SAR S-92 Missing Ireland
crab
Radar operates at the speed of light, 300,000,000 kilometres per second, or 300 metres every millionth of a second. Modern electronic components have reduced the size and increased the power of radar systems, but there is a physical limit to how much functionality can be built into a single system. Low power motion detector systems can operate at ranges of a few feet, but a system capable of picking up targets at a range of 40 miles will have its' performance compromised at very close ranges due to the laws of physics. In order to obtain unhindered close range performance, the use of two identical physically coupled but electrically isolated scanning dishes would be required. The sheer size of such an antenna system would preclude fitting in most aircraft.
I am as certain as I can be that the crew did not see Black Rock on the radar display and that Black Rock was certainly not presented on the display or recognised as presenting imminent danger. I am also sure that such an experienced crew were lulled into a false sense of security by perhaps not appreciating the ultimate physical limits of the radar system due to weather and possibly a latent deterioration in system performance due to the ageing of components.
A flawed approach pattern, lack of detailed mapping information, radar returns that were "unreliable" at close range and a loss of situational awareness in dodgy weather all combined to create the circumstances for this tragedy to happen. The classic holes in the swiss cheese lining up analogy of many aviation accidents.
sargs
Yes, 75 metres is a theoretical minimum range with a half microsecond transmit/receive switchover. You are quite correct that quarter wave stubs of waveguide are used to provide almost perfect isolation between the transmit and receive sections and that no electronic components are necessary to provide this isolation, just careful design and adherence to mechanical tolerances. But radar waveguide alters in size due to temperature changes and/or mechanical damage when knocked so some part of the transmitter energy can end up being fed towards the receiver circuits as the transmit pulse is fired off. To cater for this, there is a very high speed schottky diode mounted across the waveguide of the receiver section designed to short out if receive signals exceed a small fraction of a volt. It is this component that ages and needs to be replaced at regular intervals. When I worked for Marconi, every radar engineer carried half a dozen replacements in their tool kit. It was standard practice to replace them on every visit for servicing or fault finding.
Radar operates at the speed of light, 300,000,000 kilometres per second, or 300 metres every millionth of a second. Modern electronic components have reduced the size and increased the power of radar systems, but there is a physical limit to how much functionality can be built into a single system. Low power motion detector systems can operate at ranges of a few feet, but a system capable of picking up targets at a range of 40 miles will have its' performance compromised at very close ranges due to the laws of physics. In order to obtain unhindered close range performance, the use of two identical physically coupled but electrically isolated scanning dishes would be required. The sheer size of such an antenna system would preclude fitting in most aircraft.
I am as certain as I can be that the crew did not see Black Rock on the radar display and that Black Rock was certainly not presented on the display or recognised as presenting imminent danger. I am also sure that such an experienced crew were lulled into a false sense of security by perhaps not appreciating the ultimate physical limits of the radar system due to weather and possibly a latent deterioration in system performance due to the ageing of components.
A flawed approach pattern, lack of detailed mapping information, radar returns that were "unreliable" at close range and a loss of situational awareness in dodgy weather all combined to create the circumstances for this tragedy to happen. The classic holes in the swiss cheese lining up analogy of many aviation accidents.
sargs
Yes, 75 metres is a theoretical minimum range with a half microsecond transmit/receive switchover. You are quite correct that quarter wave stubs of waveguide are used to provide almost perfect isolation between the transmit and receive sections and that no electronic components are necessary to provide this isolation, just careful design and adherence to mechanical tolerances. But radar waveguide alters in size due to temperature changes and/or mechanical damage when knocked so some part of the transmitter energy can end up being fed towards the receiver circuits as the transmit pulse is fired off. To cater for this, there is a very high speed schottky diode mounted across the waveguide of the receiver section designed to short out if receive signals exceed a small fraction of a volt. It is this component that ages and needs to be replaced at regular intervals. When I worked for Marconi, every radar engineer carried half a dozen replacements in their tool kit. It was standard practice to replace them on every visit for servicing or fault finding.
Join Date: Jun 2002
Location: home and abroad
Posts: 582
Likes: 0
Received 0 Likes
on
0 Posts
Originally Posted by [email protected]
I think that might be a very valid point in that after a couple of years of 'magenta' all those hard-won lessons on old technology get forgotten and reliance on the new and shiny kit becomes absolute.
