In that article in glenbrook's link, the main thing the creator of GPWS acknowledges is the limitations of the database - does that fact get included in any training?

SASless, you said

Are you sure you are correct in your statement about user data being able to be added to the EGPWS database? My understanding is that any new obstacles etc has to be notified to the equipment manufacturer for inclusion in the very infrequent updates. i don't think it is reasonable to expect the operator to check the whole egpws database for correctness.

I know the egpws database has its origins many years ago when computer memory was expensive and the database had extremely low resolution to keep the file size small. An obstacle such as a radio mast would show up as a 1km square block on the screen and the mast would not necessarily have been in the centre of the square. In fact, if the mast was exactly at the meeting point of 4 blocks on the map grid, it would show up as a 2x2 km block. I am not aware that the database has become higher resolution in recent times. Having said that, there is no excuse for Black Rock not to be in the database.

(I may be incorrect in the size of the egpws map grid blocks because, being an American product, I would have thought the units would have been non metric)

So why did no member of the crew notice a honking great lighthouse shining at them?

Even in IMC, when you get close to it, it will light up the inside of the cloud.

When you're VMC and co-alt, it should be shining right into the cockpit, and those things are BRIGHT!

Lighthouse service says it was on at the time (as in, remotely switched on), but was it actually working? Why would it not have been noticed? The first anyone seems to know of an obstacle (despite the Co-Pilot having Radar displayed in GMAP mode on his MFD), is when the crew in back ask for a right turn.

The 701 radar has pretty good ground mapping. And it will do it to a close enough range that objects don't disappear from the display until you're nearly on top of them (unlike the 660 radar), so why no visual of the light and no radar sighting of a b***dy great rock in front of you?

crab, are crew trained to appreciate that EGPWS is not approved for navigation? Or is that fact passed over and they learn to trust it more than perhaps they should?

Although, with their speed, altitude and the fact that the gear was down, warning and caution messages relayed to the crew would be limited and different to a situation where they were faster with the gear up.

Good question nooby and I think it will come back to the reliance on automation again - we get given the new gucci kit and don't expect it to go wrong.

The main difference between the Mk 3 and the Mk 3A Sea King was the AP - the Mk 3 was single channel, steam driven and it's failure modes could easily put you in the water so the crews watched it like a hawk. The 3A had a duplex digital AFCS which hardly ever went wrong, had benign failure modes and was very good so crews trusted it - sometimes more than they should have.

The S92 is in another dimension as far as capability goes so it is very easy to see how complacency can creep in.

Yes, and the report states that in the quote from Honeywell. It's done by the operator providing the data to Honeywell and they include it in the update cycle.

I asked that and another poster pointed out that R118, out of Sligo, was much more familiar with the area and would have known of Blackrock and the need for a higher altitude in that part of the approach.

That's what I thought. I read your statement as the operator having the ability to add data directly to the aircraft database without having to involve Honeywell.

I think the discussion about EGPWS is a bit of a red herring as in the described aircraft configuration, no warning would have been given even if Black Rock had been in the database as the system would have assumed the crew were making an approach to land on the helipad - low speed, gear down.

SAS

"One Man's Opinion here.....this Crew were handed a Time Bomb by Management, Training, and Safety. The Operator failed to change from old ways and failed to embrace new technology that requires a change of Mindset and Safety Culture."

That's a pretty damning statement about the management of this SAR operation. All those involved will no doubt have pride and strive to be professional, like the flight crew. What information are you relying upon to conclude that?

Did the crewman, who spotted the Island, give a strong enough warning ? Is there a language protocol that means you are giving an immediate instruction rather than an advisory ?

Alphanumeric wrote;-

Yet another uneducated conclusion. How do you know there was not a proper two crew approach brief? They both acknowledged they were unfamiliar with the area. The report does indicate there was an approach brief but acknowledges there was no mention of the presence of Black Rock nor the lighthouse. The big question was why? I cannot believe the crew were anything but conscientious in carrying out there task so one can only assume, there was a deficiency in the way the terrain information was presented or the way the procedure was designed. As I mentioned previously, I don't understand why the prescribed track went directly overhead a 300' high lump of granite and brick.

