GPS APPROACHES
Crab,
The concentration of my last post was on a change that will happen (DGPS with EGNOS); whether industry will fund further development of a device which is primarily used for weather detection and avoidance, I somewhat doubt.
However, en passant you have raised an important issue:
and it is therefore necessary - in the interest of safety - to correct the impression that you may have given (certainly with regard to civilian ARA procedures). As you will see from the text below, all procedures that I am aware of provide a heading offset to ensure that the helicopter is not "pointing at or near to" the target (in JAR-OPS the target miss is set to 300m). The 'offset' is an important part of the procedure; this, and the accuracy issues discussed below, led to the review of the ARA procedure and establishment of the MAPt (at least in the JAA) to 0.75 miles. The stark fact is that, unlike runways procedures, the largest obstacle in the flight path is the target itself.
The elimination/reduction of some of the positioning and tracking inaccuracies and the reduction of pilot workload that will come with the use of DGPS, will be the key to reducing the minima.
Rather than to repeat a previous thread, here is a previous post which explains the issue of RADAR errors and heading offsets:
What was also missed from this original post was the recognition that the CASA ARA system requires a heading offset of 15 degree at the OIP.
Jim
The concentration of my last post was on a change that will happen (DGPS with EGNOS); whether industry will fund further development of a device which is primarily used for weather detection and avoidance, I somewhat doubt.
However, en passant you have raised an important issue:
But since the final element of a rig letdown IMC is going to be pointing at or near to it, will using DGPS let the operators use reduced minima or must they stay the same because the radar is still the limiting factor?
The elimination/reduction of some of the positioning and tracking inaccuracies and the reduction of pilot workload that will come with the use of DGPS, will be the key to reducing the minima.
Rather than to repeat a previous thread, here is a previous post which explains the issue of RADAR errors and heading offsets:
JAR-OPS (ICAO PansOps based) criteria has an Offset Initiation Point at 1.5 nm - at which point a 10 degree heading change is made which, if continued beyond the MAPt (not permitted by the way), would pass the rig at a mean distance of 300m - 400m. The choice of the MAPt of 0.75 nm was based (very simply) on an aggregation of the position error, RADAR (tracking and ranging) errors, pilot reaction and rate of turn (at the MAPt) to give a probability of collision with the rig at less than or equal to the ICAO Risk Model acceptable target. At the MAPt, the pilot has to fly a rate one turning missed approach of at least 30 degrees into a Missed Approach Arc (identified on RADAR by 5 nm from the target).
An Offshore Standard Approach Procedure (OSAP) based upon the FAA AC 90-80B would have one of two type of approaches: the delta 30 has a 30 degree heading change initiated at the OIP - a distance from the rig of 1.1 nm - the resulting track will take the helicopter to the MAP at a tangent of a circle with a radius of 0.6 nm from the rig; the second method requires the pilot to establish - by 2 nm from the rig - a parallel offset track which passes the rig by 0.5 nm - the MAP is set at 0.7 from the rig.
Each of these methods - JAA and FAA - has its merits but both have defined missed approach procedures established to avoid collision with the rig by a safe margin.
Subjectively, it could be observed that the JAA procedure appears to be optimized for rig acquisition at the MAPt as the rig should be seen on the RADAR, and in the front window, at that point; the OSAP procedure appears to be optimized for missed approach as, at the MAP, it is improbable that the rig will be seen in the front window.
Speaking for the JAA procedure, the MAPt could be closer to the rig if the FAP, OIP and MAPt could be established by a more accurate method than just the RADAR (differential GPS for example). This would remove some of the tracking and ranging errors which now are accounted for.
An Offshore Standard Approach Procedure (OSAP) based upon the FAA AC 90-80B would have one of two type of approaches: the delta 30 has a 30 degree heading change initiated at the OIP - a distance from the rig of 1.1 nm - the resulting track will take the helicopter to the MAP at a tangent of a circle with a radius of 0.6 nm from the rig; the second method requires the pilot to establish - by 2 nm from the rig - a parallel offset track which passes the rig by 0.5 nm - the MAP is set at 0.7 from the rig.
