Mistaken altimeter settings can kill you

[Stephen Szikora]

Here's a primer on what can go wrong with altimeter settings:

http://asrs.arc.nasa.gov/publications/d ... l9_low.htm

So why have I posted this? A friend of mine Mark Norrish once told me about an experience he had flying for Voyageur in Northern Ontario. He was landing in Sudbury at night when both pilots saw red lights out their side windows (above them) and realized that they were too low for the approach. They pulled up and later landed successfully. It turns out that they were given an altimeter setting that was off by 1" putting them 1000' below indicated altitude. They were lucky they had not flown into the radio tower.

This incident happened only 12 weeks after the loss of a Voyageur King Air and the deaths of two pilots and two paramedics at Chapleau on Nov 29, 1988 while on a air ambulance flight. A couple of years ago, Mark started thinking about his incident again and wondered what happened at Chapleau to Voyageur 796. The first thing he did was read the Canadian Aviation Safety Board final report and a couple of things didn't make sense. The report concluded that it was a CFIT accident and was not able to explain why this happened, which pointed them to pilot error. Curious to know more, Mark took the extraordinary step of filing an Access to Information request and was able to obtain the entire accident investigation file including tapes of witness interviews. In reviewing the file he quickly noticed that there were some significant errors in the investigation itself including the analysis of the weather information (the weather profile used was off by 24 hours) and there did not seem to be any investigation of the seemingly small but significant split between the pilot's and co-pilot's altimeter settings as found in the wreckage. There were also some other curious facts in the file that seemed to have lead the investigation down other paths.

Mark contacted the TSB and asked them to consider reopening the file in light of his observation of the errors and oversights in the file. At the same time, he filed a form incident report describing what happened to him around the same time suggesting that this is one area they may wish to look into. The TSB has declined citing a lack of resources and the need to set priorities. So, Mark has decided to pursue it himself. He has reviewed the file in detail, visited the crash site, interviewed relevant people, gathered interesting documentation and has posted it openly on the internet at:

http://www.voyageur796.org/

The project is far from completed but there is plenty there for anyone interested to sink their teeth into. I've been through it all and I think the most plausible theory of the case, and one that was not investigated fully, is that Voyageur 796 was given an incorrect altimeter setting from Chapleau resulting in the aircraft being too low for the approach. By the time it was discovered by the flight crew it was too late. In other words, the same thing happened to Voyageur 796 as had happened to Mark but they weren't so lucky. That said, within the file are contradictory pieces of evidence but the reliability of this evidence is questionable so it is not a clear case.

Thinking about what can go wrong is not always fun but it is a big part of being a pilot. Taking the time to sift through an actual accident investigation file is a rare opportunity and very informative. You'll have a better understanding of the process and will be able to apply your own knowledge and experience to reach a conclusion. This particular case is interesting because there are several people who think this was a botched investigation, completed at a time when the system was by many accounts not functioning well (a conclusion reached in a government report at the time.) You should at least follow Mark's progress on this case to see where it leads us.

[ant_321]

The same thing could have happened to me this past week. I was given an altimeter setting, set it and read it back. While I was in descent I randomly had a look at my flight plan where I jotted down the atis. The altimeter I had written down was 1" different than I was given. I questioned atc and found out they gave me the wrong altimeter, at night and in IMC with VV of 300'. Scary stuff.

[Stephen Szikora]

Mark has interviewed an airport manager who admitted that he had done the same thing (from the other end) giving the wrong setting to an approaching aircraft. He was questioned a few minutes later by the pilot to double check it before the aircraft got too low. The approaching pilot thought it didn't sound right based on his visual clues. In that case it was daytime in good conditions so no real problem but it could just as easily have happened at night in IMC.

[CID]

This is a good case for TAWS equipage and TC's additional requirement that it be independent of altimeter setting.

[Dash-Ate]

He should meet Widow! Compare notes.

[CpnCrunch]

This is a good case for TAWS equipage and TC's additional requirement that it be independent of altimeter setting.

Yes, I was just thinking the same thing. Even a cheap aviation GPS will have terrain warnings today. If you're flying in a commercial operation there really isn't any excuse for not spending a few bucks on this.

[Redneck_pilot86]

That was the cause of the Lockhart Air crash back in 09 as well...

[pelmet]

It is a good idea to compare the altimeter setting given to you with the one that was on the printed weather report(if there is a weather report). Depending on aircraft type, you may be able to preset the printed altimeter setting and then if you have to make a big change when given the altimeter setting via the radio, you will be suspicious.

A Learjet crashed on approach to Masset many years back due to the pilots accidentally setting their altimeter one inch off so that can happen as well.

[sanjet]

Learned to also use the GPS altimeter to aid in reference if you got a gut feeling the altimeter setting is wrong.

[Johnny#5]

Is TAWS required now for over 6 pax?

[sidestick stirrer]

Happened to me once, in the descent clearance for CYVR, at night in iFR conditions. Was only caught as the 777 has a nice feature wherein any QNH can also be displayed while the altimeter is set to STD. This way, the trend can be monitored every hour during cruise as well as making a mistake quite obvious.
Shortly thereafter, The Company changed their SOPs so that the QNH set when approaching FL180 on descent has to be derived from two, separate sources, not just the clearance, the ATIS or the METAR.