Alphanumeric, to blame the crew in such simplistic terms puts your own professionalism in doubt.

Any time I saw Rescue 115 or 118 head to Blacksod on Marine Traffic they route along the Coast Line, Rescue 115 heads up over Galway Bay to Clifden then along the coast to Blacksod, 118 departs Sligo straight along the coast then down to Blacksod, I guess with 116 coming cross country the first coastline they encounter is Blacksod

The report clearly mentions that the approach briefing notes and moving map contained data showing blackrock and its heights etc . It did not contain any comment as to whether the crew accessed such information

And this was not an approach to Blackrock as a LS - that would have required such a brief - this was part of a letdown to cloudbreak to get to Blacksod.

Therefore all the talk about databases, GPWS and especially the construction of the procedure is absolutely valid.

Do you include a brief during a procedural letdown to a rig on EVERY WP in the FMS? I suspect not - such detail would be confined to the final approach to the LS/rig itself.

All terrain is higher than sea level.

The accident report distinctly says there was an approach briefing. It can be inferred that the crew was unaware of Blackrock as a piece of terrain at 300 feet. It does seem that Capt. Fitzpatrick was aware of rocks there as she knows why the altitude warning sounds, but she thinks they are over the waypoint. So she knows the waypoint is over "terrain", but obviously doesn't know its height. In fact, even though they are only at 200 feet, she is not concerned by the alerter.

What does APBSS mean at the top of that chart we are considering some kind of buck shee or homemade approach "plate"? On looking at it, it seems that it is more likely a VFR day or night or reduced viz transit route from heliport to heliport as Blackrock and Blacksod are terminus points. There is no real indication that some approach procedure for Blacksod keys on Blackrock or the waypoint.

Inbound to Blackrock from the east, it is clear that the intention is to land at Blackrock heliport. It is a straight in approach with a presumption of VFR conditions or at worst reduced viz. There is no scenario of approaching it from the ocean to the west. That may provide an explanation at least why this seems to be a very inadequate IMC procedure. It wasn't meant to be.

Leaving Blackrock also assumes you are lifting off from the heliport for Blacksod, which does have a bit of an indirect routing. There is no scenario in which you pass over Blackrock.

For whatever that is worth, the question then is whether this "approach" is programmed into the FMS, or whether each waypoint is sequentially selected manually by a Direct To button and the waypoint. The error in that would be assuming it is an IFR approach.

Noooby

Maybe it was one important detail that most likely they didn't have the radar target (a b***dy great rock) in front of them! I may be wrong, but it is my understanding that the heading was approximately 120 degrees and due to the wind the track was 098 degrees. And if you have 10 NM scale on your radar, it is not always that easy to realise that you are drifting to an obstacle if you are pretty close (e.g. 0,3 nm) with that kind of drift angle... But I agree that the 701 radar is very good for ground mapping...

and 'children of the magenta' deride the Sea King blind arc! :=
Maybe a 'clear blind arc left' might have helped in this case? :(

No, there was never any intent to land at Blackrock - why on earth would there be? Blackrock was 2 things - firstly a WP in the company letdown route to get to Blacksod and secondly a bloody great rock on the way to do so. The fact that they were overlaid on a procedure without any clear indication of the threat is the real issue here.

Al-bert - agreed, you and I both know that it would not have happened in a SK letdown because we had a dedicated radar operator rather than an over-tasked co-pilot to stop us bumping into things.

Exactly Crab!

That would be quite impossible in a S92 due to its size. I've landed many times on the Blackrock pad, but with a much smaller Bo105 - admittedly the last time was almost 45 years ago - and from photos the pad hasn't increased in size since then.

Maybe I wasn't clear on that. I was talking about that APBSS generally, the Approach Blacksod South, I turns out is what that stands for. My point was to suggest that it seems to be a transit route from heliport to heliport at each lighthouse. Eastbound off Blacksod, the arrows going the other way lead to the heliport at Blackrock. Conversely, the route seems to be for a machine that has lifted off Blackrock heliport. Which eliminates the possibility of hitting Black Rock.