With regard to using the radar for ground mapping: I never trust the thing to show me what I need to see, so I constantly select short to medium ranges, adjust gain, tilt and mode just to make sure I don't miss anything that might harm us. Low level below MSA, especially at night/in poor visibility is a dangerous place to be. Changing scales and modes also helps in showing something hidden under a waypoint marker. Never take anything for granted ..
Join Date: Apr 2017
Location: Edinburgh
Posts: 4
Likes: 0
Received 0 Likes
on
0 Posts
Originally Posted by [email protected]
how long do you think it would take to task a UK-based aircraft (probably from Caernafon) to fly across Ireland, refuel and then go to the FV compared to tasking the Sligo aircraft to something that might technically be outside their area?
My point is not about taking Sligo out of the picture. I can see why Sligo would have been tasked, even if HMCG was the principal mover.
Having flown pre, post, and during the magenta revolution and experienced the full spectrum of technology across approx four decades worth of aircraft innovation I believe it is overly-simplistic just to blame the technology (and over-reliance on it). Crews have been routinely flying serviceable aircraft into the ground since aviation began.
The reasons vary and causes range from straightforward to vastly complex. Wherever there is human decision-making involved in a system (of any kind) there will be accidents. James Reason's 'Swiss-cheese' model is an eye-catching representation of the way that a combination of errors conspire to cause tragedy. Even this though is a simplistic view; the investigators on this one will have a hell of a job exposing all the relevant factors, and their complicated relationships, that caused this tragedy.
Those that shout "more training" and "less automation" may feel better but the picture is vastly more complicated than that. Learning from accidents like this will no doubt help safety culture to evolve but, in my opinion, the human element combined with the limits of current technology will continue to conspire and result in accidents.
I admired Bristow's optimism with their recent 'Target Zero' safety program and statement that they genuinely believe that accident rates can be reduced to zero. I'm not so sure.
The reasons vary and causes range from straightforward to vastly complex. Wherever there is human decision-making involved in a system (of any kind) there will be accidents. James Reason's 'Swiss-cheese' model is an eye-catching representation of the way that a combination of errors conspire to cause tragedy. Even this though is a simplistic view; the investigators on this one will have a hell of a job exposing all the relevant factors, and their complicated relationships, that caused this tragedy.
Those that shout "more training" and "less automation" may feel better but the picture is vastly more complicated than that. Learning from accidents like this will no doubt help safety culture to evolve but, in my opinion, the human element combined with the limits of current technology will continue to conspire and result in accidents.
I admired Bristow's optimism with their recent 'Target Zero' safety program and statement that they genuinely believe that accident rates can be reduced to zero. I'm not so sure.
Join Date: Jan 2011
Location: on the cusp
Age: 52
Posts: 217
Likes: 0
Received 0 Likes
on
0 Posts
That is a very valid question.
With the radar set to 10 NM range on the PM's (copilot) NAV screen in this case, the waypoint symbol for "BLKMO" would hide the radar return from Blackrock. If using 5 NM, or even better 2.5 NM range, the return from Blackrock would be relatively larger than the waypoint symbol, and hence get your attention.
I tried this a couple of days ago (in the S92 with the Primus 701A radar) using an oil rig as "Blackrock". With 10 NM range, GMAP2, optimal gain and tilt, and at a distance of 1-2 NM from the oil rig it was very difficult to actually spot the return because of the waypoint symbol. Blackrock might have given a slightly larger return than an oil rig though.
With the radar set to 10 NM range on the PM's (copilot) NAV screen in this case, the waypoint symbol for "BLKMO" would hide the radar return from Blackrock. If using 5 NM, or even better 2.5 NM range, the return from Blackrock would be relatively larger than the waypoint symbol, and hence get your attention.
I tried this a couple of days ago (in the S92 with the Primus 701A radar) using an oil rig as "Blackrock". With 10 NM range, GMAP2, optimal gain and tilt, and at a distance of 1-2 NM from the oil rig it was very difficult to actually spot the return because of the waypoint symbol. Blackrock might have given a slightly larger return than an oil rig though.
For those calculating radar MDR remember it is not just PW but also the Receiver Recovery Time that has to be used in the calculation.
Join Date: Dec 2013
Location: Norfolk
Age: 67
Posts: 1
Likes: 0
Received 0 Likes
on
0 Posts
FC80
The aircraft flew into a rock that clearly was not visible or obvious to the crew on the radar display. I (and others) have proposed several reasons how and why this could happen.
It is indeed a long time since I was a hands on radar engineer but the physics hasn't changed since then, only the equipment has got smaller and lighter.
