Does anyone know how much a real IRU would drift during a flight? I find it hard to believe that a real flight would operate with the IRU drift PIC simulates. By the end of any flight over about 3 hrs, I seem to be on radio navigation as the IRUs have drifted to 10 or 15 minutes off. This is espcially scary over the Atlantic when I'm on IRS nav only. I don't think this is my problem, as my IRS always starts perfectly aligned.Any thoughts?

"By the end of any flight over about 3 hrs, I seem to be on radio navigation as the IRUs have drifted to 10 or 15 minutes off."Not sure what you mean here, Smitty, but if you're saying that your FMC position (which is currently using radio updating) is different from the raw IRS positions by 10~15 minutes after a 3 hour flight, then I would consider this to be excessive.1 minute = 1 nm, if I remember correctly. Therefore your IRS has drifted 10~15nm on a 3 hour flight.Here's a chart for a 747-400, which should be similar to the 767, showing acceptable errors in green...http://www.ozemail.com.au/~iriddell/Error.gifNormally, however, I would expect errors less than 5nm on even the longest 747-400 flight (13~14hours).Sorry, but I can't explain the large errors you are experiencing.Rgds.Ian.

After many longhaul 767 flights, including overwater flights such as Syd-Hnl, and flights across Indonesia where the navaids dont work anyway, I have never seen an error greater than 1nm. Typical IRS errors at the end of a flight (this includes radio updating) are 0.1-0.4nm.

Hi,Drift is determined at random between min/max rates each time the IRU is aligned. Because we suppose each pilot flies another aircraft (thus with another IRUs) everyday. But I admit the maximal possible drift is bit too large. I programmed it this way to add a bit spice to you flights guys ;-) I've also a question for Ian. If I understand well the chart, radial error increases during the first hours of flight (which seems logical) but keep constant thereafter, as if the drift suddenly gets null. Why ? Regards,Laurent Crenier_____________________________PIC Panel co-designer

I'm not Ian but I suppose that a drift of more than 25 nm's is never acceptable, no matter how long the flight.

"If I understand well the chart, radial error increases during the first hours of flight (which seems logical) but keep constant thereafter, as if the drift suddenly gets null. Why ? "I can't say that I completely understand the chart either, Laurent, but as KB says, the chart probably doesn't represent actual drift, but represents the limits of allowable drift. 35 nm was probably an absolute limit imposed by aviation regulatory departments (FAA, CAA, etc)On a normal flight, I'd say that the most drift (on a time basis) would occur when the aircraft is taxying, turning, etc (i.e. when the aircraft is being subjected to the most accelerative forces). On a short flight, a substantial amount of the total flight time would be spent doing this, so the limits would be set a little higher (on a time basis). However, this doesn't explain the graph's strange shape.Cheers.Ian.