IRS Drift
Joined: Sep 1998
Posts: 1,615
Likes: 1
From: wherever
Reviewing the OP, it does seem that since the errors were "all at the same time and with the same amount." as if we have been looking at an FMC error all along and nothing to do with the IRUs?
I am also puzzled by underfire post#4 "BTW, there is a mechanism for the pilot to update the IRU manually with a GPS coordinate." - this is new to me, since in my day, once aligned, the IRU did not have a 'position' to update, and 'position' was purely an FMC function. Have things changed?
I am also puzzled by underfire post#4 "BTW, there is a mechanism for the pilot to update the IRU manually with a GPS coordinate." - this is new to me, since in my day, once aligned, the IRU did not have a 'position' to update, and 'position' was purely an FMC function. Have things changed?
Very much yes!
Take a look at the spec for a 10 year old IRS: http://www51.honeywell.com/aero/comm..._Micro_IRS.pdf
Per Ardua ad Astraeus
Joined: Mar 2000
Posts: 18,575
Likes: 4
From: UK
Nice bit of kit. I can see this unit has effectively a 'mini FMC' built in. I still maintain that whatever 'updates' are available, only the derived positions are updated (via the Kalman filter) and not the platform alignment from which the root accuracy derives. Again, I suspect the OP problem was alignment positional data error or some sort of FMC computational malfunction to produce a common triple system error. It would be nice to see what engineers did as spanners asked
Joined: Jan 2008
Posts: 2,173
Likes: 51
From: Blighty (Nth. Downs)
Quote from underfire:
"The IRS needs to be updated or it does drift..."
Response from me:
"Really? Would you care to be more specific?"
Reply from underfire:
"IRU fast/regular ALIGN...."
underfire (again!),
You simply cannot realign an INS or IRS when the vehicle is in motion, which is what we are discussing here.
Quote from BOAC:
I still maintain that whatever 'updates' are available, only the derived positions are updated (via the Kalman filter) and not the platform alignment from which the root accuracy derives.
Agreed. Old hands will remember that, before the days of FMSs, we had dual or triple INS systems with an associated computer, into which you entered up to about 9 waypoints manually using the Lat/Long. IIRC, you could display track and distance between selected waypoints (or to the next one), the current TRK(T) and GS, the current HDG(T) and Drift, the current W/V (if you were TAS-equipped), or PPOS (present position).
If you were curious (or desperate), there was a facility to update the displayed PPOS. Unless the INSs were all really in trouble, in those pre-GPS days you could never hope to get a fix of position accurate enough in the air to improve on the (dead-reckoning) position that an INS was giving you. (The least-poor method was when you were passing about 30 - 60 nm north or south of a VOR/DME station.)
Latest IRSs are much more accurate than the INSs we had in the 1970s, but AFAIK they remain a DR (dead-reckoning) form of position estimation based on accelerations in azimuth from stationary at the start point. They are incapable of providing a FIX of position. If they get a bad start (mis-alignment) from the point of origin, particularly if the alignment latitude is slightly wrong, the track and GS values will be inaccurate, and therefore the position error builds up proportional to time. And they cannot be realigned when the vehicle is in motion.
"The IRS needs to be updated or it does drift..."
Response from me:
"Really? Would you care to be more specific?"
Reply from underfire:
"IRU fast/regular ALIGN...."
underfire (again!),
You simply cannot realign an INS or IRS when the vehicle is in motion, which is what we are discussing here.
Quote from BOAC:
I still maintain that whatever 'updates' are available, only the derived positions are updated (via the Kalman filter) and not the platform alignment from which the root accuracy derives.
Agreed. Old hands will remember that, before the days of FMSs, we had dual or triple INS systems with an associated computer, into which you entered up to about 9 waypoints manually using the Lat/Long. IIRC, you could display track and distance between selected waypoints (or to the next one), the current TRK(T) and GS, the current HDG(T) and Drift, the current W/V (if you were TAS-equipped), or PPOS (present position).
If you were curious (or desperate), there was a facility to update the displayed PPOS. Unless the INSs were all really in trouble, in those pre-GPS days you could never hope to get a fix of position accurate enough in the air to improve on the (dead-reckoning) position that an INS was giving you. (The least-poor method was when you were passing about 30 - 60 nm north or south of a VOR/DME station.)
Latest IRSs are much more accurate than the INSs we had in the 1970s, but AFAIK they remain a DR (dead-reckoning) form of position estimation based on accelerations in azimuth from stationary at the start point. They are incapable of providing a FIX of position. If they get a bad start (mis-alignment) from the point of origin, particularly if the alignment latitude is slightly wrong, the track and GS values will be inaccurate, and therefore the position error builds up proportional to time. And they cannot be realigned when the vehicle is in motion.
Joined: Jun 2009
Aviation Qualifications: Military
Posts: 1,856
Likes: 650
From: florida
I take issue with Chris' idea of in-flight alignment.
As far back as 1968 - 1969, the A-7D had both an inertial and a doppler. When coupled, we could get a great airborne alignment given initial launch point for the flight. Then we could select an update fix and refine persent position and all was well. Sometimes we took off and practiced ACM without a full alignment, but the nav system came thru and we were golden after 15 or 20 minutes.
Our prime mode for nav was doppler-inertial, but we had another four modes below that. For example, I flew from Hickam to Guam in the "airmass" mose - TAS and estimated winds plus heading. Had a tanker to fly with, plus 5 other SLUFs. After ten hours I was about 15 miles off. Wish Amelia Earhart would have been that close.
More about the cosmic stuff on the SLUF at the F-16 net site.
As far back as 1968 - 1969, the A-7D had both an inertial and a doppler. When coupled, we could get a great airborne alignment given initial launch point for the flight. Then we could select an update fix and refine persent position and all was well. Sometimes we took off and practiced ACM without a full alignment, but the nav system came thru and we were golden after 15 or 20 minutes.
Our prime mode for nav was doppler-inertial, but we had another four modes below that. For example, I flew from Hickam to Guam in the "airmass" mose - TAS and estimated winds plus heading. Had a tanker to fly with, plus 5 other SLUFs. After ten hours I was about 15 miles off. Wish Amelia Earhart would have been that close.
More about the cosmic stuff on the SLUF at the F-16 net site.
Joined: Jun 2009
Posts: 1,344
Likes: 80
From: Bedford, UK
If the wrong level datum was input into all 3 systems then I would think the error would be the same as 'g' is incorrectly coupled into the accelerometer channel. Schuler tuning should give this an 84 minute or so period (from memory) but perhaps the observation wasn't long enough to observe. In any case, I would have thought that the system will be correcting against real time GPS with the Kalman filter estimating both position & dynamics, and also estimating the error characteristics of the accelerometers and gyros. So all should have been good but with some unusally high errors in whatever channel was compromised (eg accelerometer bias).