The aircraft flew into a rock that clearly was not visible or obvious to the crew on the radar display. I (and others) have proposed several reasons how and why this could happen.
It is indeed a long time since I was a hands on radar engineer but the physics hasn't changed since then, only the equipment has got smaller and lighter.
Join Date: Sep 2003
Location: uk
Posts: 119
Likes: 0
Received 0 Likes
on
0 Posts
FC80
The aircraft flew into a rock that clearly was not visible or obvious to the crew on the radar display. I (and others) have proposed several reasons how and why this could happen.
It is indeed a long time since I was a hands on radar engineer but the physics hasn't changed since then, only the equipment has got smaller and lighter.
The aircraft flew into a rock that clearly was not visible or obvious to the crew on the radar display. I (and others) have proposed several reasons how and why this could happen.
It is indeed a long time since I was a hands on radar engineer but the physics hasn't changed since then, only the equipment has got smaller and lighter.
Join Date: Dec 2013
Location: Norfolk
Age: 67
Posts: 1
Likes: 0
Received 0 Likes
on
0 Posts
Transcript from the CVR suggests that they were not aware of their position in relation to Black Rock. If it had been clearly visible on the radar, someone would have commented on it. It was the FLIR operator that alerted the pilot to the presence of something ahead.
This crew were betrayed by a combination of factors with inadequate navigational data and a failure of the radar to highlight the presence of Black Rock immediately ahead for some reason being primary factors.
This crew were betrayed by a combination of factors with inadequate navigational data and a failure of the radar to highlight the presence of Black Rock immediately ahead for some reason being primary factors.
Join Date: Apr 2017
Location: Dublin
Posts: 6
Likes: 0
Received 0 Likes
on
0 Posts
Transcript from the CVR suggests that they were not aware of their position in relation to Black Rock. If it had been clearly visible on the radar, someone would have commented on it. It was the FLIR operator that alerted the pilot to the presence of something ahead.
This crew were betrayed by a combination of factors with inadequate navigational data and a failure of the radar to highlight the presence of Black Rock immediately ahead for some reason being primary factors.
This crew were betrayed by a combination of factors with inadequate navigational data and a failure of the radar to highlight the presence of Black Rock immediately ahead for some reason being primary factors.
Join Date: Jul 2007
Location: Canada
Posts: 228
Likes: 0
Received 0 Likes
on
0 Posts
I dunno, ...
player104
A useful video that shows some of the pitfalls when interpreting the radar display. However note the range settings on the display and target return distances of between 20 and 40 miles. At close range settings the display can be much harder to interpret although all the points mentioned still apply.
A useful video that shows some of the pitfalls when interpreting the radar display. However note the range settings on the display and target return distances of between 20 and 40 miles. At close range settings the display can be much harder to interpret although all the points mentioned still apply.
Guest
Posts: n/a
Tezzer. They loaded the APBSS approach (well, flight plan really) which included BLKMO as it's first waypoint. They clearly did not know that BLKMO was Blackrock and that Blackrock was a 300ft tall lump of rock. Part of that was their fault for not reading the notes to the plate and the captions on the plate, but a large share of the issue rests with the operator for not making clear what this chart and approach was designed for, and more importantly what it was not designed for.
O'leary. What's becoming clear is that BLKMO didn't appear as a target (which they would have avoided), it most likely appeared as a caption for BLKMO (which they were aiming for) which happened to be obscuring the target.
O'leary. What's becoming clear is that BLKMO didn't appear as a target (which they would have avoided), it most likely appeared as a caption for BLKMO (which they were aiming for) which happened to be obscuring the target.
Last edited by puntosaurus; 18th Apr 2017 at 17:49.
Gouli,
"One point three err miles to run to eh blackmo…
and after that its bravo kilo sierra delta alpha..."
Would indicate that the PM knew where they were - what they clearly were not aware of was the vertical extent of the island or they would not have been at 200 ft asl.
"One point three err miles to run to eh blackmo…
and after that its bravo kilo sierra delta alpha..."
Would indicate that the PM knew where they were - what they clearly were not aware of was the vertical extent of the island or they would not have been at 200 ft asl.
I'm guessing there have been numerous 92 SAR crews around the bazaars (UK, Eire, Canada, US, Australia and Brunei) using their training sorties to fly towards small islands and trying to replicate situations where they can't see them on radar! If not, might be an idea!
Join Date: Jul 2007
Location: Canada
Posts: 228
Likes: 0
Received 0 Likes
on
0 Posts
I suppose, but ...
