME's to determine the current position. Even if a ground based DME should fail or send erroneous information, the FMC will notice this and 'ban' that radio station for future tuning (on that flight). Same applies for localizers and VORDME's. Now, the FMC will compare the three IRSs' position and the radio position and will determine how far the average of the IRS position has drifted. This way, as soon as you leave the radio updated region, your position will still be quite accurate for a while. The IRS will drift further and I haven't heard of FMC's that are capable of prognosing IRS drift trend. Thus, even if your route has a few IRS NAV ONLY segments as you put it, you won't notice a very big map shift. For that you would need a very long IRS NAV ONLY segment, such as crossing the atlantic.Regards,Mark

Hi Tero, You are right, today mainly Laser Ring Gyros are used. To explain briefly how they work imagine a tube that has been shaped like a triangle. At each end (or where any two legs meet), a mirror has been put at such an angle that any ray travelling within the tube will be reflected into the next segment of the triangle. This way, a single ray of light can travel the whole way through the triangle. The trick is then to send a pulse of laser light one way through the tube while another pulse is sent the other direction. If the LRG is submitted to a angular acceleration in the plane that LRG is in, one of the waves will have a longer distance to travel (we have to realize that the differences in distance is excessively small). When the two pulses return to the point where they originated, they will be made to interfere with each other. This causes a standing wave (forgive me if this is not the proper english term - it may be a direct translation of the term I learned in university here in Copenhagen). The frequency (I think, 29 summers is doing a lot to a man's memory) is directly proportional to the angular acceleration. Laser Ring Gyros are much more reliable than the old kind of IRUs where you had accelerometres on the gyro stabilized platform. However, you still have the problem with the error introduced when you integrate the accelerations measured to get a solution fix.Boaz

"The most accurate is DME - DME."I think you'll find that LOC updating is more accurate, Mark ;-)On aircraft fitted with GPS, the prioritized navigational modes for the Flight Management Computer System (FMCS) to determine its position are: IRS/LOC/(GPS) or IRS/MLS/(GPS) * IRS/LOC/(DME/DME) or (IRS/MLS/(DME/DME) Reserved IRS/LOC/(VOR/DME) or IRS/MLS/(VOR/DME) IRS/LOC or IRS/MLS Reserved IRS/GPS IRS/DME/DME IRS/VOR/DME IRS only GPS only DME/DME only VOR/DME only Or at least that's what the man from Honeywell told us a while ago. Note that IRS is mentioned in all but the very lowest position fixes, even though it may not actually be used to compute the current FMC position. This is because IRS position and radio position are regularly compared. If the two positions vary by more than certain amount, the FMC switches to IRS position only.Cheers.Ian.P.S. It may not be clear that IRU's have both accelerometers and laser gyros. Accelerometers are basically weights on springs. These measure aircraft acceleration in three axes (basically, up/down, forwards/backwards and left/right)... whereas the laser gyros measure angular changes in three axes (basically, pitch, roll and yaw). You cannot compute aircraft position without the two.

"On aircraft fitted with GPS, the prioritized navigational modes for the Flight Management Computer System (FMCS) to determine its position are:"Let me just add that this is applicable to the Honeywell FMC. I discovered, only today, that the Smiths FMC, which is fitted to 737's, handles the priorities differently ;-)Cheers.Ian.

"I think you'll find that LOC updating is more accurate, Mark ;-)"Thanks Ian, I mixed the two up!On a sidenote, do you know how high up localizers are available for position updates? Also, what's the range of a loc? 16 nm?Thanks,Mark

Localiser coverage extends out to 25nm 10* either side of the normal localiser axis...or 17nm 35* either side of the localiser axis.