There is nothing on this "approach" that takes place east of Black Rock. So the part of the flight we know did go east seems to be some letdown procedure APPS1 at the crew's option, following which they would join into a procedure which is based on a helicopter taking off from Blackrock light and going to Blacksod, likely the normal procedure in servicing Blackrock light, which is unmanned.

Someone, perhaps you Crab, has already pointed out that this seems not to be a real IFR approach. Others have said it seems stupid to base a letdown point around a 300 feet rock. It is. But this chart does not show any procedure east of Black Rock.

Is there a APBR? page in the route guide, some approach for Black Rock? Or is it just the reverse of the Black Sod approach, or what seems like it may be the missed for Blacksod.

It is not unusual that this crew is not familiar with Black Rock. They are an S-92 crew in an aircraft that as I understand cannot use Black Rock heliport. So they have never been out there. They haven't even been to Blacksod for quite a while.

If you know Blackrock is 300 feet high, and you know APBSS is a chart limited to use for helicopters transiting between Blackrock and Blacksod, then you aren't considering it an approach for a machine that doesn't initially lift off Blackrock. Sure, it can be used from Black Rock eastbound, but only if you cross BLKMO at say 1500 and step down going through the other fixes to 200 at wherever makes sense to be at that altitude on the final into Blacksod.

I don't see why a flight needs to go that far out, when a Trans Down to VFR at 200 at say BKSDA works better.

Thanks Democritus, for confirming what I said in the post above in response to Crab. I had seen somewhere that you couldn't land an S-92 there as it seems tight to the structure. I never meant to be taken to mean that R116 was attempting to land at Blackrock.

Hopefully I've made my small issue clear about that APBSS being an "approach" only in terms of flight from one helipad to the other. And, that if you know the Black Rock waypoint is also a pad, and that it's 300 feet ASL, you can cook up something which begins at Black Rock, but at a higher altitude. I'd say that is what R118 did earlier, or used only part of the APBSS.

Crab,

But did not that Radar Operator have a Blind Spot looking ahead?

The 92 FLIR Operator identified the Terrain and asked for a Turn.....not the same advanced warning as the Radar Operator might have been able to give for sure.

We can agree that there are some glaring issues that should have been addressed long before this Crew took to the Air....and few of them were under their direct control.

As another ICG S-92 Landed at Blacksod just an hour or so before this Crew crashed.....it will be interesting to hear how they managed not to hit Blackrock in the process.

What did they do....or not do....that lead to their success I wonder?

I would enjoy being on the Accident Board.....and be able to ask so many questions of numerous parties about their actions and in-actions that might have played a role in the Accident Chain for this tragedy!

The blind arc was 14 deg either side of dead ahead. It was SOP to regularly 'clear the blind arc' by a brief turn left or right (depending on drift) under the direction of the radar operator. This particular accident, as Crab said, would not have occurred in one of our SKs.

Additionally.....yours was not a dual purpose Weather Radar and probably had much better definition of surface objects too I bet.

And being raw radar on the Mk 3 it had Admiralty Charts (including lighthouse characters and elevation) or OS maps overlaid as required.

Non-SAR operators posting on here and questioning the wisdom of this kind of approach need to remember that it should be a simple procedure for a well trained SAR crew. Imagine for one second that Blacksod was not a refuel point but the position of an injured person. Would there have been a procedure? Of course not, they'd have used the kit on board to ease their way in. As it was, they followed and trusted a procedure that it seems like none of them had flown and in doing so, missed the basics, namely checking what's on the radar.

It was not a Survivor....and was a planned destination for refueling.

One that was used by a an earlier flight that same evening.

The Operator had every opportunity to create a known safe IMC procedure for use by Crews....even those un-familar with the site. The Route Guide failed to rise to the level needed to accomplish that.

If it were a Survivor and thus could be anywhere within the Area of Operations then your comment would be valid.....but it was not.

Shortly after the crash, I went to Google Earth and took a gander at the area around Blackrock and Blacksod and noticed two bits of Rock near Blackrock and the thought that went through my mind was that it would be possible for a crew to perhaps over look them or confuse them with Blackrock somehow if the weather precluded visual contact with the Lighthouse illumination.