O'leary. What's becoming clear is that BLKMO didn't appear as a target (which they would have avoided), it most likely appeared as a caption for BLKMO (which they were aiming for) which happened to be obscuring the target.
Join Date: Jan 2005
Location: Norway
Posts: 9
Likes: 0
Received 0 Likes
on
0 Posts
Thanks, my beer mat calculations had Blackrock as sub tending 900m at a couple of miles (including antenna beamwidth). At 10 nm display across, then each pixel would be about 50m and so 18 pixels. Symbols are often between 16 and 32 pixels. My concern was that if a symbol was coincident with the return then it might not have been as compelling an indication as it should. Particularly if not being looked for. All built on assumptions of what display ranges were selected so just wild conjecture. Thanks for the answer.
For those calculating radar MDR remember it is not just PW but also the Receiver Recovery Time that has to be used in the calculation.
For those calculating radar MDR remember it is not just PW but also the Receiver Recovery Time that has to be used in the calculation.
Yes, it is, as it turns out. That does not mean the crew were aware that it was a waypoint that was located with a rock sticking 300 feet out of the ocean. They very clearly were not.
In answer to a question posed by Rotorspeed about why they would select an APP1 mode for descent, and whether that was imposed by the operator, it would be helpful to have 5 more minutes of the CVR before what is in the report.
I think the report makes it clear that the APBSS diagram was being used. It is not clear if the waypoints in that diagram are structured into a selectable approach in the FMS. If they are, then that supports a view that this is "imposed by the operator". The next question is if the operator imposes it as an IFR approach. If that is so, it is clearly inadequate, lacking vertical guidance.
There are a couple of features of this diagram that have the potential for the crew being led into the danger of presuming it is an IFR procedure for IMC conditions. The first is the lighter blue shading on either side of the leg lines. That shading only appears over water, except at BLCKMO. It does not appear over any terrain, or so it appears. I expect most would think it indicates an obstacle free safe space. Which it does. Even at 200 feet. Except for Blackrock, or over land around Blacksod.
The second feature is that it cannot be flown visually, although it obviously can be flown VFR. However to go from waypoint to waypoint, the flight must utilize instrumentation in the cockpit. Other than the helipads/lights, the other waypoints cannot be identified visually.
So the waypoints must be in the nav databases of helicopters which use this route. That would have to include the helicopters which use it for its primary purpose, transiting between Blacksod and Blackrock. Those aren't S-92s. But they are helicopters. We know an Air Corps machine was used to place the investigators there. So some other operator has a version of that diagram, and somehow CHC has it in its route guide. Legacy procedure seems reasonable.
It seems this crew has little to no local knowledge. That fact alone, and the availability of APBSS in the aircraft, underly this accident to a considerable degree.
Great discussion on radar. One thing from the CVR that stands out is that the crew operated on an assumption about the environment they were in. They were wrong to believe 200 feet was a safe altitude. When Capt. Duffy says something large to the right, it is obvious that assumption is challenged and that 200 feet, where they are and at 75 knots, may be/was a dangerous place to be and that a climb to MSA was a priority, followed by a different lookdown with the radar. Radar guys, they will see the picture then, or no? SAR pilots, yes or no on the immediate climb?
I would be interested on hearing how synthetic vision gear may have avoided this. Is it true that like EGPWS, if Blackrock isn't in the database, same problem?
If the purpose of the preliminary report is to immediately advise other operators and aircrew of issues, I think another five minutes of immediately preceeding CVR would be very enlightening for that purpose.
In answer to a question posed by Rotorspeed about why they would select an APP1 mode for descent, and whether that was imposed by the operator, it would be helpful to have 5 more minutes of the CVR before what is in the report.
I think the report makes it clear that the APBSS diagram was being used. It is not clear if the waypoints in that diagram are structured into a selectable approach in the FMS. If they are, then that supports a view that this is "imposed by the operator". The next question is if the operator imposes it as an IFR approach. If that is so, it is clearly inadequate, lacking vertical guidance.
There are a couple of features of this diagram that have the potential for the crew being led into the danger of presuming it is an IFR procedure for IMC conditions. The first is the lighter blue shading on either side of the leg lines. That shading only appears over water, except at BLCKMO. It does not appear over any terrain, or so it appears. I expect most would think it indicates an obstacle free safe space. Which it does. Even at 200 feet. Except for Blackrock, or over land around Blacksod.