Each of these methods - JAA and FAA - has its merits but both have defined missed approach procedures established to avoid collision with the rig by a safe margin.
Subjectively, it could be observed that the JAA procedure appears to be optimized for rig acquisition at the MAPt as the rig should be seen on the RADAR, and in the front window, at that point; the OSAP procedure appears to be optimized for missed approach as, at the MAP, it is improbable that the rig will be seen in the front window.
Speaking for the JAA procedure, the MAPt could be closer to the rig if the FAP, OIP and MAPt could be established by a more accurate method than just the RADAR (differential GPS for example). This would remove some of the tracking and ranging errors which now are accounted for.
What was also missed from this original post was the recognition that the CASA ARA system requires a heading offset of 15 degree at the OIP.
Jim
Jim, thanks for the detail, I was aware that there was a heading offset, hence the wording of my post, but not how much.
I come back to the point that if you had a radar with better resolution you wouldn't need such a large safety margin in the procedure as the position of the obstacle (and also therefore the landing area) would be more accurately portrayed in the cockpit. If you trust DGPS sufficiently to reduce the minima because the tracks will be more accurately flown you must trust it to get closer because the position of the obstacle is, theoretically, more accurately known. Personally I would rather rely on real time radar info than a possibly misplotted GPS position.
I come back to the point that if you had a radar with better resolution you wouldn't need such a large safety margin in the procedure as the position of the obstacle (and also therefore the landing area) would be more accurately portrayed in the cockpit. If you trust DGPS sufficiently to reduce the minima because the tracks will be more accurately flown you must trust it to get closer because the position of the obstacle is, theoretically, more accurately known. Personally I would rather rely on real time radar info than a possibly misplotted GPS position.
Aser,
I have PMed you.
Crab,
I'm not sure a RADAR certificated for this type of operations is an option because the of the limited market.
However, you have raised one or two interesting points; unless industry can provide a RADAR that is certificated for these type of operations it is a non-starter; DGPS without obstacle detection and avoidance is also a non-starter.
The revised ARA (or whatever name it takes in the future) will have to meet the hazard analysis required for the introduction of a new procedure. In my view a single piece of equipment will not be able meet the requirement and the combination of DGPS and obstacle-detection-and-avoidance (combined with RADALT) will be required.
The issue of data-base integrity is one which has troubled the industry for some time. Clearly any risk assessed instrument procedure cannot have pilot-entered waypoints; it was this that led to my remark on one of the early posts:
Unless there is menu selection which can provide a 360 degree approach capability for any location, pilots might be tempted to circumvent the formalism of the system. However, as with RADAR certification, such a software addition will not be provided for a limited market because of the level of software certification that will be required (being a safety related modification).
For all of these reasons, the CAA research is interesting; let's see the result when the research is concluded.
It could be that onshore procedures are a much simpler issue and are concerned mainly with system integrity issues (which will probably be resolved with EGNOS and Galileo). I still consider that 170's suggestion of a combined DGPS and INS provides the best solution for locations without access to other aids.
Jim
I have PMed you.
Crab,
I'm not sure a RADAR certificated for this type of operations is an option because the of the limited market.
However, you have raised one or two interesting points; unless industry can provide a RADAR that is certificated for these type of operations it is a non-starter; DGPS without obstacle detection and avoidance is also a non-starter.
The revised ARA (or whatever name it takes in the future) will have to meet the hazard analysis required for the introduction of a new procedure. In my view a single piece of equipment will not be able meet the requirement and the combination of DGPS and obstacle-detection-and-avoidance (combined with RADALT) will be required.
The issue of data-base integrity is one which has troubled the industry for some time. Clearly any risk assessed instrument procedure cannot have pilot-entered waypoints; it was this that led to my remark on one of the early posts:
One of the continuing issues is the loss of the ability to approach from any direction - once that has to be decided by the software, it will require extensive modification.
For all of these reasons, the CAA research is interesting; let's see the result when the research is concluded.
It could be that onshore procedures are a much simpler issue and are concerned mainly with system integrity issues (which will probably be resolved with EGNOS and Galileo). I still consider that 170's suggestion of a combined DGPS and INS provides the best solution for locations without access to other aids.
Jim
Thanks for the info Jim