I do not believe this is the time or place to see a RAF SAR vs Civvie SAR argument be renewed. Pointing out the differences in equipment and procedures is fine....but what is being discussed is how an experienced Crew in a well equipped aircraft came into contact with terrain.

mp3
http://podcast.rasset.ie/podcasts/au...59713_232_.mp3

mp3
https://cdn.radiocms.net/media/001/a...audio_file.mp3


https://www.irishtimes.com/news/irel...-116-1.3048795


The clip from RTE, Graham Liddy suggests the obstacle data probably came from a state dept, Dept of Transport possibly getting data from OSI and IAA. Where does Honeywell source that data?

Hard to believe anyone has better maps of Ireland than OSI.

Isn't the light beam columnated? The light wouldn't go up into the cloud, it's angled down towards the surface. And at one flash every 12 seconds, they might not have been below cloud for long enough before hitting the rock for the light to be visible to them. But yes, from 500' up even if you were in cloud, you'd expect to see the light, or the clouds light up around you, every 12 seconds from a few miles out.

The co-pilot did have a target on radar, he reported "target at 6 miles 11 o'clock Large" and the pilot replied "just a small little island, that's BLMO itself". This call came 15 seconds after the co-pilot reported 1.3 miles to BLKMO. This I just don't understand, it's evidence of a breakdown in SA. I mean, when you've just been told BLKMO is 1.3 miles in front of you (according to GPS), and there is a radar target 6 miles ahead of you, how could that radar target possibly be BKLMO? What that target probably was is Duvillaun More, which you'd expect to be at 11 o-clock on radar because the next segment takes you to the right of it. And the co-pilot should have identified the radar target as a possible obstacle on the next segment by looking at the map, and he didn't question the pilot when it should have been obvious that a radar target 6 miles in-front of you couldn't possibly be a GPS waypoint only 1 mile in-front of you. And another thing I don't understand, when the co-pilot reported 1.3 miles to BLKMO, there MUST have been a big red target directly in front of him on the radar (assuming the radar was set to 10nm arc and 1.3nm is still in the detectable range). Also, the crewman told the pilot the island was visible directly ahead 15 seconds before impact, and to come right. In the 6 seconds before hitting the rock the FDR shows they were not making any heading change to ensure lateral separation.

From 200' above the surface and at 3nm range, a 300' island in-front of you would probably indicate on the radar screen as an obvious red blob with little depth, probably with no other colours evident, probably with some sea clutter either side of it, with a great big arc of black for miles and miles behind it. And as you got closer to the target the red blob would be getting bigger and bigger, and the black arc behind it would stay just as black. That great big arc of black behind a red radar return should attract your attention, because that black area with no radar returns is potentially dangerous. It's dangerous of course because a big island in-front of you has bounced all the radar energy back to you, and nothing behind it can be detected. If it was only a small island with little elevation the radar image would be completely different, you should see sea clutter returns only a short distance behind the red return. I've got a hunch the radar was painting a good picture of what was in-front of them for the last 6nm, but there was no proper interpretation of what the radar was telling them.

I know it's easy to be critical from armchair comfort with the benefit of time and hind-sight and the rest of it, but the brutal reality of it is, I don't see much PIC stuff going on. The pilot said virtually nothing of substance through the whole of the end of CVR transcript, when the aircraft was getting in an imminent state of peril. And the bit the pilot did say about BLKMO was obviously wrong but it wasn't corrected by the co-pilot. Blackrock must have been detectable on the radar directly in-front of the aircraft for the whole of the track inbound to BKLMO, and it should have been identified and reported as a hazard to the pilot, but it wasn't. The warning from the rear crew about the rock in-front of them either wasn't made forcefully enough, or wasn't acted on quickly enough (no avoiding action was being taken in the 6 seconds before impact). My reading of that little bit of CVR transcript is not enough of substance was being said amongst the crew. My flight instructor used to say "silence is bad CRM" - for me, and again this is being said from the comfort of my armchair, there is just too much silence on that CVR transcript.

To add to my last...

There would be no red blob, they were in GMap which gives gives a blue/green/purple palette with less contrast. Pedantic, perhaps, but I am highlighting that so much of the comments are based on a basis of knowledge and experience not directly related to this incident.