Another procedure that can lessen the danger of this error is to view the FAF as a descent gate, in that we should not be descending below the FAF crossing altitude until we have indication of FAF passage.
One of five reasons to check the altitude when crossing the FAF...

[J31]

Is TAWS required now for over 6 pax?

Harper government takes action to enhance air safety

http://www.tc.gc.ca/eng/mediaroom/relea ... e-6777.htm

No. H076/12
For release - July 4, 2012

OTTAWA — The Honourable Denis Lebel, Minister of Transport, Infrastructure and Communities, today announced new regulations to improve aviation safety in Canada. The new regulations require private turbine-powered and commercial airplanes with six or more passenger seats to be equipped with an alert system known as the “terrain awareness and warning system” (TAWS).

“While Canada has one of the safest aviation systems in the world, we are committed to the continuous improvement of aviation safety,” said Minister Lebel. “Terrain awareness and warning systems will help save lives.”

The system provides acoustic and visual alerts to flight crews when the path of their aircraft is likely to collide with terrain, water or obstacles — a situation that can happen when visibility is low or the weather is poor. This gives the flight crew enough time to take evasive action.

The new regulations will also significantly increase safety for small aircraft, which fly into remote wilderness or mountainous areas where the danger of flying into terrain is highest.

Under the new regulations, operators will have two years to equip their airplanes with TAWS.

The regulations comply with the International Civil Aviation Organization’s standards and bring Canadian regulations closer to those of other aviation authorities, including the United States and European Union. Canada’s Transportation Safety Board also recommends the wider use of TAWS to help pilots assess their proximity to terrain.

More details at:

http://www.tc.gc.ca/eng/civilaviation/o ... y-1527.htm

[XcessEnergy]

One issue regarding altimeter settings which bothers me is the region I fly in constantly uses altimeter settings that are relayed from inexperienced observers. These private aerodrome mine operators have up to date AWOS systems however they are not connected to a automated transmitting radio (because they are not certified). And weather information received is from (in my opinion, untrained or minimally trained radio operator). Basically the operator looks at the information either written on a piece of paper (which may be hours old), or a electronic readout of the current WX. Additionally, during communication the weather advisory received is random and incomplete (such an example: Winds are from the NEN at 2 kts, ceilings unlimited broken at 5465’ (not sure what to make of that), temperature -25, altimeter 29.85) as well the lack of an observation time is very concerning. Was the observation taken 4 hours, or 5 minutes ago… the response is usually “not sure”…. Very Very frustrating, and multiple SMS reports later no change in the private operators weather briefings. I almost think that behind the scenes no one wants to upset the customer (in this case the aerodrome operator).

However, regarding this topic of wrong and grossly inaccurate altimeter settings, there is multiple mine / aerodrome sites within 50nm of each other and on some occasions each sites altimeter setting can differ by 3/4 inch (the most I have personally witnessed). Add multiple aircraft flying into different aerodromes (some with overlapping approach procedures) and different altimeter settings all within a 50nm is a recipe for disaster (I have heard numerous close calls, although have not experienced any myself), but I do find stating the altimeter we are using to other potential conflict aircraft helps. However, the bright side most of the approaches at these sites are an LPV which means we can correspond G/S indication with FAWP crossing altitudes as a last resort before a TWAS alert goes off.

Then across the lake at a civilian aerodrome the TC approach procedure refers to using the altimeter setting form the mine site, but the approach plate does not indicate a frequency to obtain this altimeter setting (which is most likely wrong). And I suspect due to lack of training, the aerodrome radio operator does not really understands the severity of passing an inaccurate altimeter to aircraft in the area. I just hope someone takes these SMS reports seriously and attempts to correct these errors before people actually pay the ultimate price.

[catan man]

This is where OCD is a good thing. I know personally, in IMC and having done our checks through 10 with Altimeter setting checked, I always find myself around sector, checking for the Altimeter setting again knowing we just went through that. There's something about shitting my pants as we crash I have a fear of.

[old_man]

I once saw from one metar to the next hourly one that the 'meter setting dropped 0.2". If someone had shot an ILS with the old one or just before the new one came out that could have been a surprise. That being said the weather was obviously not conducive to any form of flight and the metar did state PRESFR but just something to think about.

[Morav]

28.40ish" in YHZ Today!

[Sidebar]

CADORS 2012C4516
TSB A12C0143

... a Canadair CL-600-2C10 (CRJ 700), N773SK operated by Skywest Airlines as flight SKW 619R, was on final approach for an RNAV approach to Runway 31 at Regina when the crew received a GPWS "too low gear" warning. The crew climbed and then continued the approach. The flight landed without further incident. The flight crew determined the altimeters had been set to 30.12 inches, while the altimeter setting provided by air traffic services was 29.12.

[bobm]

I am attaching what happened here and ended up the Toronto Star.

Second, incorrect altimeter settings happen way too much. Especially in a two crew environment, who's is also not following SOP'?