The second feature is that it cannot be flown visually, although it obviously can be flown VFR. However to go from waypoint to waypoint, the flight must utilize instrumentation in the cockpit. Other than the helipads/lights, the other waypoints cannot be identified visually.
So the waypoints must be in the nav databases of helicopters which use this route. That would have to include the helicopters which use it for its primary purpose, transiting between Blacksod and Blackrock. Those aren't S-92s. But they are helicopters. We know an Air Corps machine was used to place the investigators there. So some other operator has a version of that diagram, and somehow CHC has it in its route guide. Legacy procedure seems reasonable.
It seems this crew has little to no local knowledge. That fact alone, and the availability of APBSS in the aircraft, underly this accident to a considerable degree.
Great discussion on radar. One thing from the CVR that stands out is that the crew operated on an assumption about the environment they were in. They were wrong to believe 200 feet was a safe altitude. When Capt. Duffy says something large to the right, it is obvious that assumption is challenged and that 200 feet, where they are and at 75 knots, may be/was a dangerous place to be and that a climb to MSA was a priority, followed by a different lookdown with the radar. Radar guys, they will see the picture then, or no? SAR pilots, yes or no on the immediate climb?
I would be interested on hearing how synthetic vision gear may have avoided this. Is it true that like EGPWS, if Blackrock isn't in the database, same problem?
If the purpose of the preliminary report is to immediately advise other operators and aircrew of issues, I think another five minutes of immediately preceeding CVR would be very enlightening for that purpose.
Last edited by cncpc; 18th Apr 2017 at 19:12.
Having flown pre, post, and during the magenta revolution and experienced the full spectrum of technology across approx four decades worth of aircraft innovation I believe it is overly-simplistic just to blame the technology (and over-reliance on it). Crews have been routinely flying serviceable aircraft into the ground since aviation began.
The reasons vary and causes range from straightforward to vastly complex. Wherever there is human decision-making involved in a system (of any kind) there will be accidents. James Reason's 'Swiss-cheese' model is an eye-catching representation of the way that a combination of errors conspire to cause tragedy. Even this though is a simplistic view; the investigators on this one will have a hell of a job exposing all the relevant factors, and their complicated relationships, that caused this tragedy.
Those that shout "more training" and "less automation" may feel better but the picture is vastly more complicated than that. Learning from accidents like this will no doubt help safety culture to evolve but, in my opinion, the human element combined with the limits of current technology will continue to conspire and result in accidents.
I admired Bristow's optimism with their recent 'Target Zero' safety program and statement that they genuinely believe that accident rates can be reduced to zero. I'm not so sure.
The reasons vary and causes range from straightforward to vastly complex. Wherever there is human decision-making involved in a system (of any kind) there will be accidents. James Reason's 'Swiss-cheese' model is an eye-catching representation of the way that a combination of errors conspire to cause tragedy. Even this though is a simplistic view; the investigators on this one will have a hell of a job exposing all the relevant factors, and their complicated relationships, that caused this tragedy.
Those that shout "more training" and "less automation" may feel better but the picture is vastly more complicated than that. Learning from accidents like this will no doubt help safety culture to evolve but, in my opinion, the human element combined with the limits of current technology will continue to conspire and result in accidents.
I admired Bristow's optimism with their recent 'Target Zero' safety program and statement that they genuinely believe that accident rates can be reduced to zero. I'm not so sure.
Simple fix for the future?
Lose leg 1 from the diagram. Make BKSDA the IAF. Procedure turn right at MSA and descending after IAF to 2000. Balance unchanged other than putting in crossing altitudes for the next waypoints, MDA in the last leg, and a missed procedure. Draft a proper chart.
Don't make all the crossing altitudes after BKSDA at 200 feet. Same question arises, why spend more time down that low than necessary.
Lose leg 1 from the diagram. Make BKSDA the IAF. Procedure turn right at MSA and descending after IAF to 2000. Balance unchanged other than putting in crossing altitudes for the next waypoints, MDA in the last leg, and a missed procedure. Draft a proper chart.
Don't make all the crossing altitudes after BKSDA at 200 feet. Same question arises, why spend more time down that low than necessary.
Guest
Posts: n/a
cnpc. The assumption is that they refer to notes in the supporting pages. And I don't think the operator should be in the market for simple fixes. They should either publish approaches that are fit for purpose or not publish approaches at all and rely on the SAR training and alleviations to allow crews to find the safest approach.
Last edited by puntosaurus; 18th Apr 2017 at 19:52.