Yes, the radar should have been painting a return but I am reminded that radar does not paint vertically. A shoal and a cliff can appear the same and we can easily see what we expect to see...

I do agree that they weren't saying much. 2:46 was the first indication of the island, calmly reported and responded to. But until 10 seconds later, 2 seconds before impact it looks like a typical sterile cockpit in minimum weather.

Most appear not to have actually read the report (not you GulliBell).

The proximate and fatal error was that the crew were unaware the waypoint was also an island. Most of the rest of the errors follow from that clearly flawed mental model of the route.

In the report, it is an assumption that the camera was how the island was seen, not confirmed fact...

And I won't bother to repeat the observations I made previously about my own experience with these routes (not procedures).

I do recognize that the UK seems to have a different practice with regard to IMC letdowns to off airport locations and I don't know if similar practice applies to Ireland. Where I have flown, I would not expect a useful IAP to Blacksod to be approved by the CAA.

the valid point about procedures in this case (without reopening the debate) is that instead of having a dedicated radar operator who controls the letdown (and can have a FLIR picture on one screen and the radar on another) while the copilot concentrates on monitoring the pilot, the AP and the nav - you have instead the co-pilot doing everything whilst the two very capable guys in the back can only interject if they see something on their screens they don't like.

Because in civilian SAR, the rearcrew are not flightcrew (no licences) they are not allowed to be responsible for the navigation of the aircraft - this is such a waste of crew resources simply because civil aviation regards anyone not in the cockpit as cabin crew.

Those rearcrew who are ex-mil will probably still try to take as much of an active part in any letdown because that is the way they are wired - those who have spent their time as 'cabin crew' are perhaps used to being 'ignored' which might explain why the request for heading change was not made more forcefully.

The ex-mil rear crewman I have had always know the track and distance to the next waypoint, and the minimum height on that sector, and on reaching that waypoint, which direction to turn, and on to what heading. They follow the procedure all the way with their finger on the map. If I deviate from anything they expect, they will bark at me.

The radar still provides useful information to determine a shoal from a cliff. With antenna tilt adjustments a shoal will be surrounded by sea clutter, and it will be painted with some depth. A cliff will be painted with virtually no depth, and nothing but black behind it. It's that black area immediately behind a strong return with little depth that should set off alarm bells in the mind. Same applies for weather interpretation.

ARA's are not approved within 30NM of land. The newer regulations (i.e. CASA exemptions in Australia) are approved within 15NM of terrain. The weather RADAR is only used for target ID as there is no terrain in the area and a RADAR return is the only way to ID the target. So, unfortunately, weather radar is not a great solution for terrain avoidance in this situation, only weather avoidance

Not sure if it has changed in the uk but when I left I found that 30nm restriction was only in australia.

Si

Yes it is for SAR, dont mix SAR with CAT.
Every country have its own regulation.

Among the many many learning points here, I would like to propose two more:

Some vertical guidance alongside a company FMS roue that uses islands as waypoints?

Fly all IMC (non navaid) letdowns in Heading or Auto hover mode? These are dynamic procedures and it is easier and quicker to react to shipping, terrain, crew instructions, wind changes, escape headings etc.

Max - :ok:

The Preliminary Report page 15 states:
In relation to Black Rock and its Lighthouse the EGPWS manufacturer informed the
Investigation that “The lighthouse obstacle is not in the obstacle database and the terrain of the island is not in our terrain database.”
and the Preliminary Report page 17:
Figure No.4 shows the Operator’s Route Guide for Blacksod (designated ‘APBSS16
(Blacksod South) Route’) which the Crew of R116 was using at the time of the accident.
The Route Guide includes an associated separate page of text setting out, inter alia,
waypoint designations and coordinates, hazards and obstacles and other general
comments. This page identified a lighthouse at Black Rock with an associated height of 310’.


However, without resort to the separate page of text, the map from the Route Guide in Figure 4 shows a spot height of 282 ft at BLKMO, which is presumably the terrain height of Black Rock (not including the height of the lighthouse)?


I attended a talk by a Honeywell representative several years ago who suggested CFIT would soon be a thing of the past because the database included every feature and building on the planet.