Thirdly, and not that is important now but I highly doubt that the cause 796 crash was incorrect altimeters. The aircraft crashed proceeding westbound to the NDB for the procedure turn. It flew low overhead (as heard by the Airport manager) the airport (from YTS) and crashed on the LEE side of the hill. That is, he was descending faster than the terrain.

I know more than I would like to remember about that night. Laurie and I had flown out of Timmins as we were standing in for 796 Air Ontario C-500 crew. Actually, we had been flying a fair bit during those last few Air Ontario days. Oh, one thing that may or may not have been mentioned, that fateful night was the crew's first trip the day they took over the contract. One other thing...the "medic's" were not happy about the changes coming being taken over by the new contractor. The Captain had transferred from the YXL base and home was in Muskoka. You have probably heard of his dad.

And the Ops Manager from Voyageur at the time? Now high up at the TSB I believe.

[Stephen Szikora]

The problem with the Voyageur 796 file is that there were a number of mistakes made in the investigation itself. There's a story in the story. The use of the wrong weather data is only one example. With respect to your comment about the crash site being on the lee side of the hill, there is now some evidence that the location of the crash site as marked in the file was simply incorrect. In the official report a topographical map is shown and an X marked on the map. I suspect this is why you concluded that the crash site was on the lee side. However, how accurately that X was marked is in question, particularly given that the wreckage was spread over a 650' path and a total of 750' from the first contact with the trees.

Yes, this was the first flight under the new contract and yes the paramedic's union was upset over the change. In fact, one of the medics on the flight had gone to Queen's Park as a union representative a week prior to raise their concerns. Among others, he met with then opposition leader Bob Rae and his concerns were reported in the legislature the day after the crash.

Yes also to the fact that the person in charge of flight ops at Voyageur at the time is now the director of investigations at the TSB. In fact, the question of potential conflict of interest in the decision not to reopen the file has been raised with the Board and dismissed.

As someone who was also familiar with operating procedures in the area at the time of the crash, I would encourage you to contact Mark and talk about it. A number of people have come forward and provided information.

What makes this file so interesting is that there are a number of overlapping stories within it. Mistakes were made both before and after the accident. To a large degree, the causes of the crash have never been determined and the official report by default points to pilot error. In the course of his reexamination of the file and new evidence gathered subsequently, Mark has discovered that the conclusions reached in the report and the process followed were flawed. There may never be conclusive evidence of the cause(s) of the crash but the story of the investigation and its problems needs to be told.

[Colonel Sanders]

I am only a moron, but I have found that the altitude
produced by my GPS is startlingly accurate.

Remember that an indicating altimeter can never
malfunction, but if it ever did, you could simply use
your GPS altitude.

My indicating altimeter has never, ever malfunctioned
and I have never, ever referred to my my GPS altitude,
for the legal beagles here, who have TC Enforcement on
speed dial

Given the above, it can be educational to crosscheck
your indicating altimeter after you punch in the setting,
with your GPS altitude, if you are in a remote location,
or if you have any doubt whatsoever as to the accuracy
of your altimeter setting (e.g. a large difference).

IIRC there was an accident a few years back at Kingston,
Ontario where a pilot dialled in an altimeter setting an
inch off. He was a thousand feet low as a result. But
you would have thought he would have been suspicious,
dialling in that huge a change in altitude, unless he
was descending out of the flight levels, which he wasn't.

[CpnCrunch]

Yes, I regularly check my altitude with a GPS, especially when flying in the uncontrolled airspace underneath Calgary's TCA. It's normally within 50ft agreement.

[NeverBlue]

Wouldn't one be inclined to look at their Radio/Radar Altimeter if they fear they're close/not close enough to terrain in cases like this?

They wouldn't have had a GPWS in 1988 but surely the King Air would have had a RadAlt...no?

[bobm]

It did go down on the Lee side of the hill. I was based in YLD before this accident and knew just about everyone. I also know the guys that were hired by Max to clean up the wreckage after the investigation...on the lee side of the hill. In fact, Hank and I were business partners for awhile.

And since they were still out bound to the beacon, they were more than a thousand feet out...

There was lots of hearsay and innuendos at the time about a medic trying to down the aircraft. Having flown with them the night before, they were stressed but I doubt it.

And GPS???? That was pretty new and I am not sure if the Voyageur aircraft was equipped yet. Approach GPS were definitely not out yet.

[JigglyBus]

Stephen, any chance you could provide a link to the report?

Also, you've mentioned several times that numerous mistakes were made.... could you list the mistakes so they're obvious when we read it?

[ditar]

... a Canadair CL-600-2C10 (CRJ 700), N773SK operated by Skywest Airlines as flight SKW 619R, was on final approach for an RNAV approach to Runway 31 at Regina when the crew received a GPWS "too low gear" warning. The crew climbed and then continued the approach. The flight landed without further incident. The flight crew determined the altimeters had been set to 30.12 inches, while the altimeter setting provided by air traffic services was 29.12.

Back in March, Skywest did the same thing in Calgary, except in this case their TCAS warned them before they descended on top of me. The reason was the same.